Download View - OhioLINK Electronic Theses and Dissertations Center

Francisella tularensis Lipopolysaccharide O-antigen Dictates the Outcome of
Human Complement Activation
Dissertation
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy
in the Graduate School of The Ohio State University
By
Corey Davis Clay, M.S.
Integrated Biomedical Science Graduate Program
The Ohio State University
2009
Dissertation Committee:
Professor Larry Schlesinger, Co-Advisor
Professor John Gunn, Co-Advisor
Professor Jennifer Edwards
Professor Susheela Tridandapani
Professor Mark Wewers
Copyright by
Corey Davis Clay
2009
Abstract
Francisella tularensis is a Gram-negative facultative intracellular bacterium that
is a potential weapon of bioterrorism when aerosolized. Macrophage infection is
necessary for disease progression and efficient phagocytosis by human macrophages
requires serum opsonization by complement. Microbial complement activation leads to
surface deposition of highly regulated multimeric protein complexes that can promote
opsonization or membrane lysis depending on the nature of the complexes formed.
Outcomes of complement activation by bacteria largely depend upon the fate of
complement component C3 following deposition. Functional cleavage fragments derived
from C3 include C3b, which promotes both opsonization and microbial lysis, and C3bi,
which specifically promotes opsonization.
Here, we study interactions between F. tularensis and the human complement
cascade to gain a better understanding of the processes of immune evasion and cellular
infection employed by this deadly bacterium. We examine mechanisms of resistance to
complement-mediated lysis, the nature of C3 component surface deposition, and
mechanisms of complement activation. We show that, upon incubation in fresh nonimmune human serum, Schu S4 (F. tularensis subsp. tularensis), F. tularensis subsp.
novicida, and LVS (F. tularensis subsp. holarctica live vaccine strain) are resistant to
ii
complement-mediated lysis. LVSG and LVSR are variant strains derived from LVS that
have altered surface carbohydrate structures and are susceptible to complement-mediated
lysis in serum. C3b deposition, however, occurs on each strain tested, indicating that
complement is not solely activated by variant strains. Complement-susceptible strains fix
markedly increased amounts of total C3-derived fragments. Specifically, the presence of
C3b is persistent compared to C3bi only on susceptible strains and the deposition of
downstream complement components C5 and C7 is significantly greater. These results
indicate that upon binding to wildtype strains, C3b becomes rapidly cleaved to form
C3bi, which facilitates opsonization and evasion of downstream lytic components of
complement. Characterization of differences in the production of important surface
glycans between resistant and susceptible strains and employment of targeted mutant
strains allowed us to determine that LPS O-antigen plays a significant role in dictating the
outcome of complement activation and the nature of C3 deposition on F. tularensis.
Both O-antigen producing and O-antigen-deficient strains rely heavily on the
classical complement activation pathway. C1, a component of the classical pathway, is
required for optimal lysis of complement-susceptible strains, and for optimal C3
deposition on all strains. Furthermore, we show that wildtype and O-antigen-deficient
strains activate the classical pathway in an uncommon manner that is independent of
antibody. The direct binding of C1 is reduced in the presence of O-antigen, which limits
the activation of downstream components including C3. We conclude that F. tularensis
activates complement in an unusual manner such that the rate of C3b deposition is
restricted allowing for the efficient conversion of C3b to C3bi only on virulent strains. In
iii
the absence of O-antigen, however, increased activation of C1 leads to a C3b deposition
rate that is greater than the rate of C3b to C3bi conversion, which ultimately leads to
bacterial lysis by downstream components of the cascade.
iv
Dedicated to my wife Nalynne, whose love and continuous support
keep me motivated to go the extra “bug” mile.
v
Acknowledgments
It is my sincere pleasure to first acknowledge Drs. Larry Schlesinger and John
Gunn for their support during these graduate research years. They guided my
development as a scientist in such a way as to foster excitement and determination toward
the process of discovery. It is because of their mentorship, their example of teamwork,
and their generosity of time and spirit that I have reached this milestone. I also wish to
thank my thesis committee: Drs. Jennifer Edwards, Susheela Tridandapani, and Mark
Wewers for their counseling and continued endorsement.
I thank the faculty and staff of the Center for Microbial Interface Biology for their
help, support, and encouragement over the years. They helped expand my knowledge
and comprehension of host pathogen interactions and piqued my imagination on a daily
basis. The CMIB promotes a collaborative spirit between laboratories that synergizes the
scientific productivity and inspiration of its collective constituents. It is a spirit that I can
only hope to experience again in the upcoming phases of my career.
I would like to acknowledge, specifically, the contributions and suggestions of
Drs. Chad Rappleye, William Lafuse, Daniel Wozniak, and Brian Ahmer during
laboratory meetings and following public presentations. I thank the “Francisella” team
for helpful and thought-provoking discussions as well as experimental guidance and
assistance; and these individuals especially include Dr. Nrusingh Mohapatra, Shilpa Soni,
vi
Dr. Brian Bell, Dr. Ashwin Balagopal, Dr. Jordi Torrelles, and Heather Curry.
As the former director of the Medical Scientist Program, I thank Dr. Allan Yates
for his vision, mentorship and dedication to its students. I thank Dr. Virginia Sanders for
mentoring me as a trainee with the NIH-sponsored Integrative Immunobiology Training
Program, which partially funded this work. Dr. Sanders, also the Director of the
Integrated Bioscience Graduate Program, is an enlightening teacher and a reliable
advocate whose door is always open for her students. I thank our collaborator, Dr. Dara
Frank at the Medical College of Wisconsin who provided mutant bacterial strains used in
our studies. I thank Dr. Scott Ferguson (University of Iowa) and Dr. Michael Pangburn
(University of Texas Health Science Center at Tyler) for imparting their wisdom and for
their willingness to respond to lengthy emails. Finally, I acknowledge funding from the
NIH/NIAID Regional Center of Excellence (Region V) for Bio-defense and Emerging
Infectious Diseases Research.
To the extended laboratory family, in addition to our shared love for complement
biology and for science as a whole; I will fondly remember our efforts on the softball
field, our post mock exam toasts, our July 4th celebrations on 10, our White elephant
parties, and above all, our friendships. Let us not become strangers. In particular, I wish
to thank the Gunn laboratory for putting up with Cake and the often ensuing excitement,
and thanks to Dr. Robert Crawford for often providing the Cake.
vii
Vita
November 8, 1977 . . . . . . . . . . . . . . . . . . . Born – Denver, CO
1996. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Graduated Carroll High School
Southlake, TX
1996-2000. . . . . . . . . . . . . . . . . . . . . . . . . B.S. Pre-Professional Studies
University of Notre Dame
2001-2003. . . . . . . . . . . . . . . . . . . . . . . . . .M.S. Molecular Toxicology
University of Cincinnati
2003-2005. . . . . . . . . . . . . . . . . . . . . . . . . Medical Scientist Fellow
The Ohio State University
2005-2006. . . . . . . . . . . . . . . . . . . . . . . . . University Fellow
The Ohio State University
2006-2007. . . . . . . . . . . . . . . . . . . . . . . . . Graduate Research Assistant
The Ohio State University
2007-2008. . . . . . . . . . . . . . . . . . . . . . . . . Graduate Research Fellow
The Ohio State University
2008-present. . . . . . . . . . . . . . . . . . . . . . . . Presidential Dissertation Fellow
The Ohio State University
Publications
1.
Premanandan C, Storozuk CA, Clay CD, Lairmore MD, Schlesinger LS, Phipps
AJ. Complement protein C3 binding to Bacillus anthracis spores enhances
phagocytosis by human macrophages. Microb Pathog. 2009 Jun;46(6):306-14.
2.
Clay CD, Soni S, Gunn JS, Schlesinger LS. Evasion of complement-mediated
viii
lysis and complement C3 deposition are regulated by Francisella tularensis LPS
O-antigen. J Immunol. 2008. Oct 15;181(8):5568-78.
3.
Butchar JP, Cremer TJ, Clay CD, Gavrilin MA, Wewers MD, Marsh CB,
Schlesinger LS, Tridandapani S. Microarray analysis of human monocytes
infected with Francisella tularensis identifies new targets of host response
subversion. PLoS ONE. 2008 Aug 13;3(8):e2924.
4.
Butchar JP, Rajaram MV, Ganesan LP, Parsa KV, Clay CD, Schlesinger LS,
Tridandapani S. Francisella tularensis induces IL-23 production in human
monocytes. J Immunol. 2007 Apr 1;178(7):4445-54.
5.
Genter MB, Clay CD, Dalton TP, Dong H, Nebert DW, Shertzer HG.
Comparison of mouse hepatic mitochondrial versus microsomal cytochromes
P450 following TCDD treatment.
Biochem Biophys Res Commun. 2006 Apr 21;342(4):1375-81.
6.
Shertzer HG, Clay CD, Genter MB, Chames MC, Schneider SN, Oakley GG,
Nebert DW, Dalton TP. Uncoupling-mediated generation of reactive oxygen by
halogenated aromatic hydrocarbons in mouse liver microsomes. Free Radic Biol
Med. 2004 Mar 1;36(5):618-31.
7.
Shertzer HG, Clay CD, Genter MB, Schneider SN, Nebert DW, Dalton TP.
Cyp1a2 protects against reactive oxygen production in mouse liver microsomes.
Free Radic Biol Med. 2004 Mar 1;36(5):605-17.
8.
Tsuneoka Y, Dalton TP, Miller ML, Clay CD, Shertzer HG, Talaska G,
Medvedovic M, Nebert DW. 4-aminobiphenyl-induced liver and urinary bladder
DNA adduct formation in Cyp1a2(-/-) and Cyp1a2(+/+) mice. J Natl Cancer Inst.
2003 Aug 20; 95(16): 1227-37.
Fields of Study
Major Field: Integrated Biomedical Science
ix
Table of Contents
Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Dedication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Vita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
List of Tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv
List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
List of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Chapters:
1.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Tularemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 1
Historical background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . 1
Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 1
Clinical presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Vaccine development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Relevant aspects of pulmonary innate immunity . . . . . . . . . . . . . . . . . . 5
extracellular immunity in the lung . . . . . . . . . . . . . . . . . . . . . . . . 6
cell-mediated immunity in the lung . . . . . . . . . . . . . . . . . . . . . . 7
Important pathogenic features of tularemia . . . . . . . . . . . . . . . . . . . . . . 10
Characteristics of the host response to F. tularensis . . . . . . . . . . . . . . . . 13
protective role of an early proinflammatory response. . . . . . . . . . .15
pulmonary innate immune suppression by F. tularensis . . . . . . . .17
F. tularensis uptake by macrophages . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
Constitution of the cell wall of F. tularensis . . . . . . . . . . . . . . . . . . . . . 24
F. tularensis and phase variation . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . 28
Virulence effects of altered F. tularensis surface structures . . . . . . . . . .30
x
1.2 Complement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Potential outcomes of complement activation . . . . . . . . . . . . . . . . . . . . .
The classical pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The classical complement pathway in innate
immunity and homeostasis . . . .. . . . . . . . . . . . . . . . . . . . . . . . .
The lectin and alternative pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The terminal lytic pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Negative regulation of complement activity . . . . . . . . . . . . . . . . . . . . . .
Complement receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Complement activity in the airway . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bacterial compelement evasive strategies . . . . . . . . . . . . . . . . . . . . . . .
32
32
35
36
38
41
42
43
47
51
52
1.3 Specific Aims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.
Evasion of complement-mediated lysis and complement C3 deposition are
regulated by Francisella tularensis LPS O-antigen
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Bacterial strains used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Human sera, complement components, and reagents . . . . . . . . . . . . . . . 64
Bronchoalveolar lavage (BAL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Bactericidal assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
C3 deposition assays and Western blotting . . . . . . . . . . . . . . . . . . . . . . . 66
ELISA to detect complement component deposistion of
F. tularensis strains . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 67
Determination of the nature of C3 bound to F. tularensis . . . . . . . . . . . 68
Transmission electron microscopy . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . 68
LPS expression analysis by silver stain and Western blot . . . . . . . . . . . 69
Microscopy assay of F. tularensis uptake by AMs . . . . . . . . . . .. . . . . . 70
Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Results
Complement mediated lysis of F. tularensis in human serum . . . . . . . . 72
Fixation of complement components C3, C5 and C7 . . . . . . . . . . . . . . 75
Temporal analysis of the nature of C3-derived fragments that
bind to F. tularensis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
O-antigen expression is a major determinant of susceptibility to
complement-mediated lysis and C3b to C3bi conversion . . . . . . 84
Complement activity in bronchoalveolar lavage fluid and the
effect of opsonization on F. tularensis uptake by human
alveolar macrophages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
xi
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97
3.
Francisella tularensis principally activates the classical complement
pathway in the presence and absence of LPS O-antigen
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bacterial strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Human sera, complement components, and reagents . . . . . . . . . . . . . .
Bactericidal assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C3 deposition and Western blotting . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
105
105
106
107
108
108
Results
Complement-mediated lysis of susceptible F. tularensis strains
occurs by more than one activation pathway . . . . . . . . . . .. . . . 109
The role of C1q in mediating C3 deposition on complement-resistant
and complement–susceptible strains . . . . . . . . . . . . . . . . . . . . . 116
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4.
Francisella tularensis LPS O-antigen restricts direct binding and activation
of complement component C1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Bacterial strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Human sera, complement components, and reagents . . . . . . . . . . . . . . . 125
Complement hemolytic (CH50) assays. . .. . . . . . . . . . . . . . . . . . . . . . . . . 126
Determination of antibody in non-immune donor serum to F. tularensis 127
Bactericidal assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
C1q, C3, and C4 deposition assays and Western blotting . . . . . . . . . . . . 128
ELISA to detect deposition of C1 subcomponent, C4, Factor H and
C4 binding protein on F. tularensis strains . . . . . . . . . . . . . . . . 129
Results
F. tularensis LPS O-antigen production negatively influences the
consumption of complement hemolytic activity . . . . . . . . . . . . 130
Activation of the classical pathway by F. tularensis occurs
independently of antibody . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 133
O-antigen mediated regulation of complement occurs upstream of C3
deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
F. tularensis associated O-antigen limits binding of C1 in serum . . . . . . 143
xii
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
5.
Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
xiii
List of Tables
2.1
Summary of F. tularensis strains used. . . . . . . . . . . . . . . . . . . . . . . .73
xiv
List of Figures
1.1
Structure of lipid A, core, and O-antigen molecules synthesized
by F. tularensis species tularensis, holarctica, and novicida . . . . . . 26
1.2
The complement activation and terminal pathways . . . . . . . . . . . . . .34
1.3
Cleavage products of native C3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.1
Susceptibility to complement-mediated lysis differs among
F. tularensis strains.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2.2
Complement component C3 deposition occurs in greater amounts
on complement-susceptible strains of F. tularensis . . . . . . . . . . . . . 76
2.3
Quantitative analysis of complement components C3, C5, and C7
fixed by F. tularensis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.4
The nature of bacteria-bound C3 fragments for different
F. tularensis strains over time.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
2.5
Transmission electron microscopic images of LVS and LVSG
show differences in the outer membrane. .. . . . . . . . . . . . . . . . . . . . . 85
2.6
Complement susceptibility and surface C3b stability are
determined by F. tularensis LPS O-antigen expression. . . . . . . . . . . 87
2.7
Western blot showing LPS O-antigen production by different
F. tularensis strains. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2.8
Restoration of O-antigen expression on a mutant strain results in
complement resistance and C3b inactivation . . . . . . . . . . . . . . . . . . . 90
2.9
Human concentrated bronchoalveolar lavage fluid (cBAL) is
deficient of complement lytic activity. . . . . . . . . . . . . . . . . . . . . . . . 93
xv
2.10
Complement mediated uptake of Schu S4 and LVS by human
alveolar macrophages (AMs). . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .95
3.1
Complement activation by susceptible F. tularensis strains occurs
via more than one pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.2
Optimal lysis of variant F. tularensis strains is dependent
upon C1q . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
3.3
Optimal lysis of LVSΔwbtA, an O-antigen mutant strain, is
dependent upon C1q . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
3.3
Deposition of C3 on both wildtype and variant strains is
C1q-dependent . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . .. . . . . . . . . 117
3.5
C3 fixation by LVSΔwbtA is predominantly C1q-dependent. . . . . . .118
4.1
Consumption of complement hemolytic activity by LVS and
LVSΔwbtM, the latter an isogenic O-antigen mutant strain . . . . . . . 131
4.2
Immunoglobulin binding to LVS and LVSΔwbtM in human
non-immune serum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
4.3
Complement activation by both susceptible and resistant strains
of F. tularensis occurs independently of antibody . . . . .. . . . . . . . . . 136
4.4
LVSΔwbtM binds greater amounts of Factor H (FH) and C4
binding protein (C4bp) compared with LVS. . . . . . . . . . . . . . . . . . . 139
4.5
C4 activation and deposition on F. tularensis occurs in an
antibody-independent manner. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .141
4.6
C1q binding to various strains of F. tularensis is affected by
O-antigen expression and by uncharacterized components of serum 144
4.7
C1q and C1s bind in greater amounts to LVSΔwbtM compared to
LVS in serum. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 147
xvi
List of Abbreviations
AG:
agammaglobulinemic serum
AMs:
alveolar macrophages
ASF:
airway surface fluid
BAL:
bronchoalveolar lavage
C1inh:
C1-esterase inhibitor
C1qd:
C1q-depleted serum
C1qr:
C1q-replete serum
C4bp:
C4 binding protein
C5d:
C5-depleted serum
C8d:
C8-depleted serum
cBAL:
concentrated bronchoalveolar lavage fluid
CCP:
complement control protein
CD:
cluster of difference antigen
cfu:
colony forming units
CH50:
50% hemolytic titrate
CRIg:
complement receptor of the immunoglobulin superfamily
CRs:
complement receptors
DAF:
decay accelerating factor
xvii
DAMPs:
damage associated molecular patterns
DCs:
dendritic cells
ECM:
extracellular matrix
FB:
Factor B
FBd:
Factor B-depleted serum
FBr:
Factor B-replete serum
FD:
Factor D
FH:
Factor H
FI:
Factor I
Fc:
fragment crystallizable
FcγRs:
IgG Fc receptors
GVB:
gelatin veronal buffer
HI:
heat inactivated
HRP:
horseradish peroxidase
IFN:
interferon
Ig:
immunoglobulin
IL:
interleukin
iNOS:
inducible nitric oxide synthase
kDa:
kilodalton
Kdo:
2-keto-3-deoxy-D-manno-octulosonic acid
LBP:
LPS-binding protein
LD50:
50% lethal dose
xviii
LPS:
lipopolysaccharide
LVS:
live vaccine strain
LVSG:
grey LVS phase variant
LVSR:
rough mutant derived from LVS
MAC:
membrane attack complex
MAPK:
mitogen activated protein kinase
MASP:
MBL associated serine protease
MBL:
mannose binding lectin
MCP:
membrane cofactor protein
MOI:
multiplicity of infection
MR:
mannose receptor
PAMPs:
pathogen associated molecular patterns
PI3K:
phosphoinositide-3 kinase
PRRs:
pathogen recognition receptors
RCA:
regulators of complement activation
S100A9:
S100 calcium binding protein A9
SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis
SP-A:
surfactant protein A
SR-A:
class A scavenger receptor
spp.:
species
subsp.:
subspecies
TGF:
tumor growth factor
xix
TH1:
T helper cell type 1
TH2:
T helper cell type 2
TLRs:
toll-like receptors
TNF:
tumor necrosis factor
xx
Chapter 1: Introduction
1.1 Tularemia
Historical background
Francisella tularensis was first isolated and characterized in Tulare County,
California in 1911 during an outbreak of plague-like disease in ground squirrels (1).
Originally, the organism was classified within the genus Bacterium and was later
reclassified as Pasteurella. In the United States, the first bona fide case of tularemia with
concomitant isolation of the causative organism occurred two years later in Ohio (2).
Fifteen years later, in 1928, Edward Francis published a comprehensive summary of over
600 cases and he later coined the term “tularemia” because of the bacteremia associated
with most cases (3, 4). In 1947, the new genus, Francisella, was named in his honor.
Epidemiology
Francisella are small, aerobic, pleiotrophic, Gram-negative coccobacilli. Four
subspecies have been described including subspecies tularensis (type A), holarctica (type
B), novicida, and mediasiatica. Subspecies novicida and mediasiatica do not cause
disease in immunocompetent humans, but novicida maintains virulence in some animals
and has proven value as a safe laboratory model organism for pathogenesis studies (5).
1
Subspecies holarctica exists throughout North America, Europe, and Asia; whereas,
subspecies tularensis is exclusive to North America. For unknown reasons, tularemia
only occurs between latitudes 30° and 70° of the Northern Hemisphere (6). Type A F.
tularensis is the most deadly subspecies and may, in fact, be the most infectious bacterial
pathogen known. The LD50 in mice is as few as 1-4 organisms subcutaneously (7). Type
B F. tularensis causes milder disease with lower mortality rates (8).
Tularemia is naturally a zoonotic disease and F. tularensis can infect a large
number of vertebrates and invertebrates. A smaller number of animals, however, plays a
significant role in the organism’s ecological life cycle in defined geographical loci (9,
10). The terrestrial lifecycle of F. tularensis depends on wild rabbits as “amplifying”
hosts and on arthropod vectors that primarily include ticks, biting flies, and mosquitoes.
The aquatic lifecycle depends on muskrats, beavers and voles that shed microbes directly
into aquatic habitats so that the cycle of re-infection occurs independently of arthropod
vectors. Humans become infected through insect bites or by direct contact with the
organism. Handling of infected animal carcasses (e.g. skinning and meat processing) can
result in the aerosolization of bacteria and is the most common route of human
transmission. Persons with increased risk to exposure generally include laboratory
workers, farmers, landscapers, veterinarians, hunters, trappers, cooks, and meat handlers
(11).
The highest incidence of tularemia in the United States occurred in 1939 when
2291 cases were reported (11). This number has decreased significantly with time and in
the 1990s, a total of 1368 cases were reported. This may be because of a decreased rate
of infections or to an increased success rate for the empirical treatment of undiagnosed
2
infections. Nonetheless, in the year 2000, tularemia was reinstated as a nationally
notifiable disease because of its potential use as a weapon of bioterrorism.
During the cold war, both the United States and the Union of Soviet Socialist
Republics stockpiled antibiotic and vaccine resistant strains of F. tularensis for use in
biowarfare (12). Bioweapon development continued in the U.S. until the late 1960s and,
reportedly, in the U.S.S.R. until the 1990s (12, 13). Accounts indicate that Japan may
have promoted the development of this weapon more intensely. Documented human
experiments were performed by Japanese scientists using Chinese detainees before and
during World War II (14). F. tularensis can be easily disseminated, causes a high
mortality rate, and has the potential to cause mass panic among the public; and thus, is
given the highest priority classification by the Centers for Disease Control as a category
A select agent.
Clinical presentation
Pneumonic and non-pneumonic forms of tularemia occur. The mortality rate for
untreated cases of ulceroglandular and typhoidal (without the development of pneumonia
via hematological dissemination) tularemia is between 5% and 15% (15). Maculopapular
lesions of the skin that ulcerate centrally are characteristic for cutaneous tularemia.
When mucous membranes are involved, severe local inflammation can have exudative
and purulent qualities with or without the presence of subepidermal or regional lymph
node abscesses. Lymphadenopathy can occur independently of other symptoms, but
regional lymphadenopathy can be profound when associated with ulcerative cutaneous
disease or with conjunctivitis. Involved lymph nodes are generally tender to palpation
3
and measure between 0.5 and 10 cm. They may spontaneously resolve or may persist for
as long as three years. Typhoidal tularemia is characterized by bacteremia and the spread
of infection from the inoculation site to distal organs (12).
Pneumonic tularemia occurs upon inhalation of as few as 1-10 F. tularensis
bacilli or upon hematological spread of bacteria to the lungs from distal sites of infection
(12). Approximately 30% of cases of ulceroglandular tularemia progress to involve
pneumonia. The mortality rate for untreated pneumonic tularemia is between 30% and
60%. (15). After inoculation of the lung, patients are asymptomatic for an average of five
days (16, 17). The clinical presentation ranges from mild flu-like symptoms to the
involvement of high fever, general malaise, chills, dry cough, pleuritic chest pain,
delirium, and pulse-temperature dissociation (high temperature without expected
tachycardia) (18). After patients become symptomatic, chest x-rays reveal uni- or
multilobular parenchymal infiltrates with variable severity and severe hilar
lymphadenopathy.
Vaccine development
Currently, no licensed vaccines exist for public use. In humans, the predominant
antibody response targets lipopolysaccharide (LPS) (18). In mice, pretreatment with LPS
is protective against subsequent inoculation with strains of low virulence that are derived
from Type B F. tularensis. However, the immunoprotective effect of LPS is decreased
against inoculation with Type A bacteria (19-21). Immunoreactive proteins have been
characterized but do not elicit protection against Type A bacteria (22-25). A live vaccine
strain (LVS) was developed in the 1950s by serial passage of subspecies holarctica in
4
mice (26-28). LVS is used to vaccinate at-risk personnel and is under review as an
“investigational new drug” by the FDA. Like subspecies novicida, LVS is avirulent in
humans, but causes fulminant disease (at higher doses compared to wildtype subspecies)
in animal models (8). Efficacy of LVS is promising because, in humans, vaccination was
shown to provide good protection against subsequent inoculation with 10 infectious doses
of Type A bacteria and partial protection against 100 infectious doses (29). However,
LVS has not been licensed because the genetic mechanism of attenuation has not been
uncovered. Also, it has mixed colony morphological features resulting from its capacity
for phase variation and this may reduce its efficacy. Finally, LVS confers variable levels
of immunogenicity in animal models (5).
Relevant aspects of pulmonary innate immunity
A general review of pulmonary innate immunity should be appreciated before
reviewing the pathogenesis of pneumonic tularemia. The innate immune system in the
lung is unique due to a persistent exposure to environmental pathogens, allergens, and
toxins, which must be neutralized without compromising the delicate architecture of the
airspace. Most potential pathogens are cleared by the muco-ciliary escalator and other
mechanical elements of host defense (30). Those that bypass the ciliated epithelium to
reach small bronchioles and alveoli encounter extracellular antimicrobial molecules,
epithelium with immune signaling capabilities, and alveolar macrophages (AMs) within
the airspace as well as dendritic cells (DCs) within the parenchyma. A successful
immune response can involve the immediate destruction, or incapacitation, of microbes
by extracellular mediators or by resident leukocytes. Antimicrobial effectors may be
5
increased or enhanced with the induction of an immediate and localized inflammatory
response, which must subsequently be resolved in order to maintain physiological
functionality.
Extracellular immunity in the lung
The functionally diverse array of proteins that exists within the airway surface
fluid (ASF) lining bronchioles and alveoli includes: complement proteins (discussed in
detail below), antibodies, collectins, defensins, lactoferrin, lysozyme, cathelicidins, and
protease inhibitors (31). In concert, these molecules act synergistically to disrupt
microbial cell membranes. Furthermore, their activity is enhanced within the low pH
associated with ASF. IgA and IgG, respectively, cause bacterial agglutination and
opsonization. Surfactant proteins (SP), SP-A and SP-D, are collectins that have a major
role in pulmonary innate immunity by agglutinating and directly lysing microbes as well
as by acting as environmental signals to modulate leukocyte activity (32, 33). Defensins
are short cationic peptides with a broad spectrum of antimicrobial activity. Their size and
charge enable intercalation into foreign cell membranes causing pore formation and
disruption of osmotic homeostasis. Lactoferrin sequesters iron to inhibit microbial
growth and also has a disruptive effect on membranes. Lysozyme hydrolyses
peptidoglycan, a component of microbial surfaces. LL-37, a cathelicidin, acts
synergistically with lactoferrin and lysozyme to disrupt microbial membranes, and it acts
as a stimulant for the degranulation and the proinflammatory capacity of leukocytes.
Elafin and SLPI (secretory leukoprotease inhibitor) are serine protease inhibitors with
dual functions (34). They protect the mucosal surface from damage mediated by potent
6
proteases (e.g. neutrophil elastase and cathepsin G) secreted during an inflammatory
reaction, and they contain cationic domains that permeate membranes in the same manner
as do defensins.
Cell-mediated immunity in the lung
Lung parenchyma including epithelium, endothelium and fibroblasts are active
mediators of pulmonary innate immunity (31). Epithelial cells are responsible for the
constitutive secretion of, and can be induced to secrete increased levels of, the
extracellular modulators listed above. Both epithelial and endothelial cells express
pattern recognition receptors (PRRs), which respond to conserved microbial epitopes
called pathogen associated molecular patterns (PAMPs). These cells additionally express
cytokine and chemokine receptors involved in the coordination of an appropriate, local,
immune response. Epithelial cells also produce reactive oxygen species and
proinflammatory cytokines (e.g. IL-1β, TNFα, and IL-6) that each enhance the
microbicidal environment. Tissue remodeling and repair require fibroblasts that
modulate the extracellular matrix by secreting components such as collagen as well as
degradative proteases known as matrix metalloproteases that enhance leukocyte
recruitment. Necrosis of any of these cell types can result in the release of damage
associated molecular patterns (DAMPs), which are intracellular proteins that act in cell
homeostasis but also act as ligands to activate and recruit leukocytes to sites of tissue
disrepair.
Innate immune cells in the airway include resident cell populations
(macrophages, DCs and mast cells) and recruited itinerant cells (monocytes and
7
neutrophils) that respond to infection. The professional phagocytes include neutrophils,
undifferentiated monocytes, DCs and macrophages. They migrate towards sites of
infection in response to chemical stimuli. The expression of a variety of phagocytic
receptors aids in pathogen uptake. Neutrophils represent the largest population of
leukocytes in circulation including the capillary bed of the lung (30). They exhibit a
greater degree of phagocytic potential compared to monocyte-derived cells and produce
potent antimicrobial factors that include oxidants and hydrolytic enzymes. Monocytes
are also recruited to sites of active inflammation, albeit less rapidly when compared with
neutrophils, and effectively destroy phagocytosed pathogens via the hydrolytic and
oxidative contents of phagolysosomes (35).
AMs account for 95% of cells in the alveolar lumen and are most likely the first
phagocytic cells to encounter pathogens (30). To effectively maintain sterility of the
airspace, these cells are in a constant state of active phagocytosis. Phagocytic receptors
expressed by AMs include complement receptors (CRs) (CR1, CR3, CR4, and CRIg),
Fcγ receptors (FcγRs), and PRRs; for example, the mannose receptor (MR), Dectin-1, and
class A scavenger receptor (SR-A) (36-40). The specific nature of phagocytic receptor
expression by AMs is largely dependent upon, and regulated by, the constituent
components of ASF. Unique features of AMs include their ability to secrete IL-10 and
TGFβ (anti-inflammatory cytokines), produce an attenuated oxidative response to particle
ingestion, and express increased MR levels; all indicative of an “alternative activation”
state (41-43). However, AMs can also be stimulated to produce TNFα, IFNγ, G-CSF,
IL-1 and IL-12 (proinflammatory cytokines) as well as chemokines that effectively
recruit and activate circulating neutrophils to aid in pathogen elimination (44). Thus, the
8
AM response to infection is a key determinant in the outcome of the overall pulmonary
innate immune response, and it is mediated in part by the specific repertoire of cell
surface receptors that become activated. Interstitial DCs share some elements of surface
receptor expression with AMs (e.g. CRs, FcγRs, and select PRRs), but the comparative
levels of expression differ. Also, DCs have an increased capacity for antigen
presentation and leukotaxis and respond to infection by producing TH1- or TH2-type
cytokines (45). Importantly, phenotypic differences between monocyte-derived cells,
such as AMs and DCs, are not dogmatic because of their inherent adaptability (46).
Phenotypic parameters of monocyte-derived cells; such as surface marker expression,
cytokine production, and migratory capacity; are highly dependent upon, and change
rapidly, with environmental cues.
Abundant immunostimulatory PRRs that do not mediate phagocytosis and that
are found prominently on most of the cell types listed above include the Toll-like
receptors (TLRs). Several distinct TLRs have been characterized, and a discussion of
major transmembrane TLRs is germane to F. tularensis-associated pathogenesis.
Generally, recognition of bacterial LPS involves the LPS receptor; a complex composed
of TLR4, MD2, and CD14 (47-49). Binding of LPS to this receptor complex also
requires LPS-binding protein (LBP), which recognizes the lipid A portion of LPS. CD14,
an extracellular protein with no cytoplasmic signaling domain, interacts directly with
LPS-bound LBP. This interaction enhances recognition by TLR4, which contains leucine
rich repeat (LRR) cytoplasmic domains that activate intracellular signaling pathways.
TLR2 heterodimerizes with TLR1 or TLR6 to recognize a variety of microbial
lipopeptides (e.g. zymosan and remnants of peptidylglycan) (50). On monocyte-derived
9
cells, TLR2 expression is dependent upon the degree of cell differentiation with the
highest amount present on monocytes compared to matured macrophages and DCs (51,
52). Activation of TLR4 and TLR2 ultimately leads to a signaling cascade that elicits
enhanced microbicidal efficacy and the production of proinflammatory cytokines.
Important pathogenic features of tularemia
Classic virulence factors, such as exotoxin, are not produced by F. tularensis (53,
54). Furthermore, due to well characterized modifications of its structure (discussed in
detail below), F. tularensis associated LPS is not recognized by LBP and does not elicit
the hyperinflammatory systemic response associated with endotoxins of other Gramnegative bacteria (55-58). Virulence associated with F. tularensis is, thus, most likely
due to its ability to replicate rapidly in vivo and to proliferate in multiple organs. The
extensive inflammatory response that occurs during late stages of tularemia is the most
likely primary cause of multi-organ system failure and mortality.
F. tularensis is a fast-growing, facultative intracellular organism that has the
capacity to infect multiple cell types including monocytes, macrophages, DCs,
neutrophils, lymphocytes, epithelial cells, many types of laboratory cell lines, and other
types of cells (59-65). Immediately upon inhalation, F. tularensis is found in situ
predominantly in AMs and subsequently in both macrophages and recruited neutrophils
(66). However, bacilli are also located within epithelial and endothelial cells. Infection
of non-immune cell types may provide local protection from antibodies and other
components of immune surveillance and may also act as reservoirs for local bacterial
replication during the early stages of disease. Alveolar type II cell infection can also
10
result in an enhanced immune response by activating chemokine expression and the
recruitment of naive leukocytes (67).
Although it is likely that incubation occurs in epithelial and endothelial cells as
well as in leukocytes, bacilli migrate systemically within macrophages (66, 68).
Pulmonary infection results in the recruitment of macrophages (CD11b+/CD11c+ cells) to
the airway, which phagocytose bacteria and subsequently migrate within the
reticuloendothelial system (69). Carriage of bacteria to mediastinal lymph nodes can be
inhibited by blocking the migratory capacity of respiratory macrophages. CD11c is
traditionally considered as a marker for DCs and multiple studies have identified
migratory cells containing F. tularensis as CD11c+; thereby, it has been reported that
DCs are the primary cell type responsible for the systemic spread of bacteria from the
lung to the periphery (69-71). However, resident AMs also present CD11c on their
surface and may adopt a migratory phenotype upon ingesting F. tularensis (72, 73).
Thus, the true identity of the predominant cell type responsible for dissemination of
bacteria is, currently, unknown but may be important for the design of therapeutic
interventions that target phagocytosis or cell signaling responses.
Early animal studies reported conflicting data with regard to free F. tularensis in
the bloodstream (68, 74). However, recent evidence of a role for sepsis in the
pathogenesis of tularemia is convincing. The inoculation of mice with both virulent (F.
tularensis Type A, and subsp. novicida) and attenuated (LVS) strains results in
significant extracellular bacteremia (75-77). Subsp. novicida mutants lacking the
FN_0444 gene, which encodes a 58 kDa protein with no known function or homology to
characterized proteins from unrelated bacteria, are attenuated and do not elicit significant
11
host inflammatory responses, including hypercytokinemia and leukocyte recruitment
(77). The authors of this study concluded that attenuation was a result of the inability of
mutants to cause significant sepsis because S100A9, a marker for severe sepsis that is
released primarily from neutrophils, was less detectable when compared to infections
with wildtype subsp. novicida. Interestingly, the homolog in F. tularensis, encoded by
FT_0918, is also necessary for virulence in mice, and a mutant is protective as a vaccine
against lethal doses of wildtype strains (78).
Undoubtedly, both extracellular and intracellular survival mechanisms are
required for the pathogenesis of tularemia. A well-characterized pathogenicity island is
present, in single or duplicate copies, on the F. tularensis genome (79). Mutation of the
genes encoded in the pathogenicity island results in attenuation. Also, studies that
employ “shot-gun” strategies to produce random genomic mutations consistently identify
pathogenicity island associated genes as major virulence factors. Each gene, with the
exception of pdpD (absent in Type B strains), is required for growth within macrophages
(80). Select pathogenicity island genes share homology with genes encoding a protein
cluster that can form a Type VI secretion apparatus in other species (79), but neither the
assembly nor function of a directed secretion apparatus nor secreted factors have
conclusively been demonstrated for F. tularensis. Nonetheless, specific gene products,
such as IglC, are clearly necessary to confer the ability of bacilli to escape from
phagosomes, as discussed below, and to survive intracellularly to maintain virulence.
Interestingly, Bosio et al. reported that the pretreatment of mice with clodronate, which
depletes AMs, enhanced their resistance to infection with LVS (71). To further illustrate
the importance of intracellular survival in pathogenesis, it has been shown that infection
12
of rats with subsp. novicida is not lethal, which is likely because this subspecies does not
grow efficiently in rat macrophages (59).
Characteristics of the host response to F. tularensis
The immune response to F. tularensis is highly dependent on the route of
infection. Intradermal infection results in a robust and immediate inflammatory response
that includes the early production of proinflammatory cytokines including INFγ, TNFα,
IL-12 and IL-1β (81, 82). Conversely, the immune response to inhalation of F. tularensis
is characterized by an early asymptomatic phase without significant tissue pathology and
a late hyperinflammatory phase. During the initial stage of respiratory infection,
containment of bacteria by the innate immune system fails, and dissemination occurs via
the reticuloendothelial system leading to inoculation of distal organs such as lymph
nodes, spleen, and liver.
Gross and histopathological studies of aerosol infections using multiple animal
models are consistent with this pattern of progression (83-85). In summary, the degree
of morbidity for models of tularemia correlates with pathological changes in the spleen
and liver, which may not be apparent until the day of death. Both organs may become
significantly enlarged, or may even atrophy, but are consistently mottled with
hemorrhagic infarcts and foci of necrotic tissue within poorly organized lesions.
Interestingly, in many cases, the lungs are not remarkably changed in terms of gross
pathology and retain buoyancy. Histopathological changes in the lung are also
unremarkable until the day of, or the day prior to, death. Although less dramatic when
compared to the liver and spleen, well-demarcated, and sometime granulomatous, foci
13
containing liquified tissue, bacteria, and infiltrating monocytes and neutrophils are
apparent. Necrotic lesions in the lungs tend to occur perivascularly, adjacent to smalland medium-sized vessels, with occasional vasculitis and destruction of the vessel wall.
Generalized parenchymal lesions are less frequent with mildly thickened alveolar septal
walls. Thus, the severity of pneumonic tularemia, compared with other routes of
infection, is most likely due to an ineffective local innate immune response that is
permissive to bacterial replication and to carriage of infected cells away from the site of
primary infection.
