Download Isolation and expression of recombinant antibody

Journal of Immunological Methods 276 (2003) 185 – 196
www.elsevier.com/locate/jim
Recombinant Technology
Isolation and expression of recombinant antibody fragments to the
biological warfare pathogen Brucella melitensis
Andrew Hayhurst a,b,1, Scott Happe a,1,2, Robert Mabry a,c, Zephyr Koch a,
Brent L. Iverson a,c,*, George Georgiou a,b,d,*
a
Institute for Cellular and Molecular Biology, University of Texas at Austin, Austin, TX 78712-1095, USA
b
Chemical Engineering Department, University of Texas at Austin, Austin, TX 78712-1062, USA
c
Department of Chemistry and Biochemistry, University of Texas at Austin, Austin, TX 78712, USA
d
Biomedical Engineering Program, University of Texas at Austin, Austin, USA
Received 28 August 2002; received in revised form 15 January 2003; accepted 15 January 2003
Abstract
Brucella melitensis is a highly infectious animal pathogen able to cause a recurring debilitating disease in humans and
is therefore high on the list of biological warfare agents. Immunoglobulin genes from mice immunized with gammairradiated B. melitensis strain 16M were used to construct a library that was screened by phage display against similarly
prepared bacteria. The selected phage particles afforded a strong enzyme-linked immunosorbent assay (ELISA) signal
against gamma-irradiated B. melitensis cells. However, extensive efforts to express the respective single chain antibody
variable region fragment (scFv) in soluble form failed due to: (i) poor solubility and (ii) in vivo degradation of the c-myc
tag used for the detection of the recombinant antibodies. Both problems could be addressed by: (i) fusing a human kappa
light chain constant domain (Ck) chain to the scFv to generate single chain antibody fragment (scAb) antibody fragments
and (ii) by co-expression of the periplasmic chaperone Skp. While soluble, functional antibodies could be produced in this
manner, phage-displaying scFvs or scAbs were still found to be superior ELISA reagents for immunoassays, due to the
large signal amplification afforded by anti-phage antibodies. The isolated phage antibodies were shown to be highly
Abbreviations: BCA, bicinchoninic acid; BSA, bovine serum albumin; Ck, human kappa light chain constant domain; Cm,
chloramphenicol resistance gene; ELISA, enzyme-linked immunosorbent assay; FPLC, fine performance liquid chromatography; HRP,
horseradish peroxidase; IMAC, immobilized metal affinity chromatography; g3p, M13 gene III protein; g8p, M13 gene VIII protein; IPTG,
isopropyl h-D-thiogalactopyranoside; LPS, lipopolysaccharide; phAb, phage displayed antibody fragment; MALDI/TOF, matrix-assisted laser
desorption ionization/time of flight; M-MLV, Moloney-murine leukemia virus; Mab, monoclonal antibody; PBS, phosphate-buffered saline;
RT-PCR, reverse transcriptase-polymerase chain reaction; scAb, single chain antibody fragment; scFv, single chain antibody variable region
fragment; SDS-PAGE, sodium dodecyl sulfate polyacrylamide electrophoresis; TB, terrific broth; Tween-20, polyoxyethylenesorbitan
monolaurate.
* Corresponding authors. B.L. Iverson, Department of Chemistry and Biochemistry, University of Texas at Austin, Austin, TX 78712-1062,
USA. Tel.: +1-512-471-5053; fax: +1-512-471-8696. G. Georgiou, Chemical Engineering Department, University of Texas at Austin, Austin,
TX 78712-1095, USA. Tel.: +1-512-471-6975; fax: +1-512-471-7963.
E-mail addresses: [email protected] (B.L. Iverson), [email protected] (G. Georgiou).
1
These authors contributed equally to this work.
2
Present address: Bio Crest/Stratagene, Cedar Creek, TX 78612, USA.
0022-1759/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0022-1759(03)00100-5
186
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
specific to B. melitensis and did not recognize Yersinia pseudotuberculosis in contrast to the existing diagnostic monoclonal YST 9.2.1.
D 2003 Elsevier Science B.V. All rights reserved.
Keywords: Biological agent detection; Brucella melitensis; Phage display; scFv; 9E10; c-myc tag
1. Introduction
The Brucellae are Gram-negative facultative intracellular pathogens that are responsible for zoonoses
worldwide (Corbel, 1997; Boschiroli et al., 2001;
Moreno and Moriyon, 2002). Infection causes abortion in wild and domestic animals resulting in
enormous financial losses especially in developing
parts of the world. Contact with infected animals can
result in human infection, which while rarely fatal,
can cause long-term recurring illness marked by
debilitating malaise and fever. Brucellae are easily
dispersed in aerosols, have a low dose to infectivity
ratio, and can survive extended periods of time
separated from their host. As a result of these
features, the Brucellae, particularly Brucella melitensis, Brucella abortus, and Brucella suis, were among
the first organisms to be considered as offensive
biological warfare weapons (Franz et al., 1997).
Indeed, these three of six very closely related species
remain on the Centers for Disease Control and
Prevention select agent list.