Global microarray studies using mice infected with Type A strains are consistent
with this pattern of early inflammatory suppression followed by a sudden onset of
systemic inflammation (86). The mean time to death in mice is 4-6 days. Few prominant
changes in host gene expression occur until the fourth day post infection, at which point
significant increases in TH1 type cytokine (INFγ, TNFα, and many known INFγ- and
TNFα-regulated genes) expression occurs. In lung homogenates, INFγ is undetectable
until day 3 post infection (87). In bronchoalveolar lavage (BAL) fluid of mice infected
with subsp. novicida, neutrophil specific chemokines (ELR+ CXC chemokines) are
expressed at baseline levels during the course of immune suppression, then rise
dramatically concurrent with INFγ induction (76). Interestingly, Malik et al. showed that
neutrophil migration into airways inoculated with F. tularensis was attenuated in mice
deficient in matrix metalloproteinase 9, although bacterial loads were equal to those in
wildtype mice. This resulted in decreased histopathology and significantly longer
survival rates for mutants (88).
14
The delayed hyperinflammatory systemic response to pulmonary infection is
likely the primary cause of mortality associated with tularemia. However, the precise
signals that lead to a switch from inflammatory suppression to hyperactivity are largely
unknown. The delayed inflammatory response may be at least partially explained by
sepsis since cytokines that are diagnostic for severe sepsis, including CXCL10, IL-6, and
IL-8, increase systemically by 1,000-fold beginning at 3 days post infection (76). Sepsis
is also generally associated with a shift towards the expression of TH2 related cytokines,
and increases in IL-10 and IL-4 also occur within a delayed time course (76). Another
factor that may promote a hyperinflammatory response is the release of DAMPs from
necrotic host cells. For example, HMGB-1 is an abundant nuclear and cytoplasmic
protein with homeostatic functions under normal conditions, but has proinflammatory
signaling capabilities when released from cells.
Protective role of an early proinflammatory response
A rapid proinflammatory response to F. tularensis infection is protective.
Compared to pneumonic tularemia, subcutaneous infections require a much larger lethal
dose, partially due to a rapid induction of IL-12, TNFα and INFγ (89). Mice treated with
neutralizing antibodies against INFγ or TNFα and mutant murine strains deficient in
INFγ production succumb to normally sublethal intradermal doses of LVS (90, 91). An
early and appropriate inflammatory response in the lung may also be protective. It was
shown that the survival of mice treated exogenously with IL-12 surpassed untreated mice
after intranasal inoculation with LVS (92). The protective role of IL-12 was dependent
on its ability to induce INFγ. The mechanism of INFγ or TNFα protection is not entirely
15
known. For example, protection cannot completely be explained by the induction of
nitric oxide because iNOS-/- mouse macrophages treated with INFγ restrict the
intracellular growth of LVS (93). In human monocytes, INFγ also limits the intracellular
growth of LVS independently of NO and other reactive oxygen species (68).
Alternatively, INFγ may inhibit the ability of engulfed bacteria to escape the phagosome
(94-96).
F. tularensis stimulates proinflammatory responses predominantly via stimulation
of TLR2 and the inflammasome. Induction of TNFα (which commonly requires TLR2
activation) is enhanced when thioglycollate-treated peritoneal macrophages are infected
with mutant LVS strains that do not escape the phagosome, probably since TLR2 is also
contained within and induces signal transduction from the phagosomal membrane (97).
Stimulation of TLR2 is likely mediated by the bacterial surface lipoproteins, TUL4 and
FTT_1103 (98). Notably, F. tularensis LPS is known not to bind and activate TLR2
(99). Interestingly, TLR2 activation in the airways of mice by LVS may enhance
immunosuppression by inducing alternatively activated macrophages that produce TGFβ
(100). However, TLR2-/- mice are highly susceptible to normally sublethal infectious
doses (101). Systemic release of IL-1β and IL-18 is dependent upon inflammasome
activation within the cytosol of leukocytes. Upon phagosomal escape, recognition of F.
tularensis by pyrin results in inflammasome activation in human monocytes (102). Also,
mice deficient in key inflammasome components are susceptible to normally sublethal
infectious doses (103).
16
Pulmonary innate immune suppression by F. tularensis
F. tularensis does not simply fail to elicit a proinflammatory immune response
early during the course of infection but actively inhibits proinflammatory signaling by
incompletely understood mechanisms. A study by Bosio et al. used flow cytometry to
analyze infected cells isolated from murine lungs. They found that Schu S4, a Type A
strain, infects predominantly CD11c+ cells (AMs and DCs), but that they fail to become
activated (70). As expected, considering the microarray studies discussed above,
proinflammatory cytokines such as TNFα, IL-12 and IL-1β are not induced in infected
cells nor do they increase the expression of MHCII or CD86 (antigen presentation
effectors). In contrast, a significant increase in TGFβ, an anti-inflammatory cytokine that
decreases the microbicidal capacity of leukocytes, occurred. Indicative of active immune
suppression, the cotreatment of infected mice with a normally potent proinflammatory
molecule, LPS isolated from Escherichia coli, failed to induce lung histopathology or
activation of CD11c+ cells (70). However, although blocking the activity of TGFβ
reduced the rate of bacterial growth in the lung, complete restoration of an effective
inflammatory response did not occur. This suggests that additional factors contribute to
immune suppression in the lung by F. tularensis.
F. tularensis also inhibits the in vitro stimulatory effects of INFγ and of agonists
for TLR2 and TLR4. Parsa et al. showed that the treatment of human or murine
monocytes with INFγ resulted in attenuated STAT1 activation when cells were already
infected with subsp. novicida or LVS (104). STAT1 is a component of the IFNγ
signaling pathway and must be phosphorylated during the course of INFγ-mediated gene
regulation. They found that SOCS3, an inhibitor of STAT1 phosphorylation, was
17
upregulated in response to an unknown heat-stabile bacterial factor that is most likely
either lipophilic or that interacts with unidentified host cell surface receptors. In studies
using human monocytes and monocyte-derived DCs, it has been recently reported that
challenge with subsp. novicida, Type A F. tularensis or conditioned growth media results
in the inhibition of cell activation by TLR agonists that include LPS, zymosan, or
PAM3CSK4 (105, 106). Chase et al. showed that inhibition occurred in both infected
cells and noninfected bystander cells. They also found that the inhibitory effect was
mediated by an unknown heat-stable bacterial factor but ruled out a potential contribution
by sloughed LPS molecules. LVS has a similar capacity to inhibit TLR-mediated
activation in a murine macrophage cell line, and this effect was dependent on iglC
expression, which is a component of the pathogenicity island (107).
F. tularensis uptake by macrophages
The mechanical processes and downstream effects of phagocytosis are highly
complex events involving several potential ligand-receptor interactions and potential
influences by non-receptor-mediated events (108). Multiple phagocytic receptors may
simultaneously recognize microbes resulting in potentially cooperative or noncooperative downstream signaling events. As a further complication, non-phagocytic
receptors that do not have a role in microbial uptake, per se, interact with microbial
ligands within the context of the phagosomal cup and can also modulate signaling events
initiated by phagocytic receptors. The cumulative effect of receptor interactions results
in a specific downstream signaling cascade that affects the ultimate fate of ingested
18
particles (e.g., phagosome trafficking and maturation) and the cellular response to
infection (e.g., inflammatory cytokine release).
Efficient uptake of F. tularensis by human macrophages requires opsonization in
complement-sufficient media and occurs via a novel macropinocytic process involving
“pseudopod loops” (109). Pseudopod extension is dependent upon actin polymerization
and PI3K-mediated signaling. The contributing cellular receptors and microbial
mediators initiating this unique process have only partially been defined. Several studies
using various F. tularensis subspecies and strains have all shown that the removal of
complement dramatically reduces phagocytosis by mouse macrophages and by human
monocytes, macrophages and DCs (60, 109-111). Furthermore, complement receptors
co-localize with F. tularensis-containing phagosomes, and complement-mediated uptake
is inhibited by anti-CR3 and anti-CR4 antibodies. In the absence of functional
complement, F. tularensis uptake by phagocytes is relatively inefficient and requires
higher multiplicities of infection. However, non-complement dependent receptors that
may also have a role in enhancing F. tularensis uptake by human macrophages include
the MR and FcγRs (110, 111). Preopsonization with SP-A can also enhance uptake via
an unknown mechanism (110). Opsonization with an unknown heat-labile serum
component also enhances SR-A-mediated uptake of LVS in a mouse macrophage cell
line and in transfected epithelial cells (112). Finally, it was recently shown that
elongation factor Tu, which is a cytosolic protein that is also expressed on the surface of
LVS, can act as a ligand for nucleolin and this may enhance phagocytosis (113).
Signaling pathways downstream of cell surface receptors do not propagate
independently, and the effect of crosstalk may be enhanced when receptors are associated
19
with membrane microdomains (114). On mouse macrophages, phagocytosis of LVS was
shown to occur in association with caveolin-1-containing lipid raft membrane
microdomains (115). It is currently not clear which, if any, of the above mentioned hostpathogen interactions contribute to the immunosuppressive capacity of F. tularensis or its
ability to survive intracellularly; this is a major focus of ongoing research. The unusual
mechanism of uptake involving pseudopod loops may be a significant clue that
unidentified receptors or the lectin-binding domain of CR3 interacts with F. tularensis.
Furthermore, it is possible that unidentified receptors, in addition to CRs, have a role in
suppressing an appropriate immunostimulatory response to uptake.
Several intracellular pathogens are known to use CR-mediated phagocytosis to
gain access into host cells, including Mycobacteria (116-119), Neisseria (120),
Leishmania (121, 122), Legionella (123), Histoplasma (124), Listeria (125),
Trypanosoma (126). The mechanical process of CR-mediated phagocytosis is distinct
from conventional opsonic phagocytosis, which is typified by FcγR-mediated uptake. In
the case of the latter, pseudopods extend from the phagocyte and advance across the
surface of an antibody-decorated target as increasing numbers of IgG-FcγR interactions
occur (127-129). Ligation of the receptor results in alteration of its cytoplasmic domain,
which further results in the activation of secondary signaling molecules leading to actin
polymerization and advancement of the pseudopod. This is known as the “zippering”
model (130). Engulfment mediated by complement has been described as a modified
“zippering” event that microscopically appears as a “sinking phagosome” (131-133).
Phagocytosis of Mycobacteria and Salmonella can occur by this process, which does not
utilize pseudopod extensions. Opsonized particles attach and form a depression in the
20
cell surface with fewer attachment points between opsonic complement components and
cognate receptors resulting in a more loosely adherent phogocytic cup. Some
intracytosolic signaling molecules that have a role in actin remodeling and other signaling
events are shared between FcγR- and CR-mediated uptake pathways, and some signaling
molecules are unique to each pathway (134, 135). Receptor cooperativity is largely
dependent upon the spaciotemporal location of such secondary molecules that may
positively or negatively influence one another.
A third major mechanism of engulfment by cells is macropinocytosis, or
membrane ruffling. This is known as the “triggering” mechanism of phagocytosis (130,
136). Macropinosomes do not rely on opsonin-receptor interactions to shape the
phagosome as in a purely zippering mechanism. Rather, sheet-like ruffles with
microfilamentous cores project from the cell membrane and nonspecifically engulf
extracellular material. Triggered ruffling can occur globally on the cell membrane when,
for example, cells are treated with growth hormones (137, 138). Using a specialized
secretion apparatus that transports effector molecules into targeted cells, pathogens such
as Salmonella, Shigella, and Legionella induce localized membrane ruffling and
engulfment by both professional and non-professional phagocytes (139-143).
Internalization of Legionella can occur through a unique process called “coiling”
phagocytosis (144). Coiling is characteristic of simultaneous triggering and zippering
because a tightly juxtaposed pseudopod repeatedly envelops a bacillus before the
occurrence of membrane fusion and cell entry.
Both triggering and zippering mechanisms of CR3-mediated phagocytosis are
described. C3bi-mediated internalization of Neisseria gonorrhoeae into primary cervical
21
cells (a type of non-professional phagocyte) elicits membrane ruffling by a triggering
mechanism (120). Conversely, whereas complement-opsonized zymosan appears to sink
into CR3-expressing CHO (chinese hamster ovary) cells, neutrophils, or monocytes
(suggestive of zippering); engagement of these same cells by unopsonized zymosan
(capable of adhering to the CR3 lectin-binding domain) results in pseudopodia extensions
that are reminiscent of membrane ruffles and a triggering mechanism (145). Thus, the
specific mechanism by which CR3-mediated phagocytosis ensues is likely dependent
upon the precise nature of the particle in question as well as the region(s) within CR3 to
which the particle is bound. Furthermore, whether opsonic or nonopsonic CR3-mediated
phagocytosis involves membrane triggering may rely on the intracellular locations of
secondary GTPase signaling molecules (146), which may be cell-type dependent or
dependent upon the binding of cooperative receptors.
Ultrastructurally, pseudopods that engulf F. tularensis do not tightly adhere to the
bacterium indicating that receptors do not zipper around the organism. Despite the
potential expression of a secretory apparatus by F. tularensis, neither formalin fixation,
heat killing, nor protease treatment have an effect on the rate and nature of phagocytosis.
So, it is unlikely that pseudopod formation occurs by a mechanism like that employed by
Salmonella and Shigella, involving the injection of effector molecules into the cytosol of
ingesting phagocytes. Rather, these results suggest important microbe-receptor
interactions that occur in addition to C3bi ligation and likely also have major downstream
effects on phagosome trafficking and intracellular signaling. Peroxidation of bacterial
surface moieties by periodate treatment results in uptake by a process resembling
conventional CR-mediated phagocytosis (147). This indicates a role for preformed
22
surface glycans or glycolipids rich in carbohydrates interact directly with CR#,
unidentified receptors, or that modify the nature of complement activation and degree of
opsonin deposition. Mutant strains that do not produce LPS-associated O-antigen initiate
a phagocytic process more akin to conventional phagocytosis involving pseudopod loops
that are less spacious and adhere more tightly to bacteria (147). This may result from
increased complement activation and a greater number of component-CR interactions
(see Chapter 2).
Once inside the cell, maturation of F. tularensis-containing phagosomes is
arrested at a late endosomal-like stage. A modest degree of phagosomal acidification
occurs within 15 to 30 minutes of phagocytosis (148). Microscopically, the F. tularensis
containing vacuole takes on an unusual appearance with a dense fibrillar coat surrounding
it (149). Within one hour, the phagosomal membrane is degraded by an unknown
mechanism (that may or may not require phagosomal acidification) and rapid
intracellular replication within the cytosol ensues (95, 148, 150-152). In lung epithelial
cells, as in macrophages, F. tularensis escapes the endocytic vacuole and replicates upon
gaining access to the cytosolic compartment (153). F. tularensis is also unique among
facultative intracellular pathogens because it inhibits the oxidative burst upon uptake by
neutrophils in order to survive (154). At 20 hours post-infection in mouse bone marrowderived macrophages, bacteria re-enter the endocytic pathway by an autophagy-mediated
process and reside in double membrane-bound vacuoles (150). The role of autophagy
remains unclear because bacteria continue to grow and replicate intracellularly until
viability of the host cell is compromised. The nature of F. tularensis-induced cell death
23
may be either necrotic or apoptotic, involving MAPK-associated and caspase-9dependent pathways (155, 156).
Constitution of the cell wall of F. tularensis
The cell surface of Gram-negative bacteria consists of extracellular proteins,
glycans, and lipoglycans. On F. tularensis, the outermost cell surface consists of a
glycocalyx, or a capsule, and LPS O-antigen. The innermost components include
extracellular membrane proteins, LPS core polysaccharide, and LPS lipid A. The
specific nature of the F. tularensis capsule has not been determined conclusively. An
electron lucent material, typical of a loose glycocalyx, surrounds bacilli that are grown in
defined media, but its expression likely requires specific growth parameters (157, 158).
Hood reported that decapsulation of F. tularensis occurs in hypertonic saline, and that
isolated capsular material is biochemically distinct from the cell wall of decapsulated
bacilli (159). A putative capsule locus in the LVS genome, containing orthologous genes
to capB and capC of Bacillus anthracis, was reported, but it is unknown whether their
gene products contribute to capsule formation on F. tularensis (160).
LPS, associated with Gram-negative bacteria, constitutes the outer leaflet of the
outer membrane [reviewed in ref. (161)]. Lipophilic acyl conjugates associated with the
lipid A portion of LPS anchor it to the membrane. Lipid A is the biologically active
component of LPS that elicits pathological responses to sepsis. The canonical structure
for lipid A, expressed by other Gram-negative bacteria such as Escherichia coli, is a
disaccharide moiety that is biphosphorylated and hexa-acylated (162, 163). Lipid A
bound to LBP, which is secreted by host cells, is recognized by TLR4 and stimulates a
24
proinflammatory response. However, Gram-negative bacteria that produce tetra- or
penta- acylated lipid A species have a decreased immunostimulatory capacity (164).
Other modifications to lipid A similarly decrease its TLR4-activating capacity and these
include the addition of various carbohydrates, altered phosphorylation, and modifications
in conjugate acyl chain length.
The structures of lipid A isolated from F. tularensis subsp. holarctica and
novicida are similar and contain a β-(1,6)-linked glucosamine disaccharide backbone
each with amide-linked fatty acids (Fig. 1.1). Fatty acid conjugation occurs at the 2
((18:0)-3-OH) and 2’ (branched acyloxyacyl group: (18:0)-3-(16:0)) positions and esterlinked fatty acids at the 3 ((18:0)-3-OH) position, but not the 3’ position (165-167). This
acylation pattern contrasts with the shorter (12-14 carbons per chain) hexa-acylated lipid
A species associated with inflammatory enteric bacteria. A phosphate is attached to the
diglucosamine backbone on all strains tested except for LVS, in which a phosphate is
located on the reducing moiety at position 1. The 4’ phosphate is absent due to the
phosphatase activity of LpxF (168). Absence of the 4’ phosphate is known to reduce
stimulation of TLR4 (169). Instead, the addition of mannose at the 4’ position can occur.
A recent finding is that over 95% of the LPS expressed by subspecies novicida is in the
form of free lipid A with galactosamine, carrying a net positive charge, attached to the
position 1 phosphate and glucose substituted for core region sugars at the 6’ position of
the non-reducing lipid A galactosamine (165, 168). Mass spectrometric analyses of lipid
A from clinical isolates of subsp. tularensis and holarctica suggest the expression of
common structures among all F. tularensis subspecies (99).
Core- and O-antigen-associated polysaccharides are also conserved among
25
Figure 1.1. Structure of lipid A, core, and O-antigen molecules synthesized by F.
tularensis subspecies tularensis, holartica, and novicida. The lipid A structure consists
of a β-(1,6)-linked glucosamine disaccharide with amide-linked fatty acids at the 2- and
2’-positions, and ester-linked fatty acids at the 3-position. Lipid A carbohydrate
modifications include the addition of galactosamine through the 1-position phosphate,
mannose at the 4’-position and glucose at the 6’-position. Lipid A molecules that have
glucose in their structure would not be modified by the addition of Kdo-core-O-antigen.
Unless noted on the structure, modifications are present in all F. tularensis subspecies.
Linkages of individual carbohydrate residues are shown. Core: Kdo = 2-keto-3-deoxy-Dmanno-octulosonic acid; Man = mannose; Glc = Glucose; GalNAc = N-acetyl
galactosamine. O-Antigen: QuiN4Fm = 4,6-dideoxy-4-formamido-D-glucose;
GalNAcAN = 2-acetamino-2-deoxy-D-galacturonamide; QuiNAc = 2-acetamino2,6,dideoxy-D-glucose; Qui2NAc4NAc = 2,4,-diacetamino-2,4,6-trideoxy-D-glucose.
From Gunn et al., Ann NY Acad Sci 1105:202 (161).
26
27
F. tularensis subspecies with the exception of novicida. For each, the core region is
attached to the 6’ position of lipid A-associated galactosamine by 2-keto-3-deoxy-Dmanno-octulosonic acid (Kdo) (167). Unlike the core regions from other Gram-negative
bacteria, that of F. tularensis lacks phosphate modifications and contains a single Kdo
sugar. The novicida core differs subtly from other species with the addition of glucose
residues. Subspecies tularensis and holarctica produce identical O-antigen
polysaccharide chains (21, 170-172). It is a repeating tetramer containing β-DQui4NFm-α-D-GalNacAN-α-D-GalNAcAN-β-D-QuiNAc. O-antigen associated with
novicida contains α-D-GalNAcAN substituted at the first position and β-DQui2NAc4NAc at the fourth position (173).
F. tularensis and phase variation
The capacity for phase variation by F. tularensis was first reported in 1951 (174).
Based on the appearance of colonies illuminated by opaque lighting, two phase variants
were described for subsp. tularensis. Selected colonies could be subcultured with a
relatively stable phenotype, but reversion did occur. So-called “grey” variants, as
opposed to the predominant “blue” wildtype variants, produced small colonies with rough
morphology. The grey variants were also less virulent, less immunostimulatory, and
grew more slowly compared to blue variants.
Since the initial description of Type A F. tularensis phase variants, the capacity
for holarctica strains and for LVS, but not for novicida, to phase vary has been shown.
Grey LVS variants are also less virulent than are blue variants (subcutaneous LD50 values
28
of 109 and 105, respectively) (175). The relative decrease in virulence may contribute to
the reduced efficacy of grey variant strains as a vaccine, compared with LVS, against
subsequent challenge with subsp. tularensis. The rate of phenotypic variation increases
in conditions that cause stress to bacteria, notably by nutrient exhaustion with extended
periods of stationary phase culture, in vivo passage, and intracellular growth. The
potential for blue to grey variation in large commercial preparations of LVS produced for
public vaccination is a major cause of FDA scrutiny, particularly because the causative
genetic mechanism for variation is unknown. Furthermore, the phenotypic and molecular
differences between variants in toto are unknown (161).
It is clear that major differences in LPS structure exist between blue
and grey variants. Based on an increased capacity to elicit nitric oxide from rat peritoneal
macrophages, the LPS derived from LVSG is predicted to have an altered lipid A
composition when compared to wildtype LVS (176). Definitive studies showing
structural lipid A differences have not, however, been reported. Structural differences in
O-antigen composition between blue and grey LVS variants have also been implicated.
In one study, monoclonal antibodies were used to demonstrate differences between
variants (176). The antibodies were specific for either the tularensis-type O-antigen
polysaccharide or the novicida-type O-antigen polysaccharide. Indicative of an Oantigen structural change occurring upon phase variation, blue variant O-antigen was
detected only by the tularensis-type specific antibody, but grey variant O-antigen was
detected by both. Furthermore, upon reversion from grey to blue colony morphology,
detection by the novicida-type specific antibody was lost.
Grey variant isolates have also been described that are completely deficient in O29
antigen expression (177). As with previously described grey variants, these isolates also
had a reduced capacity to survive in macrophages and provided decreased vaccination
efficacy. A third class of grey variants has also been described that produces an
intermediate amount of O-antigen (161). These isolates are also susceptible to
macrophage killing but, surprisingly, were resistant to treatment with H2O2 and
polymixin B compared with LVS. Thus, based on the given deviation in grey variant
LPS characteristics, F. tularensis phase variants may not represent dichotomous
phenotypes, per se. Instead, it may be more appropriate to consider the process of phase
variation as a bacterial response to stress, potentially involving multiple signaling and
genetic pathways, and potentially leading to multiple bacterial phenotypes. Importantly,
with regard to phase variation, changes to LPS have not been linked directly to changes
in gene transcription or enzyme modifications. It is equally important that coexistent and
undiscovered changes (occurring in addition to, and independently of, LPS alterations)
between wildtype and variant phenotypes may confer environment-specific survival
advantages in unknown ways.
Virulence effects of altered F. tularensis surface structures
Glycocalyx and O-antigen are well-characterized virulence factors for many
Gram-negative pathogens (163, 178). They, generally, confer resistance to host innate
immune effector molecules like antimicrobial peptides that disrupt bacterial membranes.
They also enhance intracellular survival by buffering membrane active antimicrobials
like nitric oxide. LVS and subsp. novicida O-antigen mutant strains as well as a capsule
mutant strain derived from LVS were found to be susceptible to lysis in serum, unlike
30
their parent strains (179-184). Similarly, LVS grey variants that completely lack Oantigen are also susceptible to lysis in serum (177). Interestingly, Sandstrom et al.
reported that deposition of complement component C3bi, an opsonin that is required for
efficient CR-mediated phagocytosis, did not occur on wildtype LVS (183).
Indicative of the importance of complement activation and regulation in the
pathogenesis of tularemia are the abilities of F. tularensis to survive extracellularly as
well as to efficiently infect and modulate the inflammatory response of macrophages
through a process mediated in part by CRs. Evidence exists that surface components of
the F. tularensis cell wall, namely capsule and LPS O-antigen, have a role in resistance to
the lytic effects of complement and in the regulation of opsonin deposition. However,
little research has focused on specific interactions between complement components and
cell surface molecules. Also, little is known about potential differences regarding the
nature of complement activation by various pathogenic and attenuated F. tularensis
subspecies. Fundamentals of the complement system will be reviewed briefly followed
by a description of the specific aims of the research contained within this thesis.
31
1.2 Complement
Overview
During the late 1800’s, two competing theories existed regarding the mechanism
by which microorganisms are killed in serum. Elie Metchnikoff championed the
“cellular theory” and demonstrated that cells in blood were capable of ingesting
microbes (185). The “humoral theory” was proposed by Fodor, Nuttall and Buchner who
demonstrated the presence of a heat-labile component of cell-free serum with the capacity
to lyse bacteria (186). It was later, soon after the discovery of antibodies, that Bordet
showed that the temperature-sensitive lytic component (present in non-immune sera)
could be added to heat-treated immune sera to restore its lytic capacity (187). He later
described the first complement fixation test to demonstrate the role of complement in cell
lysis in a quantitative manner. Of course, both theories of innate immunity proved to be
true and it is now known that immune cells, antibodies, and complement work in concert
to both enhance and regulate the activities of one another in the course of an immune
response to infection.
The complement cascade uses approximately 30 highly evolutionarily conserved
proteins some of which are found soluble in blood and tissue, and some are found on cell
membranes. Activation of the complement cascade occurs in the presence of nonprotected, or ‘foreign’, surfaces lacking in complement regulatory components (discussed
32
below). Such surfaces include large antibody-antigen complexes, altered host cells, and
microbes. Upstream sensory components of the cascade bind to non-protected surfaces
and, generally, become functional (or activated) as proteases upon cleavage. The
cleavage event induces conformational changes in native components that expose
intrinsic functional domains for which specific downstream complement components are
substrates. The result is a sequential chain of activating events as naive components are
activated by converted upstream components.
The entirety of the cascade can be divided into four distinct pathways (Fig. 1.2)
that may or may not be activated independently of one another (188). There are three
activation pathways; the classical, lectin, and alternative pathways. So-called “sensory”
components of the three activation pathways detect non-protected surfaces. Subsequent
generation of C3- and C5-convertases (composed of the activated products of upstream
components) can result in the activation of the fourth, or terminal lytic, pathway. The
classical pathway normally becomes activated upon antibody binding to a non-protected
surface. However, it can also be activated in the absence of antibody. The lectin
pathway uses the same downstream components as the classical pathway, but employs
soluble lectins (e.g. mannose-binding lectin) as upstream sensory components to
recognize specific sugar moieties common to the surfaces of microbes. Activation of the
alternative pathway occurs upon direct covalent binding of complement component C3b
to pathogens or altered self (non-protected) surfaces.
33
Figure 1.2. The Complement activation and terminal pathways. Deposition of C3b
via the classical and lectin-mediated pathways or via the alternative pathway is mediated
by C3-convertases (C4b2a and C3bBb, respectively). C3b deposition can also occur via
the alternative pathway via the amplification loop. C3b can form complexes with parent
C3-convertases to form novel complexes with C5-convertase activity (C4b2a3b and
C3bBb3b). Fragments derived from C3b (C3bi, C3dg, and smaller fragments) cannot
participate in C5-convertase formation (indicated by the black bar). C3b and C3bi are the
major complement-associated opsonins.
34
Potential outcomes of complement activation
Complement is a highly complex system that is both multifunctional and tightly
regulated (189). A major function of complement is the release of smaller activation
products, cleaved fragments derived from native components, which can act as
proinflammatory signals localized to sites of complement activity. Such fragments have
anaphylactic and chemotactic capabilities by interacting directly with leukocyte
receptors. Their release rapidly increases the local population of immune cells capable of
combating infection. If unregulated, these components can also cause disease pathology
resulting from collateral tissue destruction mediated by recruited inflammatory
leukocytes. Another major function of complement is opsonization. Specific, larger,
cleaved complement component fragments bind to particulate surfaces and act as
opsonins for phagocytic receptors on leukocyte surfaces. Opsonization increases the
efficiency of particle uptake by phagocytes. This enhances clearance of potential
pathogens, cellular debris (homeostatically important for wound healing and tissue
remodeling), and circulating antibody-antigen complexes. The third major function of
complement is membrane lysis. The membrane attack complex (MAC) is composed of
terminal components of the cascade. The MAC produces pores in a membrane, thereby,
disrupting intracellular osmotic equilibrium. Thus, MAC formation is a powerful and
cell-independent process contributing to innate immune defense. However, unregulated
MAC activity can also be detrimental to the host. For example, many autohemolytic
syndromes occur when inappropriate complement activation on erythrocyte surfaces
results in their lysis and anemia for the host.
35
The classical pathway
The classical pathway of complement activation is initiated by antibody (IgM or
clusters of IgG) binding to a foreign surface and subsequent recruitment of C1. C1 is a
pentameric complex formed by C1q, which recognizes antigen-associated antibody, and
two molecules each of C1s and C1r, which have serine proteolytic enzymatic activity.
The structure of C1q is described as “a bunch of tulips” (190). It is composed of eighteen
polypeptide chains that form six, collagen-like, N-terminal, triple-helices and six, Cterminal, trimeric, globular heads, which bind to the Fc region of antibodies. Binding to
antibody causes a conformational shift in C1q, and this leads to a shift in the relative
positions of proenzymatic C1r and C1s (191). As a result of this shift, C1r becomes autoproteolytically active and cleaves itself. In turn, activated C1r rapidly cleaves native C1s,
another serine protease that activates downstream components.
Native C4 and C2 are substrates for C1s. Cleavage of C4 causes release of C4a,
a relatively small fragment with anaphylactic properties. The larger C4b fragment
contains a highly labile thioester group that is exposed upon cleavage of C4. The
thioester group is rapidly hydrolysed in solution, but in the presence of an activating
surface rich in hydroxyl and/or amine groups, it becomes covalently bound within the
immediate vicinity of C1. C1s also cleaves C2 to form C2a, which is a serine protease
that forms a complex with covalently bound C4b. The result is C4b2a, a complex with
C3-convertase activity. C3 is the central complement component that is activated by
each of the three activation pathways (Fig. 1.3). A large degree of amplification occurs
after the initial C1 binding event and C3-convertase formation. Given optimal
36
Figure 1.3. Cleavage products of native C3. Native C3 is a heterodimer composed of
an α and β chain. The reactive thioester moiety (*), which forms covalent bonds with
acceptor molecules, is protected within the 3-dimensional structure of native C3, but
becomes available upon conversion to C3b. C3a is cleaved from the α-chain of native
C3 by C3-convertases to form C3b. Factor I-mediated cleavage of C3b can result in the
formation of C3bi (shaded regions) or C3dg. Further fragmentation of C3dg to C3d can
occur by hydrolysis (e.g. via the action of trypsin). Numbers signify the molecular
weights of each fragment in kDa.
37
conditions, approximately 240 C3b molecules adhere to an activating surface for each
molecule of C1 (192). Cleavage of C3 by C3-convertases produces C3b and C3a, the
latter of which is a more potent anaphlyatoxin than C4a. There are three potential
outcomes for C3b. It can interact with the parent C4bC2a C3-convertase to form a new
complex (C4b2a3b) with C5-convertase activity. The other two potential outcomes are
discussed below, but include nucleation of novel complexes that augment C3b deposition
via the amplification loop of the alternative pathway and inactivation (further cleavage)
of C3b to form smaller fragments (including C3bi, C3dg, and yet smaller inert fragments)
that include ligands for CRs.
The classical complement pathway in innate immunity and homeostasis
In the context of infection of a naive host, or one that has not yet enabled
development of adaptive immunity to the pathogen, the classical pathway may be
activated by preformed nonspecific antibodies or by the direct binding of C1q to the
pathogen in an antibody-independent manner. Natural antibodies exist that are not
necessarily produced in response to foreign antigen. These antibodies are constitutively
produced throughout life. They commonly have low affinity to multiple, self and nonself, antigens that can include bacterial, viral, and apoptotic cell components (193).
Evidence exists that natural antibodies have major roles in maintaining self-tolerance and
in homeostasis (194). Often, antibodies that recognize cell breakdown products also
recognize pathogens and have a role in innate immunity (195, 196). Pre-existing
antibodies, but not natural antibodies, are produced following exposure to an organism
possessing cross-reacting epitopes, which are similar to those associated with the newly
38
introduced pathogen. Thus, both natural and pre-existing antibodies in non-immune
serum can, potentially, contribute to activation of the classical pathway by a novel
pathogen.
As an example of antibody-independent activity, direct C1q binding occurs on
apoptotic cells via its globular head domains (197). Potential ligands for this interaction
are numerous, but recent studies suggest phosphatidylserine as a major target (198). The
trimeric, globular, heads of C1q are composed of modular domains each with distinct
ligand specificities (199). However, due to the presence of key cationic residues in the
binding regions of each, ligands for direct C1q binding are commonly polyanionic.
Complement activation by apoptotic cells is important in the processes of normal tissue
remodeling and in homeostasis because phagocytes engulf and digest complementopsonized cell remnants. The importance of this process is exemplified in C1q-deficient
mice that fail to efficiently eliminate apoptotic cellular debris, which ultimately causes
severe glomerulonephritis (200).
Evidence also exists that the globular heads of C1q can bind directly to microbes
and activate complement. This was first shown in 1978 by Cooper et al. who incubated
radiolabeled purified components of C1 with various LPS preparations (201). They
showed that C1q bound with increasing affinity to LPS molecules containing decreasing
amounts of O-antigen and that the greatest amount of binding occurred on free lipid A.
In some cases, direct C1q-binding does not result in C1s activation. Steriochemical
interactions, in such cases, between the binding surface and the globular heads of C1q
likely do not achieve the critical shift within the C1 complex that causes C1r
autoactivation (202). Binding, without activation, of C1 would negatively influence
39
complement function. However, Cooper et al. showed that the binding of purified C1 to
bacteria resulted in cleavage of C1s. It has been shown that purified C1 can be activated
by rough strains derived from several Gram-negative bacterial species, including
Escherichia, Klebsiella, and Salmonella (203-205). In addition to binding LPS, C1q has
been shown to bind to various negatively charged outer membrane proteins of Gramnegative bacteria (206-211). However, as discussed above, binding does not
automatically confer function and activation of the classical pathway in an antibodyindependent manner by outer membrane proteins has not been demonstrated directly.
A major drawback of early studies that showed purified C1 activation by rough
LPS and rough strains was the absence of C1-esterase inhibitor (C1inh) in these assays.
C1inh belongs to a family of serine protease inhibitors (serpins), and it targets multiple
proteases, including C1r and C1s (212). It is considered a suicide inhibitor because it
acts as a substrate for proteolytic cleavage by targeted enzymes. Upon cleavage, a
conformational change in the hinge region of C1inh occurs that allows it to covalently
bind to the functional domain of the target protease to form an inert complex. C1inh-C1r
and C1inh-C1s complexes typically dissociate from C1q, which can cause a change in the
affinity of the C1q-ligand interaction and in the release of C1q as well (213, 214). C1inh
is also loosely associated with the native C1 complex in serum and inhibits spontaneous
C1 activation while it is in solution (215). Thus, studies of antibody-independent C1
activation by pathogens should include C1inh as a control in assays performed using
purified components. A study by Tenner et al. of antibody-independent C1 activation by
rough Escherichia strains illustrates this concept (216). Although purified C1 remained
40
bound to the surfaces of rough bacteria in the presence of C1inh, consumption of C2 (a
measure of functional C1s activity) was effectively blocked.
The lectin and alternative pathways
The lectin-associated activation pathway also activates C3 via formation of C4b2a
but uses surface recognition molecules distinct from C1 (217, 218). Mannose-binding
lectin (MBL) and the ficolin family of proteins bind to conserved carbohydrate patterns
and recruit MASP (MBL-associated serine proteases) proteins. MASP-1 and MASP-2
act in an analogous way to C1r and C1s to activate C4 and C2, which results in C4b2a
deposition.
The alternative pathway is initiated directly by C3, which occurs in the fluid
phase independently of surface identity and is known as C3 “tickover” (219). At a very
low rate, C3 can spontaneously interact with H2O to form an intermediate C3(H2O)
molecule that retains C3a (see Fig.1.3). Hydrated C3 activates the alternative pathway
because C3(H2O) binds to Factor B (FB), a complex that recruits Factor D (FD). FD is a
serine protease that releases Ba, which results in C3(H2O)Bb formation. Bb is analogous
to C2a and C3(H2O)Bb, which is unstable, can activate C3 (220). In its native form, the
reactive thioester moiety of C3 is highly protected. The C3 crystal structure has been
recently published and illustrates that the reactive group is sequestered within a core of
eight homologous macroglobulin domains (221). The exposed thioester associated with
C3b is highly labile and is rapidly hydrolysed, but it binds promiscuously to nucleophilic
hydroxyl or amine groups in the presence of an activating surface.
41
Once C3b is bound to an activating surface, regardless of the upstream activation
pathway involved, a positive feedback amplification loop can be initiated. Formation of
C3bBb (the alternative pathway-associated C3-convertase) is again dependent on FDmediated FB cleavage. The stability of surface C3bBb is enhanced by properdin.
Recently, properdin was also described as having sensory potential, particularly when
released from neutrophils, by binding first to activating surfaces and then inducing
activation of C3 (222, 223). Binding of an additional C3b molecule to C3bBb results in
formation of the alternative pathway C5-convertase (C3bBb3b) and activation of the
terminal lytic pathway.
The terminal lytic pathway
The terminal pathway is initiated by C5-convertases, which activate native C5 to
release C5a (the most potent anaphylatoxin) and C5b (224). Although C3, C4 and C5 are
homologous proteins within the α2-macroglobulin family; C5b does not contain a
reactive thioester group and forms a complex directly with surface-bound C3b (225).
This complex acts as a nidus for MAC and initiates the lytic pathway, beginning with the
exposure of a hydrophobic region associated with C3b-bound C5b that recruits C6. In
turn, C6 undergoes a conformational change to expose a binding site for C7. C5b-7 then
dissociates from C3b and a hydrophobic region of C7 penetrates the cell surface. C8 is a
trimeric protein with a β chain that binds to C7 and an α chain that inserts into the lipid
bilayer. Maturation of MAC is completed by the recruitment of several (from 12-19)
molecules of C9 that form an ion-permeable membrane pore.
42
Negative regulation of complement activity
Multiple regulatory mechanisms exist that negatively influence the complement
cascade. These include complement components intrinsic to host cell membranes as well
as soluble components. The outcome of complement activation, thus, depends on the
relative activity of these inhibitors in conjunction with active upstream sensory
components. Regulators of complement activation (RCA), whether membrane-bound or
soluble, share varying numbers of a repeating motif known as the complement control
protein (CCP) module (226). Additional regulatory mechanisms include: C1inh
regulation of the classical pathway (as discussed above), serum carboxypeptidase N
catabolism of anaphylatoxins, and inhibition of MAC formation on most cell surfaces due
to the binding capacity of CD59 for C5b-8 and C5b-9 (227). Finally, it should be noted
that C3- and C5-convertases are relatively unstable complexes that rapidly dissociate and
that C4b, C3b, and MAC components are only transiently able to bind target surfaces
(227).
RCA proteins are composed of strings of four to thirty CCP modules joined by
short linking sections (228). CCP domains are globular units of approximately 60 amino
acids that each contain two, highly conserved, pairs of disulfide-bonded cysteine residues
(229). The specific location of cyteines and the pattern of disulfide bond formation result
in a common three-dimensional globular structure consisting of a hydrophobic core with
the N- and C- termini at opposing ends of the long axis of the domain. Proline residues,
at the second position before the fourth cysteine (CIV-2) and at CI+3, or CI+4; a
tryptophan, between cysteines III and IV; a glycine residue, at CII+3; and 4 hydrophobic
peptides, within the 10 residues that precede CII, are highly conserved among CCP
43
domains. All of the RCA proteins are transcribed from the same genetic locus on the
long arm of chromosome 1, which is indicative of evolutionary relatedness resulting from
gene duplication (229). However, distinct CCP modules do not always function in the
same way, and degrees of sequence identity between separate modules can range from
100% to undetectable (228). A region exists that is, generally, highly divergent among
CCP modules, and it projects from the elongated three-dimensional surface. This region
is known as the “hypervariable” region and likely confers differences in function among
CCP modules within various RCA proteins.
Binding sites composed of between two and four CCP modules allow RCA
proteins to bind to C3b, C4b or both. The result can be inhibition of the formation of new
C3- or C5-convertases, acceleration of the dissociation of existing convertases, or
cofactor activity for the proteolytic degradation of C3b, or C4b, by Factor I (FI) (219).
Indiscriminant deposition of C3b could potentially cause tissue damage if not for the
surface expression of membrane-bound RCA proteins on host cells, which include
membrane cofactor protein (MCP, CD46), decay accelerating factor (DAF, CD55), and
CR1 (CD35). Each of these membrane-bound RCA proteins captures C3b and C4b,
which causes C3-convertase decay and enhanced FI-mediated inactivation of C3b and
C4b. CR1 also acts as a CR (see below).
Factor H (FH) is the major soluble RCA protein that inhibits C3b-mediated
complement activity. Deficiencies in FH are clinically associated with atypical hemolytic
uremic syndrome (aHUS) and age-related macular degeneration (ARMD), which are
apparently complement-mediated diseases (226). FH has been described as the primary
sensory molecule of the alternative pathway that is responsible for self and non-self
44
discrimination because it recognizes specific host associated molecular patterns
(HAMPs) (226). Surfaces that effectively recruit FH are, thus, protected in a similar
manner to surfaces expressing membrane-bound RCA proteins. FH is a glycoprotein
composed predominantly 20 CCP domains (230). CCP1-CCP4 are necessary for fluidphase complement regulation because they bind to C3b (231). This causes electrostatic
repulsion of Bb resulting in dissociation of C3bBb and inhibition of the formation of new
C3bBb complexes (232). Upon binding to C3b, FH also provides a binding platform for
FI, which is a serine protease that cleaves C3b to form C3bi and C3dg. The surface
specificity of FH is conferred by CCP5-CCP20. FH selectively binds to polyanionic
carbohydrates such as glycosaminoglycans and sialic acid on host cell surfaces (231).
Factor H-like protein (FHL-1) is an alternatively spliced form of FH that contains CCP1CCP7 and an additional four hydrophobic residues at its C-terminus (233).
C4-binding protein (C4bp) is another soluble RCA protein that, at physiologically
relevant concentrations, specifically inhibits ativity of the C3-convertase formed via the
classical and lectin pathways (234). C4bp is an oligomer composed of 6-8 α-chains and 1
β-chain that has a spider-like structure by electron microscopy (235). Each α-chain
contains a recognition site for C4b within CCP1-CCP3, and each molecule of C4bp can
bind up to four, surface-attached, molecules of C4b (236, 237). In serum, the β-chain of
C4bp forms a complex with protein S, a molecule that acts as a bridge for binding to
polyanionic phospholipid membranes (234). Indicative of the importance of C4bp to
development and homeostasis, C4bp-deficiency is not associated with described disease
states and is likely embryonic lethal. Interestingly, expression is related to hormone
status, and serum levels of C4bp increase particularly during pregnancy and hormone45
replacement therapy (238). Like FH, C4bp inhibits C3-convertase by three mechanisms
including inhibition of novel C4b2a formation, increased C4b2a dissociation, and
cofactor activity for FI mediated cleavage of C4b.
Many endogenous targets that are directly recognized by C1q also interact with
FH and C4bp (239). The inhibitory capacities of negative regulators such as C4bp are
enhanced dramatically when they bind to the same molecular targets that activate
complement (240). Other endogenous ligands that bind both directly to C1q and to either
FH or C4bp include C-reactive protein, serum amyloid protein, amyloid Aβ peptide,
prions, free DNA, and late apoptotic and necrotic cells (241-246). The direct binding of
C1q can also be inhibitory if binding occurs outside of the C1q globular head domain so
that associated C1-esterases are not activated. For example, components of the ECM,
such as fibronectin and laminin, bind C1q in this manner (247, 248). Importantly,
immunoglobulins G and M are the only known endogenous ligands for the globular heads
of C1q that do not bind to either FH or C4bp (239).
Apoptotic cells are primarily generated in the contexts of tissue remodeling and
regeneration. It stands to reason that, under physiological conditions, overt complement
activation and an ensuing inflammatory reaction during this process would be
detrimental. In fact, apoptosis is characterized by the lack of an induced inflammatory
immune response, suggesting a lack of anaphylatoxin release. However, the removal of
apoptotic cellular debris by phagocytes is clearly beneficial and necessary. Thus,
complement-mediated opsonization, with concomitant regulation of the complement
cascade, can occur via the direct binding of C1q and the simultaneous binding of FH and
C4bp. Binding targets for FH on apoptotic cells are, currently, unknown and may require
46
C-reactive protein as a bridging molecule (249). However, it is clear that
phosphatidylserine represents a high-affinity target for both C1q and C4bp (198, 250,
251). On viable cells, phosphatidylserine is a component of only the inner leaflet of the
plasma membrane, but it is transferred to the outer leaflet during the processes of
apoptosis and necrosis. Concurrent changes to the cell membrane can include decreased
amounts of RCA-family membrane proteins, indicative of an increased requirement for
the recruitment of soluble negative regulators of complement to control inflammation
(252).
Complement Receptors
Differential CR expression on distinct cell types enables great variability in
immune responses to complement activation, which can be pro- or anti-inflammatory. As
noted above, deposition complement components on microbe and/or particle surfaces
results in opsonization. CRs mediate adhesion and internalization of C1q-, C4b-, C3b- or
C3bi-coated particles by a process called opsonophagocytosis (108), but their functional
roles extend beyond the mechanical process of opsonophagocytosis due to the effects of
downstream signaling events on the regulation of adjacent signaling pathways and on
gene transcription. Non-opsonophagocytic CRs also exist including the C3a receptor
(C3aR) and the C5a receptor (C5aR), which recongnize the anaphylactic byproducts of
complement activation and are particularly potent mediators of endotoxin-related septic
shock (224).
CR1 (CD35) is a single chain transmembrane receptor with a short cytoplasmic
tail and an extracellular lectin-like recognition domain that binds C3b and, to a lesser
47
extent, C3bi (253, 254). CR1 recognizes additional ligands including C1q, C4b and
MBL (254, 255). Because it is highly expressed on erythrocytes, a major function of
CR1 is the binding of antibody-C3b complexes for shuttling to the liver and spleen where
they are removed by FcγR-mediated phagocytosis. As described above, CR1 also
contains up to 30 CCP modules and acts as a cofactor for FI leading to degradation of
bound C3b to C3bi and C3dg (256). C3dg is recognized by CR2 that is expressed as a
component of the B-cell receptor complex. The β2 integrins shown to function as
complement receptors include CR3 (αMβ2 integrin, Mac1, or CD11b/CD18) and CR4
(αxβ2 integrin, p150/95, or CD11c/CD18). Both are heterodimers that share a common
β2 chain (CD18) and differ with respect to their α chain (αM/CD11b and αx/CD11c,
respectively). A fourth opsonophagocytic CR, CRIg, shares homology with the
immunoglobulin receptor family and has only recently been shown to mediate
phagocytosis upon binding to either C3b or C3bi (37).
Both CR3 and CR4 have been shown to mediate optimal phagocytosis of F.
tularensis. Each CR recognizes C3bi via the I-domain on subunits CD11b and CD11c
(39, 257). Active and inactive conformations of the I-domain exist that depend upon
inside-out signaling mediated by the activity of alternate receptors (131, 258). Exposure
to inflammatory cytokines such as TNFα; microbial ligands, such as LPS; and receptormediated cell adhesion are examples of triggering events leading to inside-out activation
of the I-domain (259, 260). In addition to the I-domain, both the lectin-domain of CR3
(that mediates nonopsonic phagocytosis) and the CD18 subunit have the capacity to bind
alternate microbial ligands (145, 261, 262). The binding sites on CRs for F. tularensis
recognition have not been delineated. It should be clear that CR recognition of opsonin48
bound microbes can be a complex event depending on the nature of the C3 fragment, the
activation state of CRs, and potential, additional, microbial surface factors that interact
with receptors. Interaction with other receptors adds to this complexity. For example,
co-recognition of a particle by CR3 and FcγRs can produce cooperative effects (263).
Particles coated sub-optimally with IgG, for example, are efficiently internalized only
when also coated with complement. CR3 binding can enhance proinflammatory
signaling events mediated by other receptors (such as FcγR), but does not initiate
proinflammatory signaling when activated in isolation (264-266). Oleic acid and the
S100 protein MRP-14 (myeloid related protein that is 14 kDa) are examples of DAMPs
that are released from injured cells and that also have the capacity to induce activation of
CR3 (267, 268). A role for CR3 in anti-inflammatory responses has been described as
well. Marth et al. showed that CR3 engagement on human monocytes with anti-CR3
antibodies, purified C3bi, or Histoplasma capsulatum caused inhibition of IFNγ and of
IL-12 production (269). They also used anti-CR3 antibodies to successfully limit IL-12
production in a murine model of IL-12-dependent septic shock. Suppression of an
immune response to many endogenous C3bi opsonized particles, such as apoptotic cells,
would be beneficial. Intracellular pathogens, however, may take advantage of the
suppressive potential of CR3.
Less is known about receptors for C1q, but examining their roles in mediating
particle uptake and in signal regulation is an active area of current research. In an early
study by Guan et al., antibodies against C1qRp (CD93) were shown to inhibit C1qmediated phagocytosis by leukocytes (270). However, subsequent studies using
macrophages isolated from CD93-/- mice showed that C1q-induced phagocytosis
49
remained intact, suggesting the involvement of additional receptors. CR1 recognizes the
collagenous tail of C1q, but whether binding contributes to enhanced or attenuated
phagocytosis in the context of additional receptor-ligand interactions is undetermined
(270, 271). A receptor known as gC1qR recognizes the globular head of C1q and can also
have an inhibitory effect on LPS-mediated proinflammatory signaling through PI3K and
akt activation (272, 273). However, the role of gC1qR-mediated signaling has been
contested because it lacks a predictable transmembrane domain and has been shown to
localize to the cytosolic compartment (274, 275). Macrophage uptake of apoptotic cells
can occur through cC1qR (also known as calreticulin), which recognizes the collagen-like
tail of C1q (276). CD91 (LDL-related receptor protein, LRP) forms a complex with
cC1qr and mediates uptake through a process involving macropinocytosis. Particle
uptake via the cC1qR/CD91 complex was also recently shown to have a negative
influence on proinflammatory signaling pathways (277). In this study, Fraser et al.
determined the effects cC1qR/CD91 activation by C1q on the proinflammatory response
to LPS treatment in monocytes. Treatment with C1q enhanced nuclear translocation of
inhibitory NFκB complexes and of cAMP response element binding (CREB) protein, two
transcriptional regulators that would have the dual effect of inhibiting LPS-mediated
upregulation of IL-1β, TNFα, and IL-6 as well as enhancing IL-10 upregulation. It
remains possible that the pro-phagocytic and immunomodulatory effects of C1q that were
previously ascribed to C1qRp and gC1qR were mediated, at least in part, by
cC1qR/CD91.
50
Complement activity in the airway
Within the context of a respiratory inflammatory response, and the concomitant
increased vascular permeability, exudative fluid that contains native components of the
complement system can enter alveolar septae and the airspace (31). However,
complement components are also produced and secreted locally as constituents of ASF,
even in the absence of inflammatory stimuli. Thus, in collaboration with the other
soluble mediators of innate immunity in the airway, described above; complement is a
major contributor to pulmonary host defense. As evidence, human and animal studies
show that complement deficiency significantly increases susceptibility to a variety of
pulmonary infections (278-280). AMs and/or Type II alveolar epithelial cells produce
and secrete complement proteins locally (281, 282). Locally produced components are
known to include self/non-self discriminatory lectins, C1q, C2, C4, and C3; and the
functional activities of the classical and lectin pathways in ASF have been shown most
conclusively (283-286). However, activity of the alternative pathway is likely negligible
due to low FB levels in ASF compared with serum (285). Furthermore, naive C1q-/- and
C4-/- mice are susceptible to infections with group B Streptococcus and S. pneumonia
when compared with wildtype litter mates, a phenotype only partially mediated by
natural IgM (287, 288). Also indicative of a major role for these pathways in clinical
disease is a recent review by Sjoholm et al., which reports that 57% of C2-deficient
patients were identified as having positive histories for invasive pneumonic bacterial
infections (289).
51
Bacterial complement evasive strategies
Gram-negative bacterial pathogens, by definition, must counteract the
microbicidal effects of the complement system upon contact. Intracellular survival may
impart escape from complement component exposure. Under most circumstances,
however, microbial population growth and dissemination necessitate exposure to the
extracellular environment during the cycle of cellular re-infection. Bacteria employ
evasive mechanisms to counteract each step in the complement cascade. Strategies
include the application of cloaking mechanisms to evade activation pathway sensory
molecules, protease secretion, the disruption of convertase assembly via the recruitment
of host RCA proteins, the inhibition of functional MAC formation, among others.
Considering the potential microbicidal effect of complement, it is not surprising that most
successful pathogens possess redundant mechanisms targeting multiple aspects of the
cascade, thereby, allowing them to avoid complement-mediated killing (290-295).
The production of capsule and O-antigen by some Gram-negative bacteria can
limit the access of sensory components of complement to activating surface molecules.
For example, both components are produced by clinical isolates of Klebsiella and results
in limited activation of the classical pathway by prohibiting direct C1q-binding to surface
proteins (296). Transcriptional modifications (e.g. phase variation, mutation, or lateral
gene transfer) resulting in altered surface characteristics enable bacteria to evade
recognition by antibody or by complement sensory components (297-300). Another
“cloaking” mechanism involves binding of complement components in a way that limits
their function. Aeromonas salmonicida, for example, expresses a C1q-binding outer
membrane protein that inhibits activation of complement and confers increased resistance
52
to complement-mediated killing (301). In this case, the steriochemical nature of the
interaction with the globular heads of C1q, likely, does not result in C1-associated
esterase activation. It is also possible that C1q is bound via the collagenous tail, which
would also inhibit C1-esterase.
Various complement components can be inactivated directly by bacterial
proteases. Periodontitis-causing Porphyromonas gingivalis secretes a 97 kDa protease
that degrades C3 and IgG (302). Elastase and alkaline protease secreted from
Pseudomonas aeroginosa target C1q for degradation by cleaving its A- and C- chains
(303). These proteases also target C3 and completely digest the N-terminal portion of the
α-chain. Secreted proteases can also affect the immunomodulatory effects of
complement. For example, Serratia marcescens secretes a 56 kDa protease that
inactivates C5a (304). In vivo, mice intraperitoneally infected with low proteaseproducing strains demonstrate significantly enhanced neutrophil migration into the
peritoneal cavity. CRs may also be targeted by microbial products as is exemplified by
proteases secreted from P. ginigvalis that cleave C5aRs on neutrophils and limit
chemotaxis and localized inflammation (305).
A common mechanism of complement evasion employed by Gram-negative
bacteria is the acquisition of host-derived, soluble, negative regulators, such as C1inh,
FH, and C4bp. Both C1inh and C1-esterases have positively charged residues proximal
to their respective active sites. A sandwich model has been proposed to explain the
enhancement of C1-esterase-mediated inhibition in the presence of linear, polyanionic
molecules, such as heparin, dextran sulfate, and, potentially, LPS (306). C1inh has been
used as a treatment in models of septic shock because it binds directly to purified LPS,
53
although it is unknown whether this therapeutic effect is related to inhibition of C1esterase activity (307, 308). Certain strains of Bordetella pertusis have been shown to
bind to C1inh directly; however, the mechanism of binding and the direct effect on
complement sensitivity are unknown (309). Lathem et al. published a more
comprehensive characterization of an unexpected strategy employed by E. coli strain
O157:H7 to exploit C1inh (310, 311). StcE is a plasmid-encoded zinc metalloprotease
that cleaves,but does not inactivate, C1inh. Instead, StcE acts as a bridge by binding to
both the surface of the bacterium and to cleaved C1inh, which localizes its inhibitory
capacity on C1. Interestingly, antibody-sensitized erythrocytes treated with StcE are also
resistant to classical pathway-mediated lysis. This suggests that the surface binding of
the StcE-C1inh complex is not species-specific.
FH protects host cells from the lytic effects of complement by binding to
polyanionic surface molecules, mostly sialic acid-conjugated proteins, and promoting
C3bi formation. Bacteria have evolved mechanisms to mimic host cell components in
order to recruit FH and evade amplification of complement component deposition. The
expression of sialylated lipooligosaccharide and Por1A on N. gonorrheae exemplifies the
most direct form of host mimicry (312, 313). The acquistion of FH and FHL-1, but not
C4bp, is essential to Borrelia burgdorferi and B. afzelii serum-resistance (291). Up to
five Borrelia complement regulator-acquiring surface proteins (CRASPs) exist that bind
to distinct domains on FH and/or FHL-1. In addition to CRASPs, FH-binding proteins
expressed by Borrelia spp. include: outer surface protein E (OspE), OspE/F-related
protein A (ErpA), ErpC, ErpP, p21, and an unknown 35kDa protein [reviewed in (291)].
54
YadA surface expression on Yersinia enterolitica also mediates FH recruitment and
serum-resistance (314).
Bacterial proteins can capture C4bp to inhibit both the classical and lectin
pathways by enhancing C4b2a decay and inactivation of C4b. Surface proteins recruit
C4bp by binding to α-chain CCP domains using both ionic and hydrophobic interactions
in a non-strain-specific fashion. Porins 1A and 1B from N. gonorrheae form
hydrophobic and ionic bonds, respectively, with the CCP1 domain (315). Impressively,
considering the capacity of N. gonorrheae to also bind FH, inhibition of C4bp
interactions with these bacteria using Fab fragments against the CCP1 domain resulted in
complete complement-mediated lysis of normally resistant Por1A- or Por1B-positive
strains. Distinct loops of Por1A are responsible for binding FH and C4bp (316). Type
IV pili-associated pilC from N. gonorrheae can also bind C4bp via CCP1-CCP2 (317).
E. coli expresses OmpA, an outer membrane protein that interacts hydrophobically with
the CCP3 domain to confer resistance to lysis in serum and inhibition of the deposition of
downstream complement components (318, 319). Moraxella catarrhalis expresses
ubiquitous surface proteins A1 and A2 that bind ionically to the CCP2, CCP5 and CCP7
domains (320). Finally, filamentous hemaglutinin is a surface protein expressed by
clinical isolates of B. pertusis that can bind C4bp; however, protection from lysis does
not occur as a direct result (321).
Many types of LPS represent efficient activating surfaces for the alternative
complement pathway. A major feature of effective activating surfaces, such as LPS, is an
increased affinity of bound C3b for FB compared to FH (322, 323). Differences in both
composition and length of O-antigen result in diverse quantitative capacities to consume
55
complement components (324, 325). Although smooth bacterial strains that produce Oantigen are more commonly serum-resistant and rough strains that lack O-antigen are
more commonly susceptible, O-antigen does not invariably confer resistance to
complement because several smooth strains are susceptible to complement-mediated lysis
(326, 327). This suggests that O-antigen-mediated complement inhibition cannot simply
be attributed to steric hindrance and limited access to the membrane. As noted, the
composition of LPS O-antigen can result in FH recruitment and in the negative regulation
of complement at the level of C3-convertase formation. A more common mechanism of
inhibition involves the stability of MAC assembly. Serum-resistant and serum-sensitive
strains of E. coli and Salmonellae commonly fix equivalent amounts of C3 (328). In one
study using S. minnesota strains, rough serum-sensitive strains fixed significantly less C3
when compared to isogenic smooth strains (329). Furthermore, rough serum-sensitive S.
minnesota merely consumed 25% of the hemolytic activities of C5, C7, and C9, enough
to confer lysis. Unexpectedly, the isogenic serum-resistant strain completely consumed
the hemolytic activities of terminal pathway components (329). Subsequent binding
studies showed that C5b-9 complexes initially bound to resistant strains of E. coli and
Salmonellae, but were subsequently shed from their surfaces (330). Bacterial acceptor
molecules were not concomitantly shed, and stable attachment of C5b-7 was shown; thus,
shedding of MAC likely occurs as a result of its formation distal to the lipid bilayer and
ineffective membrane insertion of C5b-8. Interestingly, serum-sensitive clinical isolates
of Pseudomonas aeruginosa that do not produce O-antigen also consumed less hemolytic
complement activity compared with smooth isogenic strains, but fixed more C3b,
suggesting a similar mechanism of lytic evasion for smooth strains (331).
56
MAC release is also implicated as a major mechanism of complement evasion
employed by encapsulated bacteria. In contrast to N. gonorrheae, Group B N.
meningitidis express a sialylated bacterial capsule that is essential for virulence (332).
Although it has been shown that FH does bind to the capsule, this does not result in
significant differences in C3b deposition compared with nonencapsulated strains, and
only a minor fraction of C3b is converted to C3bi (316). As an alternative primary
mechanism of resistance, Ram et al. showed that the binding of MAC is much lower on
encapsulated strains when compared with serum-susceptible, nonencapsulated, strains
(316). Given the production of C5a did not significantly differ between these strains, the
formation of C5-convertase, likely, occurred on the encapsulated strains. However, an
inability of C5b-7 to form a stable hydrophobic interaction with capsule leads to its
release.
Bacterial surface proteins are also major inhibitory factors that limit functional
MAC deposition. TraT is a plasmid-encoded outer membrane protein expressed by E.
coli K12 that inhibits C7 binding to C5b6 (333). An example of host mimicry is the
expression of an 80 kDa surface protein by B. burgdoferi, which shares antigenic
characteristics with CD59 (334). This CD59-like molecule binds to both C8 and C9 and
inhibits the formation and insertion of MAC. E. coli and Helicobacter pylori can also
recruit host-derived CD59 under circumstances that cause it to become detached from
cells and available within the fluid phase (335, 336).
57
1.3 Specific Aims
Little is currently known about interactions between F. tularensis and components
of the complement system. Considering the multifactorial influences of complement on
overall inflammatory responses and the dynamic stages of the host response to infection
with F. tularensis, characterizing important aspects of F. tularensis-complement
interactions will undoubtedly increase our understanding of the unique pathogenesis
associated with tularemia. Such interactions have already been shown to enhance the
phagocytosis of bacilli by macrophages and DCs in vitro but may also prove critical for
modulating the inflammatory responses of infected cells and for maintaining a bacteremic
phase during the latter septic stage of tularemia. We hypothesized that complement
activation by F. tularensis leads to opsonization but not to lysis of virulent strains and,
further, that negative regulation of complement would be mediated by major surface
glycans, such as LPS and capsule.
To examine this hypothesis, we pursued the following specific aims:
1. Using several subspecies and strains of F. tularensis that differ in their surface
glycan expression, we analyzed the nature of C3 deposition in terms of
specific fragmentation patterns, susceptibility to complement-mediated lysis,
and the deposition of downstream components of MAC.
58
2. We characterized the relative contributions of complement activation
pathways for complement-mediated lysis of susceptible strains and for
opsonization of resistant strains.
3. We evaluated the role of surface glycans in mediating the negative regulation
of complement activity by exploring those mechanisms commonly used by
other bacterial species.
59
Chapter 2
Evasion of complement-mediated lysis and complement C3 deposition are regulated
by Francisella tularensis LPS O-antigen
Introduction
Complement is a highly regulated and multifunctional system that is the major
extracellular arm of innate immunity. Its activation results in three major potential
outcomes: lysis upon assembly and insertion of the terminal membrane attack complex
(MAC), opsonization, and the release of anaphylatoxins that enhance local inflammation
(219, 256, 337). Each activation pathway of complement leads to assembly of C3convertase, an enzymatically active complex formed by the cleavage fragments of
upstream components. Both C4b2a (classical and lectin pathway C3-convertase) and
C3bBb (alternative pathway C3-convertase) cleave C3a from native C3 to form C3b. A
cryptic reactive thioester group within native C3 becomes available on C3b to form
covalent linkages with either hydroxyl or amine moieties on microbial acceptor
molecules (338). Bound C3b interacts with parent C3-convertases to form C5convertases in order to initiate the terminal lytic pathway leading to insertion of the
MAC. However, negative regulation of complement activity can result in further
cleavage of C3b to smaller, inactive fragments (C3bi, C3dg, or C3d) that do not initiate
60
the terminal lytic pathway (219). The C3b and C3bi fragments mediate
opsonophagocytosis of bound organisms via their recognition by complement receptors
(CR) on professional phagocytes (339). Macrophage CR-mediated entry has long been
regarded as a mechanism by which microbes are killed less effectively following
phagocytosis (131). Thus, it is noteworthy that this pathway is utilized by several
intracellular pathogens.
Francisella tularensis is a Gram-negative facultative intracellular coccobacillus.
Two subspecies of F. tularensis exist that cause human disease. Type A, F. tularensis
subsp. tularensis, is endemic to North America. Type B, F. tularensis subsp. holarctica,
is endemic to Eurasia, Japan and North America and is less virulent than Type A (5, 340).
F. tularensis subsp. novicida is avirulent in humans, except in rare cases (341), but causes
a fulminant disease in mice. Like subsp. novicida, the live vaccine strain (LVS) derived
from holarctica is attenuated in humans but not in relevant animal models. LVS is
undergoing FDA scrutiny as a vaccine since both the induced mechanism of
immunological protection and the genetic basis of attenuation are undefined. Also,
spontaneous changes occur with respect to colony morphology resulting in phase variants
(called grey variants) with reduced virulence and immunogenicity, which may affect
vaccine utility. The genetic basis for phase variation, which occurs for Type A isolates as
well, is unknown (342). In fact, based on changes in LPS structure, multiple grey variant
phenotypes have been described including variants that completely lack O-antigen and
variants that express antigenically distinct O-antigen compared with LVS (176, 177).
Tularemia results from exposure to infected animal tissue, bites from arthropod
vectors, or the direct ingestion or inhalation of F. tularensis (12). Pneumonic tularemia is
61
the most serious form as untreated cases result in 30-60% mortality compared to 5-6%
mortality associated with cutaneous disease (343, 344). Pneumonic tularemia develops
upon inhalation of less than 10 colony forming units (cfu) or via hematogenous spread of
bacilli from peripheral sites to the lung (12, 29). Because of its highly infectious and
lethal nature, F. tularensis is designated as a Category A select agent by the CDC and is a
potential weapon of bioterrorism. F. tularensis predominantly infects and replicates
within macrophages and spreads systemically via the reticuloendothelial system (68).
Considering the extremely low infectious dose required to cause disease, it is
paradoxical that in vitro studies examining macrophage infection by F. tularensis require
high experimental multiplicities of infection (MOI) of at least 100:1 without
opsonization. However, phagocytosis of Type A F. tularensis, LVS, and subsp. novicida
by human monocyte-derived macrophages increases dramatically in the presence of
serum in a C3- and CR-dependent manner (109, 110). Complement-mediated
opsonization has also been shown for the phagocytosis of LVS by human monocytederived dendritic cells (60). Given the importance of complement-mediated opsonization
for uptake by phagocytes, and a previous report that LVS does not fix C3 (183), we
studied the nature of complement interactions with F. tularensis in non-immune human
serum. Here, we examine several F. tularensis subspecies and variant strains to
determine (i) whether complement activation leads to bacterial lysis, (ii) the nature of C3
deposition and subsequent fragmentation, (iii) the affect of surface glycans such as
capsule and LPS on the nature of complement activation, and (iv) whether complement
mediated opsonization results in increased F. tularensis uptake by human alveolar
macrophages.
62
Materials and Methods
Bacterial strains. F. tularensis subsp. tularensis strain Schu S4, a Centers for
Disease Control and Prevention clinical isolate, was provided by Rick Lyons (University
of New Mexico, Albuquerque, NM). F. tularensis subsp. holarctica LVS (ATCC 29684)
was provided by Karen Elkins (Center for Biologics Research and Evaluation, US FDA,
Bethesda, MD). F. tularensis subsp. novicida (U112; Fn), LVSG, and LVSR were
provided by Fran Nano (University of Victoria, Victoria, BC, Canada). LVSG is a
spontaneous grey phase variant that rarely reverts to LVS when grown on chocolate II
agar (176). LVSR was originally described as a capsule-negative strain (183) and was
selected for its rough colony morphology after the mutagenesis of LVS by treatment with
acridine orange. The LPS O-antigen mutants, LVSΔwbtA and LVSΔwbtM, provided by
Dara Frank (Medical College of Wisconsin, Milwaukee, WI), were created by modified
Himar1 transposon (HimarFT)-mediated mutagenesis of LVS (345). The complemented
mutant strain (LVSΔwbtM:pFTNAT-wbtM) and LVSΔwbtM containing the empty
pFTNAT complementation plasmid (LVSΔwbtM:pFTNAT) were also provided by Dr.
Frank. Experiments using Schu S4 were carried out within biosafety level 3 (BSL3)
select agent-certified laboratories with adherence to federal and institutional select agent
regulations. Bacteria were grown overnight (approximately 18 hours) on chocolate II
(chocII) agar (Becton Dickinson, Franklin Lakes, NJ) at 37°C. For experiments using
LVSΔwbtM and related strains, bacteria were grown overnight on Mueller-Hinton (MH)
agar containing 2.1% MH broth (Becton Dickinson), 0.5% NaCl, 1.6% agar, 1% protease
peptone (Becton Dickinson), 0.1% glucose, 2% Isovitalex (Becton Dickinson), 0.025%
63
ferric pyrophosphate, and 2.5% fetal bovine serum (Invitrogen, Carlsbad, CA). Strains
containing the pFTNAT plasmid were grown overnight on MH agar containing 50 µg/ml
nourseothricin (Sigma-Aldrich, St. Louis, MO). For electron microscopy, bacteria were
grown overnight in modified tryptic soy broth (Difco Laboratories, Detroit, MI)
containing 135 µg/ml ferric pyrophosphate and 0.1% cysteine hydrochloride. DH5α, a
laboratory strain of E. coli, was grown overnight on Luria-Bertani (LB) agar at 37°C
prior to use.
Human sera, complement components, and reagents. Serum was isolated
from healthy adult volunteers with no known exposure to F. tularensis according to a
protocol approved by the Ohio State University Medical College Internal Review Board.
The sera were processed to maintain optimal complement activity (346). Briefly, isolated
non-heparinized whole blood was kept at room temperature for 1 hour to allow for clot
formation and then at 4°C to allow for clot contraction. The clot was removed by
centrifugation at 500xg for 15 minutes at 4°C. The serum fraction was collected, filter
sterilized, aliquoted, and stored at -80°C. Heat inactivation (HI) was performed at 56°C
for 30 minutes. C5-depleted (C5d) and C8-depleted (C8d) sera were purchased from
Complement Technology, Inc (San Antonio, TX) and stored at -80°C. On the day of use,
fresh sera were thawed at room temperature, then immediately chilled on ice until
needed. A concentrated serine protease inhibitor cocktail (containing AEBSF, aprotinin,
elastatinal, and GGACK) was purchased from Calbiochem (Madison, WI). 50%
hydroxylamine was purchased from Alfa Aesar (Ward Hill, MA). Other chemicals were
purchased from Sigma-Aldrich.
64
Bronchoalveolar lavage (BAL). Healthy human volunteers with no known
exposure to F. tularensis and no smoking history underwent BAL with approximately
100 ml of saline according to a previously described procedure (347) that has been
approved by the Ohio State University Medical College Internal Review Board. Alveolar
macrophages (AMs) were separated from BAL fluid by centrifugation. Cells were
washed in RPMI 1640 with L-glutamine (Invitrogen), resuspended at a final
concentration of 2x105 cells/ml in RPMI containing 10% autologous serum, and placed in
monolayer culture in 24-well tissue culture plates for 1 hour at 37°C in 5% CO2. After
the removal of cellular material, EDTA was added to the lavage fluid to a final
concentration of 2 mM, and the mixture was filter-sterilized and kept at 4°C to prevent
complement activity. As previously described (283), BAL fluid was concentrated
(cBAL) 20- to 30-fold using Amicon Centriplus 10 kDa cut-off concentrator tubes
(Millipore, Bedford, MA). cBAL fluid was stored at -80°C until used. Prior to
experimentation, cBAL fluids were dialyzed at 4°C, using a membrane with a 10 kDa
cut-off (Pierce, Rockford, IL), at a 4,000/1 (vol/vol) ratio for 2 hours against Dulbeco
phosphate-buffered saline (PBS) containing calcium and magnesium ions.
Bactericidal assays. Complement-mediated killing was carried out using fresh
non-immune or HI (negative control) sera or cBAL with or without 10mM EDTA. For
experiments using serum, bacteria were suspended in gelatin veronal buffer (GVB++;
0.1% gelatin, 5.5 mM barbital, 142 mM NaCl, 0.5 mM MgCl2, 0.15 mM CaCl2; pH 7.3)
at equalized concentrations by measuring the optical density at 600 nm. For each assay,
65
2-3x106 bacteria were incubated with various serum concentrations or with 90% cBAL
for 1 hour in microcentrifuge tubes (final volume of 200µl in reaction buffer) at 37°C
with slow agitation. Reactions were stopped by placing tubes on ice for 5 minutes. For
some experiments, a serine protease inhibitor cocktail diluted in ice-cold PBS was used to
stop the reaction. 10-fold serial dilutions were plated to determine surviving cfu.
C3 deposition assays and Western blotting. Fresh donor, HI donor or C5d sera
were used to evaluate complement component C3 deposition on F. tularensis strains.
After pre-blocking microcentrifuge reaction tubes in PBS containing 0.1% human serum
albumin (HSA) (ZLB Plasma, Boca Raton, FL) for 30 minutes at 37°C, 5x108
bacteria/reaction were incubated in 10% sera for various times at 37°C. Adding an icecold serine protease inhibitor cocktail in blocking buffer terminated deposition and
fragmentation. Unbound C3 was removed from bacterial pellets by washing in blocking
buffer (x3) and PBS (x1) at room temperature (12,000xg for 3 minutes). To ensure equal
lane loading, aliquots were taken from the final PBS wash and plated to count cfu. After
final re-suspension in Laemmli’s sample buffer, samples were boiled for 2 minutes and
separated by minigel 7.5% SDS-PAGE (BioRad, Hercules, CA), followed by protein
transfer to PVDF membranes (Millipore, Billerica, MA). Membranes were blocked
overnight at 4°C in advanced ECL blocking buffer at 2% (v/w), which was also used for
antibody dilution (Amersham, Piscataway, NJ). Goat antiserum to human C3 (Quidel,
San Diego, CA) (diluted to 1:20,000) served as the primary antibody for 1 hour
incubations at room temperature. HRP-conjugated rabbit anti-goat IgG (H+L) antibody
66
(Biorad) (diluted to 1:20,000) was used as the secondary antibody for 1 hour incubations
at room temperature. Advanced ECL reagent was used for detection (Amersham).
ELISA to detect complement component deposition on F. tularensis strains.
C8d serum was used to evaluate complement component C3, C5, and C7 deposition on F.
tularensis strains. After pre-blocking microcentrifuge reaction tubes as described for
Western blotting experiments above, 3x108 bacteria/reaction were incubated in 10% fresh
serum or serum containing 10mM EDTA for 5 or 30 minutes at 37°C. Reactions were
stopped and samples were washed as described above for Western blotting experiments.
To ensure equal well loading, aliquots were taken from the final PBS wash and plated to
count cfu. 3x107 bacteria in suspension were added to medium-binding polystyrene wells
in triplicate (Costar, Corning, NY) and left to dry overnight. Wells were blocked
overnight at 4°C with 3% ovalbumin. After extensive washing with PBS, primary
antibodies were added for 1 hour incubations at room temperature, which included goat
antisera to human C3 (Quidel), human C5 (CompTech) and human C7 (CompTech)
(each diluted to 1:10,000 in blocking buffer). HRP-conjugated rabbit anti-goat IgG
(H+L) antibody (Biorad) (diluted to 1:2,000) was used as the secondary antibody for 1
hour incubations at room temperature. Substrate was added for 10 min at room
temperature (BioRad), and the reaction was stopped with 2% oxalic acid. Absorbance at
415 nm was measured on a 96-well plate reader (Molecular Devices, Sunnyvale, CA).
Values obtained from samples containing EDTA (≤0.2 in each case) were subtracted out
in each case.
67
Determination of the nature of C3 bound to F. tularensis. C3-bound bacterial
pellets were obtained as described above for Western blotting. To examine the nature of
the C3 fragments bound to each strain, hydroxylamine (NH2OH) was used to cleave
thioester bonds formed between C3 and bacterial acceptor molecules as described with
modifications (348). Briefly, after completing C3 deposition reactions and removal of
unbound C3, samples were solubilized by boiling in 1% SDS for 5 minutes. Control
samples were prepared immediately for Western blotting and paired samples were first
incubated in 2M NH2OH in 20mM Tris-buffered H2O, pH 10.5, for 1 hour at 37°C.
Western blotting was performed as described above except that 16.5 cm gels were used to
create better separation between fragments and to allow for increased sample loading.
Also, 5% powdered milk was used to block and suspend antibodies (1° at 1:2000 and 2°
at 1:4000). Antibody incubations were performed for 1 hour at room temperature. ECL
reagents were used for detection (Amersham). Band densitometry was analyzed using
Image J software available through the National Institutes of Health website
(http://rsb.info.nih.gov/ij/).
Transmission electron microscopy. After overnight culture at 37°C in tryptic
soy broth, LVS and LVSG were washed and fixed with 2.5% warmed glutaraldehyde for
5 min followed by a combination of 2.5% glutaraldehyde and 1% osmium tetroxide in 0.1
M sodium cacodylate, pH 7.3, for 15 min at 4°C (149). Bacilli were then stained with
0.25% uranyl acetate in 0.1 M sodium acetate buffer at pH 6.3 for 45 min. Pelleted
bacteria were dehydrated through a graded series of ethanol, rinsed in
hydroxypropylmethylacrylate, and infiltrated with Polybed 812. Samples were delivered
68
to The Ohio State University Campus Microscopy and Imaging Facility. Thin sections
cut with a Leica EM UC6 ultramicrotome were collected onto Formvar-coated copper
grids, stained with uranyl acetate and lead citrate, and viewed using a transmission
electron microscope (Philips CM12) at 60 kV.
LPS expression analyses by silver stain and Western blot. Bacteria, cultured
overnight on chocII agar plates, were suspended in PBS at a concentration of 3x1010
cfu/ml, as determined by optical density in order to equalize the amount of bacteria, and
subsequently washed twice in PBS by centrifugation at 12,000xg for 4 minutes with resuspension. The final pellet was re-suspended in 200µl Laemmli loading buffer, boiled
for 10 minutes, and then incubated with Proteinase K (Invitrogen) at a final concentration
of 10 mg/ml for 2 hours at 65°C. Samples were boiled again for 10 minutes and stored at
-20°C until used. LPS was separated by 12% SDS-PAGE and silver stained as described
(349). Briefly, gels were fixed overnight in solution containing 40% EtOH and 5% acetic
acid. Gels were then incubated in 0.7% periodic acid in fixing solution for 7 minutes and
subsequently washed with multiple exchanges of water. The staining solution (0.013%
concentrated ammonium hydroxide, 0.02N NaOH, and 0.67% silver nitrate (w/v) in
water) was applied with vigorous agitation for 10 minutes, then gels were washed 3x
(each 10 minutes) in water. Gels were developed in solution containing 0.275%
monohydrous citric acid (v/w) and 0.0025% formaldehyde. Upon completion,
development was stopped using 5% acetic acid.
For Western blotting, strains were grown overnight on chocII agar plates.
Bacteria were suspended at a concentration of 1x1010 cfu/ml (as determined by optical
69
density in order to equalize the amount of bacteria), pelleted, boiled in Laemmli’s sample
buffer for 10 minutes, then incubated with 10 mg/mL Proteinase K at 65°C for 2 hrs. The
samples were separated by 12% SDS-PAGE and transferred to a nitrocellulose
membrane. Membranes were blocked overnight with 5% dehydrated milk and
immunoblotting was performed using anti-subsp. tularensis FB-11 antibody (Abcam,
Cambridge, MA) (diluted to 1:1000) or anti-subsp. novicida #5 monoclonal antibody
(ImmunoPrecise, Victoria, BC) (diluted to 1:1000) for 4 hours at room temperature, using
goat anti-mouse IgG (Biorad) (diluted to 1:4000) for 2 hours at room temperature as the
secondary antibody. Blots were developed with BCIP/NBT (Sigma-Aldrich) as the
substrate.
Microscopy assay of F. tularensis uptake by AMs. AM monolayers were
formed on Chromerge-cleaned glass coverslips in 10% autologous serum in RPMI
medium at 37°C with 5% CO2 for 1 hour to allow for attachment. The cells were then
washed with warm RPMI and incubated with RHH medium (RPMI 1640 with Lglutamine, 10 mM HEPES, and 0.5% HSA) or RH medium (RPMI 1640, L-glutamine,
and 10 mM HEPES, 10% autologous fresh or HI serum). 50µl of appropriately diluted
bacterial stock was then added to each well. AMs were incubated on a rotating platform
for 30 min and then under stationary conditions for an additional 90 min, both at 37°C in
5% CO2. After incubation, the cells were washed extensively with warm media to
remove nonadherent bacteria and fixed in 2% paraformaldehyde. Some fixed AMs on
coverslips were permeabilized with 100% methanol for 5 minutes and washed. Bacteria
were counted by indirect immunofluorescence microscopy. AMs were incubated with a
70
monoclonal mouse anti-subsp. tularensis lipopolysaccharide primary antibody (Abcam)
(diluted 1:1,000 in blocking buffer composed of 5% HI human AB serum [Cambrex, East
Rutherford, NJ] and 1% bovine serum albumin [Sigma] in PBS) for 4 hours at room
temperature with gentle rotation. After extensive washing, AMs were incubated with
Alexa Fluor 488-conjugated goat anti-mouse IgG (Invitrogen) (diluted 1:1,000 in
blocking buffer) for 90 min at room temperature. Coverslips were mounted on glass
slides. The average number of bacteria per macrophage on each coverslip was determined
by counting a minimum of 200 cells per coverslip using a 100x oil immersion objective
with a wide-bandwidth 570-nm dichroic mirror on a BX51 Olympus fluorescence
microscope (Olympus, Melville, NY). Associated bacteria (attached and ingested) were
counted on AMs permeabilized with methanol. Extracellularly attached bacteria were
counted on non-permeabilized AMs. Ingestion was determined by subtracting the
number of attached bacteria from the number of associated bacteria (number ingested =
number associated – number attached). Triplicate coverslips were used for each test
group.
Statistics. To determine associations of significance between or within groups
where indicated, data were analyzed using unpaired (one-tailed) or one-sample Student’s
t tests or one-way ANOVA followed by Bonferroni’s post-tests (using Graphpad Prism 4
software). Cfu data were converted to log values to equalize variances. Differences
between groups were considered statistically significant for p values <0.05.
71
Results
Complement-mediated lysis of F. tularensis in human serum
Since complement-mediated opsonization is necessary for efficient phagocytosis
of F. tularensis by macrophages and dendritic cells, we compared the ability of five
strains to survive complement-mediated lysis in healthy non-immune human serum
(Table 1). Strains included Type A virulent F. tularensis subsp. tularensis (Schu S4), F.
tularensis subsp. novicida (Fn), F. tularensis subsp. holarctica live vaccine strain (LVS),
a grey phase variant of LVS (LVSG), and a putative capsule-negative strain derived from
LVS (LVSR). We also included DH5α, a laboratory E. coli strain, as a positive control
for complement activity in serum because of its marked sensitivity to complementmediated lysis. Bacteria were incubated with 5% or 50% (fresh or HI) donor serum for 1
hour, the reaction stopped and the bacteria subsequently plated to count surviving cfu
(Fig. 2.1). Because complement components C1, C2 and Factor B (FB) are irreversibly
denatured by mild heat, HI serum served as a negative control for complement-mediated
lysis. In each experiment, cfu obtained from incubations in GVB++ alone (reaction
buffer) or in HI serum were equivalent. Results are presented as the ratio of surviving
cfu in fresh versus HI serum. Data show that virulent strains known to cause disease in
either humans or in animal models (Schu S4, Fn, and LVS) are resistant to lysis in both
5% and 50% serum. By comparison, LVSG and LVSR were both susceptible to
complement-mediated lysis as survival decreased by up to 1,000-fold and 100,000-fold,
respectively.
72
Table 2.1. Summary of F. tularensis strains used
F.
Virulence
LPS
Corresponding
tularensis
a
(human)
antigenicity
Strain used
subspecies
tularensis +++
tularensis-type Schu S4
holarctica ++
tularensis-type LVSb
LVSGc
LVSRd
LVSΔwbtAe
LVSΔwbtM f
novicida
+/novicida-type Fn (U112)
a
Based on LPS structure and recognition by monoclonal
antibodies specific for either subsp. tularensis or subsp.
novicida
b
live vaccine strain derived from subsp. holarctica that is
avirulent in humans, but retains virulence in mice
c
LVS grey phase variant (176)
d
LVS rough mutant (176)
e
polar LVS O-antigen mutant (345)
f
nonpolar O-antigen mutant (22)
73
Figure 2.1. Susceptibility to complement-mediated lysis differs among F. tularensis
strains. 2x106 bacteria/reaction were incubated in either fresh non-immune serum
(separate donor for each experiment) or heat-inactivated (HI) serum (devoid of
complement activity) at concentrations of 5 or 50% for 1 hour at 37°C, then washed in
ice cold PBS. DH5α, a laboratory strain of E. coli that is highly susceptible to lysis by
complement, served as a positive control for serum complement activity. Data are
presented as percent survival, i.e., cfu obtained from reactions in fresh serum normalized
to cfu from reactions in HI serum. Means +/- SEM are given (N=3). *, p < 0.0001 for
one-sample Student’s t test with theoretical mean of 100. ND, not detected.
74
Fixation of complement components C3, C5 and C7
Complement component C3 is the central component of the complement cascade.
Since C3-fragment deposition is necessary for complement-mediated lysis and for
opsonization, we examined deposition on each of the five strains of F. tularensis
described above by Western blotting. Bacteria were incubated in 10% fresh or HI donor
serum for 30 minutes and then washed extensively to remove unbound C3 prior to sample
preparation. Native C3 is a heterodimer composed of a 120 kDa α-chain and a 75 kDa βchain. Upon activation, C3a is cleaved from the α-chain leaving C3bα’, a 110 kDa
fragment. Upon fixation, C3bα’ covalently binds acceptors on the bacterial surface, and
depending on the acceptor, will migrate as a band greater than 110 kDa. Also, both the
binding of multiple acceptors of varying molecular weight and the further degradation of
C3bα’ can result in a ladder-type banding pattern. The β-chain, however, is released by
reduction of C3b cystine bonds in sample buffer and is seen as a 75 kDa band.
C3 deposition was apparent on complement-resistant F. tularensis strains (Schu
S4, Fn, and LVS) indicative of complement activation (Fig. 2.2.A). C3 deposition on
complement-susceptible strains (LVSG and LVSR) was performed using C5-depleted
(C5d) serum (Fig. 2.2.B). C5 is a component of the terminal lytic pathway, and its
absence does not affect C3 deposition but does prevent bacterial lysis. To confirm that
C3 deposition is unaffected, assays using resistant strains were performed with both C5d
and fresh donor sera for comparison (Fig. 2.2.A). C3 associated with LVSG and LVSR
was markedly increased compared with resistant strains. Also, we found that LVSG
bound less C3 compared with LVSR.
75
Figure 2.2. Complement component C3 deposition occurs in greater
amounts on complement-susceptible strains of F. tularensis. 5x108 wildtype (A) or
variant strain (B) bacteria/reaction were incubated in buffer alone, or in 10% fresh, HI, or
C5-depleted (C5d) serum for 30’ at 37°C, washed in ice cold PBS containing protease
inhibitors, boiled in Laemmli’s buffer, separated by 7.5% SDS-PAGE (5x107
bacteria/lane unless diluted) and examined for C3 by Western blotting. Aliquots from
each sample were plated to count cfu just prior to lysis to equalize loading amounts.
Goat antiserum to human C3 was used for detection. Control lanes contain 2ng purified
76
native C3. The blot shown is representative of at least three independent experiments.
C3α, 120 kDa chain. C3β, 75 kDa chain.
77
We performed ELISAs to quantify total C3 deposition and to examine fixation of
the downstream complement components C5 and C7. C8 is a component of the MAC
and binds to the microbial surface subsequent to C7 deposition. Thus, we incubated each
strain in 10% C8-depleted (C8d) serum for 5 or 30 minutes at 37°C to assess fixation of
the upstream components. The above Western blotting results are supported by the
finding that LVSR bound significantly greater amounts of total C3 (Fig. 2.3). LVSG
bound an intermediate amount, but still more than each wildtype strain. At 5 minutes,
compared with LVSR, wildtype strains did not bind C5 and LVSG bound significantly
less than LVSR. At 30 minutes, C5 that bound to wildtype strains increased, but the
amount bound to LVSG increased dramatically and was comparable to the amount fixed
by LVSR. For C7 (a component of the MAC), a similar trend occurred in that there was
little binding at 5 minutes except on LVSR. Then, at 30 minutes, C7 binding to wildtype
strains remained low and binding to LVSG increased to approach the level fixed by
LVSR. Together, these findings provide strong evidence that the regulation of C3
deposition, which results in a marked reduction in the deposition of downstream
components, is crucial for F. tularensis resistance to complement-mediated lysis.
Temporal analysis of the nature of C3-derived fragments that bind to F. tularensis
The nature of bound C3 fragmentation can determine the outcome of complement
activation. C3b is necessary for activation of the terminal lytic pathway of complement,
but smaller cleavage fragments (including C3bi, C3dg, and smaller fragments) do not
initiate lysis. Of particular interest is C3bi because, like C3b, it mediates
78
Figure 2.3. Quantitative analysis of complement components C3, C5, and C7 fixed
by F. tularensis. For 5’ and 30’ at 37°C, 3x108 bacteria/reaction were incubated in 10%
fresh C8-depleted (C8d) serum or C8d serum containing 10mM EDTA to block
complement activity, washed in ice cold PBS containing protease inhibitors, and resuspended in H2O. Aliquots from each sample were plated to count cfu. 3x107
bacteria/well were applied to 96-well plates and dried overnight. Goat antisera to human
C3, C5, and C7 were used for detection. Absorbance at 415 nm was measured and values
obtained using EDTA in serum were subtracted from matched values obtained using
fresh serum. Means +/- SEM are given (N=3). For comparisons between LVS, LVSG,
and LVSR, significant mean differences were determined by ANOVA followed by
Bonferroni’s Multiple Comparison post-tests. *, p < 0.05.
79
80
opsonophagocytosis. On F. tularensis, inactivation of C3b to C3bi would account for
both resistance to lysis and C3-mediated opsonization. We assessed the nature of C3
fragmentation by Western blot analysis (Fig. 2.4). Importantly, the study of fragment
deposition by immunoblotting is confounded by the fact that C3α forms complexes via
covalent bonds, thereby affecting band migration. At physiological pH, C3α-acceptor
thioester bonds form more readily than amide bonds (293). We employed hydroxylamine
treatment, which cleaves ester linkages, to release C3α fragments from acceptor and C5convertase complexes. Treatment resulted in the disappearance of the majority of high
molecular weight bands.
Bacteria were incubated with 10% C5d serum from 1 to 60 minutes and we
compared hydroxylamine treated versus untreated samples. For resistant strains (Schu
S4, LVS, and Fn), overall C3 fragment deposition occurred rapidly and increased over
time (Fig. 2.4.A). In hydroxylamine treated groups, the appearance of C3biα1’ (68 kDa
C3biα’ fragment) occurred in excess of C3bα’ (110 kDa). For susceptible strains (LVSG
and LVSR), the overall rate of C3 fragment deposition was increased compared with
resistant strains (Fig. 2.4.B). Although C3bi was present on LVSG and LVSR, the
persistent and increasing appearance of C3bα’ was in stark contrast to its relative absence
on resistant strains. Also, the greater amounts of C3bα’ on LVSR at early time points,
compared with LVSG, correlates with its greater susceptibility to lysis. At early time
points, ratios of C3biα1’ band densities to C3bα’ band densities illustrate more rapid
conversion of C3b to C3bi on resistant strains (Fig. 2.4.C). Unfortunately, densitometry
is less quantitative as the band intensities for C3 deposition become saturating for LVSG
and LVSR. Ratios of less than 1 for LVSG and LVSR at early time
81
Figure 2.4. The nature of bacteria-bound C3 fragments for different F. tularensis
strains over time. 7.5x108 wildtype (A) or variant strain (B) bacteria/reaction were
incubated in 10% C5-depleted serum from 1-60 minutes at 37°C. After washing in ice
cold PBS containing protease inhibitors, bacterial pellets were lysed by boiling in 1%
SDS. To allow for the quantitative assessment of fragmentation, hydroxylamine
(NH2OH) was used to release C3bα’ (and smaller α-chain fragments) from covalent
thioester bonds to bacterial acceptor molecules. Immunoblotting for C3 was performed
82
as in Figure 2, except that each lane contains 1x108 bacteria and a 16.5 cm gel was used
for SDS-PAGE to allow for greater band separation. Control lanes contain 2ng each of
native C3 and C3bi. Each blot is representative of at least three independent experiments.
C3α, 120 kDa native C3 α-chain. C3bα’, 110 kDa C3b α-chain. C3β, 75 kDa native C3
β-chain. C3biα1’, 68 kDa C3bi α-chain fragment. (C) Densitometry ratios of C3biα1’ to
C3bα’ at early time points are compared for each strain. For variant strains, ratios ≤ 1
indicate a higher rate of C3b deposition compared to the rate of C3b to C3bi conversion.
Mean ratios +/- SEM for the densitometry of independent blots are given (N=2 for Schu
S4 and LVSG, N=3 for LVS, Fn and LVSR).
83
points, prior to the saturation of band intensities, indicate that new C3 deposition occurs
more rapidly than conversion of bound C3b to C3bi. For F. tularensis, these results
indicate the importance of rapid conversion of C3b to C3bi for resistance to complementmediated lysis. Also, we conclusively show that complement activation by Schu S4,
LVS, and Fn leads to opsonization characterized by persistent C3bi deposition.
O-antigen expression is a major determinant of susceptibility to complementmediated lysis and C3b to C3bi conversion
Thus far, our examinations of complement activation by LVS variant strains used
LVSG and LVSR. We predicted that differences in either capsule or LPS between LVS
and the variant strains results in complement susceptibility. We performed transmission
electron microscopy to compare LVS and LVSG (Fig 2.5). We did not detect any
appearance of a glycocalyx even for LVS, which is completely resistant to complement.
However, the majority of LVS bacilli were surrounded at least partially by an electron
dense material that was far less abundant on LVSG. We focused on the possibility that
this material is representative of LPS associated O-antigen.
Both LVSG and LVSR have been shown to express structurally altered LPS Oantigen compared with LVS (176). Furthermore, although LVSR is described in the
literature as a capsule mutant strain, a microscopic comparison of LVS and LVSR by
Sandstrom et al. also suggests that LVSR is deficient of the electron dense outer layer
that we show is associated with LVS. Here, we used an LVS wbtA mutant (LVSΔwbtA),
with deletion of a gene within the O-antigen operon and devoid of LPS O-antigen, to
directly test the effect of LPS O-antigen expression on complement activation (Table 1,
84
Figure 2.5. Transmission electron microscopic images of LVS and LVSG show
differences in the outer membrane. LVS (A-C) and LVSG (D-E) were grown
overnight in TSB at 37°C. An ill-defined, thick electron dense material (arrows) is often,
but not always, observed on the outer surface of LVS. The presence of glycocalyx
associated with LVS, as described in previous studies, could not be seen. The width of
each panel is approximately 2µm.
85
Fig. 2.6). Survival assays were performed as described above to examine effects on
complement-mediated lysis. In 5% and 50% fresh donor serum, lysis of LVSΔwbtA did
not differ significantly from lysis of LVSR (Fig. 2.6.A). Next, Western blots of C3
deposition were performed to determine the effect of O-antigen on the temporal nature
and mechanism of C3 fixation. Like LVSR, and unlike LVS, LVSΔwbtA rapidly fixed
total C3-derived fragments over time and specifically fixed a markedly increasing
amount of C3bα’ (Fig. 2.6.B). Note that even at one minute, when amounts of C3β are
similar between the three strains, amounts of C3bα’ are greater on the susceptible strains.
Based on the observed effects of a wbtA deletion on complement activation by the
mutant bacteria, we characterized O-antigen expression by both wildtype and variant
strains examined in our studies. Using whole cell Fn, LVS, LVSG, LVSR, and
LVSΔwbtA lysates, we analyzed O-antigen expression by Western blot (Fig. 2.7) and
silver stain (data not shown). Using an anti-subsp. tularensis-type O-antigen antibody for
immunoblotting, a typical laddering pattern is demonstrated for LVS with a grouping of
full-length chains near the top of the membrane and clearer delineation of bands
representing progressively shorter chains. For LVSG, an identical pattern is evident, but
total O-antigen expression is reduced compared with LVS. This implies that LVS and
LVSG express O-antigen of similar length, but that the amount of O-antigen per
bacterium (or the amount of O-antigen expressing bacteria in a population) is reduced for
LVSG. O-antigen was not detected for LVSR or, as expected, LVSΔwbtA. These results
were duplicated by silver stain in that O-antigen banding was evident for LVS, was less
evident for LVSG, and was absent for LVSR and LVSΔwbtA (data not shown). Using an
anti-subsp. novicida-type O-antigen antibody, we detected full length O-antigen
86
Figure 2.6. Complement susceptibility and surface C3b stability are determined by
F. tularensis LPS O-antigen expression. Since LVSR was found to be deficient in OAg expression, serum survival (A) and C3-fragment deposition assays (B) were repeated
using an O-Ag mutant strain, LVSΔwbtA, for comparison. Survival assays in human
non-immune fresh donor and HI sera were done as described for Figure 2.1. Mean ratios
+/- SEM are given (N=3). ns, no significance by unpaired Student’s t test. Kinetic
analyses of C3-fragments (released from bacterial acceptors using NH2OH) bound to
LVS, LVSR, and LVSΔwbtA were done as described in Figure 2.4. The blot is
representative of 2 independent experiments.
87
Figure 2.7. Western blot showing LPS O-antigen production by different F.
tularensis strains. Whole cell bacterial lysates (containing 1x1010 cfu) were subjected to
12% SDS PAGE and immunoblotting was performed using an anti-F. tularensis Oantigen antibody. As in Figure 2.6, a laddering pattern indicative of full-length O-antigen
production is evident for LVS and LVSG, with less O-antigen production by LVSG. Oantigen was not detected on LVSR or LVSΔwbtA. The blot shown is representative of 2
independent experiments.
88
expression by Fn, but not by LVS or any of the LVS-derived variant strains (data not
shown). Thus, wildtype strains, which resist complement-mediated lysis and rapidly
mediate conversion of surface-bound C3b to C3bi, express abundant, full-length Oantigen. LVSR and LVSΔwbtA, which are highly susceptible to complement-mediated
lysis and rapidly fix high amounts of C3bα’, do not express O-antigen. LVSG, which is
moderately susceptible to lysis and fixes persistent C3bα’ (albeit less rapidly compared
to LVSR and LVSΔwbtA), expresses a relatively intermediate amount of O-antigen.
To confirm the regulatory role of O-antigen in mediating resistance to
complement-mediated lysis and in mediating conversion of C3b to C3bi, we employed an
additional mutant strain devoid of wbtM (LVSΔwbtM), which is a gene downstream of
wbtA in the O-antigen operon. In addition, we used the complemented strain containing
the pFTNAT plasmid expressing functional wbtM as described (22). We found that
compared to wildtype LVS and to the complemented mutant (LVSΔwbtM:pFTNATwbtM), LVSΔwbtM and LVSΔwbtM that contained an empty plasmid
(LVSΔwbtM:pFTNAT) were susceptible to complement mediated lysis in both 5 and
50% fresh donor serum at 1 hour (Fig. 2.8.A). The degree of susceptibility correlated
strongly with both LVSΔwbtA and LVSR. By ELISA, we also found that in 10% C8d
serum, LVSΔwbtM fixed relatively high amounts of C3, C5, and C7 (comparable to
LVSR in Fig. 2.3), but that LVSΔwbtM:pFTNAT-wbtM fixed less total C3 and minimal
C5 and C7 (comparable to wildtype strains in Fig. 2.3) (data not shown). Finally, we
determined the nature of C3-derived fragments fixed in 10% C5d serum at 10 minutes.
We found high amounts of C3bα’ on LVSΔwbtM and LVSΔwbtM:pFTNAT, but much
higher amounts of C3biα1’ compared to C3bα’ fixed to LVS and to
89
Figure 2.8. Restoration of O-antigen expression on a mutant strain results in
complement resistance and C3b inactivation. (A) Survival assays in human nonimmune fresh donor and HI sera were done as described for Figure 2.1 using LVS,
LVSΔwbtM, LVSΔwbtM containing an empty complementation plasmid
(LVSΔwbtM:pFTNAT), and the complemented mutant strain expressing O-antigen
(LVSΔwbtM:pFTNAT-wbtM). Mean ratios +/- SEM are given (N=3). (B) Analysis of
C3-fragments (released from bacterial acceptors using NH2OH) fixed by each strain
90
described in A. C3 fixation occurred in 10% C5d serum at 37°C for 10’. Western blots
were performed as described in Figure 2.4 and the blot shown is representative of 2
independent experiments.
91
LVSΔwbtM:pFTNAT-wbtM (Fig. 2.8.B). We conclude that relative O-antigen
expression correlates strongly with the degree of complement activation by each strain.
Complement activity in bronchoalveolar lavage fluid and the effect of opsonization
on F. tularensis uptake by human alveolar macrophages
Complement activation and resistance to complement-mediated lysis in serum
would be important both for the intramacrophage life cycle of F. tularensis and the
occurence of late bacteremia associated with tularemia. However, we are also interested
in the potential for complement activation in the airway, which may occur immediately
upon exposure. We tested bacterial survival in bronchoalveolar lavage (BAL) fluid
isolated from a human volunteer (Fig. 2.9). To account for dilution effects that occur
during the process of alveolar lavage, isolated BAL fluid was concentrated (cBAL) to
better analyze the effects of complement as a component of airway surface fluid (ASF).
We tested LVS, which is resistant to complement in serum; and we tested LVSG and
DH5α, which are susceptible to complement in serum. We added EDTA as a negative
control because it inhibits any potential lytic effects of complement. We found that all
the bacterial strains tested, including DH5α, were completely resistant to lysis in cBAL.
Since the relative concentrations of individual complement components differs in ASF
compared with serum (see Chapter 1), it is likely that the terminal pathway is not a
functional constituent of ASF. However, it is quite possible that complement activity in
BAL fluid leads to C3 deposition on F. tularensis strains, but this was not tested.
Several laboratories, including ours, have conclusively shown that complement is
a powerful opsonin in mediating efficient F. tularensis uptake by human monocytes,
92
Figure 2.9. Human concentrated bronchoalveolar lavage fluid (cBAL) is deficient of
complement lytic activity. 3x106 bacteria/reaction were incubated in PBS, cBAL, or
cBAL containing 10mM EDTA for 1 hour at 37°C. Reaction tubes were then set on ice
for 5’. Bacteria were washed in cold PBS containing HSA and then plated to count cfu.
DH5α, a highly susceptible strain of E. coli, was used as a positive control for
complement activity. Bacterial survival was not significantly affected by the additions of
cBAL or EDTA. Means +/- SD are given for triplicate measurements in 1 experiment.
93
monocyte-derived dendritic cells, and monocyte-derived macrophages. Furthermore,
studies using animal models indicate that macrophages in the airway are the primary in
vivo targets for intracellular infection leading to bacterial dissemination. However, to our
knowledge, the uptake of virulent or avirulent strains of F. tularensis by human alveolar
macrophages (AMs) has not been previously studied. We isolated AMs and tested the
effect of complement-mediated opsonization on uptake of both the highly virulent Schu
S4 strain and the avirulent (for humans) LVS strain (Fig. 2.10). As a source of
complement, we used 10% autologous donor serum or used HI serum as a negative
control. By microscopic analysis, we quantified the total number of associated bacteria
per AM (attached and ingested) (Fig. 2.10.A) and the number of attached bacteria per
AM (Fig. 2.10.B), and then calculated the number of ingested (phagocytosed) bacteria
per AM by subtracting the number of attached bacteria from the number of associated
bacteria (Fig. 2.10.C). We found that bacterial association and attachment increased for
both strains with the addition of HI serum compared with media alone suggesting a
potential role for heat stable opsonins such as Ig that bind to Fcγ receptors. Compared to
HI serum, bacterial association and attachment also increased for both strains with fresh
serum. As a preliminary study, data suggest that in the case of LVS, the addition of fresh
serum leads to primarily increased attachment of bacteria to AMs whereas in the case of
Schu S4, the addition of fresh serum allows for greater attachment and ingestion
(phagocytosis) of bacteria.
94
Figure 2.10. Complement mediated uptake of Schu S4 and LVS by human alveolar
macrophages (AMs). Monolayers of AMs were infected with Schu S4 or LVS at an
MOI of 30:1 in the presence of 10% autologous donor whole or HI serum (or culture
media containing 0.5% HSA) for 2 hours at 37°C. Cells were washed twice in media and
then fixed in 2% paraformaldehyde. Some monolayers were permeabilized using MeOH.
Bacteria were immunofluorescently labeled and counted by microscopy. (A)
Permeabilized monolayers were used to count total associated bacteria per macrophage
(attached and ingested). (B) Non-permeabilized monolayers were used to count the
number of attached bacteria per macrophage. (C) Ingestion was quantified by subtracting
the number of attached bacteria from the number of associated bacteria. Bars represent
means +/- SD for triplicate measurements in 1 experiment (except in the case of
association for Schu S4 where only two measurements were obtained).
95
96
Discussion
The rapid onset and disease progression of pneumonic tularemia, despite low
inoculation doses, implicates a failure of innate immune responses to control infection
with F. tularensis. That F. tularensis survives and replicates within phagocytes is
particularly indicative of an ineffective microbicidal reaction to infection. However,
bacilli must also survive exposure to extracellular mediators of innate immunity,
including complement, as a prerequisite for cellular invasion. It has been established that
F. tularensis survives in whole blood both in vivo and ex vivo (74, 75). Importantly,
pneumonic tularemia, the most severe form of tularemia, can develop secondary to
cutaneous or mucosal infections. Clearly, F. tularensis-complement interactions in
serum or interstitial fluid would impact the outcome of secondary pneumonic disease.
Complement is also abundant in bronchoalveolar fluid, and thus, likely affects primary
pneumonic disease as well (281, 282, 286). Direct evidence of potent classical pathway
activity in isolated human bronchoalveolar lavage fluid was demonstrated by C3 fixation
by Mycobacterium tuberculosis and group B streptococcus (283, 285). The importance
of complement in pulmonary immunity is exemplified in patients who are genetically
deficient in either complement components or complement receptors, since they are at
significantly increased risk for respiratory infection (279).
In addition to the direct mediation of pathogen lysis, complement modulates
innate immune responses via the release of component fragments with anaphylactic
activity and via opsonophagocytosis, which would affect cell-mediated responses.
Matrix metalloproteinase 9 (MMP9) deficient mice have increased resistance to Schu S4
97
infection, likely due to an inability to recruit highly active neutrophils to the lung (88).
Anaphylactic complement fragments may have a similar role to MMP9 in recruiting
neutrophils to the site of F. tularensis infection. In addition, since the rate of F. tularensis
phase variation increases upon intracellular or in vivo growth, increased complement
activation by grey variants might compound local inflammation. Unfortunately, the role
of complement in mediating disease in animal models caused by Type A strains or Fn
(the most virulent strains for mice) has not been studied. Opsonization is influential
beyond simply increasing the rate of particulate uptake. Depending on the identity of the
opsonin (e.g. IgG versus C3bi) and associated surface receptors, downstream signaling
events differ. Phagosomal trafficking, cytokine responses, and production of reactive
oxygen intermediates are all influenced by the exact nature of receptor-ligand interactions
(108).
Our data demonstrate complement activation by each F. tularensis strain tested.
Several studies previously addressed susceptibility of Fn, LVS, and respective derivative
strains to complement-mediated lysis (discussed below). We extended these studies to
include Schu S4 and we provide an analysis of complement activation at the level of
surface component deposition. Since it was shown that CR3 and CR4 have a major role
in opsonophagocytosis of F. tularensis by human monocyte-derived phagocytes (60, 109,
110) and the primary ligand for these receptors is C3bi, we hypothesized that F.
tularensis would fix C3b and that conversion to C3bi would ensue. Conversion to C3bi
and smaller fragments would also account for the ability of bacilli to survive
extracellularly in vivo. We characterize the nature of F. tularensis-bound C3-derived
fragments and show directly that opsonization with C3bi occurs. Furthermore, we show
98
preliminary data indicating that complement mediated opsonization significantly
increases the attachment of Schu S4 and LVS to isolated human AMs and potentially
increases the efficiency of Schu S4, but not LVS, uptake by AMs. Our finding that LVS
fixes C3bi contradicts an earlier study, which reported that LVS did not bind
radioactively labeled C3-derived fragments when both were added to human serum (183).
Since we find that rapid C3b conversion occurs on the surface of LVS, it is possible that
C3 cleavage adversely affected radioactive labeling in that study.
We also compare the nature of C3 deposition on strains resistant to complementmediated lysis to its nature on susceptible strains. We found that, on resistant strains,
conversion of C3bα’ to C3biα’ occurred more rapidly than deposition of new C3bα’.
Finally, we found that O-antigen expression is a major determinant of complement
activation. The strength of this finding is increased by the fact that multiple LVS variant
strains, generated by distinct methodologies, were used. These include the selection of a
spontaneous colony morphology variant (LVSG), the creation of a strain by random
mutagenesis of LVS followed by the selection of a rough colony morphology variant
(LVSR), and the construction of targeted gene mutants (LVSΔwbtA and LVSΔwbtM).
A role for capsule has previously been implicated in the serum resistance of F.
tularensis (350). We chose to study LVSR, a putative capsule negative (Cap-) strain
derived from LVS by Sandstrom et al. (183) and subsequently studied by Cowley et al.
(176), in order to characterize the effect of capsule production on complement component
deposition. For other bacteria, encapsulation has been shown to reduce C3 deposition
and to protect against subsequent lysis (351-353). The specific nature of the F. tularensis
capsule has not been determined conclusively. Hood reported that decapsulation of F.
99
tularensis occurs in hypertonic saline, and that capsular material is biochemically distinct
from the cell wall of decapsulated bacilli (159). Recently, a putative capsule locus in the
LVS genome, containing orthologous genes to capB and capC of Bacillus anthrasis, was
reported (160). In some studies, an electron lucent material typical of a loose capsule can
clearly be seen surrounding bacilli grown in defined media (157, 158). Despite these
reports, several laboratories, including ours (Fig. 2.5), have been unable to conclusively
identify capsular material by microscopy, possibly due to the use of rich media for
culturing (109, 180). In studies of serum resistance, Cap- strains (including LVSR) were
shown to activate complement (183, 184). Sorokin et al. reported, however, that the
Cap- strain used in their study exhibited a truncated O-antigen, which may have been the
true determinant of complement susceptibility. To substantiate the designation of LVSR
as a Cap- mutant, electromicrographs comparing LVSR to LVS were presented by
Sandstrom et al. (183). However, the represented LVS capsule does not resemble the
images of F. tularensis capsule shown in the above cited studies. Furthermore, O-antigen
expression by LVSR was not studied. We conclude, based on our studies, that mutations
in LVSR affect O-antigen expression, and not capsule production. However, that Oantigen is a constituent of capsular polysaccharide cannot be ruled out. Also, the
possibility exists that encapsulation occurs only under specific growth conditions and
provides an additional measure of protection against complement.
Since we do not detect encapsulation on serum-resistant wildtype strains, we
explored alternative mechanisms of resistance, including the effect of LPS expression.
Compared with other Gram-negative pathogens, F. tularensis LPS is unique (161). The
F. tularensis LPS O-antigen is composed of a repeating carbohydrate tetramer that is
100
identical for all subspecies tested thus far except for subsp. novicida, which has two novel
sugars in the tetramer (21, 173). It is unlikely, however, that this structural difference in
O-antigen affects complement activation since Schu S4 and LVS do not differentially fix
C3-fragments compared to Fn (Fig. 2.2). Previously, O-antigen mutant strains derived
from both LVS and Fn have been found to be susceptible to lysis in serum, unlike the
parent strains (179-182). Also, LVS grey variants that lack O-antigen are similarly
susceptible to lysis in serum (177). We repeated these results using LVSΔwbtA (Fig.
2.6.A) and LVSΔwbtM (Fig. 2.8.A). Furthermore, we show similarities relating to serum
resistance and C3 fixation between LVSR, LVSΔwbtA, and LVSΔwbtM (Figs. 2.6.B and
2.8.B). Since the absence of O-antigen expression is also similar for each strain, and
since complementation of the LVSΔwbtM mutant restores resistance to lysis and efficient
C3b to C3bi conversion, we conclude that loss of O-antigen expression is the primary
cause of serum susceptibility.
In addition to identifying a role for O-antigen, we show that complement
resistance is due, in part, to a reduction of C3 on wildtype strains. Our data indicate that
this is the result of rapid conversion of fixed C3b to C3bi, which would limit C3convertase-mediated amplification of C3b deposition (Fig. 2.4). C3b inactivation is a
common resistance mechanism employed by Gram-negative bacteria that are human
pathogens and C3bi fixation enables efficient uptake by host phagocytes (291-293, 354).
We are currently evaluating two hypotheses to explain C3b cleavage based on known
microbial mechanisms of complement evasion. The first involves direct bacterial
protease expression and the second involves the recruitment of host-derived negative
regulators of complement. Importantly, a complete characterization of the mechanisms
101
of C3b inactivation would include the evaluation of both hypotheses, since neither is
mutually exclusive, and either might be affected by O-antigen expression. O-antigen
expression is reduced or absent on each complement-susceptible strain used in the present
study and each was derived by a distinct methodology. Thus, the existence of
unidentified F. tularensis serum resistance factors, unaffected by O-antigen, is unlikely.
O-antigen cannot act as an acceptor, at least not as the only acceptor, for serum-derived
negative regulators (involved in C3b cleavage) because C3bi is abundant on LVSR,
LVSΔwbtA, and LVSΔwbtM (Figs. 2.6 and 2.8). That C3b conversion occurs on
complement-susceptible strains, albeit at a slower rate than C3b deposition, signifies the
relative importance of O-antigen expression for survival. Other potential mechanisms of
resistance to complement-mediated lysis cannot be ruled out such as steric hindrance to
the formation of the MAC or the fixation of complement components by moieties distal
to the outer membrane. Distal component fixation would likely result in the consumption
of overall serum hemolytic activity, a possibility discussed further in Chapter 4.
In summary, we present conclusive evidence that complement is activated, based
on C3 fixation, by virulent F. tularensis strains including Schu S4, LVS, and Fn. Rapid
conversion of C3b to C3bi on these strains contributes to their ability to resist
complement-mediated lysis. LPS variant strains derived from LVS were susceptible to
complement-mediated lysis, due in part to limited C3b to C3bi conversion that led to
striking increases in C3bα’ fragment deposition and to increased binding of components
of the terminal lytic pathway (including C5 and C7). Finally, we identify O-antigen as a
key determinant for the outcome of complement activation by F. tularensis.
102
Chapter 3
Francisella tularensis principally activates the classical complement pathway in the
presence and absence of LPS O-antigen
Introduction
Francisella tularensis causes a fulminate disease known as tularemia, which is
particularly deadly with pulmonary involvement. Untreated pneumonic tularemia results
in mortality rates of up to 30-60% (12). This facultative intracellular Gram-negative
bacterium primarily infects macrophages and disseminates both within cells and
extracellularly via the reticuloendothelial system (53, 69). Studies using animal models
and human volunteers determined that as few as 10 bacteria can cause pneumonic
disease, however, rapid intracellular population growth and bacterial dissemination lead
to a late disease state characterized by severe tissue necrosis and sepsis. Death can occur
from within days to weeks following an infection, and thus, an ineffective innate immune
response is primarily responsible for the failure to control bacterial growth. Although
pathogenic characteristics of tularemia have been described, little is currently known
about specific molecular interactions between this organism and components of the
innate immune system. We are interested in describing interactions between F.
tularensis and the complement cascade, which is the major arm of humoral immunity for
103
the non-immune host.
The bacterium’s ability to survive extracellulary and to efficiently infect
macrophages is dependent in part on its ability to regulate complement. There are three
major pathways of complement activation (188): classical, lectin and alternative. Surface
antigen recognition by IgM or clusters of IgG initiates the classical pathway. C1 is a
heteromultimeric complex composed of one C1q molecule that typically binds
immunoglobulin and two molecules each of C1r and C1s. Upon binding, C1 is activated
so that C1s, an esterase, cleaves the downstream components C4 and C2. The major
fragments of cleaved C4 and C2 are C4b and C2a, which interact to form a complex that
becomes covalently bound to the surface. This complex, C4b2a, is a C3-convertase that
cleaves the central complement component C3. C3b can then covalently bind to the
activating surface in order to initiate the terminal lytic pathway of complement.
The lectin-mediated pathway is the second major complement activation pathway
and it depends on surface carbohydrate recognition by soluble lectins, including mannose
binding lectin (MBL) and members of the ficolin protein family. Bound lectins recruit
MBL-associated serine proteases (MASPs) that are homologous to C1s and that activate
C4 and C2. Thus, the classical and lectin pathways converge at the level of C4b2a C3convertase formation.
Random H2O-catalyzed “tick-over” of C3 initiates the alternative pathway, the
third major activation pathway. C3(H2O) reacts with alternative pathway components in
solution at a low rate to form soluble C3-convertases. C3 contains a reactive thioester
moiety that forms a covalent bond with bacterial surface acceptor molecules, but that is
shielded by the three-dimensional structure of native C3 (221). However, cleavage of C3
104
uncovers the thioester moiety on C3b. Once C3b becomes surface bound, it can form a
heteromultimeric complex with Factor B (FB) and Factor D (FD). FD is a serine protease
that releases FBa from native FB, resulting in C3bBb formation, the alternative pathway
C3-convertase that in turn cleaves native C3 molecules. Amplification of C3b deposition
occurs via the amplification loop (regardless of which activation pathway initially
produces C3b) due to alternative pathway-mediated formation of C3bBb. This
potentially leads to exponentially increased amounts of surface C3b in comparison to the
amounts of surface-bound upstream components. C3b is also important because further
cleavage by Factor I produces C3bi, the major complement-associated opsonin that
mediates increased phagocytosis of F. tularensis by macrophages.
In Chapter 2, we demonstrated that complement activation by F. tularensis occurs
in the presence and absence of surface lipopolysaccharide (LPS) O-antigen. However,
complement activation resulted only in the lysis of strains producing reduced amounts of
O-antigen or that were completely O-antigen-deficient. For virulent O-antigen-producing
strains that were resistant to lysis, activation resulted predominantly in surface-associated
C3bi and a marked reduction in surface C3b compared to susceptible strains. Here we
examine the relative contribution of the three major complement activation pathways
leading to the lysis of complement-susceptible strains and to the opsonization of resistant
strains.
Materials and Methods
Bacterial strains. F. tularensis subsp. tularensis strain Schu S4, a Centers for
Disease Control and Prevention clinical isolate, was provided by Rick Lyons (University
105
of New Mexico, Albuquerque, NM). F. tularensis subsp. holarctica LVS (ATCC 29684)
was provided by Karen Elkins (Center for Biologics Research and Evaluation, US FDA,
Bethesda, MD). F. tularensis subsp. novicida (U112; Fn), LVSG, and LVSR were
provided by Fran Nano (University of Victoria, Victoria, BC, Canada). LVSG is a
spontaneous grey phase variant that rarely reverts to LVS when grown on chocolate II
agar (176). LVSR was originally described as a capsule-negative strain (183) and was
selected for its rough colony morphology after the mutagenesis of LVS by treatment with
acridine orange. The LPS O-antigen mutant, LVSΔwbtA, provided by Dara Frank
(Medical College of Wisconsin, Milwaukee, WI), was created by modified Himar1
transposon (HimarFT)-mediated mutagenesis of LVS (345). Experiments using Schu S4
were carried out within biosafety level 3 (BSL3) select agent-certified laboratories with
adherence to federal and institutional select agent regulations. Bacteria were grown
overnight (approximately 18 hours) on chocolate II agar (Becton Dickinson, Franklin
Lakes, NJ) at 37°C.
Human sera, complement components, and reagents. Serum was isolated
from healthy adult volunteers with no known exposure to F. tularensis according to a
protocol approved by the Ohio State University Medical College Internal Review Board.
The sera were processed to maintain optimal complement activity (346). Briefly, isolated
non-heparinized whole blood was kept at room temperature for 1 hour to allow for clot
formation and then at 4°C to allow for clot contraction. The clot was removed by
centrifugation at 500xg for 15 minutes at 4°C. The serum fraction was collected, filter
sterilized, aliquoted, and stored at -80°C. Heat inactivation (HI) was performed at 56°C
106
for 30 minutes. C5-depleted (C5d), C1q-depleted (C1qd), and Factor B-depleted (FBd)
sera, and purified C1q and Factor B (FB) were purchased from Complement Technology,
Inc (San Antonio, TX) and stored at -80°C. On the day of use, fresh sera were thawed at
room temperature, then immediately chilled on ice until needed. A concentrated serine
protease inhibitor cocktail (containing AEBSF, aprotinin, elastatinal, and GGACK) was
purchased from Calbiochem (Madison, WI). Other chemicals were purchased from
Sigma-Aldrich.
Bactericidal assays. Complement-mediated killing was carried out using fresh
non-immune or HI (negative control) sera. Bacteria were suspended in gelatin veronal
buffer (GVB++; 0.1% gelatin, 5.5 mM barbital, 142 mM NaCl, 0.5 mM MgCl2, 0.15 mM
CaCl2; pH 7.3) at equalized concentrations by measuring the optical density at 600 nm.
For each assay, 2x106 bacteria were incubated with various serum concentrations for 1
hour in microcentrifuge tubes (final volume of 200µl in reaction buffer) at 37°C with
slow agitation. For some experiments, 10mM EDTA or 10mM EGTA with 7mM MgCl2
was added to fresh serum in order to differentiate potential complement activation
pathways. In other experiments, C1qd, C1q–replete (C1qr; achieved by adding purified
C1q to C1qd at a final concentration of 200µg/ml), FBd, and FB–replete (FBr; achieved
by adding purified Factor B to FBd at a final concentration of 400µg/ml) sera were used.
Reactions were stopped by placing tubes on ice for 5 minutes. 10-fold serial dilutions
were plated to determine surviving colony forming units (cfu).
107
C3 deposition and Western blotting. C5d, C1qd, and C1qr sera were used to
evaluate complement component C3 deposition on F. tularensis strains mediated by the
classical pathway. After pre-blocking microcentrifuge reaction tubes in PBS containing
0.1% human serum albumin (ZLB Plasma, Boca Raton, FL) for 30 minutes at 37°C,
5x108 bacteria/reaction were incubated in 10% sera for various times at 37°C. Adding an
ice-cold serine protease inhibitor cocktail in blocking buffer terminated deposition and
fragmentation. Unbound C3 was removed from bacterial pellets by washing in blocking
buffer (x3) and PBS (x1) at room temperature (12,000xg for 3 minutes). To ensure equal
lane loading, aliquots were taken from the final PBS wash and plated to count cfu. After
final resuspension in Laemmli’s sample buffer, samples were boiled for 2 minutes and
separated by minigel 7.5% SDS-PAGE (BioRad, Hercules, CA), followed by protein
transfer to PVDF membranes (Milipore, Billerica, MA). Membranes were blocked
overnight at 4°C in advanced ECL blocking buffer at 2% (v/w), which was also used for
antibody dilution (Amersham, Piscataway, NJ). Goat antiserum to human C3 (Quidel,
San Diego, CA) (diluted to 1:20,000) served as the primary antibody for 1 hour
incubations at room temperature. HRP-conjugated rabbit anti-goat IgG (H+L) antibody
(Biorad) (diluted to 1:20,000) was used as the secondary antibody for 1 hour incubations
at room temperature. Advanced ECL reagent was used for detection (Amersham).
Statistics. To determine associations of significance between or within groups
where indicated, data were analyzed using unpaired (one-tailed) or one-sample Student’s
t tests or one-way ANOVA followed by Dunnett Multiple Comparison post-tests (using
Graphpad Prism 4 software). Cfu data were converted to log values to equalize
108
variances. Differences between groups were considered statistically significant for p
values <0.05.
Results
Complement-mediated lysis of susceptible F. tularensis strains occurs by more than
one activation pathway
We began to differentiate the relevant complement activation pathways targeting
susceptible F. tularensis strains using chelators in serum viability assays. Initiation of the
classical pathway is typically dependent upon surface recognition by IgM or IgG and
subsequent recruitment of C1 (a complex formed by C1q, C1r, and C1s). The interaction
between C1q and C1r is stabilized by Ca++ and the serine protease activity associated
with C1s requires Mg++ (355). Similarly, lectin pathway-associated MASP proteases
require Ca++ for activity. There is no Ca++ requirement associated with the alternative
pathway; however, Mg++ is necessary for the interaction between FB and C3b. We used
EDTA, which chelates both Ca++ and Mg++, to block all three activation pathways.
EGTA, which chelates Ca++, was used to specifically block the classical and lectin
mediated pathways (Fig. 3.1). 7mM Mg++ was added to buffers containing EGTA to
ensure optimal alternative pathway function. In serum concentrations as low as 1%
where the classical and lectin, but not alternative, pathways are active, the inclusion of
EDTA and EGTA inhibited complement activity (restored viability) toward LVSG and
LVSR. Bacterial lysis in 10% or higher serum concentrations was not completely
inhibited by EGTA, indicative of some alternative pathway activity. The implications of
these data are that C3bBb (alternative pathway C3-convertase) formation occurs on both
109
Figure 3.1. Complement activation by susceptible F. tularensis strains occurs via
more than one pathway. LVSG and LVSR were incubated in increasing concentrations
of fresh or HI serum, or fresh serum containing 10mM EDTA or 10mM EGTA with
7mM Mg++ for 60 minutes at 37°C. EDTA inhibits all complement activation pathways
110
by chelating both Ca++ and Mg++. EGTA (with Mg++) inhibits the classical and lectinmediated pathways by specific Ca++ chelation. Viable bacteria were plated to count cfu.
Results indicate that activation of the alternative pathway occurs at concentrations above
10%. Also, Ca++-dependent (classical and/or lectin-mediated) pathways are activated in a
concentration dependent manner in serum concentrations as low as 1%. Bars represent
means +/- SEM of independent reactions in sera from separate donors (N=3). For each
serum test group, significant mean differences from controls (no serum) were determined
by ANOVA. *, p < 0.01 for each Dunnett Multiple Comparison post-test. ND, not
detected.
111
LVSG and LVSR, but that C4b2a (classical and lectin pathway C3-convertase) formation
is predominant. At higher serum concentrations, once C4b2a initiates C3b deposition,
amplification of C3b deposition via C3bBb likely occurs. Amplification via C3bBb
formation is a possible explanation for the relatively high levels of C3b deposition shown
for susceptible strains (see Chapter 2).
To further explore mechanisms of complement activation, we used C1qd serum
(C1q-depleted serum deficient of classical pathway activity; Fig. 3.2.A) or FBd serum
(Factor B-depleted serum deficient of alternative pathway activity; Fig. 3.2.B) to test
survival of complement-susceptible strains. Lysis of LVSG did not occur in 5, 10, or
20% C1qd serum, but was restored in C1q-replete (C1qr) serum. Lysis of LVSR did not
occur in 5% C1qd serum, and occurred to only a small extent in 10% and 20% C1qd
serum, but was completely restored in C1qr serum. In FBd and FB-replete (FBr) sera,
lysis occurred that was comparable with C1qr serum for both strains. Also, compared
with FBd serum, lysis appears to increase in 20% FBr serum (significant only for LVSR),
which supports only a minor role for alternative pathway activation. Survival assays
using fresh serum with the addition of chelators (data not shown) and using component
deficient sera (Fig. 3.3) were repeated using LVSΔwbtA, a targeted mutant derived from
LVS that does not produce O-antigen. Results correlated with those previously described
for LVSR in that a similar degree of lysis occurred for each serum concentration and that
the classical pathway was necessary for optimal lysis.
Taken together, the data presented here demonstrate two important findings.
First, the classical pathway, but not the alternative pathway, is necessary for the optimal
lysis of complement susceptible strains at both low and high serum concentrations.
112
Figure 3.2. Optimal lysis of variant F. tularensis strains is dependent upon C1q.
LVSG and LVSR were incubated in 5-20% (A) C1q-depleted (C1qd) and C1q-replete
C1qd (C1qr) or (B) Factor B-depleted (FBd) and Factor B-replete FBd (FBr) sera for 1
113
hour at 37°C. Viable bacteria were plated to count cfu. Minimal, but significant, lysis of
LVSR occurs in 10% or 20% C1qd serum, which implicates a minor role for the
alternative or lectin pathway. Repletion with C1q, however, restores optimal lysis of
both strains. Conversely, repletion of FBd serum has no effect on lysis compared with
FBd serum (except for LVSR at 20%). Bars are the means +/- SEM of independent
reactions (N=3). For each serum test group, significant mean differences from controls
(no serum) were determined by ANOVA. *, p < 0.01 for each Dunnett Multiple
Comparison post-test. **, p < 0.05 for one-tailed Student’s t tests for significance
between depleted and replete sera. ND, not detected.
114
Figure 3.3. Optimal lysis of LVSΔwbtA, an O-antigen mutant strain, is dependent
upon C1q. Bacteria were incubated in 5-20% C1q-depleted (C1qd), C1q-replete C1qd
(C1qr), Factor B-depleted (FBd) or Factor B-replete FBd (FBr) serum for 1h at 37°C and
then plated to count cfu. Means +/- SEM of independent reactions (N=3) are given. For
each serum test group, significant mean differences from controls (no serum; white bars)
were determined by ANOVA. *, p < 0.01 for each Dunnett Multiple Comparison posttest. **, p < 0.05 for one-tailed Student’s t tests for significance between depleted and
replete sera. ND, not detected.
115
Second, in the absence of an active classical pathway (either by Ca++ chelation or by
depletion of C1q) we observed that complement-susceptible strains can activate a low
level of the alternative pathway to mediate bacterial killing.
The role of C1q in mediating C3 deposition on complement-resistant and
complement–susceptible strains
The above data demonstrate that that C1q is required for the optimal lysis of
LVSG and LVSR. We next determined the role of C1q in mediating C3 deposition on
both complement-resistant and complement-susceptible strains. We examined C3
deposition on Schu S4, LVS, Fn, LVSG, LVSR, and LVSΔwbtA by Western blotting
(Figs. 3.4 and 3.5). Bacteria were incubated in 10% C5d (positive control), C1qd, and
C1qr sera as sources of complement (Fig. 3.4). Since susceptible bacteria are lysed in
C1q-sufficient serum within 1 hour (Fig. 3.2), we tested survival at earlier time points
and found that significant lysis did not occur in 15 minute assays (data not shown). Thus,
15 minute serum incubations were performed in order to evaluate C3 deposition on
complement-susceptible strains in C1qr serum. The 120 kDa α-chain and 75 kDa βchain associated with native C3 are shown. C3 activation results in C3a release from the
α-chain leaving C3bα’, a 110 kDa fragment. Covalently bound C3bα’ fragments
associated with bacterial acceptors are present at high MWs but cropped from the blot
shown in Fig. 3.4. Although C3 binding occurs in C1qd serum, it does so at a level just
above the threshold for detection for some strains. These results clearly demonstrate that
C1q is required for optimal C3 deposition on each of the strains tested, including
human/mouse virulent strains and avirulent strains. Our cumulative data provide strong
116
Figure 3.4. Deposition of C3 on both wildtype and variant strains is C1q-dependent.
5x108 bacteria/reaction were incubated in 10% C5-depleted (C5d), C1q-depleted (C1qd),
or C1q-replete C1qd (C1qr) serum for 15 minutes at 37°C, washed in ice cold PBS
containing protease inhibitors, boiled in Laemmli’s buffer, separated by 7.5% SDSPAGE (5x107 bacteria/lane unless diluted) and examined for C3 by Western blotting.
Aliquots from each sample were plated to count cfu just prior to lysis to equalize loading
amounts. Goat antiserum to human C3 was used for detection. Control lanes contain 2ng
purified native C3. The blots shown are representative of at least three independent
experiments. C3α, 120 kDa chain. C3β, 75 kDa chain.
117
Figure 3.5. C3 fixation by LVSΔwbtA is predominantly C1q-dependent. 5x108
bacteria/reaction were incubated in 10% C5-depleted (C5d), C1q-depleted (C1qd), or
C1q-replete C1qd (C1qr) serum for 15 minutes at 37°C, and immunoblotting was
performed as in Fig. 3.4. Control lanes contain 2ng purified native C3. The blot is
representative of 3 independent experiments.
118
evidence that the classical pathway has a dominant role in mediating complement
activation by F. tularensis.
Discussion
The important role of the classical pathway in F. tularensis killing by normal
human serum was previously demonstrated by Sorokin et al. who reported lysis of
surface variant strains in serum (184). In their study, variant strains were described as
lacking in capsule production; however, evidence exists that these were in fact O-antigen
mutant strains. Consumption of complement activity, indicative of activation, in pooled
human serum was mediated by whole bacteria, isolated outer membrane, and purified
LPS. Significantly less complement consumption occurred in pre-absorbed serum, which
was depleted of specific bactericidal antibodies by incubating pooled sera with acetonedried bacteria at 4°C. Repletion of pre-absorbed serum with anti-whole cell or anti-LPS
antibodies restored complement consumption. With respect to LVSR, Sandstrom et al.
reported that depletion of both IgM and C4 in human serum dramatically decreased lysis
(183).
In the present study, we show that variant strains deficient in O-antigen
production are highly susceptible to lysis in serum in the presence of an intact classical
pathway. When function of the classical pathway is blocked in serum (either by
chelating Ca++ or depleting C1q), variant strain lysis is greatly reduced. Furthermore,
analysis of C3 deposition in C1q-depleted and in C1q-replete serum establishes the
importance of the classical pathway not only for the lysis of variant strains, but also for
complement activation by virulent strains.
119
Previously, we reported the identification of natural or pre-existing anti-F.
tularensis antibodies (using subsp. novicida) in human non-immune serum (110).
Commonly, such antibodies are produced in response to prior Gram-negative bacterial
infections because of shared epitopes among LPS O-antigen species. We do not expect,
however, that anti-O-antigen antibodies play a significant role in classical pathway
activation by F. tularensis since they would not bind to O-antigen mutant strains.
Purified LPS derived from variant strains most likely lacking O-antigen were shown by
Sorokin et al. to consume complement. Thus, we predict that natural antibodies target
separate components of LPS (i.e. core sugars or lipid A).
In the context of pneumonic tularemia, the finding that the classical pathway
predominantly mediates opsonization of wildtype F. tularensis is of particular interest to
us. As a constituent of airway surface fluid (ASF), the relative concentrations of
individual complement components differ compared to serum. Alveolar macrophages
(AMs) and alveolar Type II cells secrete certain components such as C1q locally (97,
282). In human brochoalveolar lavage (BAL) fluid, Watford et al. report no significant
alternative pathway activity largely because Factor B cannot be detected (285).
Components of the classical pathway are readily available, however, and have been
shown to mediate opsonization of Group B streptococci and Mycobacterium tuberculosis
in BAL fluid (283, 285). Therefore, it is likely that F. tularensis is readily opsonized in
ASF leading to enhanced phagocytosis by AMs in vivo. We are currently examining C3
fixation by F. tularensis in isolated human BAL fluid.
In Chapter 2, we demonstrate that the nature of C3-fragmentation differs on Oantigen-producing F. tularensis strains compared to O-antigen-deficient strains. As a
120
result, resistant strains are effectively opsonized upon activation of complement whereas
susceptible strains are lysed. Since LPS O-antigen is widely considered to be a strong
activator of the alternative complement pathway (323), we considered the possibility that
wildtype strains would preferentially activate the alternative pathway, but that Oantigen-mutant strains would not. Activation of complement by separate pathways could
account for deviations in C3-fragmentation. However, any definitive evidence of
alternative pathway activation by F. tularensis is demonstrated here for variant, but not
wildtype strains. Despite the possibility that alternative pathway-mediated C3b
amplification occurs on susceptible strains, we consider the evidence to be overwhelming
in regards to the importance of the classical pathway for both resistant and susceptible
strains. In Chapter 4, we examine alternative mechanisms to explain the negative impact
of O-antigen on the persistence of surface-bound C3b.
121
Chapter 4
Francisella tularensis LPS O-antigen restricts direct binding and activation of
complement component C1
Introduction
Francisella tularensis is an extremely infectious Category A select agent that
causes tularemia. This facultative intracellular Gram-negative bacterium primarily
infects macrophages and can spread via the reticuloendothelial system both within
circulating cells and extracellularly (53, 69). When bacilli are inhaled or spread to the
lung hematogenously, pneumonic tularemia develops that has an untreated mortality rate
of 30-60% (12). Antibiotic-resistant strains of F. tularensis have been weaponized and
represent a significant risk to public health in the current age of bioterrorism.
The ability of F. tularensis to cause bacteremia (75-77), coupled with recent
evidence that complement-mediated opsonization is necessary for efficient bacterial
phagocytosis by human macrophages (60, 109-111); signifies that F. tularensis must
interact with the complement system in a highly controlled fashion. The complement
system is composed of more than 30, mostly soluble, proteins in serum and extracellular
fluids (188). It is a multifunctional system that is critically important for optimal innate
immunity as well as normal tissue homeostasis. Activation of complement is necessary
122
to achieve opsonization, however, negative regulation of complement is also necessary
for F. tularensis to escape complement-mediated lysis.
We have shown in Chapter 2 that lipopolysaccharide (LPS) O-antigen production
by wildtype F. tularensis strains is a primary determinant for resistance to complementmediated lysis. The amount of O-antigen expression by variant strains was shown to
inversely correlate with the degree of sensitivity to complement-mediated lysis for each
strain. We found that C3 deposition, which is the central component of complement,
occurred on all of the F. tularensis subspecies and variant strains tested, but occurred in
greater amounts on sensitive O-antigen deficient strains. Furthermore, the ratio of C3b
(required for activation of the terminal lytic pathway as well as amplification of C3
deposition via the alternative pathway) to C3bi (required for efficient opsonization) was
greater on sensitive O-antigen deficient strains. The purpose of this chapter is to further
characterize the mechanism by which F. tularensis O-antigen negatively regulates
complement.
For other Gram-negative species, the production of LPS O-antigen confers
resistance to complement-mediated lysis by three major mechanisms. The first
mechanism involves potent alternative pathway activation by O-antigen itself (323). In
this case, activated complement components bind to O-antigen distal to the outer
membrane, which results in unstable MAC formation. Upon release from parent C5convertases, immature MAC complexes composed of C5-7 and C5-8 contain
hydrophobic domains that must intercalate within a lipid bilayer to remain surface-bound.
When MAC formation does not occur within proximity to the Gram-negative outer
membrane, immature complexes are shed from the surface. Subsequently, because
123
activating bacteria remain intact, unchecked complement activation ensues resulting in
the rapid consumption of native complement components and the overall complement
hemolytic activity of the system.
The second major mechanism of O-antigen-mediated complement resistance
involves the recruitment of regulators of complement activity (RCA) proteins, including
Factor H (FH) and C4 binding protein (C4bp). FH and C4bp contain domains that bind
directly to C3b and C4b, respectively, as well as domains that can interact with surface
components of bacteria such as sialylated glycolipids (230, 234). Both of these RCA
proteins have dual functionality by causing dissociation of C3-convertases and by
providing a binding platform for Factor I, a serine protease that cleaves C3b and C4b to
form C3bi and C4bi. Thus, bacterial recruitment of FH and C4bp can lead to an overall
decrease in C3b deposition and subsequent lysis by terminal pathway components. The
third major mechanism involves the inhibition, by either steric hindrance or competitive
binding, of molecular interactions between bacterial surface acceptor molecules and
sensory components of complement such as antibody (297, 298). In this chapter we
explored all three potential mechanisms to explain complement resistance and
susceptibility to complement and find that regulation is primarily upstream of component
C3 activation. We also demonstrate the binding of RCA proteins to F. tularensis, but
present evidence that binding is not affected by O-antigen.
Materials and Methods
Bacterial strains. F. tularensis subsp. tularensis strain Schu S4, a Centers for
Disease Control and Prevention clinical isolate, was provided by Rick Lyons (University
124
of New Mexico, Albuquerque, NM). A live vaccine strain derived from F. tularensis
subsp. holarctica (LVS; ATCC 29684) was provided by Karen Elkins (Center for
Biologics Research and Evaluation, US FDA, Bethesda, MD). LVSG was provided by
Fran Nano (University of Victoria, Victoria, BC, Canada). LVSG is a spontaneous grey
phase variant that rarely reverts to LVS when grown on chocolate II (chocII) agar (176).
The LPS O-antigen mutants, LVSΔwbtA and LVSΔwbtM, and the complemented mutant
strain (LVSΔwbtM:wbtM) provided by Dara Frank (Medical College of Wisconsin,
Milwaukee, WI), were created by modified Himar1 transposon (HimarFT)-mediated
mutagenesis of LVS and use of the complementation plasmid pFTNAT (345).
Experiments using Schu S4 were carried out within biosafety level 3 (BSL3) select agentcertified laboratories with adherence to federal and institutional select agent regulations.
Bacteria were grown overnight (approximately 18 hours) on chocolate II agar (Becton
Dickinson, Franklin Lakes, NJ) at 37°C.
Human sera, complement components, and reagents. Serum was isolated
from healthy adult volunteers with no known exposure to F. tularensis according to a
protocol approved by the Ohio State University Medical College Internal Review Board.
The sera were processed to maintain optimal complement activity (346). Briefly, isolated
non-heparinized whole blood was kept at room temperature for 1 hour to allow for clot
formation and then at 4°C to allow for clot contraction. The clot was removed by
centrifugation at 500xg for 15 minutes at 4°C. The serum fraction was collected, filter
sterilized, aliquoted, and stored at -80°C. Heat inactivation (HI) was performed at 56°C
for 30 minutes. C5-depleted (C5d) and C1q-depleted (C1qd) sera and purified C1, C1q,
125
C1-esterase inhibitor (C1inh), and C4 were purchased from Complement Technology, Inc
(San Antonio, TX) and stored at -80°C. On the day of use, fresh sera were thawed at
room temperature, then immediately chilled on ice until needed. 50% hydroxylamine
was purchased from Alfa Aesar (Ward Hill, MA). Other chemicals were purchased from
Sigma-Aldrich.
Complement hemolytic (CH50 ) assays. Bacteria were suspended in gelatin
veronal buffer (GVB++; 0.1% gelatin, 5.5 mM barbital, 142 mM NaCl, 0.5 mM MgCl2,
0.15 mM CaCl2; pH 7.3) at equalized concentrations by measuring the optical density at
600 nm. Various concentrations of bacteria were incubated in 10% fresh human nonimmune donor serum for 60 minutes at 37°C, and the reaction was stopped by placing
tubes on ice for 5 minutes. Bacteria were removed by centrifugation (12,000xg for 4
minutes), and supernatants were filtered using CoStar Spin-X Centrifuge Tube filters
(Corning, Corning, NY). Residual complement hemolytic activity was determined as
described with modifications (356). Briefly, supernatants were serially diluted in GVB++
and added to microtiter polypropylene round-bottom wells (Corning). Antibodysensitized sheep erythrocytes (Complement Technology) were added at a final
concentration of 2.5x107 cells/ml in 100µl. At least 8 serial dilutions for each supernatant
were tested. Plates were incubated for 30 minutes at 37°C. Hemolysis was stopped with
150µl ice-cold GVB++ and plates were centrifuged at 1000xg for 5 minutes at 4°C.
Supernatants containing different amounts of released hemoglobin were transferred to
fresh polystyrene flat-bottom microtiter plates (Corning) and absorbance at 412 nm was
measured on a 96-well plate reader (Molecular Devices, Sunnyvale, CA). Fractional
126
lysis (z) was calculated for each well by inducing 100% lysis (using H2O) in control
wells and then normalizing. Serial concentrations for each test group were then plotted
against (z/1-z) to determine the 50% hemolytic titrate (equal to 1 CH50 unit). The amount
of complement consumption in bacterial preparations was calculated by comparison to
CH50 units calculated for control serum incubated in GVB++ alone ([1-CH50 test / CH50
control] x 100%).
Determination of antibody in non-immune donor serum to F. tularensis. To
compare binding of pre-existing antibodies against Ft in the presence and absence of Oantigen, 3x108 LVS and LVSΔwbtM were incubated for 30 min at 37°C in fresh sera
from multiple donors, or C5d serum at different concentrations (1.25%, 2.5%, 5%, and
10%), or in GVB++ alone. After vigorous washing, bacteria were dried overnight onto
medium-binding polystyrene Costar 96-well plates (Corning), and wells were then
blocked using 3% ovalbumin in veronal buffer at 4°C overnight. Antibodies used for
detection were diluted into 0.3% ovalbumin and included HRP-conjugated goat antihuman IgG and goat anti-human IgM (1:10,000 and 1:1,000 respectively) (Sigma).
Antibody incubations were 3 hours at room temperature. HRP substrate (BioRad,
Hercules, CA) was added for 10 min at room temperature, and the reaction was stopped
with 2% oxalic acid. Absorbance at 415 nm was measured on a 96-well plate reader.
Bactericidal assays. Complement-mediated killing was carried out using C5d
serum (negative control), fresh serum obtained from a patient with X-linked
agammaglobulinemia (AG) that was provided by Dr. Marcus Horwitz (UCLA, Los
127
Angeles, CA), or AG replete with added physiological concentrations of purified IgG (30
mg/ml) or IgM (4 mg/ml) obtained from pooled human serum (Sigma). Bacteria (1x106)
were suspended in GVB++ at equalized concentrations by measuring the optical density at
600 nm and then incubated with 5% serum for 1 hour at 37°C with slow agitation. Tubes
were placed on ice for 5 minutes to stop the reactions. 10-fold serial dilutions were
plated to determine surviving colony forming units (cfu).
C1q, C3, and C4 deposition assays and Western blotting. Fresh donor, AG,
and C5d sera were used to evaluate complement component C3 deposition on Ft strains.
After pre-blocking microcentrifuge reaction tubes in PBS containing 0.1% human serum
albumin (HSA) (ZLB Plasma, Boca Raton, FL) for 30 minutes at 37°C, bacteria were
incubated in various concentrations of sera, or sera with added 10mM EDTA, for given
times at 37°C. Deposition was stopped by placing tubes on ice for 5 minutes, followed
by washing by sequential centrifugation (12,000xg for 4 minutes) and vigorous
resuspension in blocking buffer or PBS. To ensure equal lane loading, aliquots were
taken from the final PBS wash and plated to count cfu. To analyze fragmentation
patterns, hydroxylamine (NH2OH) was used to cleave thioester bonds formed between
C4 and bacterial acceptor molecules as described with modifications (348). Briefly, after
completing C4 deposition reactions and removal of unbound C4, samples were
solubilized by boiling in 1% SDS for 5 minutes. Control samples were prepared
immediately for Western blotting and paired samples were first incubated in 2M NH2OH
in 20mM Tris-buffered H2O, pH 10.5, for 1 hour at 37°C. After final re-suspension in
Laemmli’s sample buffer, all samples were boiled for 2 minutes and separated by minigel
128
7.5% (C3- and C4-derived fragments) or 15% (C1q) SDS-PAGE, followed by protein
transfer to PVDF membranes (Milipore, Billerica, MA). Membranes were blocked
overnight at 4°C in advanced ECL blocking buffer at 2% (w/v), which was also used for
antibody dilution (Amersham, Piscataway, NJ). Goat antiserum to human C3 (Quidel,
San Diego, CA) (diluted to 1:20,000), goat antiserum to human C4 (Complement
Technology) (diluted to 1:1,000), and goat antiserum to C1q (Complement Technology)
(diluted to 1:20,000) served as the primary antibodies for 1 hour incubations at room
temperature. HRP-conjugated rabbit anti-goat IgG (H+L) antibody (Biorad) (diluted to
1:20,000, 1:10,000, and 1:10,000, respectively) was used as the secondary antibody for 1
hour incubations at room temperature. Advanced ECL reagent was used for detection
(Amersham).
ELISA to detect deposition of C1 subcomponent, C4, Factor H and C4
binding protein on F. tularensis strains. C5d, HI C5d, and C1qd sera were used to
evaluate complement component deposition on LVS and LVSΔwbtM. After pre-blocking
microcentrifuge reaction tubes as described for Western blotting experiments above, 35x108 bacteria/reaction were incubated in given serum concentrations for 5 (C1
components) or 30 minutes at 37°C. Reactions were stopped and samples were washed
as described above for Western blotting experiments. To ensure equal well loading,
aliquots were taken from the final PBS wash and plated to count cfu. 3-8x107 bacteria in
suspension were added to medium-binding polystyrene wells in triplicate (Corning) and
left to dry overnight. Wells were blocked overnight at 4°C with 3% ovalbumin. After
extensive washing with PBS, primary antibodies (diluted in 0.3% ovalbumin) were added
129
for 2 hour incubations at room temperature, which included goat antisera to human C1q
(Complement Technology) (diluted 1:3,000), mouse polyclonal anti-human C1s
(Abnova, Taipei City, Taiwan) (diluted 1:1,000), goat antisera to human C4
(Complement Technology) (diluted 1:2,000), goat antisera to human FH (Complement
Technology) (diluted 1:2,000), and murine monoclonal anti-human C4bp (Quidel)
(diluted 1:2,000). HRP-conjugated rabbit anti-goat IgG (H+L) antibody (Biorad) was
used as a secondary antibody to detect primary anti-C1q, C4, and FH antibodies (diluted
to 1:5,000) and HRP-conjugated goat anti-mouse IgG (H+L) antibody (Biorad) was used
to detect primary anti-C1s and C4bp antibodies (diluted 1:1,000 and 1:2,000,
respectively). Secondary antibody incubations were 1 hour at room temperature. HRP
substrate (BioRad) was added for 10 min at room temperature, and the reaction was
stopped with 2% oxalic acid. Absorbance at 415 nm was measured on a 96-well plate
reader.
Results
F. tularensis LPS O-antigen production negatively influences the consumption of
complement hemolytic activity
We compared overall complement activation by LVS, which expresses full-length
LPS O-antigen, with activation by LVSΔwbtM, a mutant strain that does not produce Oantigen (Fig. 4.1). If O-antigen produced by F. tularensis causes overt complement
activation distal to the outer membrane, then LVS would be expected to consume a
distinctly higher percentage of serum hemolytic activity compared to LVSΔwbtM. For
these assays, different concentrations of bacteria were incubated in 10% fresh human
130
Figure 4.1. Consumption of complement hemolytic activity by LVS and
LVSΔwbtM, the latter an isogenic O-antigen mutant strain. Bacteria were grown
overnight on chocII agar, then suspended at the given concentrations in GVB++ with 10%
fresh non-immune donor serum for 30 minutes at 37°C. After removing bacteria by
centrifugation and filtration, supernatants were used to determine residual CH50 activity.
A series of eight dilutions for each test serum was added to individual wells containing
antibody-sensitized sheep erythrocytes in 96-well plates and incubated for 30 minutes at
37°C. H2O was added for 100% lysis controls. Hemolysis was terminated by adding ice
cold GVB++, and unlysed erythrocytes were pelleted (1000xg, 5 minutes, 4°C).
Supernatants were transferred to fresh 96-well plates to measure absorbance at 412 nm.
Fractional lysis (z = Abstest / Abs100%) was calculated for each well. Serial concentrations
for each test group were then plotted against (z/1-z) to determine the 50% hemolytic
titrate (equal to 1 CH50 unit). Remaining CH50 activity was compared to that of serum
131
incubated with GVB++ alone to determine the percentage of complement activity
consumed in the presence of bacteria ([1 - CH50 units in test serum/CH50 units in control
serum] x 100%). Bars are values derived from single consumption reactions for each
condition tested and are representative of 2 independent experiments with sera from
separate donors.
132
non-immune donor serum for 1 hour at 37°C. Bacteria were subsequently removed by
centrifugation followed by filtration of the supernatant. CH50 levels (50% hemolytic
titration) were determined for each supernatant and compared with the CH50 of control
serum. At 5x1010 cfu/ml, both strains consumed nearly 100% of the hemolytic activity
associated with control serum. At 1x1010 cfu/ml, however, LVSΔwbtM consumed nearly
100% of the hemolytic activity in serum and consumption by LVS was reduced to about
63%. At lower bacterial concentrations, consumption by LVSΔwbtM was reduced and
remained comparable to consumption by LVS. These data are consistent with our
proposal that F. tularensis LPS O-antigen is an inhibitory determinant for complement
activation rather than a mediator of uncontrolled complement activation distal to the outer
membrane and primarily affects interactions with the complement cascade upstream of
the terminal lytic pathway.
Activation of the classical pathway by F. tularensis occurs independently of
antibody
We previously reported the occurrence of natural or pre-existing antibodies in
human non-immune serum against F. novicida (110). Since we have also shown that all
of the F. tularensis subspecies and variant strains tested to date predominantly activate
the classical pathway, we predicted that increased antibody binding would occur on Oantigen deficient strains. We tested this by ELISA in order to quantify IgG and IgM
binding to LVS and LVSΔwbtM in various concentrations of either donor non-immune or
C5-depleted (C5d) serum (Fig. 4.2). Data show that the saturation of immunoglobulin
(both IgG and IgM) binding to LVSΔwbtM occurs in lower concentrations of serum
133
Figure 4.2. Immunoglobulin binding to LVS and LVSΔwbtM in human nonimmune serum. Bacteria (3x108 cfu/ml) were incubated with given concentrations of
fresh serum for 30 minutes at 37°C, washed, and resuspended in H2O. 4x107
bacteria/well were applied to 96-well plates and dried overnight. HRP-conjugated goat
anti-human IgG and goat anti-human IgM antibodies were used for detection. After the
addition of HRP substrate, absorbance was read at 415nm. Data were normalized to
control wells containing no bacteria. Bars represent means +/- SD (triplicates) for one of
three independent experiments using sera from two separate donors and commercially
available C5-depleted serum.
134
compared to LVS. For LVSΔwbtM, saturated binding occurs in 5% serum or less. For
LVS, however, binding of both IgG and IgM increases from 5% to 10% serum resulting
in a greater amount of bound antibody compared to LVSΔwbtM in 10% serum. We
concluded that since LVS has a greater binding capacity for potential complement
activating antibodies, yet activates less complement than LVSΔwbtM, that activation of
the classical pathway by F. tularensis might occur independently of antibody.
In order to test complement activation in the absence of antibody, we used serum
obtained from a patient with X-linked agammaglobulinemia, a genetic disease
characterized by an inability to produce immunoglobulins. First, we determined whether
complement mediated lysis of LVSΔwbtM occurs in agammaglobulinemic (AG) serum
(Fig. 4.3.A). Bacteria were incubated in 5% C5d (negative control) or AG serum for 1
hour at 37°C, then plated to determine surviving cfu. Data show that complementmediated lysis of LVSΔwbtM in AG serum did occur resulting in a nearly 100-fold
decrease in survival compared to controls. Furthermore, reconstitution of AG serum with
physiological concentrations of IgG or IgM, isolated from pooled human serum, did not
appreciably enhance bacterial lysis. In a preliminary assay, we also determined whether
C3 deposition on LVS occurs independently of antibody by Western blotting (Fig. 4.3.B).
Bacteria were incubated in 5% C5d or AG serum for 30 minutes at 37°C, followed by
extensive washing to remove unbound C3 prior to sample preparation. The amount of C3
binding and the degree of fragmentation appeared equivalent. Together with data
reported in Chapter 3 showing the importance of C1q, these data provide evidence that
complement activation by both susceptible and resistant strains of F. tularensis results
from antibody-independent activation of the classical pathway.
135
Figure 4.3. Complement activation by both susceptible and resistant strains of F.
tularensis occurs independently of antibody. (A) LVSΔwbtM (1x106) was incubated in
5% C5-depleted serum (C5d; deficient in lytic activity), serum obtained from a patient
with X-linked agammaglobulinemia (AG; deficient in immunoglobulins), or AG serum
replete with physiologically relevant amounts of purified human IgG (1.5 mg/ml) or IgM
(0.2 mg/ml) for 1 hour at 37°C, washed in ice cold PBS, and plated to count surviving
colony forming units (cfu). Bars are means +/- SD (triplicates) for one of two
independent experiments. (B) As a preliminary experiment, LVS (5x108) was incubated
136
in 5% C5d or AG serum for 1h at 37°C, washed, resuspended in Laemmli’s buffer,
separated by 7.5% SDS-PAGE (5x107 bacteria/lane) and examined for C3 by Western
blotting. Aliquots from each sample were plated to count cfu just prior to lysis to
equalize loading amounts. Goat antiserum to human C3 was used for detection.
137
O-antigen mediated regulation of complement occurs upstream of C3 deposition
We previously analyzed the kinetics of C3 deposition and fragmentation on
several strains of F. tularensis and found that an increased rate of C3b conversion to C3bi
occurred in the presence of O-antigen. Factor I cleaves C3b to form C3bi, but C3b must
first be recognized and bound by a cofactor within the RCA protein family. Since Factor
H (FH) is a soluble RCA protein that facilitates Factor I activity and is commonly
recruited by Gram-negative bacteria to inhibit complement, we hypothesized that Oantigen producing F. tularensis strains would bind greater amounts of FH compared to
mutant strains.
To test this, we incubated LVS and LVSΔwbtM in 10% fresh C5d serum, washed
bacteria to remove unbound FH, and quantified FH binding by ELISA (Fig. 4.4).
Unexpectedly, we found that markedly higher amounts of FH bound to LVSΔwbtM
compared to LVS. C4 binding protein (C4bp) is another RCA family member that is a
cofactor for Factor I and facilitates conversion of C4b to C4bi. An increased rate of C4b
inactivation on O-antigen producing bacilli could also explain decreased C3 deposition.
However, using the same experimental protocol that was used to test for FH binding, we
found that C4bp also preferentially associates with LVSΔwbtM compared to LVS. FH
and C4bp are composed of several distinct complement control protein (CCP) domains
that can recognize a variety of surface components and separate domains that recognize
binding sites on C3b and C4b, respectively. Thus, we reasoned that increased binding to
LVSΔwbtM could be explained by higher amounts of surface C3b and C4b. In order to
measure binding in the absence of C3b/C4b, we used 10% heat inactivated (HI) C5d and
10% C1q-depleted (C1qd) sera as sources of RCA proteins. C3b/C4b fixation does not
138
Figure 4.4. LVSΔwbtM binds greater amounts of Factor H (FH) and C4 binding
protein (C4bp) compared with LVS. For 30 minutes at 37°C, 3x108 bacteria/reaction
were incubated in 10% fresh C5-depleted (C5d), HI C5d, or C1q-depleted (C1qd) serum,
then washed, and resuspended in H2O. Aliquots from each sample were plated to count
cfu in order to equate input of bacteria. 3x107 bacteria/well were applied to 96-well
plates and dried overnight. Goat antisera to human FH and mouse polyclonal anti-human
C4bp were used for detection. Absorbance at 415 nm was measured and values were
normalized to controls containing no bacteria. Means +/- SD are given (triplicates) for a
representative experiment (n=2).
139
efficiently occur without C1q, which is irreversibly denatured by mild heat (see Chapter 3
and data not shown). In the absence of surface-bound C3b, data show nearly equivalent
binding of FH between LVSΔwbtM and LVS (Fig. 4.4). Also, in the absence of surfacebound C4b, C4bp binding to both bacteria is abolished. These results indicate that F.
tularensis recruits soluble RCA proteins in serum, but that the presence of O-antigen does
not have a major impact on bacterial surface recognition in the absence of bound C3b and
C4b.
Despite the finding that greater amounts of FH bind to LVSΔwbtM compared to
LVS in fresh serum, a markedly higher ratio of C3b to C3bi is associated with O-antigen
deficient strains (see Chapter 2). Thus, O-antigen-deficient bacilli must fix C3b at a
greater rate compared to wildtype strains. We examined C4b deposition as an indicator
of the magnitude of C3-convertase formation via the classical pathway, which in turn,
influences the rate and amount of C3b deposition. LVS and LVSΔwbtM were each
incubated in 10% C5d serum for 30 minutes at 37°C, washed, and used in ELISA (Fig.
4.5.A, left half). Data show greater amounts of C4b deposition on LVSΔwbtM compared
to LVS. These results were not surprising since C4bp binds to fixed C4b, which was
found in greater amounts on the surface of LVSΔwbtM. We also examined C4 deposition
in the absence of antibody by incubating bacteria in buffer containing purified human C4
and physiologically equivalent concentrations of purified human C1 and C1-esterase
inhibitor (C1inh) (Fig. 4.5.A, right half). C1inh has been previously shown to block the
esterase activity of C1 when directly bound to a rough Escherichia strain (216). Our
results indicate that, even in the presence of C1inh, F. tularensis activates C1 leading to
C4b deposition, and similar to the results seen in 10% C5d serum, activation is greater for
140
Figure 4.5. C4 activation and deposition on F. tularensis occurs in an antibodyindependent manner. (A) LVS and LVSΔwbtM (5x108 bacteria/reaction) were
incubated in 10% C5-depleted serum or in buffer containing purified C1, C1-esterase
inhibitor (C1inh), and C4 for 30 minutes at 37°C. Bacteria were washed and resuspended
in H2O. Aliquots from each sample were plated to count cfu to equate input of bacteria.
3x107 bacteria/well were applied to 96-well plates and dried overnight. Goat antisera to
human C4 was used for detection and absorbance at 415 nm was measured. Values were
normalized to controls containing no bacteria. Means +/- SD are given (triplicates) for a
representative experiment (n=2). (B) LVS and LVSΔwbtM (5x108 bacteria/reaction)
were incubated in 5% agammaglobulinemic serum (AG) for 60 minutes at 37°C, washed,
resuspended in Laemmli’s buffer, separated by 7.5% SDS-PAGE (7x107 bacteria/lane)
and examined for C4 by Western blotting. Aliquots from each sample were plated to
141
count cfu just prior to lysis to equalize loading amounts. Goat antiserum to human C4
was used for detection.
142
LVSΔwbtM. In a preliminary experiment, LVS and LVSΔwbtM were also incubated in
5% AG serum for 1 hour at 37°C followed by removal of covalently acceptor-bound C4derived fragments using NH2OH (Fig.4.5.B). Samples were examined by Western
blotting and the results demonstrate greater C4b binding to LVSΔwbtM compared to
LVS. Also, some degree of C4b inactivation is indicated by the appearance of bands
corresponding to C4bi on each strain, further evidence for C4bp recruitment. Together,
these results provide strong evidence that F. tularensis can bind and activate the C1
complex directly, independent of antibody, leading to C4b inactivation as well as to
functional C3 convertase (C4b2a) formation.
F. tularensis associated O-antigen limits binding of C1 in serum
We used several F. tularensis wildtype and variant strains to analyze the effect of
O-antigen production on the binding of C1. Bacteria were incubated in 20% C5d serum
for 10’ at 37°C, washed, and prepared for Western blotting to detect bound C1q. In
serum, C1q binding was greatest on LVSG, which expresses less O-antigen compared
with LVS (Fig. 4.6.A). Interestingly, a greater amount of C1q bound to Schu S4, an F.
tularensis subsp. tularensis clinical isolate that is highly virulent in humans compared to
LVS but that was shown (see Chapter 2) to fix equivalent amounts of C3 compared to
LVS. EDTA dissociates C1r and C1s from C1q because interactions are Ca++-dependent.
In the presence of EDTA, C1q binding was enhanced on LVS and Schu S4 and was
equivalent to LVSG. To confirm that increased binding to LVSG is due to differences in
O-antigen expression, we also compared C1q binding between LVS and LVSΔwbtA (an
O-antigen mutant strain) (Fig. 4.6.B) and between LVSΔwbtM and LVSΔwbtM:wbtM
143
Figure 4.6. C1q binding to various strains of F. tularensis is affected by Oantigen expression and by uncharacterized components of serum. Bacteria (5.5x109)
were incubated in 20% C5-depleted (C5d) serum with or without EDTA for 10 minutes
(A-C) or in 2.5µg/ml or 0.5µg/ml C1q (in GVB++) for 1 minutes (D) at 37°C, washed,
boiled in Laemmli’s sample buffer, separated by 15% SDS-PAGE (8x108 bacteria/well)
144
and examined for C1q binding by Western blotting. Aliquots from each sample were
plated to count cfu just prior to lysis to equalize loading amounts and goat antiserum to
human C1q was used for detection. Control lanes containing purified C1q (a heteromer
composed of 3 distinct polypeptides) are shown. Blots shown are representative of at
least two independent experiments.
145
(the complemented LVSΔwbtM strain that expresses plasmid-encoded wbtM) (Fig.
4.6.C). In each experiment, C1q binding was enhanced on O-antigen deficient strains.
To further analyze C1q binding to F. tularensis in the absence of antibody, we
replaced serum with GVB++ containing purified human C1q in our Western blotting assay
(Fig. 4.6.D). The amount of C1q binding to LVS and LVSΔwbtA in buffer containing
2.5µg/ml and 0.5µg/ml of purified protein was equivalent. Altogether, these data suggest
that C1q binds to both O-antigen-producing and O-antigen-deficient strains of F.
tularensis. However, when associated with other components of C1 (including C1r, C1s,
and C1inh), or in the presence of unidentified components of serum that utilize Ca++, C1q
binding to F. tularensis is limited by the presence of O-antigen. Experiments were done
to determine if C1q binding to LVS and LVSΔwbtM differs in comparison with C1s, the
serine protease associated with C1 that activates native C4 (Fig. 4.7). Bacteria were
incubated in various concentrations of serum for 5 minutes at 37°C, washed, and
transferred to 96-well plates in order to quantify binding by ELISA. Data indicate that
binding of both C1q and C1s to LVSΔwbtM approaches saturation in lower serum
concentrations compared to LVS. Altogether, these data suggest that the magnitude of
C1q binding to F. tularensis is reduced in the presence of O-antigen only when C1q is
associated with intact C1 complexes.
Discussion
In previous chapters, we show evidence that virulent wildtype F. tularensis
subspecies and strains express LPS O-antigen, and this is a key determinant for resistance
to complement-mediated lysis. In this chapter, we explore potential mechanisms to
146
Figure 4.7. C1q and C1s bind in greater amounts to LVSΔwbtM compared to LVS
in serum. For 5 minutes at 37°C, 3x108 bacteria/reaction were incubated in increasing
concentrations of fresh C5-depleted serum, washed, and resuspended in H2O. Aliquots
from each sample were plated to count cfu to equate input of bacteria. 3x107
bacteria/well were applied to 96-well plates and dried overnight. Goat antisera to human
C1q and mouse polyclonal anti-human C1s were used for detection. Absorbance at 415
147
nm was measured and values were normalized to controls containing no bacteria. Means
+/- SD are given (triplicates) for a representative experiment (n=3).
148
account for the negative regulation of complement by F. tularensis O-antigen. These
preliminary studies suggest that O-antigen primarily affects activation of the classical
pathway by inhibiting the direct binding of C1 to unknown surface components. As a
result of decreased C1 binding and activation in the presence of O-antigen, the rates of
activation and deposition of downstream complement components that include C4- and
C3-derived fragments are also decreased in the presence of O-antigen. Furthermore, both
O-antigen-producing and O-antigen-deficient strains bind RCA proteins that potentially
further limit the deposition of components of the terminal lytic pathway. However, in the
absence of O-antigen, we propose that activation of the classical pathway causes rates of
C4b and C3b deposition that surpass the rates of C4b and C3b inactivation mediated by
RCA protein recruitment.
As described above, three major mechanisms commonly account for O-antigen
mediated resistance to complement: complement consumption distal to the bacterial outer
membrane, recruitment of RCA proteins, and complement sensory component inhibition.
In order to determine whether these mechanisms are also important for F. tularensis, we
largely used LVS as the model smooth complement-resistant strain because it produces
full-length tularensis-type LPS O-antigen. We used LVSΔwbtA and LVSΔwbtM, LVSderived rough mutants that do not produce O-antigen to model complement-susceptible
strains.
We first examined the amount of hemolytic consumption by smooth and rough F.
tularensis strains. Contrary to studies using other Gram-negative bacteria, our results
show that LVSΔwbtM consumes equivalent or greater levels of serum hemolytic activity
compared to LVS. Although not directly tested, these data support the findings shown in
149
Chapter 3 and suggest that F. tularensis O-antigen represents a less effective activator of
the alternative pathway compared to O-antigen derived from other species. By
comparison, Joiner et al. reported that incubation of approximately 5x108 cfu/ml of a
rough, complement-susceptible, Salmonella minnesota strain in 10% serum caused a 26%
depletion of C9 activity; whereas, equivalent concentrations of an isogenic smooth,
complement-resistant, strain caused nearly 95% depletion (329). Similarly, Schiller et al.
compared clinical isolates of Pseudomonas aeruginosa that either were resistant to
complement mediated lysis and expressed full-length O-antigen or were susceptible and
expressed truncated or no O-antigen (331). They showed that less than 5x108 cfu/ml of
many of the smooth isolates were required to completely consume hemolytic activity
(CH50) in 50% non-immune serum, but that equivalent concentrations of rough isolates
consumed ≤ 50%. Data presented here are consistent with results shown in Chapters 2
and 3 demonstrating that less C3b is fixed by wildtype strains compared to rough strains
and that C3b deposition on LVS is not mediated by the alternative pathway in 10% C1qdepleted serum.
We predicted that the major negative regulatory influence of O-antigen was either
on the stability of C3b upon binding to the bacterial surface or the complement cascade
upstream of C3 activation. There are many examples of Gram-negative bacteria that
recruit RCA proteins in serum in order to decrease further C3 activation (see Chapter 1
for references). We found that both FH and C4bp preferentially bound LVSΔwbtM
compared to LVS, but that differences in binding did not occur in the absence of surface
C3b and C4b deposition, respectively (Fig. 4.4). This signifies that binding is affected to
a greater extent by C3b/C4b binding domains (CCP1-CCP4 of FH and CCP1-CCP3 of
150
C4bp) than by surface binding domains (CCP5-CCP20 of FH, CCP1-CCP7 of C4bp αchains, or the β-chain of C4bp). FH binding to both strains is consistent with results
shown in Chapter 2 demonstrating that C3b inactivation occurs on both smooth and
rough strains. These data provide stronger evidence that a more important role for Oantigen is in inhibiting the classical pathway upstream of C3 activation, however, we
cannot dismiss the possibility that specific surface binding sites for RCA proteins differ
between smooth and rough strains and that binding to smooth strains results in more
effective recruitment of Factor I.
Next, we focused on potential differences in bacterial surface recognition between
smooth and rough strains by sensory components of the classical pathway. In both nonimmune donor serum and C5-depleted (C5d) serum, which is commercially prepared
from pooled donor serum, we determined that neither IgG nor IgM bind in greater
amounts to LVSΔwbtM compared to LVS (Fig. 4.2); thus, we considered the possibility
that the classical pathway is activated by F. tularensis in an antibody-independent
manner. Using agammaglobulinemic serum, we demonstrate that complement activation
by both smooth and rough F. tularensis strains remains intact in the absence of antibody
(fig 4.3). Furthermore, the addition of purified immunoglobulins to
agammaglobulinemic serum does not dramatically enhance the killing of LVSΔwbtM.
Finally, C4 activation and the deposition of C4b are greater on LVSΔwbtM compared to
LVS even in the absence of antibody (Fig. 4.5). Thus, taken together, these data provide
strong evidence that O-antigen expression dictates the degree of classical pathway
activation by F. tularensis, which occurs in an antibody-independent manner.
C1q typically recognizes and binds to the Fc portions of surface-bound IgM or
151
clusters of IgG via its globular head domain. This causes a conformational shift within
the C1 complex that is necessary for the activation of C1-associated esterases and for
downstream classical pathway component activation. The direct binding of C1q to
bacteria via its globular head domain has previously been described for other Gramnegative species (201, 203-205). It is imperative, however, that binding occurs in such a
fashion that esterases become activated, which does not always occur when bacteria
directly bind C1 (202). Furthermore, C1inh is associated with C1 in serum and has been
shown to inhibit the direct activation of C1 by bacteria (216). Nonetheless, potential
bacterial acceptor molecules recognized by C1q globular heads include LPS-associated
core sugars and lipid A as well as negatively charged outer membrane proteins (206-211).
We determined whether direct binding of C1 by F. tularensis results in esterase
activation using purified components and found that, even in the presence of C1inh, C1
binding resulted in C4 activation and C4b deposition (Fig. 4.5). In Fig. 4.5, it appears
that there is little binding of C4b to the wildtype LVS. However, in a similarly designed
preliminary experiment using a low-ionic strength washing buffer for the ELISA, C4b
binding to wildtype LVS was more apparent (although still less than C4b binding to
LVSΔwbtM) indicating that the binding and activation of purified C1 also occur on
wildtype strains (data not shown). We found that greater amounts of C1q in serum bind
to O-antigen deficient strains (LVSG, LVSΔwbtA, and LVSΔwbtM) compared to
wildtype strains (LVS and Schu S4) and the complemented mutant strain
(LVSΔwbtM:wbtM) (Fig. 4.6). We also quantified C1 deposition in various
concentrations of serum by ELISA and found that higher serum concentrations were
necessary to achieve equivalent C1q and C1s binding for smooth strains compared to
152
rough strains (fig 4.7). Based on the similarity between the C1q and C1s binding curves
for both strains, we conclude that in serum, C1 binds to F. tularensis as a complex.
Interestingly, when EDTA was added to serum, which results in dissociation of
C1 into its subcomponents, C1q binding to LVS and to Schu S4 became equivalent to
LVSG (Fig. 4.6.A). Also, equivalent levels of purified C1q bind to LVS and to
LVSΔwbtA (Fig. 4.6.D). These data suggest that the number C1q acceptor molecules is
equivalent when comparing rough and smooth strains. On the other hand, we speculate
that, compared to C1q alone, C1-associated C1q may have a lower affinity for bacterial
acceptor molecules in the presence of O-antigen. Another possibility is that distinct
acceptor molecules exist for smooth and rough strains that are differentially recognized
by C1, but not by C1q. Perhaps most provocative is the preliminary finding that greater
amounts of C1q bind to Schu S4 compared to LVS, particularly since both strains fix
equivalent amounts of C3-derived fragments (Fig. 4.6.A and Chapter 2). Further testing
is needed to determine if increased Schu S4 recognition by C1q has functional
significance for the deposition of downstream components, and if so, whether surface
regulation of downstream complement activity differs between LVS and Schu S4.
It will be important to identify bacterial surface components that are recognized
by C1. Planned experiments include ELISA-based assays using both whole serum and
purified C1 to compare binding to purified LPS derived from rough and smooth strains.
If increased binding to rough strain-derived LPS occurs, then C1 may bind to exposed
core sugar or lipid A. Alternatively, outer membrane preparations from each strain that
are treated or untreated with Proteinase K can be used to determine if C1 binds to outer
membrane proteins. More experimentation is also required to definitively determine if
153
C1 binding to rough and smooth strains differs in antibody-deficient serum because, to
date, we have measured C1q and C1s binding only in fresh serum. We plan to remove
IgG and IgM from serum by adsorption with agarose beads attached to anti-human
immunoglobulin antibodies as described previously (357). Antibody-depleted serum can
then be used to compare C1-component binding between strains. Finally, we plan to test
whether purified C1 binds equally to rough and smooth strains in the presence and
absence of C1inh.
To summarize, we provide further evidence that the major mechanism of
complement activation by F. tularensis involves the classical pathway and that activation
occurs independently of antibody. These studies show that C1q binds directly to LVS,
which produces full-length LPS O-antigen, and to LVSΔwbtM, an O-antigen mutant
strain. The direct binding of purified C1 results in its functional activation and in the
subsequent deposition of C4b, which is greater for LVSΔwbtM. Furthermore, as a means
for negatively controlling complement-mediated lysis, both smooth and rough F.
tularensis strains bind RCA proteins in serum. However, FH and C4bp binding occurs in
greater amounts on rough strains and binding appears to largely be dependent on surfacebound C3b and C4b, respectively, rather than on the presence of O-antigen. We conclude
that O-antigen dictates complement activation predominantly by limiting the direct
binding of C1 and, thereby, slowing the rate of downstream C3b deposition; thus,
allowing for relatively more efficient conversion of bound C3b to C3bi.
154
Chapter 5: Synthesis
Francisella tularensis has been called the deadliest human pathogen. Inhalation
of as few as 10 bacilli can result in a highly fulminant disease characterized by the
systemic dissemination of bacteria, multiorgan failure, and death within days of the initial
infection. In this age of bioterrorism, the potential use of weaponized antibiotic-resistant
strains of F. tularensis represents a grave risk to public safety (12). However, little is
known regarding the molecular mechanisms employed by the bacterium to successfully
evade key elements of the innate immune system. To date, F. tularensis has not been
shown to produce exotoxins or other classic virulence factors associated with other
pathogens (54).
The course of pneumonic disease can be divided into separate phases. For three
to five days post-infection, both humans and experimentally infected animals experience
an asymptomatic period characterized by a suppressed immune cytokine response. This
period is followed by the rapid onset of systemic hyperinflammation, which is the major
cause of morbidity. Hepatic and splenic failure, high fever, and sepsis are salient features
of the late hyperinflammatory stage of tularemia (18, 76, 77, 84).
F. tularensis is a fast growing, facultative intracellular organism and has the
capacity to infect multiple cell types that include leukocytes and parenchymal cells.
Virulence is most likely associated with the ability of F. tularensis to replicate rapidly in
155
vivo while evading potentially microbicidal elements of the immune system. An
intracellular lifestyle is a survival mechanism employed by many pathogens as a means
to escape immune detection. In the lung, alveolar macrophages represent the principal
target for inhaled F. tularensis (66), and mutant strains incapable of intra-macrophage
survival are attenuated (80). These cells act as a reservoir for bacterial population growth
and also serve as a vehicle for systemic dissemination via the reticuloendothelial system
(68). Since gross pathology of the lung is not a major feature of pneumonic disease, and
since bacteremia becomes apparent only after distal organs become infected, the severity
of pneumonic tularemia is fundamentally dependent upon the carriage of infected
macrophages, or potentially, dendritic cells away from the lung. Development of an early
immunosuppressive environment may also depend upon macrophage infection. In vivo,
macrophages do not respond to infection by producing an appropriate proinflammatory
cytokine array, but instead produce TGFβ, an anti-inflammatory cytokine (70). In vitro,
F. tularensis also inhibits cellular responses to proinflammatory stimuli; including, INFγ
and TLR agonists; by an unknown mechanism (104-106).
Several laboratories, including our own, have shown that efficient phagocytosis of
F. tularensis by human macrophages requires complement-mediated opsonization (109111). The complement cascade represents the major arm of non-immune humoral innate
immunity. Successful pathogens must employ strategies to counteract its microbicidal
effects. We chose to study interactions between F. tularensis and components of
complement to better understand its role in the pathogenesis of tularemia. The primary
goals of the research presented in this thesis were to characterize the nature of
complement activation by F. tularensis and to identify influential bacterial surface
156
components that could potentially be targeted therapeutically. Disruption of such surface
components could alter the detrimental cycle of macrophage infection during the course
of tularemia and could also potentially render bacteria susceptible to lysis when
bacteremia arises.
Our hypothesis was that complement activation by F. tularensis subspecies leads
to opsonization but not to lysis of virulent strains, and that negative regulation of
complement would be mediated by major surface glycans such as lipopolysaccharide
(LPS) and capsule. To begin, we studied several F. tularensis subspecies and variant
strains. Virulent strains included Type A subsp. tularensis (Schu S4), subsp. novicida,
and the live vaccine strain (LVS) that was derived from subsp. holarctica. Variant strains
included a grey phase variant of LVS (LVSG) and a putative capsule-negative strain
derived from LVS by random mutagenesis (LVSR). Initial studies revealed that each of
the virulent strains was completely resistant to complement-mediated lysis in human nonimmune serum, but that lysis of variant strains occurred. However, we showed that the
virulent strains are also capable of activating complement by examining C3 deposition in
serum. Hydrolytic removal of covalently bound C3bα’ allowed us to more closely
examine the nature of the C3 fragments bound on each strain. In this way, we showed
definitively that C3bi, the major complement-associated opsonin, is fixed by virulent
strains in serum. The nature of C3 deposition differed, however, on variant strains.
Compared to virulent strains, the total amount of C3-derived fragment binding to LVSG
and LVSR increased dramatically, and there was a higher percentage of C3b binding
compared to C3bi. Downstream components of the terminal lytic pathway, including C5
and C7, were also shown to bind to variant strains in much greater amounts.
157
We selected LVSR for our studies because capsule-negative strains of F.
tularensis were previously reported to be susceptible to complement-mediated lysis
compared to parent strains (183, 184). In the course of replicating these data in our
studies, we performed a microscopic analysis to compare cell wall morphology
associated with wildtype and variant strains. Encapsulation of F. tularensis is reported to
appear as an amorphous electron-lucent substance that is loosely attached to the outer
membrane, however, this has only been shown for bacteria grown in defined media (157,
158). We did not identify capsular material even on virulent strains, possibly because we
cultured bacteria on rich media. This led us to suspect alternative differences in surface
composition between LVSR and virulent strains that were important determinants for
complement resistance.
Recent reports that LVS- and subsp. novicida-derived LPS O-antigen mutant
strains are killed in serum (179-182) and that O-antigen-deficient LVS grey variants
(unrelated to LVSG) are killed in serum (177) prompted us to examine O-antigen
production by LVSG and LVSR. Compared to LVS, we found that LVSG produces fulllength O-antigen, but in decreased amounts. This finding may indicate population
heterogeneity for LVSG or simply a reduction of O-antigen surface expression per
bacterium. O-antigen production by LVSR was undetectable. We confirmed that Oantigen deficiency accounts for complement susceptibility and enhanced C3b deposition
by testing LVS O-antigen mutant strains (LVSΔwbtA and LVSΔwbtM). Furthermore,
complementation of the LVSΔwbtM mutant with plasmid-encoded wbtM restored
complement resistance and reduced C3b persistence. The greatest amount of total C3
deposition occurred on LVSR and mutant strains. Also, the temporal nature of C3158
fragmentation was directly comparable between LVSR and the mutant strains. Our
studies show that LVSG expresses an intermediate amount of O-antigen and is less
susceptible to complement-mediated lysis compared to LVSR. The amount of total C3
deposition on LVSG was greater compared to virulent strains, but less compared to
LVSR and LVSΔwbtA. Furthermore, terminal pathway component deposition was
delayed on LVSG compared to LVSR, such that significantly greater levels of deposition
occurred only at later time points compared to virulent strains. Taken together, these data
provide convincing evidence that O-antigen is the major bacterial surface component that
dictates complement resistance for wildtype F. tularensis.
The capacity for phase variation has been described for virulent Type A F.
tularensis strains and for LVS. The bacterial molecular mechanisms that cause shifts in
frequencies of variant growth are unknown, but phenotypic differences between variant
types have been described. So-called “grey” variants, spontaneously derived from
wildtype cultures, replicate slowly, typically form small colonies on agar, and express
varying amounts of O-antigen (174, 175). When used to challenge naive animals in
experimental models, grey variants induce an attenuated immune response and are less
virulent compared to “blue”, or wildtype, variants. Conditions that result in increased
rates of phenotypic conversion between grey and blue variants include prolonged
stationary phase culture, growth in nutrient depleted media, and passage in animals or
cell cultures. It is unknown if phase variation plays a significant role during the course of
tularemia. At late stages of disease, when fulminant bacterial growth occurs in multiple
organs and bacterial nutrients may become limiting, it is likely that high rates of phase
variation occur in vivo. Based on our studies showing an increased capacity for LVSG to
159
activate complement and produce anaphylatoxins, conversion of large populations of F.
tularensis towards a grey phenotype may contribute to a hyperinflammatory
environment.
Our examination of the mechanisms of bacterial complement activation and
regulation demonstrates commonalities in the binding of both positive and negative
complement effector components between smooth and rough F. tularensis strains. We
show that an intact classical pathway is a requirement for the efficient lysis of susceptible
rough strains and for efficient C3 fixation by both rough and smooth strains.
Furthermore, we demonstrate that C3 fixation by smooth F. tularensis and lysis of Oantigen mutant strains are equally efficient in agammaglobulinemic serum that is
deficient in antibodies. In the absence of antibodies, the direct binding of C1 to microbial
surface components can activate the classical pathway, which has been demonstrated for
several rough Gram-negative species (203-205). However, in the presence of C1-esterase
inhibitor (C1inh), which is bound to C1 in serum and which negatively regulates its
function, the direct binding of C1 was shown not to result in further activation of
downstream classical pathway components (216). To our knowledge, the studies
presented here are the first to show bacterial binding and activation of C1 in the presence
of C1inh.
We also present evidence that Factor H (FH) and C4 binding protein (C4bp),
negative effectors of complement associated with the RCA family of proteins (226, 234),
bind to both smooth and rough F. tularensis strains. FH and C4bp catalyze cleavage of
C3b and C4b, respectively, by acting as cofactors for Factor I (FI), a serine protease in
serum. FH is composed of 20 complement control protein (CCP) domains, some of
160
which recognize cell surface components such as sialylated moieties on eukaryotic cells
and some of which bind directly to C3b. C3b binding can result in C3bBb C3-convertase
dissociation or inactivation of C3b, by recruited FI, to produce C3bi. C4bp is a
heteromeric protein composed of 6-8 α-chains containing binding domains for C4b and a
β-chain that, when coupled with serum Protein S, binds to polyanionic cell surfaces.
Similarly to FH, C4bp binding can result in C4b2a C3-convertase dissociation or in C4b
inactivation via FI. Binding of FH and C4bp to F. tularensis is an attractive explanation
for our findings of fixed C3b and C4b conversion to C3bi and C4bi on smooth and rough
strains. However, we show that both proteins preferentially bind rough, but not smooth,
F. tularensis strains in serum and that binding is equivalent between rough and smooth
strains in the absence of C3b/C4b fixation. Based on these results, we excluded the
prediction that preferential RCA protein recruitment by smooth F. tularensis is the reason
for increased rates of C3b to C3bi conversion compared to rough strains.
O-antigen can be a potent activator of the alternative pathway (323), which can
result in complement component fixation distal to the outer membrane (329-331). For
other Gram-negative organisms, distal component fixation results in ineffective MAC
membrane insertion and ultimately in consumption of complement hemolytic activity in
serum. We tested this potential mechanism of O-antigen-mediated complementresistance, but found that consumption of hemolytic activity was equivalent or greater for
rough F. tularensis, depending on the bacterial concentration used in this assay.
Taken together, these results provide strong evidence that the negative regulatory
influence of F. tularensis O-antigen affects the degree of classical complement pathway
activity upstream of C3b deposition. We have performed preliminary studies to
161
characterize C1 binding to smooth and rough F. tularensis. In serum, we show increased
C1q and C1s binding to O-antigen-deficient strains compared to wildtype strains. Also,
in the presence of purified C1, O-antigen mutants fix a greater amount of C4b, which
requires functional activation of C1-associated esterase upon C1 binding. Interestingly,
when bacteria are incubated in purified C1q or when C1q is dissociated from C1 in serum
using EDTA, smooth and rough F. tularensis strains bind equivalent amounts of C1q.
Based on these results, we can speculate that smooth and rough F. tularensis strains
express an equivalent amount of C1q acceptor molecules, but that C1-associated C1q has
a lower binding affinity in the presence of O-antigen. Future studies will examine the
influence of O-antigen on the direct binding of purified C1 and on C1 binding in
antibody-depleted serum. We will also identify C1 acceptor molecules to better
understand the negative influence of O-antigen on acceptor recognition. To begin, we
will perform comparative C1 binding analyses using isolated outer membrane proteins
and purified LPS from smooth and rough F. tularensis. Negatively charged outer
membrane proteins are common targets for C1q binding to other Gram-negative
organisms (206-211). Although C1q has been shown to preferentially recognize the core
sugar and/or lipid A components of LPS, we consider it a potential target of C1 binding
even to wildtype F. tularensis strains. Up to 90% of the LPS expressed by subsp.
novicida was recently shown to be in the form of free lipid A (168), and it is possible that
incomplete surface coverage by O-antigen occurs on more virulent F. tularensis
subspecies as well.
It is also of particular interest to our laboratory to determine whether classical
pathway-mediated opsonization occurs in the airway. Preliminary experiments indicate
162
that complement-resistant and complement-susceptible F. tularensis strains survive in
isolated human bronchoalveolar lavage (BAL) fluid. The significance of this result is
two-fold. The lytic capacity of complement in BAL fluid is decreased compared to
serum, probably due to deficiencies in terminal lytic pathway components. Second, it
implies that F. tularensis is resistant to soluble microbicidal effectors, such as lysozyme
and lactoferin that exist in airway surface fluid (ASF) in addition to complement.
Importantly, components of the classical pathway are functional in BAL fluid and have
been shown to mediate the opsonization of other pneumonic pathogens such as
Mycobacterium tuberculosis and Group B Streptococcus (283, 285). It is likely that F.
tularensis is also readily opsonized in ASF, which may enhance phagocytosis by alveolar
macrophages (AMs) in vivo.
CR3- and CR4-mediated opsonophagocytosis of F. tularensis results in unusual
phagosome formation involving “pseudopod loops” that is distinct from the “sinking”
phagosome normally associated with CR-mediated cellular entry (109, 131). We believe
this to be indicative of the involvement of additional important ligand-receptor
interactions occurring in concert with C3bi recognition by CR3 and/or CR4. This may be
important since signaling pathways downstream of cell surface receptors do not
propagate independently, but intercommunicate in order to cumulatively affect the
cellular response to infection. Additional receptors involved in macrophage phagocytosis
of F. tularensis, such as the mannose receptor, SR-A and Fcγ receptors have been
identified (110-112). However, the identification of as yet uncharacterized F. tularensisreceptor interactions, particularly ones that may potentially contribute to the
163
immunosuppressive effects of F. tularensis on macrophage cytokine production, could
reveal much about the pathogenic mechanisms associated with tularemia.
As we consider candidate receptors that may have an immunosuppressive role, we
are particularly struck by the similarities between F. tularensis and apoptotic cells in
terms of complement interactions and phagocytic mechanisms of macrophage ingestion.
Complement-mediated opsonization of apoptotic cells also occurs via the direct binding
of C1q (197). However, the degree that complement activation leads to anaphylatoxin
development, inflammation, and tissue damage is limited due to concurrent binding of
RCA proteins (198, 250, 251).
Furthermore, efficient macrophage uptake of apoptotic cells was shown to involve
cC1qR, a phagocytic C1q receptor that functions in concert with CD91 (276). Particle
uptake via the cC1qR/CD91 complex occurs via macropinocytosis, a form of “triggering”
phagocytosis akin to the formation of pseudopod loops. C1q binding to cC1qR/CD91
also catalyzes an anti-inflammatory signaling response characterized by nuclear
translocation of inhibitory NFκB complexes and of cAMP response element binding
(CREB) protein (277). Based on these similarities, we propose that surface-bound C1q
potentially functions as an opsonin to enhance the macrophage phagocytosis of F.
tularensis. This proposal is particularly provocative considering our dual finding that the
highly virulent Schu S4 strain binds increased amounts of C1q compared to LVS and that
human AM phagocytosis of Schu S4, but not LVS, is increased in fresh versus heatinactivated serum.
In summary, we provide strong evidence that F. tularensis activates the classical
complement pathway in an uncommon manner that is independent of antibody, but is
164
dependent on the direct binding and activation of complement component C1 by bacteria.
In the presence of LPS O-antigen surface expression, we propose that decreased C1
binding occurs, which results in decreased deposition of downstream complement
components including C4- and C3-derived fragments. Furthermore, both O-antigenproducing and O-antigen-deficient strains bind RCA proteins that, potentially, further
limit the deposition of components of the terminal lytic pathway by mediating the
inactivation of surface-bound C3b and C4b. However, due to increased C1 binding in the
absence of O-antigen, we propose that activation of the classical pathway causes rates of
C4b and C3b deposition that surpass the rates of C4b and C3b inactivation mediated by
RCA protein recruitment. Thus, complement mediated opsonization with concomitant
regulation of the complement cascade can occur via the direct binding of C1 and the
simultaneous binding of FH and C4bp.
165
References
1.
McCoy, G. W., and C.W. Chapin. 1912. Further observations on a plague-like
disease of rodents with a preliminary note on the causative agent, Bacterium
tularense. J Infect Dis 10:61-72.
2.
Wherry, W. B., and B.H. Lamb. 1914. Infection of man with Bacterium tularense.
J Infect Dis 15:331-340.
3.
Francis, E. 1928. A summary of the present knowledge of tularemia. Medicine
(Baltimore) 7:411-432.
4.
Ohara, H. 1925. Human inoculation experiment with a disease of wild rabbits,
with a bacteriologic study. Kinsei Igaku 12:401.
5.
Oyston, P. C., A. Sjostedt, and R. W. Titball. 2004. Tularaemia: bioterrorism
defence renews interest in Francisella tularensis. Nat Rev Microbiol 2:967-978.
6.
Weinberg, A. N. 2004. Commentary: Whrry W.B., and B.H. Lamb. Infection of
man with Bacterium tularense. J Infect Dis 1914;15:331-340. J Infect Dis
189:1317-1320.
7.
Bell, J. F., C. R. Owen, and C. L. Larson. 1955. Virulence of Bacterium tularense.
I. A study of the virulence of Bacterium tularense in mice, guinea pigs, and
rabbits. J Infect Dis 97:162-166.
8.
Metzger, D. W., C. S. Bakshi, and G. Kirimanjeswara. 2007. Mucosal
immunopathogenesis of Francisella tularensis. Ann N Y Acad Sci 1105:266-283.
9.
Morner, T. 1992. The ecology of tularaemia. Rev Sci Tech 11:1123-1130.
10.
Petersen, J. M., and M. E. Schriefer. 2005. Tularemia: emergence/re-emergence.
Vet Res 36:455-467.
11.
Nigrovic, L. E., and S. L. Wingerter. 2008. Tularemia. Infect Dis Clin North Am
22:489-504, ix.
12.
Dennis, D. T., T. V. Inglesby, D. A. Henderson, J. G. Bartlett, M. S. Ascher, E.
Eitzen, A. D. Fine, A. M. Friedlander, J. Hauer, M. Layton, S. R. Lillibridge, J. E.
McDade, M. T. Osterholm, T. O'Toole, G. Parker, T. M. Perl, P. K. Russell, and
166
K. Tonat. 2001. Tularemia as a biological weapon: medical and public health
management. Jama 285:2763-2773.
13.
Alibek, K. 1999. In Biohazard. Random House, New York, NY. 29-38.
14.
Gold, H. 2004. Unit 731 Testimony. Tuttle Publishing, North Clarendon, VT.
15.
Francis, E. 1928. Symptoms, diagnosis and pathology of tularemia. Jama
91:1155-1161.
16.
Evans, M. E., D. W. Gregory, W. Schaffner, and Z. A. McGee. 1985. Tularemia:
a 30-year experience with 88 cases. Medicine (Baltimore) 64:251-269.
17.
Saslaw, S., H.T. Eigelsbach, J.A. Prior, H.E. Wilson and S. Carhart. 1961.
Tularemia vaccine study II. Respiratory challenge. Arch Int Med 107:702-714.
18.
Stuart, B. M., and R.I. Pullen. 1945. Tularemic pneumonia: review of American
literature and reprt of 15 additional cases. Am J Med Sci 210:223-236.
19.
Conlan, J. W., R. KuoLee, H. Shen, and A. Webb. 2002. Different host defences
are required to protect mice from primary systemic vs pulmonary infection with
the facultative intracellular bacterial pathogen, Francisella tularensis LVS. Microb
Pathog 32:127-134.
20.
Fulop, M., P. Mastroeni, M. Green, and R. W. Titball. 2001. Role of antibody to
lipopolysaccharide in protection against low- and high-virulence strains of
Francisella tularensis. Vaccine 19:4465-4472.
21.
Prior, J. L., R. G. Prior, P. G. Hitchen, H. Diaper, K. F. Griffin, H. R. Morris, A.
Dell, and R. W. Titball. 2003. Characterization of the O antigen gene cluster and
structural analysis of the O antigen of Francisella tularensis subsp. tularensis. J
Med Microbiol 52:845-851.
22.
Janovska, S., I. Pavkova, M. Hubalek, J. Lenco, A. Macela, and J. Stulik. 2007.
Identification of immunoreactive antigens in membrane proteins enriched fraction
from Francisella tularensis LVS. Immunol Lett 108:151-159.
23.
Havlasova, J., L. Hernychova, M. Brychta, M. Hubalek, J. Lenco, P. Larsson, M.
Lundqvist, M. Forsman, Z. Krocova, J. Stulik, and A. Macela. 2005. Proteomic
analysis of anti-Francisella tularensis LVS antibody response in murine model of
tularemia. Proteomics 5:2090-2103.
24.
Fulop, M., R. Manchee, and R. Titball. 1995. Role of lipopolysaccharide and a
major outer membrane protein from Francisella tularensis in the induction of
immunity against tularemia. Vaccine 13:1220-1225.
167
25.
Golovliov, I., M. Ericsson, L. Akerblom, G. Sandstrom, A. Tarnvik, and A.
Sjostedt. 1995. Adjuvanticity of ISCOMs incorporating a T cell-reactive
lipoprotein of the facultative intracellular pathogen Francisella tularensis. Vaccine
13:261-267.
26.
Burke, D. S. 1977. Immunization against tularemia: analysis of the effectiveness
of live Francisella tularensis vaccine in prevention of laboratory-acquired
tularemia. J Infect Dis 135:55-60.
27.
Sandstrom, G. 1994. The tularaemia vaccine. J Chem Technol Biotechnol 59:315320.
28.
Tigertt, W. D. 1962. Soviet viable Pastereurella tularensis vaccines. A review of
selected articles. Bact Rev 26:354-373.
29.
McCrumb, F. R. 1961. Aerosol Infection of Man with Pasteurella Tularensis.
Bacteriol Rev 25:262-267.
30.
Zhang, P., W. R. Summer, G. J. Bagby, and S. Nelson. 2000. Innate immunity and
pulmonary host defense. Immunol Rev 173:39-51.
31.
Suzuki, T., C. W. Chow, and G. P. Downey. 2008. Role of innate immune cells
and their products in lung immunopathology. Int J Biochem Cell Biol 40:13481361.
32.
Ferguson, J. S., and L. S. Schlesinger. 2000. Pulmonary surfactant in innate
immunity and the pathogenesis of tuberculosis. Tuber Lung Dis 80:173-184.
33.
Wright, J. R. 2005. Immunoregulatory functions of surfactant proteins. Nat Rev
Immunol 5:58-68.
34.
Sallenave, J. M. 2000. The role of secretory leukocyte proteinase inhibitor and
elafin (elastase-specific inhibitor/skin-derived antileukoprotease) as alarm
antiproteinases in inflammatory lung disease. Respir Res 1:87-92.
35.
Serbina, N. V., T. Jia, T. M. Hohl, and E. G. Pamer. 2008. Monocyte-mediated
defense against microbial pathogens. Annu Rev Immunol 26:421-452.
36.
Beharka, A. A., C. D. Gaynor, B. K. Kang, D. R. Voelker, F. X. McCormack, and
L. S. Schlesinger. 2002. Pulmonary surfactant protein A up-regulates activity of
the mannose receptor, a pattern recognition receptor expressed on human
macrophages. J Immunol 169:3565-3573.
37.
Helmy, K. Y., K. J. Katschke, Jr., N. N. Gorgani, N. M. Kljavin, J. M. Elliott, L.
Diehl, S. J. Scales, N. Ghilardi, and M. van Lookeren Campagne. 2006. CRIg: a
macrophage complement receptor required for phagocytosis of circulating
pathogens. Cell 124:915-927.
168
38.
Kuronuma, K., H. Sano, K. Kato, K. Kudo, N. Hyakushima, S. Yokota, H.
Takahashi, N. Fujii, H. Suzuki, T. Kodama, S. Abe, and Y. Kuroki. 2004.
Pulmonary surfactant protein A augments the phagocytosis of Streptococcus
pneumoniae by alveolar macrophages through a casein kinase 2-dependent
increase of cell surface localization of scavenger receptor A. J Biol Chem
279:21421-21430.
39.
Myones, B. L., J. G. Dalzell, N. Hogg, and G. D. Ross. 1988. Neutrophil and
monocyte cell surface p150,95 has iC3b-receptor (CR4) activity resembling CR3.
J Clin Invest 82:640-651.
40.
Skold, C. M., A. Eklund, G. Hallden, and J. Hed. 1990. Different cell surface and
phagocytic properties in mononuclear phagocytes from blood and alveoli. A
comparative study of blood monocytes and alveolar macrophages from human
nonsmokers using flow cytofluorometry. Apmis 98:415-422.
41.
Hempel, S. L., M. M. Monick, B. He, T. Yano, and G. W. Hunninghake. 1994.
Synthesis of prostaglandin H synthase-2 by human alveolar macrophages in
response to lipopolysaccharide is inhibited by decreased cell oxidant tone. J Biol
Chem 269:32979-32984.
42.
Wewers, M. D., S. I. Rennard, A. J. Hance, P. B. Bitterman, and R. G. Crystal.
1984. Normal human alveolar macrophages obtained by bronchoalveolar lavage
have a limited capacity to release interleukin-1. J Clin Invest 74:2208-2218.
43.
Munder, M., K. Eichmann, and M. Modolell. 1998. Alternative metabolic states
in murine macrophages reflected by the nitric oxide synthase/arginase balance:
competitive regulation by CD4+ T cells correlates with Th1/Th2 phenotype. J
Immunol 160:5347-5354.
44.
Lohmann-Matthes, M. L., C. Steinmuller, and G. Franke-Ullmann. 1994.
Pulmonary macrophages. Eur Respir J 7:1678-1689.
45.
Iwasaki, A. 2007. Mucosal dendritic cells. Annu Rev Immunol 25:381-418.
46.
Stout, R. D., and J. Suttles. 2004. Functional plasticity of macrophages: reversible
adaptation to changing microenvironments. J Leukoc Biol 76:509-513.
47.
Miller, S. I., R. K. Ernst, and M. W. Bader. 2005. LPS, TLR4 and infectious
disease diversity. Nat Rev Microbiol 3:36-46.
48.
Miyake, K. 2004. Innate recognition of lipopolysaccharide by Toll-like receptor
4-MD-2. Trends Microbiol 12:186-192.
49.
Parker, L. C., M. K. Whyte, S. K. Dower, and I. Sabroe. 2005. The expression and
roles of Toll-like receptors in the biology of the human neutrophil. J Leukoc Biol
77:886-892.
169
50.
Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda,
and S. Akira. 1999. Differential roles of TLR2 and TLR4 in recognition of gramnegative and gram-positive bacterial cell wall components. Immunity 11:443-451.
51.
Henning, L. N., A. K. Azad, K. V. Parsa, J. E. Crowther, S. Tridandapani, and L.
S. Schlesinger. 2008. Pulmonary surfactant protein A regulates TLR expression
and activity in human macrophages. J Immunol 180:7847-7858.
52.
Visintin, A., A. Mazzoni, J. H. Spitzer, D. H. Wyllie, S. K. Dower, and D. M.
Segal. 2001. Regulation of Toll-like receptors in human monocytes and dendritic
cells. J Immunol 166:249-255.
53.
Oyston, P. C. 2008. Francisella tularensis: unravelling the secrets of an
intracellular pathogen. J Med Microbiol 57:921-930.
54.
Wayne Conlan, J., and P. C. Oyston. 2007. Vaccines against Francisella
tularensis. Ann N Y Acad Sci 1105:325-350.
55.
Ancuta, P., T. Pedron, R. Girard, G. Sandstrom, and R. Chaby. 1996. Inability of
the Francisella tularensis lipopolysaccharide to mimic or to antagonize the
induction of cell activation by endotoxins. Infect Immun 64:2041-2046.
56.
Sandstrom, G., A. Sjostedt, T. Johansson, K. Kuoppa, and J. C. Williams. 1992.
Immunogenicity and toxicity of lipopolysaccharide from Francisella tularensis
LVS. FEMS Microbiol Immunol 5:201-210.
57.
Telepnev, M., I. Golovliov, and A. Sjostedt. 2005. Francisella tularensis LVS
initially activates but subsequently down-regulates intracellular signaling and
cytokine secretion in mouse monocytic and human peripheral blood mononuclear
cells. Microb Pathog 38:239-247.
58.
Barker, J. H., J. Weiss, M. A. Apicella, and W. M. Nauseef. 2006. Basis for the
failure of Francisella tularensis lipopolysaccharide to prime human
polymorphonuclear leukocytes. Infect Immun 74:3277-3284.
59.
Anthony, L. D., R. D. Burke, and F. E. Nano. 1991. Growth of Francisella spp. in
rodent macrophages. Infect Immun 59:3291-3296.
60.
Ben Nasr, A., J. Haithcoat, J. E. Masterson, J. S. Gunn, T. Eaves-Pyles, and G. R.
Klimpel. 2006. Critical role for serum opsonins and complement receptors CR3
(CD11b/CD18) and CR4 (CD11c/CD18) in phagocytosis of Francisella tularensis
by human dendritic cells (DC): uptake of Francisella leads to activation of
immature DC and intracellular survival of the bacteria. J Leukoc Biol.
61.
Buddingh, G. J. a. F. C. W. J. 1941. Observations on the infection of chick
embryos with Bacterium tularense, Brucella, and Pasteurella pestis. J Exp Med
74:213-222.
170
62.
Councilman, W. T. a. R. P. S. 1921. Plague-like infections in rodents. Trans
Assoc Am Physicians 36:135-143.
63.
Francis, E. 1927. Microscopic changes of tularemia in the tick Dermacentor
andersoni and the bedbug Cimex lectularius. Public Health Rep 42:2763-2772.
64.
Lindemann, S. R., M. K. McLendon, M. A. Apicella, and B. D. Jones. 2007. An
in vitro model system used to study adherence and invasion of Francisella
tularensis live vaccine strain in nonphagocytic cells. Infect Immun 75:3178-3182.
65.
Shepard, C. C. 1959. Nonacid-fast bacteria and HeLa cells: their uptake and
subsequent intracellular growth. J Bacteriol 77:701-714.
66.
Hall, J. D., M. D. Woolard, B. M. Gunn, R. R. Craven, S. Taft-Benz, J. A.
Frelinger, and T. H. Kawula. 2008. Infected-host-cell repertoire and cellular
response in the lung following inhalation of Francisella tularensis Schu S4, LVS,
or U112. Infect Immun 76:5843-5852.
67.
Gentry, M., J. Taormina, R. B. Pyles, L. Yeager, M. Kirtley, V. L. Popov, G.
Klimpel, and T. Eaves-Pyles. 2007. Role of primary human alveolar epithelial
cells in host defense against Francisella tularensis infection. Infect Immun
75:3969-3978.
68.
Fortier, A. H., Green, S.J., Polsinelli, T., Jones, T.R., Crawford R.M., Leiby, D.A.
Elkins, K.L., Meltzer, M.S., Nacy, C.A. 1994. Life and death of an intracellular
pathogen: Francisella tularensis and the macrophage. Immunol Ser 60:349-361.
69.
Bar-Haim, E., O. Gat, G. Markel, H. Cohen, A. Shafferman, and B. Velan. 2008.
Interrelationship between dendritic cell trafficking and Francisella tularensis
dissemination following airway infection. PLoS Pathog 4:e1000211.
70.
Bosio, C. M., H. Bielefeldt-Ohmann, and J. T. Belisle. 2007. Active suppression
of the pulmonary immune response by Francisella tularensis Schu4. J Immunol
178:4538-4547.
71.
Bosio, C. M., and S. W. Dow. 2005. Francisella tularensis induces aberrant
activation of pulmonary dendritic cells. J Immunol 175:6792-6801.
72.
Vermaelen, K., and R. Pauwels. 2004. Accurate and simple discrimination of
mouse pulmonary dendritic cell and macrophage populations by flow cytometry:
methodology and new insights. Cytometry A 61:170-177.
73.
Wikstrom, M. E., and P. A. Stumbles. 2007. Mouse respiratory tract dendritic cell
subsets and the immunological fate of inhaled antigens. Immunol Cell Biol
85:182-188.
171
74.
White, J. D., J. R. Rooney, P.A. Prickett, E.B. Darrenbacher, C.W. Beard, W.R.
Griffith. 1964. Pathogenesis of experimental respiratory tularemia in monkeys.
Journal of Infectious Diseases 114:277-283.
75.
Forestal, C. A., M. Malik, S. V. Catlett, A. G. Savitt, J. L. Benach, T. J. Sellati,
and M. B. Furie. 2007. Francisella tularensis has a significant extracellular phase
in infected mice. J Infect Dis 196:134-137.
76.
Mares, C. A., S. S. Ojeda, E. G. Morris, Q. Li, and J. M. Teale. 2008. Initial delay
in the immune response to Francisella tularensis is followed by hypercytokinemia
characteristic of severe sepsis and correlating with upregulation and release of
damage-associated molecular patterns. Infect Immun 76:3001-3010.
77.
Sharma, J., Q. Li, B. B. Mishra, C. Pena, and J. M. Teale. 2009. Lethal pulmonary
infection with Francisella novicida is associated with severe sepsis. J Leukoc Biol.
78.
Twine, S., M. Bystrom, W. Chen, M. Forsman, I. Golovliov, A. Johansson, J.
Kelly, H. Lindgren, K. Svensson, C. Zingmark, W. Conlan, and A. Sjostedt. 2005.
A mutant of Francisella tularensis strain SCHU S4 lacking the ability to express a
58-kilodalton protein is attenuated for virulence and is an effective live vaccine.
Infect Immun 73:8345-8352.
79.
Nano, F. E., and C. Schmerk. 2007. The Francisella pathogenicity island. Ann N Y
Acad Sci 1105:122-137.
80.
Barker, J. R., and K. E. Klose. 2007. Molecular and genetic basis of pathogenesis
in Francisella tularensis. Ann N Y Acad Sci 1105:138-159.
81.
Golovliov, I., K. Kuoppa, A. Sjostedt, A. Tarnvik, and G. Sandstrom. 1996.
Cytokine expression in the liver of mice infected with a highly virulent strain of
Francisella tularensis. FEMS Immunol Med Microbiol 13:239-244.
82.
Stenmark, S., D. Sunnemark, A. Bucht, and A. Sjostedt. 1999. Rapid local
expression of interleukin-12, tumor necrosis factor alpha, and gamma interferon
after cutaneous Francisella tularensis infection in tularemia-immune mice. Infect
Immun 67:1789-1797.
83.
Conlan, J. W., W. Chen, H. Shen, A. Webb, and R. KuoLee. 2003. Experimental
tularemia in mice challenged by aerosol or intradermally with virulent strains of
Francisella tularensis: bacteriologic and histopathologic studies. Microb Pathog
34:239-248.
84.
Rick Lyons, C., and T. H. Wu. 2007. Animal models of Francisella tularensis
infection. Ann N Y Acad Sci 1105:238-265.
172
85.
Twenhafel, N., D. Alves, and B. Purcell. 2009. Pathology of Inhalational
Francisella tularensis spp tularensis SCHU S4 Infection in African Green
Monkeys (Chlorocebus aethiops). Vet Pathol (Accepted article).
86.
Andersson, H., B. Hartmanova, E. Back, H. Eliasson, M. Landfors, L. Naslund, P.
Ryden, and A. Sjostedt. 2006. Transcriptional profiling of the peripheral blood
response during tularemia. Genes Immun 7:503-513.
87.
Duckett, N. S., S. Olmos, D. M. Durrant, and D. W. Metzger. 2005. Intranasal
interleukin-12 treatment for protection against respiratory infection with the
Francisella tularensis live vaccine strain. Infect Immun 73:2306-2311.
88.
Malik, M., C. S. Bakshi, K. McCabe, S. V. Catlett, A. Shah, R. Singh, P. L.
Jackson, A. Gaggar, D. W. Metzger, J. A. Melendez, J. E. Blalock, and T. J.
Sellati. 2007. Matrix metalloproteinase 9 activity enhances host susceptibility to
pulmonary infection with type A and B strains of Francisella tularensis. J
Immunol 178:1013-1020.
89.
Elkins, K. L., S. C. Cowley, and C. M. Bosio. 2007. Innate and adaptive
immunity to Francisella. Ann N Y Acad Sci 1105:284-324.
90.
Chen, W., R. KuoLee, H. Shen, and J. W. Conlan. 2004. Susceptibility of
immunodeficient mice to aerosol and systemic infection with virulent strains of
Francisella tularensis. Microb Pathog 36:311-318.
91.
Leiby, D. A., A. H. Fortier, R. M. Crawford, R. D. Schreiber, and C. A. Nacy.
1992. In vivo modulation of the murine immune response to Francisella tularensis
LVS by administration of anticytokine antibodies. Infect Immun 60:84-89.
92.
Pammit, M. A., V. N. Budhavarapu, E. K. Raulie, K. E. Klose, J. M. Teale, and B.
P. Arulanandam. 2004. Intranasal interleukin-12 treatment promotes antimicrobial
clearance and survival in pulmonary Francisella tularensis subsp. novicida
infection. Antimicrob Agents Chemother 48:4513-4519.
93.
Lindgren, H., L. Stenman, A. Tarnvik, and A. Sjostedt. 2005. The contribution of
reactive nitrogen and oxygen species to the killing of Francisella tularensis LVS
by murine macrophages. Microbes Infect 7:467-475.
94.
Lindgren, H., I. Golovliov, V. Baranov, R. K. Ernst, M. Telepnev, and A.
Sjostedt. 2004. Factors affecting the escape of Francisella tularensis from the
phagolysosome. J Med Microbiol 53:953-958.
95.
Santic, M., M. Molmeret, J. R. Barker, K. E. Klose, A. Dekanic, M. Doric, and Y.
Abu Kwaik. 2007. A Francisella tularensis pathogenicity island protein essential
for bacterial proliferation within the host cell cytosol. Cell Microbiol 9:23912403.
173
96.
Santic, M., M. Molmeret, and Y. Abu Kwaik. 2005. Modulation of biogenesis of
the Francisella tularensis subsp. novicida-containing phagosome in quiescent
human macrophages and its maturation into a phagolysosome upon activation by
IFN-gamma. Cell Microbiol 7:957-967.
97.
Cole, L. E., A. Santiago, E. Barry, T. J. Kang, K. A. Shirey, Z. J. Roberts, K. L.
Elkins, A. S. Cross, and S. N. Vogel. 2008. Macrophage proinflammatory
response to Francisella tularensis live vaccine strain requires coordination of
multiple signaling pathways. J Immunol 180:6885-6891.
98.
Thakran, S., H. Li, C. L. Lavine, M. A. Miller, J. E. Bina, X. R. Bina, and F. Re.
2008. Identification of Francisella tularensis lipoproteins that stimulate the tolllike receptor (TLR) 2/TLR1 heterodimer. J Biol Chem 283:3751-3760.
99.
Hajjar, A. M., M. D. Harvey, S. A. Shaffer, D. R. Goodlett, A. Sjostedt, H.
Edebro, M. Forsman, M. Bystrom, M. Pelletier, C. B. Wilson, S. I. Miller, S. J.
Skerrett, and R. K. Ernst. 2006. Lack of in vitro and in vivo recognition of
Francisella tularensis subspecies lipopolysaccharide by Toll-like receptors. Infect
Immun 74:6730-6738.
100.
Shirey, K. A., L. E. Cole, A. D. Keegan, and S. N. Vogel. 2008. Francisella
tularensis live vaccine strain induces macrophage alternative activation as a
survival mechanism. J Immunol 181:4159-4167.
101.
Malik, M., C. S. Bakshi, B. Sahay, A. Shah, S. A. Lotz, and T. J. Sellati. 2006.
Toll-like receptor 2 is required for control of pulmonary infection with Francisella
tularensis. Infect Immun 74:3657-3662.
102.
Gavrilin, M. A., S. Mitra, S. Seshadri, J. Nateri, F. Berhe, M. W. Hall, and M. D.
Wewers. 2009. Pyrin critical to macrophage IL-1beta response to Francisella
challenge. J Immunol 182:7982-7989.
103.
Mariathasan, S., D. S. Weiss, V. M. Dixit, and D. M. Monack. 2005. Innate
immunity against Francisella tularensis is dependent on the ASC/caspase-1 axis. J
Exp Med 202:1043-1049.
104.
Parsa, K. V., J. P. Butchar, M. V. Rajaram, T. J. Cremer, J. S. Gunn, L. S.
Schlesinger, and S. Tridandapani. 2008. Francisella gains a survival advantage
within mononuclear phagocytes by suppressing the host IFNgamma response.
Mol Immunol 45:3428-3437.
105.
Chase, J. C., J. Celli, and C. M. Bosio. 2009. Direct and indirect impairment of
human dendritic cell function by virulent Francisella tularensis Schu S4. Infect
Immun 77:180-195.
174
106.
Butchar, J. P., M. V. Rajaram, L. P. Ganesan, K. V. Parsa, C. D. Clay, L. S.
Schlesinger, and S. Tridandapani. 2007. Francisella tularensis induces IL-23
production in human monocytes. J Immunol 178:4445-4454.
107.
Telepnev, M., I. Golovliov, T. Grundstrom, A. Tarnvik, and A. Sjostedt. 2003.
Francisella tularensis inhibits Toll-like receptor-mediated activation of
intracellular signalling and secretion of TNF-alpha and IL-1 from murine
macrophages. Cell Microbiol 5:41-51.
108.
Underhill, D. M., and A. Ozinsky. 2002. Phagocytosis of microbes: complexity in
action. Annu Rev Immunol 20:825-852.
109.
Clemens, D. L., B. Y. Lee, and M. A. Horwitz. 2005. Francisella tularensis enters
macrophages via a novel process involving pseudopod loops. Infect Immun
73:5892-5902.
110.
Balagopal, A., A. S. MacFarlane, N. Mohapatra, S. Soni, J. S. Gunn, and L. S.
Schlesinger. 2006. Characterization of the receptor-ligand pathways important for
entry and survival of Francisella tularensis in human macrophages. Infect Immun
74:5114-5125.
111.
Schulert, G. S., and L. A. Allen. 2006. Differential infection of mononuclear
phagocytes by Francisella tularensis: role of the macrophage mannose receptor. J
Leukoc Biol 80:563-571.
112.
Pierini, L. M. 2006. Uptake of serum-opsonized Francisella tularensis by
macrophages can be mediated by class A scavenger receptors. Cell Microbiol
8:1361-1370.
113.
Barel, M., A. G. Hovanessian, K. Meibom, J. P. Briand, M. Dupuis, and A.
Charbit. 2008. A novel receptor - ligand pathway for entry of Francisella
tularensis in monocyte-like THP-1 cells: interaction between surface nucleolin
and bacterial elongation factor Tu. BMC Microbiol 8:145.
114.
Cheng, P. C., A. Cherukuri, M. Dykstra, S. Malapati, T. Sproul, M. R. Chen, and
S. K. Pierce. 2001. Floating the raft hypothesis: the roles of lipid rafts in B cell
antigen receptor function. Semin Immunol 13:107-114.
115.
Tamilselvam, B., and S. Daefler. 2008. Francisella targets cholesterol-rich host
cell membrane domains for entry into macrophages. J Immunol 180:8262-8271.
116.
Bermudez, L. E., L. S. Young, and H. Enkel. 1991. Interaction of Mycobacterium
avium complex with human macrophages: roles of membrane receptors and
serum proteins. Infect Immun 59:1697-1702.
117.
Schlesinger, L. S., C. G. Bellinger-Kawahara, N. R. Payne, and M. A. Horwitz.
1990. Phagocytosis of Mycobacterium tuberculosis is mediated by human
175
monocyte complement receptors and complement component C3. J Immunol
144:2771-2780.
118.
Schlesinger, L. S., and M. A. Horwitz. 1990. Phagocytosis of leprosy bacilli is
mediated by complement receptors CR1 and CR3 on human monocytes and
complement component C3 in serum. J Clin Invest 85:1304-1314.
119.
Schlesinger, L. S., and M. A. Horwitz. 1991. Phagocytosis of Mycobacterium
leprae by human monocyte-derived macrophages is mediated by complement
receptors CR1 (CD35), CR3 (CD11b/CD18), and CR4 (CD11c/CD18) and IFNgamma activation inhibits complement receptor function and phagocytosis of this
bacterium. J Immunol 147:1983-1994.
120.
Edwards, J. L. 2008. The role of complement in gonococcal infection of cervical
epithelia. Vaccine 26 Suppl 8:I56-61.
121.
Blackwell, J. M., R. A. Ezekowitz, M. B. Roberts, J. Y. Channon, R. B. Sim, and
S. Gordon. 1985. Macrophage complement and lectin-like receptors bind
Leishmania in the absence of serum. J Exp Med 162:324-331.
122.
Mosser, D. M., and P. J. Edelson. 1985. The mouse macrophage receptor for C3bi
(CR3) is a major mechanism in the phagocytosis of Leishmania promastigotes. J
Immunol 135:2785-2789.
123.
Payne, N. R., and M. A. Horwitz. 1987. Phagocytosis of Legionella pneumophila
is mediated by human monocyte complement receptors. J Exp Med 166:13771389.
124.
Bullock, W. E., and S. D. Wright. 1987. Role of the adherence-promoting
receptors, CR3, LFA-1, and p150,95, in binding of Histoplasma capsulatum by
human macrophages. J Exp Med 165:195-210.
125.
Drevets, D. A., and P. A. Campbell. 1991. Roles of complement and complement
receptor type 3 in phagocytosis of Listeria monocytogenes by inflammatory
mouse peritoneal macrophages. Infect Immun 59:2645-2652.
126.
Rimoldi, M. T., A. J. Tenner, D. A. Bobak, and K. A. Joiner. 1989. Complement
component C1q enhances invasion of human mononuclear phagocytes and
fibroblasts by Trypanosoma cruzi trypomastigotes. J Clin Invest 84:1982-1989.
127.
Cox, D., and S. Greenberg. 2001. Phagocytic signaling strategies:
Fc(gamma)receptor-mediated phagocytosis as a model system. Semin Immunol
13:339-345.
128.
Griffin, F. M., Jr., J. A. Griffin, J. E. Leider, and S. C. Silverstein. 1975. Studies
on the mechanism of phagocytosis. I. Requirements for circumferential
176
attachment of particle-bound ligands to specific receptors on the macrophage
plasma membrane. J Exp Med 142:1263-1282.
129.
Griffin, F. M., Jr., J. A. Griffin, and S. C. Silverstein. 1976. Studies on the
mechanism of phagocytosis. II. The interaction of macrophages with antiimmunoglobulin IgG-coated bone marrow-derived lymphocytes. J Exp Med
144:788-809.
130.
Swanson, J. A., and S. C. Baer. 1995. Phagocytosis by zippers and triggers.
Trends Cell Biol 5:89-93.
131.
Aderem, A., and D. M. Underhill. 1999. Mechanisms of phagocytosis in
macrophages. Annu Rev Immunol 17:593-623.
132.
Allen, L. A., and A. Aderem. 1996. Molecular definition of distinct cytoskeletal
structures involved in complement- and Fc receptor-mediated phagocytosis in
macrophages. J Exp Med 184:627-637.
133.
Kaplan, G. 1977. Differences in the mode of phagocytosis with Fc and C3
receptors in macrophages. Scand J Immunol 6:797-807.
134.
Caron, E., A. J. Self, and A. Hall. 2000. The GTPase Rap1 controls functional
activation of macrophage integrin alphaMbeta2 by LPS and other inflammatory
mediators. Curr Biol 10:974-978.
135.
Hall, A. B., M. A. Gakidis, M. Glogauer, J. L. Wilsbacher, S. Gao, W. Swat, and
J. S. Brugge. 2006. Requirements for Vav guanine nucleotide exchange factors
and Rho GTPases in FcgammaR- and complement-mediated phagocytosis.
Immunity 24:305-316.
136.
Swanson, J. A. 2008. Shaping cups into phagosomes and macropinosomes. Nat
Rev Mol Cell Biol 9:639-649.
137.
Bar-Sagi, D., and J. R. Feramisco. 1986. Induction of membrane ruffling and
fluid-phase pinocytosis in quiescent fibroblasts by ras proteins. Science 233:10611068.
138.
Racoosin, E. L., and J. A. Swanson. 1989. Macrophage colony-stimulating factor
(rM-CSF) stimulates pinocytosis in bone marrow-derived macrophages. J Exp
Med 170:1635-1648.
139.
Alpuche-Aranda, C. M., E. L. Racoosin, J. A. Swanson, and S. I. Miller. 1994.
Salmonella stimulate macrophage macropinocytosis and persist within spacious
phagosomes. J Exp Med 179:601-608.
177
140.
Francis, C. L., T. A. Ryan, B. D. Jones, S. J. Smith, and S. Falkow. 1993. Ruffles
induced by Salmonella and other stimuli direct macropinocytosis of bacteria.
Nature 364:639-642.
141.
Kerr, M. C., and R. D. Teasdale. 2009. Defining Macropinocytosis. Traffic.
142.
Pizarro-Cerda, J., and P. Cossart. 2006. Bacterial adhesion and entry into host
cells. Cell 124:715-727.
143.
Watarai, M., I. Derre, J. Kirby, J. D. Growney, W. F. Dietrich, and R. R. Isberg.
2001. Legionella pneumophila is internalized by a macropinocytotic uptake
pathway controlled by the Dot/Icm system and the mouse Lgn1 locus. J Exp Med
194:1081-1096.
144.
Horwitz, M. A. 1984. Phagocytosis of the Legionnaires' disease bacterium
(Legionella pneumophila) occurs by a novel mechanism: engulfment within a
pseudopod coil. Cell 36:27-33.
145.
Le Cabec, V., S. Carreno, A. Moisand, C. Bordier, and I. Maridonneau-Parini.
2002. Complement receptor 3 (CD11b/CD18) mediates type I and type II
phagocytosis during nonopsonic and opsonic phagocytosis, respectively. J
Immunol 169:2003-2009.
146.
Caron, E., and A. Hall. 1998. Identification of two distinct mechanisms of
phagocytosis controlled by different Rho GTPases. Science 282:1717-1721.
147.
Clemens, D. L., and M. A. Horwitz. 2007. Uptake and intracellular fate of
Francisella tularensis in human macrophages. Ann N Y Acad Sci 1105:160-186.
148.
Santic, M., R. Asare, I. Skrobonja, S. Jones, and Y. Abu Kwaik. 2008.
Acquisition of the vacuolar ATPase proton pump and phagosome acidification are
essential for escape of Francisella tularensis into the macrophage cytosol. Infect
Immun 76:2671-2677.
149.
Clemens, D. L., B. Y. Lee, and M. A. Horwitz. 2004. Virulent and avirulent
strains of Francisella tularensis prevent acidification and maturation of their
phagosomes and escape into the cytoplasm in human macrophages. Infect Immun
72:3204-3217.
150.
Checroun, C., T. D. Wehrly, E. R. Fischer, S. F. Hayes, and J. Celli. 2006.
Autophagy-mediated reentry of Francisella tularensis into the endocytic
compartment after cytoplasmic replication. Proc Natl Acad Sci U S A 103:1457814583.
151.
Clemens, D. L., B. Y. Lee, and M. A. Horwitz. 2009. Francisella tularensis
phagosomal escape does not require acidification of the phagosome. Infect Immun
77:1757-1773.
178
152.
Golovliov, I., V. Baranov, Z. Krocova, H. Kovarova, and A. Sjostedt. 2003. An
attenuated strain of the facultative intracellular bacterium Francisella tularensis
can escape the phagosome of monocytic cells. Infect Immun 71:5940-5950.
153.
Craven, R. R., J. D. Hall, J. R. Fuller, S. Taft-Benz, and T. H. Kawula. 2008.
Francisella tularensis invasion of lung epithelial cells. Infect Immun 76:28332842.
154.
McCaffrey, R. L., and L. A. Allen. 2006. Francisella tularensis LVS evades
killing by human neutrophils via inhibition of the respiratory burst and
phagosome escape. J Leukoc Biol 80:1224-1230.
155.
Hrstka, R., J. Stulik, and B. Vojtesek. 2005. The role of MAPK signal pathways
during Francisella tularensis LVS infection-induced apoptosis in murine
macrophages. Microbes Infect 7:619-625.
156.
Lai, X. H., and A. Sjostedt. 2003. Delineation of the molecular mechanisms of
Francisella tularensis-induced apoptosis in murine macrophages. Infect Immun
71:4642-4646.
157.
Cherwonogrodzky, J. W., M. H. Knodel, and M. R. Spence. 1994. Increased
encapsulation and virulence of Francisella tularensis live vaccine strain (LVS) by
subculturing on synthetic medium. Vaccine 12:773-775.
158.
Geisbert, T. W., P. B. Jahrling, and J. W. Ezzell, Jr. 1993. Use of immunoelectron
microscopy to demonstrate Francisella tularensis. J Clin Microbiol 31:1936-1939.
159.
Hood, A. M. 1977. Virulence factors of Francisella tularensis. J Hyg (Lond)
79:47-60.
160.
Su, J., J. Yang, D. Zhao, T. H. Kawula, J. A. Banas, and J. R. Zhang. 2007.
Genome-wide identification of Francisella tularensis virulence determinants.
Infect Immun 75:3089-3101.
161.
Gunn, J. S., and R. K. Ernst. 2007. The structure and function of Francisella
lipopolysaccharide. Ann N Y Acad Sci 1105:202-218.
162.
Raetz, C. R., and C. Whitfield. 2002. Lipopolysaccharide endotoxins. Annu Rev
Biochem 71:635-700.
163.
Trent, M. S., C. M. Stead, A. X. Tran, and J. V. Hankins. 2006. Diversity of
endotoxin and its impact on pathogenesis. J Endotoxin Res 12:205-223.
164.
Brandenburg, K., and A. Wiese. 2004. Endotoxins: relationships between
structure, function, and activity. Curr Top Med Chem 4:1127-1146.
179
165.
Phillips, N. J., B. Schilling, M. K. McLendon, M. A. Apicella, and B. W. Gibson.
2004. Novel modification of lipid A of Francisella tularensis. Infect Immun
72:5340-5348.
166.
Shaffer, S. A., M. D. Harvey, D. R. Goodlett, and R. K. Ernst. 2007. Structural
heterogeneity and environmentally regulated remodeling of Francisella tularensis
subspecies novicida lipid A characterized by tandem mass spectrometry. J Am Soc
Mass Spectrom 18:1080-1092.
167.
Vinogradov, E., M. B. Perry, and J. W. Conlan. 2002. Structural analysis of
Francisella tularensis lipopolysaccharide. Eur J Biochem 269:6112-6118.
168.
Wang, X., A. A. Ribeiro, Z. Guan, S. C. McGrath, R. J. Cotter, and C. R. Raetz.
2006. Structure and biosynthesis of free lipid A molecules that replace
lipopolysaccharide in Francisella tularensis subsp. novicida. Biochemistry
45:14427-14440.
169.
Baldridge, J. R., and R. T. Crane. 1999. Monophosphoryl lipid A (MPL)
formulations for the next generation of vaccines. Methods 19:103-107.
170.
Thirumalapura, N. R., D. W. Goad, A. Mort, R. J. Morton, J. Clarke, and J.
Malayer. 2005. Structural analysis of the O-antigen of Francisella tularensis
subspecies tularensis strain OSU 10. J Med Microbiol 54:693-695.
171.
Vinogradov, E. V., A. S. Shashkov, Y. A. Knirel, N. K. Kochetkov, N. V.
Tochtamysheva, S. F. Averin, O. V. Goncharova, and V. S. Khlebnikov. 1991.
Structure of the O-antigen of Francisella tularensis strain 15. Carbohydr Res
214:289-297.
172.
Conlan, J. W., H. Shen, A. Webb, and M. B. Perry. 2002. Mice vaccinated with
the O-antigen of Francisella tularensis LVS lipopolysaccharide conjugated to
bovine serum albumin develop varying degrees of protective immunity against
systemic or aerosol challenge with virulent type A and type B strains of the
pathogen. Vaccine 20:3465-3471.
173.
Vinogradov, E., W. J. Conlan, J. S. Gunn, and M. B. Perry. 2004.
Characterization of the lipopolysaccharide O-antigen of Francisella novicida
(U112). Carbohydr Res 339:649-654.
174.
Eigelsbach, H. T., W. Braun, and R. D. Herring. 1951. Studies on the variation of
Bacterium tularense. J Bacteriol 61:557-569.
175.
Eigelsbach, H. T., and C. M. Downs. 1961. Prophylactic effectiveness of live and
killed tularemia vaccines. I. Production of vaccine and evaluation in the white
mouse and guinea pig. J Immunol 87:415-425.
180
176.
Cowley, S. C., S. V. Myltseva, and F. E. Nano. 1996. Phase variation in
Francisella tularensis affecting intracellular growth, lipopolysaccharide
antigenicity and nitric oxide production. Mol Microbiol 20:867-874.
177.
Hartley, G., R. Taylor, J. Prior, S. Newstead, P. G. Hitchen, H. R. Morris, A. Dell,
and R. W. Titball. 2006. Grey variants of the live vaccine strain of Francisella
tularensis lack lipopolysaccharide O-antigen, show reduced ability to survive in
macrophages and do not induce protective immunity in mice. Vaccine 24:989996.
178.
Costerton, J. W., R. T. Irvin, and K. J. Cheng. 1981. The bacterial glycocalyx in
nature and disease. Annu Rev Microbiol 35:299-324.
179.
Cowley, S. C., C. J. Gray, and F. E. Nano. 2000. Isolation and characterization of
Francisella novicida mutants defective in lipopolysaccharide biosynthesis. FEMS
Microbiol Lett 182:63-67.
180.
Raynaud, C., K. L. Meibom, M. A. Lety, I. Dubail, T. Candela, E. Frapy, and A.
Charbit. 2007. Role of the wbt locus of Francisella tularensis in
lipopolysaccharide O-antigen biogenesis and pathogenicity. Infect Immun 75:536541.
181.
Sebastian, S., S. T. Dillon, J. G. Lynch, L. T. Blalock, E. Balon, K. T. Lee, L. E.
Comstock, J. W. Conlan, E. J. Rubin, A. O. Tzianabos, and D. L. Kasper. 2007. A
defined O-antigen polysaccharide mutant of Francisella tularensis live vaccine
strain has attenuated virulence while retaining its protective capacity. Infect
Immun 75:2591-2602.
182.
Li, J., C. Ryder, M. Mandal, F. Ahmed, P. Azadi, D. S. Snyder, R. D. Pechous, T.
Zahrt, and T. J. Inzana. 2007. Attenuation and protective efficacy of an Oantigen-deficient mutant of Francisella tularensis LVS. Microbiology 153:31413153.
183.
Sandstrom, G., S. Lofgren, and A. Tarnvik. 1988. A capsule-deficient mutant of
Francisella tularensis LVS exhibits enhanced sensitivity to killing by serum but
diminished sensitivity to killing by polymorphonuclear leukocytes. Infect Immun
56:1194-1202.
184.
Sorokin, V. M., N. V. Pavlovich, and L. A. Prozorova. 1996. Francisella
tularensis resistance to bactericidal action of normal human serum. FEMS
Immunol Med Microbiol 13:249-252.
185.
Metschnikoff, E. 1884. Arb. Zool. Inst. Univ. wien. u. Zool. Stat. Triest 5:141-168.
186.
Buchner, H. 1889. Zbl. Bakt. (Naturwiss.) 5:817.
187.
Bordet, J. 1909. Studies in Immunity. J. Wiley and Sons, New York, NY.
181
188.
1998. The Human Complement System in Health and Disease. Marcel Dekker,
New York, NY.
189.
1999. Immunobiology: The Immune System in Health and Disease. Current
Biology and Garland Publishing, New York, NY.
190.
Reid, K. B., and R. R. Porter. 1976. Subunit composition and structure of
subcomponent C1q of the first component of human complement. Biochem J
155:19-23.
191.
Sim, R. B., and K. B. Reid. 1991. C1: molecular interactions with activating
systems. Immunol Today 12:307-311.
192.
Ollert, M. W., J. V. Kadlec, K. David, E. C. Petrella, R. Bredehorst, and C. W.
Vogel. 1994. Antibody-mediated complement activation on nucleated cells. A
quantitative analysis of the individual reaction steps. J Immunol 153:2213-2221.
193.
Fleming, S. D. 2006. Natural antibodies, autoantibodies and complement
activation in tissue injury. Autoimmunity 39:379-386.
194.
Boes, M., T. Schmidt, K. Linkemann, B. C. Beaudette, A. Marshak-Rothstein,
and J. Chen. 2000. Accelerated development of IgG autoantibodies and
autoimmune disease in the absence of secreted IgM. Proc Natl Acad Sci U S A
97:1184-1189.
195.
Briles, D. E., M. Nahm, K. Schroer, J. Davie, P. Baker, J. Kearney, and R.
Barletta. 1981. Antiphosphocholine antibodies found in normal mouse serum are
protective against intravenous infection with type 3 streptococcus pneumoniae. J
Exp Med 153:694-705.
196.
Haynes, B. F., J. Fleming, E. W. St Clair, H. Katinger, G. Stiegler, R. Kunert, J.
Robinson, R. M. Scearce, K. Plonk, H. F. Staats, T. L. Ortel, H. X. Liao, and S.
M. Alam. 2005. Cardiolipin polyspecific autoreactivity in two broadly
neutralizing HIV-1 antibodies. Science 308:1906-1908.
197.
Nauta, A. J., L. A. Trouw, M. R. Daha, O. Tijsma, R. Nieuwland, W. J.
Schwaeble, A. R. Gingras, A. Mantovani, E. C. Hack, and A. Roos. 2002. Direct
binding of C1q to apoptotic cells and cell blebs induces complement activation.
Eur J Immunol 32:1726-1736.
198.
Paidassi, H., P. Tacnet-Delorme, V. Garlatti, C. Darnault, B. Ghebrehiwet, C.
Gaboriaud, G. J. Arlaud, and P. Frachet. 2008. C1q binds phosphatidylserine and
likely acts as a multiligand-bridging molecule in apoptotic cell recognition. J
Immunol 180:2329-2338.
199.
Kishore, U., R. Ghai, T. J. Greenhough, A. K. Shrive, D. M. Bonifati, M. G.
Gadjeva, P. Waters, M. S. Kojouharova, T. Chakraborty, and A. Agrawal. 2004.
182
Structural and functional anatomy of the globular domain of complement protein
C1q. Immunol Lett 95:113-128.
200.
Botto, M., C. Dell'Agnola, A. E. Bygrave, E. M. Thompson, H. T. Cook, F. Petry,
M. Loos, P. P. Pandolfi, and M. J. Walport. 1998. Homozygous C1q deficiency
causes glomerulonephritis associated with multiple apoptotic bodies. Nat Genet
19:56-59.
201.
Cooper, N. R., and D. C. Morrison. 1978. Binding and activation of the first
component of human complement by the lipid A region of lipopolysaccharides. J
Immunol 120:1862-1868.
202.
Peitsch, M. C., T. J. Kovacsovics, J. Tschopp, and H. Isliker. 1987. Antibodyindependent activation of C1. II. Evidence for two classes of nonimmune
activators of the classical pathway of complement. J Immunol 138:1871-1876.
203.
Loos, M., B. Wellek, R. Thesen, and W. Opferkuch. 1978. Antibody-independent
interaction of the first component of complement with Gram-negative bacteria.
Infect Immun 22:5-9.
204.
Betz, S. J., and H. Isliker. 1981. Antibody-independent interactions between
Escherichia coli J5 and human complement components. J Immunol 127:17481754.
205.
Clas, F., and M. Loos. 1981. Antibody-independent binding of the first
component of complement (C1) and its subcomponent C1q to the S and R forms
of Salmonella minnesota. Infect Immun 31:1138-1144.
206.
Merino, S., M. M. Nogueras, A. Aguilar, X. Rubires, S. Alberti, V. J. Benedi, and
J. M. Tomas. 1998. Activation of the complement classical pathway (C1q
binding) by mesophilic Aeromonas hydrophila outer membrane protein. Infect
Immun 66:3825-3831.
207.
Alberti, S., G. Marques, S. Camprubi, S. Merino, J. M. Tomas, F. Vivanco, and
V. J. Benedi. 1993. C1q binding and activation of the complement classical
pathway by Klebsiella pneumoniae outer membrane proteins. Infect Immun
61:852-860.
208.
Mintz, C. S., P. I. Arnold, W. Johnson, and D. R. Schultz. 1995. Antibodyindependent binding of complement component C1q by Legionella pneumophila.
Infect Immun 63:4939-4943.
209.
Alberti, S., G. Marques, S. Hernandez-Alles, X. Rubires, J. M. Tomas, F.
Vivanco, and V. J. Benedi. 1996. Interaction between complement subcomponent
C1q and the Klebsiella pneumoniae porin OmpK36. Infect Immun 64:4719-4725.
183
210.
Loos, M., and F. Clas. 1987. Antibody-independent killing of gram-negative
bacteria via the classical pathway of complement. Immunol Lett 14:203-208.
211.
Stemmer, F., and M. Loos. 1985. Evidence for direct binding of the first
component of complement, C1, to outer membrane proteins from Salmonella
minnesota. Curr Top Microbiol Immunol 121:73-84.
212.
Wagenaar-Bos, I. G., and C. E. Hack. 2006. Structure and function of C1inhibitor. Immunol Allergy Clin North Am 26:615-632.
213.
Ziccardi, R. J., and N. R. Cooper. 1979. Active disassembly of the first
complement component, C-1, by C-1 inactivator. J Immunol 123:788-792.
214.
Chen, C. H., C. F. Lam, and R. J. Boackle. 1998. C1 inhibitor removes the entire
C1qr2s2 complex from anti-C1Q monoclonal antibodies with low binding
affinities. Immunology 95:648-654.
215.
Schumaker, V. N., P. Zavodszky, and P. H. Poon. 1987. Activation of the first
component of complement. Annu Rev Immunol 5:21-42.
216.
Tenner, A. J., R. J. Ziccardi, and N. R. Cooper. 1984. Antibody-independent C1
activation by E. coli. J Immunol 133:886-891.
217.
Endo, Y., M. Matsushita, and T. Fujita. 2007. Role of ficolin in innate immunity
and its molecular basis. Immunobiology 212:371-379.
218.
Fujita, T., M. Matsushita, and Y. Endo. 2004. The lectin-complement pathway-its role in innate immunity and evolution. Immunol Rev 198:185-202.
219.
Gros, P., F. J. Milder, and B. J. Janssen. 2008. Complement driven by
conformational changes. Nat Rev Immunol 8:48-58.
220.
Nicol, P. A., and P. J. Lachmann. 1973. The alternate pathway of complement
activation. The role of C3 and its inactivator (KAF). Immunology 24:259-275.
221.
Janssen, B. J., E. G. Huizinga, H. C. Raaijmakers, A. Roos, M. R. Daha, K.
Nilsson-Ekdahl, B. Nilsson, and P. Gros. 2005. Structures of complement
component C3 provide insights into the function and evolution of immunity.
Nature 437:505-511.
222.
Xu, W., S. P. Berger, L. A. Trouw, H. C. de Boer, N. Schlagwein, C. Mutsaers,
M. R. Daha, and C. van Kooten. 2008. Properdin binds to late apoptotic and
necrotic cells independently of C3b and regulates alternative pathway
complement activation. J Immunol 180:7613-7621.
223.
Hourcade, D. E. 2008. Properdin and complement activation: a fresh perspective.
Curr Drug Targets 9:158-164.
184
224.
Guo, R. F., and P. A. Ward. 2005. Role of C5a in inflammatory responses. Annu
Rev Immunol 23:821-852.
225.
Muller-Eberhard, H. J. 1986. The membrane attack complex of complement.
Annu Rev Immunol 4:503-528.
226.
Pangburn, M. K., V. P. Ferreira, and C. Cortes. 2008. Discrimination between
host and pathogens by the complement system. Vaccine 26 Suppl 8:I15-21.
227.
Law, S. K. A. a. R., K.B.M. 1995. Complement. IRL Press, Oxford.
228.
Soares, D. C., Barlow, P.N. . 2005. Complement Control Protein Modules in the
Regulators of Complement Activation. In Structural Biology of the Complement
System. D. M. a. J. D. Lambris, ed. CRC Press, Boca Raton, FL. 19-62.
229.
Campbell, R. D., S. K. Law, K. B. Reid, and R. B. Sim. 1988. Structure,
organization, and regulation of the complement genes. Annu Rev Immunol 6:161195.
230.
Schmidt, C. Q., A. P. Herbert, H. G. Hocking, D. Uhrin, and P. N. Barlow. 2008.
Translational mini-review series on complement factor H: structural and
functional correlations for factor H. Clin Exp Immunol 151:14-24.
231.
Schmidt, C. Q., A. P. Herbert, D. Kavanagh, C. Gandy, C. J. Fenton, B. S. Blaum,
M. Lyon, D. Uhrin, and P. N. Barlow. 2008. A new map of glycosaminoglycan
and C3b binding sites on factor H. J Immunol 181:2610-2619.
232.
Wu, J., Y. Q. Wu, D. Ricklin, B. J. Janssen, J. D. Lambris, and P. Gros. 2009.
Structure of complement fragment C3b-factor H and implications for host
protection by complement regulators. Nat Immunol 10:728-733.
233.
Zipfel, P. F., C. Skerka, J. Hellwage, S. T. Jokiranta, S. Meri, V. Brade, P.
Kraiczy, M. Noris, and G. Remuzzi. 2002. Factor H family proteins: on
complement, microbes and human diseases. Biochem Soc Trans 30:971-978.
234.
Blom, A. M., B. O. Villoutreix, and B. Dahlback. 2004. Complement inhibitor
C4b-binding protein-friend or foe in the innate immune system? Mol Immunol
40:1333-1346.
235.
Dahlback, B., C. A. Smith, and H. J. Muller-Eberhard. 1983. Visualization of
human C4b-binding protein and its complexes with vitamin K-dependent protein
S and complement protein C4b. Proc Natl Acad Sci U S A 80:3461-3465.
236.
Barlow, P. N., A. Steinkasserer, D. G. Norman, B. Kieffer, A. P. Wiles, R. B.
Sim, and I. D. Campbell. 1993. Solution structure of a pair of complement
modules by nuclear magnetic resonance. J Mol Biol 232:268-284.
185
237.
Ziccardi, R. J., B. Dahlback, and H. J. Muller-Eberhard. 1984. Characterization of
the interaction of human C4b-binding protein with physiological ligands. J Biol
Chem 259:13674-13679.
238.
Malm, J., M. Laurell, and B. Dahlback. 1988. Changes in the plasma levels of
vitamin K-dependent proteins C and S and of C4b-binding protein during
pregnancy and oral contraception. Br J Haematol 68:437-443.
239.
Sjoberg, A. P., L. A. Trouw, and A. M. Blom. 2009. Complement activation and
inhibition: a delicate balance. Trends Immunol 30:83-90.
240.
Blom, A. M., K. S. Nandakumar, and R. Holmdahl. 2009. C4b-binding protein
(C4BP) inhibits development of experimental arthritis in mice. Ann Rheum Dis
68:136-142.
241.
Bristow, C. L., and R. J. Boackle. 1986. Evidence for the binding of human serum
amyloid P component to Clq and Fab gamma. Mol Immunol 23:1045-1052.
242.
Agrawal, A., A. K. Shrive, T. J. Greenhough, and J. E. Volanakis. 2001.
Topology and structure of the C1q-binding site on C-reactive protein. J Immunol
166:3998-4004.
243.
Sjoberg, A. P., S. Nystrom, P. Hammarstrom, and A. M. Blom. 2008. Native,
amyloid fibrils and beta-oligomers of the C-terminal domain of human prion
protein display differential activation of complement and bind C1q, factor H and
C4b-binding protein directly. Mol Immunol 45:3213-3221.
244.
Tacnet-Delorme, P., S. Chevallier, and G. J. Arlaud. 2001. Beta-amyloid fibrils
activate the C1 complex of complement under physiological conditions: evidence
for a binding site for A beta on the C1q globular regions. J Immunol 167:63746381.
245.
Paidassi, H., P. Tacnet-Delorme, T. Lunardi, G. J. Arlaud, N. M. Thielens, and P.
Frachet. 2008. The lectin-like activity of human C1q and its implication in DNA
and apoptotic cell recognition. FEBS Lett 582:3111-3116.
246.
Navratil, J. S., S. C. Watkins, J. J. Wisnieski, and J. M. Ahearn. 2001. The
globular heads of C1q specifically recognize surface blebs of apoptotic vascular
endothelial cells. J Immunol 166:3231-3239.
247.
Barilla, M. L., and S. E. Carsons. 2000. Fibronectin fragments and their role in
inflammatory arthritis. Semin Arthritis Rheum 29:252-265.
248.
Bohnsack, J. F., A. J. Tenner, G. W. Laurie, H. K. Kleinman, G. R. Martin, and E.
J. Brown. 1985. The C1q subunit of the first component of complement binds to
laminin: a mechanism for the deposition and retention of immune complexes in
basement membrane. Proc Natl Acad Sci U S A 82:3824-3828.
186
249.
Gershov, D., S. Kim, N. Brot, and K. B. Elkon. 2000. C-Reactive protein binds to
apoptotic cells, protects the cells from assembly of the terminal complement
components, and sustains an antiinflammatory innate immune response:
implications for systemic autoimmunity. J Exp Med 192:1353-1364.
250.
Webb, J. H., A. M. Blom, and B. Dahlback. 2003. The binding of protein S and
the protein S-C4BP complex to neutrophils is apoptosis dependent. Blood Coagul
Fibrinolysis 14:355-359.
251.
Trouw, L. A., A. A. Bengtsson, K. A. Gelderman, B. Dahlback, G. Sturfelt, and
A. M. Blom. 2007. C4b-binding protein and factor H compensate for the loss of
membrane-bound complement inhibitors to protect apoptotic cells against
excessive complement attack. J Biol Chem 282:28540-28548.
252.
Elward, K., M. Griffiths, M. Mizuno, C. L. Harris, J. W. Neal, B. P. Morgan, and
P. Gasque. 2005. CD46 plays a key role in tailoring innate immune recognition of
apoptotic and necrotic cells. J Biol Chem 280:36342-36354.
253.
Ross, G. D., S. L. Newman, J. D. Lambris, J. E. Devery-Pocius, J. A. Cain, and P.
J. Lachmann. 1983. Generation of three different fragments of bound C3 with
purified factor I or serum. II. Location of binding sites in the C3 fragments for
factors B and H, complement receptors, and bovine conglutinin. J Exp Med
158:334-352.
254.
Ghiran, I., S. F. Barbashov, L. B. Klickstein, S. W. Tas, J. C. Jensenius, and A.
Nicholson-Weller. 2000. Complement receptor 1/CD35 is a receptor for mannanbinding lectin. J Exp Med 192:1797-1808.
255.
Klickstein, L. B., S. F. Barbashov, T. Liu, R. M. Jack, and A. Nicholson-Weller.
1997. Complement receptor type 1 (CR1, CD35) is a receptor for C1q. Immunity
7:345-355.
256.
Muller-Eberhard, H. J. 1988. Molecular organization and function of the
complement system. Annu Rev Biochem 57:321-347.
257.
Ross, G. D., and J. D. Lambris. 1982. Identification of a C3bi-specific membrane
complement receptor that is expressed on lymphocytes, monocytes, neutrophils,
and erythrocytes. J Exp Med 155:96-110.
258.
Shimaoka, M., J. Takagi, and T. A. Springer. 2002. Conformational regulation of
integrin structure and function. Annu Rev Biophys Biomol Struct 31:485-516.
259.
Pommier, C. G., S. Inada, L. F. Fries, T. Takahashi, M. M. Frank, and E. J.
Brown. 1983. Plasma fibronectin enhances phagocytosis of opsonized particles by
human peripheral blood monocytes. J Exp Med 157:1844-1854.
187
260.
Wright, S. D., and F. M. Griffin, Jr. 1985. Activation of phagocytic cells' C3
receptors for phagocytosis. J Leukoc Biol 38:327-339.
261.
Thornton, B. P., V. Vetvicka, M. Pitman, R. C. Goldman, and G. D. Ross. 1996.
Analysis of the sugar specificity and molecular location of the beta-glucanbinding lectin site of complement receptor type 3 (CD11b/CD18). J Immunol
156:1235-1246.
262.
Wright, S. D., and M. T. Jong. 1986. Adhesion-promoting receptors on human
macrophages recognize Escherichia coli by binding to lipopolysaccharide. J Exp
Med 164:1876-1888.
263.
Ehlenberger, A. G., and V. Nussenzweig. 1977. The role of membrane receptors
for C3b and C3d in phagocytosis. J Exp Med 145:357-371.
264.
Graham, I. L., J. B. Lefkowith, D. C. Anderson, and E. J. Brown. 1993. Immune
complex-stimulated neutrophil LTB4 production is dependent on beta 2 integrins.
J Cell Biol 120:1509-1517.
265.
Ravetch, J. V., and R. A. Clynes. 1998. Divergent roles for Fc receptors and
complement in vivo. Annu Rev Immunol 16:421-432.
266.
Wright, S. D., and S. C. Silverstein. 1983. Receptors for C3b and C3bi promote
phagocytosis but not the release of toxic oxygen from human phagocytes. J Exp
Med 158:2016-2023.
267.
Mastrangelo, A. M., T. M. Jeitner, and J. W. Eaton. 1998. Oleic acid increases
cell surface expression and activity of CD11b on human neutrophils. J Immunol
161:4268-4275.
268.
Newton, R. A., and N. Hogg. 1998. The human S100 protein MRP-14 is a novel
activator of the beta 2 integrin Mac-1 on neutrophils. J Immunol 160:1427-1435.
269.
Marth, T., and B. L. Kelsall. 1997. Regulation of interleukin-12 by complement
receptor 3 signaling. J Exp Med 185:1987-1995.
270.
Bohlson, S. S., D. A. Fraser, and A. J. Tenner. 2007. Complement proteins C1q
and MBL are pattern recognition molecules that signal immediate and long-term
protective immune functions. Mol Immunol 44:33-43.
271.
Kohl, J. 2006. Self, non-self, and danger: a complementary view. Adv Exp Med
Biol 586:71-94.
272.
Waggoner, S. N., M. W. Cruise, R. Kassel, and Y. S. Hahn. 2005. gC1q receptor
ligation selectively down-regulates human IL-12 production through activation of
the phosphoinositide 3-kinase pathway. J Immunol 175:4706-4714.
188
273.
Ghebrehiwet, B., and E. I. Peerschke. 2004. cC1q-R (calreticulin) and gC1qR/p33: ubiquitously expressed multi-ligand binding cellular proteins involved in
inflammation and infection. Mol Immunol 41:173-183.
274.
van den Berg, R. H., F. Prins, M. C. Faber-Krol, N. J. Lynch, W. Schwaeble, L.
A. van Es, and M. R. Daha. 1997. Intracellular localization of the human receptor
for the globular domains of C1q. J Immunol 158:3909-3916.
275.
McGreal, E., and P. Gasque. 2002. Structure-function studies of the receptors for
complement C1q. Biochem Soc Trans 30:1010-1014.
276.
Ogden, C. A., A. deCathelineau, P. R. Hoffmann, D. Bratton, B. Ghebrehiwet, V.
A. Fadok, and P. M. Henson. 2001. C1q and mannose binding lectin engagement
of cell surface calreticulin and CD91 initiates macropinocytosis and uptake of
apoptotic cells. J Exp Med 194:781-795.
277.
Fraser, D. A., M. Arora, S. S. Bohlson, E. Lozano, and A. J. Tenner. 2007.
Generation of inhibitory NFkappaB complexes and phosphorylated cAMP
response element-binding protein correlates with the anti-inflammatory activity of
complement protein C1q in human monocytes. J Biol Chem 282:7360-7367.
278.
Alper, C. A., N. Abramson, R. B. Johnston, Jr., J. H. Jandl, and F. S. Rosen. 1970.
Increased susceptibility to infection associated with abnormalities of complementmediated functions and of the third component of complement (C3). N Engl J
Med 282:350-354.
279.
Figueroa, J. E., and P. Densen. 1991. Infectious diseases associated with
complement deficiencies. Clin Microbiol Rev 4:359-395.
280.
Gross, G. N., S. R. Rehm, and A. K. Pierce. 1978. The effect of complement
depletion on lung clearance of bacteria. J Clin Invest 62:373-378.
281.
Cole, F. S., W. J. Matthews, Jr., T. H. Rossing, D. J. Gash, N. A. Lichtenberg, and
J. E. Pennington. 1983. Complement biosynthesis by human bronchoalveolar
macrophages. Clin Immunol Immunopathol 27:153-159.
282.
Strunk, R. C., D. M. Eidlen, and R. J. Mason. 1988. Pulmonary alveolar type II
epithelial cells synthesize and secrete proteins of the classical and alternative
complement pathways. J Clin Invest 81:1419-1426.
283.
Ferguson, J. S., J. J. Weis, J. L. Martin, and L. S. Schlesinger. 2004. Complement
protein C3 binding to Mycobacterium tuberculosis is initiated by the classical
pathway in human bronchoalveolar lavage fluid. Infect Immun 72:2564-2573.
284.
Liu, Y., Y. Endo, D. Iwaki, M. Nakata, M. Matsushita, I. Wada, K. Inoue, M.
Munakata, and T. Fujita. 2005. Human M-ficolin is a secretory protein that
activates the lectin complement pathway. J Immunol 175:3150-3156.
189
285.
Watford, W. T., A. J. Ghio, and J. R. Wright. 2000. Complement-mediated host
defense in the lung. Am J Physiol Lung Cell Mol Physiol 279:L790-798.
286.
Bolger, M. S., D. S. Ross, H. Jiang, M. M. Frank, A. J. Ghio, D. A. Schwartz, and
J. R. Wright. 2007. Complement levels and activity in the normal and LPS-injured
lung. Am J Physiol Lung Cell Mol Physiol 292:L748-759.
287.
Wessels, M. R., P. Butko, M. Ma, H. B. Warren, A. L. Lage, and M. C. Carroll.
1995. Studies of group B streptococcal infection in mice deficient in complement
component C3 or C4 demonstrate an essential role for complement in both innate
and acquired immunity. Proc Natl Acad Sci U S A 92:11490-11494.
288.
Brown, J. S., T. Hussell, S. M. Gilliland, D. W. Holden, J. C. Paton, M. R.
Ehrenstein, M. J. Walport, and M. Botto. 2002. The classical pathway is the
dominant complement pathway required for innate immunity to Streptococcus
pneumoniae infection in mice. Proc Natl Acad Sci U S A 99:16969-16974.
289.
Sjoholm, A. G., G. Jonsson, J. H. Braconier, G. Sturfelt, and L. Truedsson. 2006.
Complement deficiency and disease: an update. Mol Immunol 43:78-85.
290.
Rooijakkers, S. H., and J. A. van Strijp. 2007. Bacterial complement evasion. Mol
Immunol 44:23-32.
291.
Kraiczy, P., and R. Wurzner. 2006. Complement escape of human pathogenic
bacteria by acquisition of complement regulators. Mol Immunol 43:31-44.
292.
Lambris, J. D., D. Ricklin, and B. V. Geisbrecht. 2008. Complement evasion by
human pathogens. Nat Rev Microbiol 6:132-142.
293.
Joiner, K. A. 1988. Complement evasion by bacteria and parasites. Annu Rev
Microbiol 42:201-230.
294.
Wurzner, R. 1999. Evasion of pathogens by avoiding recognition or eradication
by complement, in part via molecular mimicry. Mol Immunol 36:249-260.
295.
Zipfel, P. F., R. Wurzner, and C. Skerka. 2007. Complement evasion of
pathogens: common strategies are shared by diverse organisms. Mol Immunol
44:3850-3857.
296.
Alberti, S., D. Alvarez, S. Merino, M. T. Casado, F. Vivanco, J. M. Tomas, and
V. J. Benedi. 1996. Analysis of complement C3 deposition and degradation on
Klebsiella pneumoniae. Infect Immun 64:4726-4732.
297.
Lerouge, I., and J. Vanderleyden. 2002. O-antigen structural variation:
mechanisms and possible roles in animal/plant-microbe interactions. FEMS
Microbiol Rev 26:17-47.
190
298.
Lukacova, M., I. Barak, and J. Kazar. 2008. Role of structural variations of
polysaccharide antigens in the pathogenicity of Gram-negative bacteria. Clin
Microbiol Infect 14:200-206.
299.
Roberts, I. S. 1996. The biochemistry and genetics of capsular polysaccharide
production in bacteria. Annu Rev Microbiol 50:285-315.
300.
Hill, S. A., and J. K. Davies. 2009. Pilin gene variation in Neisseria gonorrhoeae:
reassessing the old paradigms. FEMS Microbiol Rev 33:521-530.
301.
Merino, S., S. Vilches, R. Canals, S. Ramirez, and J. M. Tomas. 2005. A C1qbinding 40 kDa porin from Aeromonas salmonicida: cloning, sequencing, role in
serum susceptibility and fish immunoprotection. Microb Pathog 38:227-237.
302.
Schenkein, H. A., H. M. Fletcher, M. Bodnar, and F. L. Macrina. 1995. Increased
opsonization of a prtH-defective mutant of Porphyromonas gingivalis W83 is
caused by reduced degradation of complement-derived opsonins. J Immunol
154:5331-5337.
303.
Hong, Y. Q., and B. Ghebrehiwet. 1992. Effect of Pseudomonas aeruginosa
elastase and alkaline protease on serum complement and isolated components C1q
and C3. Clin Immunol Immunopathol 62:133-138.
304.
Oda, T., Y. Kojima, T. Akaike, S. Ijiri, A. Molla, and H. Maeda. 1990.
Inactivation of chemotactic activity of C5a by the serratial 56-kilodalton protease.
Infect Immun 58:1269-1272.
305.
Jagels, M. A., J. Travis, J. Potempa, R. Pike, and T. E. Hugli. 1996. Proteolytic
inactivation of the leukocyte C5a receptor by proteinases derived from
Porphyromonas gingivalis. Infect Immun 64:1984-1991.
306.
Beinrohr, L., J. Dobo, P. Zavodszky, and P. Gal. 2008. C1, MBL-MASPs and C1inhibitor: novel approaches for targeting complement-mediated inflammation.
Trends Mol Med 14:511-521.
307.
Liu, D., F. Lu, G. Qin, S. M. Fernandes, J. Li, and A. E. Davis, 3rd. 2007. C1
inhibitor-mediated protection from sepsis. J Immunol 179:3966-3972.
308.
Liu, D., S. Cai, X. Gu, J. Scafidi, X. Wu, and A. E. Davis, 3rd. 2003. C1 inhibitor
prevents endotoxin shock via a direct interaction with lipopolysaccharide. J
Immunol 171:2594-2601.
309.
Marr, N., R. A. Luu, and R. C. Fernandez. 2007. Bordetella pertussis binds human
C1 esterase inhibitor during the virulent phase, to evade complement-mediated
killing. J Infect Dis 195:585-588.
191
310.
Lathem, W. W., T. Bergsbaken, and R. A. Welch. 2004. Potentiation of C1
esterase inhibitor by StcE, a metalloprotease secreted by Escherichia coli
O157:H7. J Exp Med 199:1077-1087.
311.
Lathem, W. W., T. E. Grys, S. E. Witowski, A. G. Torres, J. B. Kaper, P. I. Tarr,
and R. A. Welch. 2002. StcE, a metalloprotease secreted by Escherichia coli
O157:H7, specifically cleaves C1 esterase inhibitor. Mol Microbiol 45:277-288.
312.
Ram, S., D. P. McQuillen, S. Gulati, C. Elkins, M. K. Pangburn, and P. A. Rice.
1998. Binding of complement factor H to loop 5 of porin protein 1A: a molecular
mechanism of serum resistance of nonsialylated Neisseria gonorrhoeae. J Exp
Med 188:671-680.
313.
Ram, S., A. K. Sharma, S. D. Simpson, S. Gulati, D. P. McQuillen, M. K.
Pangburn, and P. A. Rice. 1998. A novel sialic acid binding site on factor H
mediates serum resistance of sialylated Neisseria gonorrhoeae. J Exp Med
187:743-752.
314.
China, B., M. P. Sory, B. T. N'Guyen, M. De Bruyere, and G. R. Cornelis. 1993.
Role of the YadA protein in prevention of opsonization of Yersinia enterocolitica
by C3b molecules. Infect Immun 61:3129-3136.
315.
Ram, S., M. Cullinane, A. M. Blom, S. Gulati, D. P. McQuillen, B. G. Monks, C.
O'Connell, R. Boden, C. Elkins, M. K. Pangburn, B. Dahlback, and P. A. Rice.
2001. Binding of C4b-binding protein to porin: a molecular mechanism of serum
resistance of Neisseria gonorrhoeae. J Exp Med 193:281-295.
316.
Ram, S., F. G. Mackinnon, S. Gulati, D. P. McQuillen, U. Vogel, M. Frosch, C.
Elkins, H. K. Guttormsen, L. M. Wetzler, M. Oppermann, M. K. Pangburn, and P.
A. Rice. 1999. The contrasting mechanisms of serum resistance of Neisseria
gonorrhoeae and group B Neisseria meningitidis. Mol Immunol 36:915-928.
317.
Blom, A. M., A. Rytkonen, P. Vasquez, G. Lindahl, B. Dahlback, and A. B.
Jonsson. 2001. A novel interaction between type IV pili of Neisseria gonorrhoeae
and the human complement regulator C4B-binding protein. J Immunol 166:67646770.
318.
Prasadarao, N. V., A. M. Blom, B. O. Villoutreix, and L. C. Linsangan. 2002. A
novel interaction of outer membrane protein A with C4b binding protein mediates
serum resistance of Escherichia coli K1. J Immunol 169:6352-6360.
319.
Wooster, D. G., R. Maruvada, A. M. Blom, and N. V. Prasadarao. 2006.
Logarithmic phase Escherichia coli K1 efficiently avoids serum killing by
promoting C4bp-mediated C3b and C4b degradation. Immunology 117:482-493.
320.
Nordstrom, T., A. M. Blom, A. Forsgren, and K. Riesbeck. 2004. The emerging
pathogen Moraxella catarrhalis interacts with complement inhibitor C4b binding
192
protein through ubiquitous surface proteins A1 and A2. J Immunol 173:45984606.
321.
Berggard, K., E. Johnsson, F. R. Mooi, and G. Lindahl. 1997. Bordetella pertussis
binds the human complement regulator C4BP: role of filamentous hemagglutinin.
Infect Immun 65:3638-3643.
322.
Fearon, D. T. 1978. Regulation by membrane sialic acid of beta1H-dependent
decay-dissociation of amplification C3 convertase of the alternative complement
pathway. Proc Natl Acad Sci U S A 75:1971-1975.
323.
Pangburn, M. K., D. C. Morrison, R. D. Schreiber, and H. J. Muller-Eberhard.
1980. Activation of the alternative complement pathway: recognition of surface
structures on activators by bound C3b. J Immunol 124:977-982.
324.
Galanos, C., and O. Luderitz. 1976. The role of the physical state of
lipopolysaccharides in the interaction with complement. High molecular weight as
prerequisite for the expression of anti-complementary activity. Eur J Biochem
65:403-408.
325.
Grossman, N., and L. Leive. 1984. Complement activation via the alternative
pathway by purified Salmonella lipopolysaccharide is affected by its structure but
not its O-antigen length. J Immunol 132:376-385.
326.
Taylor, P. W. 1974. Sensitivity of some smooth strains of Escherichia coli to the
bactericidal action of normal human serum. J Clin Pathol 27:626-629.
327.
Reynard, A. M., and M. E. Beck. 1976. Plasmid-mediated resistance to the
bactericidal effects of normal rabbit serum. Infect Immun 14:848-850.
328.
Fierer, J., and F. Finley. 1979. Lethal effect of complement and lysozyme on
polymyxin-treated, serum-resistant gram-negative bacilli. J Infect Dis 140:581589.
329.
Joiner, K. A., C. H. Hammer, E. J. Brown, R. J. Cole, and M. M. Frank. 1982.
Studies on the mechanism of bacterial resistance to complement-mediated killing.
I. Terminal complement components are deposited and released from Salmonella
minnesota S218 without causing bacterial death. J Exp Med 155:797-808.
330.
Joiner, K. A., C. H. Hammer, E. J. Brown, and M. M. Frank. 1982. Studies on the
mechanism of bacterial resistance to complement-mediated killing. II. C8 and C9
release C5b67 from the surface of Salmonella minnesota S218 because the
terminal complex does not insert into the bacterial outer membrane. J Exp Med
155:809-819.
193
331.
Schiller, N. L., R. A. Hatch, and K. A. Joiner. 1989. Complement activation and
C3 binding by serum-sensitive and serum-resistant strains of Pseudomonas
aeruginosa. Infect Immun 57:1707-1713.
332.
Vogel, U., S. Hammerschmidt, and M. Frosch. 1996. Sialic acids of both the
capsule and the sialylated lipooligosaccharide of Neisseria meningitis serogroup
B are prerequisites for virulence of meningococci in the infant rat. Med Microbiol
Immunol 185:81-87.
333.
Pramoonjago, P., M. Kaneko, T. Kinoshita, E. Ohtsubo, J. Takeda, K. S. Hong, R.
Inagi, and K. Inoue. 1992. Role of TraT protein, an anticomplementary protein
produced in Escherichia coli by R100 factor, in serum resistance. J Immunol
148:827-836.
334.
Pausa, M., V. Pellis, M. Cinco, P. G. Giulianini, G. Presani, S. Perticarari, R.
Murgia, and F. Tedesco. 2003. Serum-resistant strains of Borrelia burgdorferi
evade complement-mediated killing by expressing a CD59-like complement
inhibitory molecule. J Immunol 170:3214-3222.
335.
Rautemaa, R., H. Rautelin, P. Puolakkainen, A. Kokkola, P. Karkkainen, and S.
Meri. 2001. Survival of Helicobacter pylori From complement lysis by binding of
GPI-anchored protectin (CD59). Gastroenterology 120:470-479.
336.
Rautemaa, R., G. A. Jarvis, P. Marnila, and S. Meri. 1998. Acquired resistance of
Escherichia coli to complement lysis by binding of glycophosphoinositolanchored protectin (CD59). Infect Immun 66:1928-1933.
337.
Walport, M. J. 2001. Complement. First of two parts. N Engl J Med 344:10581066.
338.
Pangburn, M. K., and H. J. Muller-Eberhard. 1980. Relation of putative thioester
bond in C3 to activation of the alternative pathway and the binding of C3b to
biological targets of complement. J Exp Med 152:1102-1114.
339.
van Lookeren Campagne, M., C. Wiesmann, and E. J. Brown. 2007. Macrophage
complement receptors and pathogen clearance. Cell Microbiol 9:2095-2102.
340.
Ellis, J., P. C. Oyston, M. Green, and R. W. Titball. 2002. Tularemia. Clin
Microbiol Rev 15:631-646.
341.
Hollis, D. G., R. E. Weaver, A. G. Steigerwalt, J. D. Wenger, C. W. Moss, and D.
J. Brenner. 1989. Francisella philomiragia comb. nov. (formerly Yersinia
philomiragia) and Francisella tularensis biogroup novicida (formerly Francisella
novicida) associated with human disease. J Clin Microbiol 27:1601-1608.
342.
Eigelsbach, H. T., Braun, W., Herring, R.D. 1951. Studies on the variation of
Bacterium tularense. J Bacteriol 61:557-569.
194
343.
2000. Tularemia. In Control of Communicable Diseases Manual. J. Chin, ed.
American Public Health Association, Washington, DC. 532-535.
344.
Cross, J. T., Penn, R.L. 2000. Francisella tularensis (tularemia). In Principles and
Practice of Infectious Diseases. G. L. Mandell, et al., ed. Churchill Livingstone,
Philadelphia, PA. 2393-2402.
345.
Maier, T. M., M. S. Casey, R. H. Becker, C. W. Dorsey, E. M. Glass, N. Maltsev,
T. C. Zahrt, and D. W. Frank. 2007. Identification of Francisella tularensis
Himar1-based transposon mutants defective for replication in macrophages. Infect
Immun 75:5376-5389.
346.
Horwitz, M. A., and S. C. Silverstein. 1980. Influence of the Escherichia coli
capsule on complement fixation and on phagocytosis and killing by human
phagocytes. J Clin Invest 65:82-94.
347.
Hunninghake, G. W., J. E. Gadek, O. Kawanami, V. J. Ferrans, and R. G. Crystal.
1979. Inflammatory and immune processes in the human lung in health and
disease: evaluation by bronchoalveolar lavage. Am J Pathol 97:149-206.
348.
Berger, M. 1990. Third component of human complement: C3. Methods Enzymol
184:619-628.
349.
Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain for detecting
lipopolysaccharides in polyacrylamide gels. Anal Biochem 119:115-119.
350.
McLendon, M. K., M. A. Apicella, and L. A. Allen. 2006. Francisella tularensis:
Taxonomy, Genetics, and Immunopathogenesis of a Potential Agent of
Biowarfare. Annu Rev Microbiol.
351.
Reckseidler-Zenteno, S. L., R. DeVinney, and D. E. Woods. 2005. The capsular
polysaccharide of Burkholderia pseudomallei contributes to survival in serum by
reducing complement factor C3b deposition. Infect Immun 73:1106-1115.
352.
Stevens, P., S. N. Huang, W. D. Welch, and L. S. Young. 1978. Restricted
complement activation by Escherichia coli with the K-1 capsular serotype: a
possible role in pathogenicity. J Immunol 121:2174-2180.
353.
Cunnion, K. M., J. C. Lee, and M. M. Frank. 2001. Capsule production and
growth phase influence binding of complement to Staphylococcus aureus. Infect
Immun 69:6796-6803.
354.
Ben Nasr, A., and G. R. Klimpel. 2008. Subversion of complement activation at
the bacterial surface promotes serum resistance and opsonophagocytosis of
Francisella tularensis. J Leukoc Biol.
195
355.
James, K. 1982. Complement: activation, consequences, and control. Am J Med
Technol 48:735-742.
356.
Morgan, B. P. 2000. Measurement of Complement hemolytic Activity,
Generation of Complement-Depleted Sera, and Production of Hemolytic
Intermediates. In Complement Methods and Protocols. B. P. Morgan, ed. Humana
Press Inc, Totowa, NJ. 61-72.
357.
Abdullah, M., I. Nepluev, G. Afonina, S. Ram, P. Rice, W. Cade, and C. Elkins.
2005. Killing of dsrA mutants of Haemophilus ducreyi by normal human serum
occurs via the classical complement pathway and is initiated by immunoglobulin
M binding. Infect Immun 73:3431-3439.
358.
Edwards, J. L., E. J. Brown, K. A. Ault, and M. A. Apicella. 2001. The role of
complement receptor 3 (CR3) in Neisseria gonorrhoeae infection of human
cervical epithelia. Cell Microbiol 3:611-622.
359.
Edwards, J. L., E. J. Brown, S. Uk-Nham, J. G. Cannon, M. S. Blake, and M. A.
Apicella. 2002. A co-operative interaction between Neisseria gonorrhoeae and
complement receptor 3 mediates infection of primary cervical epithelial cells. Cell
Microbiol 4:571-584.
196