Rapid approaches for detecting Brucellae bypass
the requirement for lengthy and fastidious culture
conditions (Doern, 2000; Sumerkan et al., 2001) and
include real-time PCR (Sreevatsan et al., 2000; Redkar et al., 2001), fluorescence whole cell nucleic acid
hybridization (Fernandez-Lago et al., 2000), or immunoassays utilizing antibodies able to specifically recognize Brucellae cells. Current immunoassay methods
employ antibodies to lipopolysaccharide (LPS), the
immunodominant surface antigen (Berger et al.,
1969). However, the immunologically distinguishing
feature in the LPS of the Brucellae, the O:9 epitope,
is also shared by certain Yersiniae (Caroff et al.,
1984). As a result, many existing anti-Brucellae sera
exhibit cross-reactivity with the Yersiniae (Bundle et
al., 1984). It is important to be able to distinguish
between these two organisms as double infections can
be common (Kittelberger et al., 1995) and since
certain Yersiniae are also considered potential biological warfare agents.
Recombinant antibody fragments are gaining
ground as supplements to monoclonal and polyclonal
animal sera in the biological agent diagnostic and
detection arena (Emanuel et al., 2000). Genetic
manipulation can be employed to tailor affinity and
specificity (Chen et al., 1999; Short et al., 2001),
stability (Reiter et al., 1994), avidity (Iliades et al.,
1997), or to generate chimeras to reporter molecules
(Kerschbaumer et al., 1997; Pearce et al., 1997; Casey
et al., 2000). Antibody fragments can be readily
expressed in Escherichia coli allowing low cost
production and purification, an important advantage
for many applications including proteomics (Hayhurst
and Georgiou, 2001).
Phage display is routinely used for the isolation of
antibody fragments to a wide variety of antigens
including haptens, carbohydrates, peptides, proteins,
and surface antigens of intact mammalian cells (Griffiths et al., 1993; Hoogenboom et al., 1998, 1999).
Panning of phage display libraries on intact pathogens
has also been employed to generate diagnostic and
therapeutic antibodies to a wide range of disease
causing agents including bacteria. Single chain antibody variable region fragment (scFv) specific for
Gram-negative, complement-resistant Moraxella
catarrhalis were isolated by competitive selection of
a semi-synthetic library with complement-sensitive
strains (Boel et al., 1998). Similarly, scFv specific to
a surface component of Gram-positive, pathogenic
Streptococcus suis were generated by subtractive
panning (de Greeff et al., 2000). Finally, scFv directed
toward surface associated antigens of Chlamydia
trachomatis, including some host cell components,
were derived from a naı̈ve scFv library (Lindquist et
al., 2002).
Here, we used phage display to isolate scFv to
gamma-irradiated B. melitensis strain 16M. After
panning a biased library on immobilized antigen,
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
two phage clones were shown to be specific for the
antigen and not cross-reactive with Yersiniae.
Efforts to express scFv antibodies revealed several
complications with common epitope tags used for
detection of scFv. These problems were successfully
addressed by converting the scFv into the single
chain antibody fragment (scAb) format (McGregor
et al., 1994; Hayhurst, 2000) and co-expression of
the chaperone Skp (Bothmann and Pluckthun, 1998;
Hayhurst and Harris, 1999) to alleviate protein
aggregation in the bacterial periplasm. This strategy
will be useful for the preparative production of
other recombinant antibodies including those specific for biological warfare agents especially since
scAb have recently been shown to have improved
pharmacokinetic properties over scFv (Maynard et
al., 2002).
2. Materials and methods
2.1. Antigens
Inactivated B. melitensis strain 16M was provided
by Edgewood Chemical and Biological Center
(ECBC, Aberdeen Proving Ground, MD). Prior to
inactivation, cells were grown in soy trypticase broth,
washed, and treated by one of three methods: cobalt
gamma-irradiation (3.0 106 rad total dose). Concentration of cells prior to inactivation was 4.5 109
cfu/ml; inactivation was confirmed by culture of 1.0
ml of suspension on solid media. The following
Yersinia species were provided by ECBC to test
for scFv cross-reactivity: Yersinia enterocolitica
(ATCC 23715), Yersinia pseudotuberculosis (ATCC
6905, 29833, and NADC 5561, 5560), Yersinia
rhodei (ATCC 43380), Yersinia ruckeri (ATCC
29473).
2.2. Antibodies
Mouse monoclonal antibody YST 9.2.1 specific for
LPS was provided by ECBC. Anti-mouse-HRP was
from BioRad (Hercules, CA) and anti-M13 HRP was
from Amersham Pharmacia (Piscataway, NJ). AntiFLAG M1, anti-His6 clone-1, anti-myc clone 9E10
and anti-His6 clone-1 HRP conjugate were all from
Sigma (St. Louis, MO).
187
2.3. Immunizations
Five-week-old female BALB/c mice were purchased from Jackson Laboratories (Bar Harbor,
ME). Mice were housed at the Texas Center for
Infectious Diseases (TCID, San Antonio, TX) on a
12h light/dark cycle and given food and water ad
libitum. All animal handling procedures were performed according to TCID guidelines. For the first
immunization, bacterial suspensions in saline were
mixed with an equal volume of Freund’s complete
adjuvant (Sigma) to form a stable emulsion. Mice
were anesthetized via inhalation of Metofane and
injected intramuscularly at 0.1 ml per site with each
animal receiving a total of 1 109 bacteria per
immunization. The mice were boosted with bacterial
cells emulsified in Freund’s incomplete adjuvant 1, 4
and 8 weeks later. Control mice were inoculated with
saline in place of bacterial suspension following the
same schedule. Six days after the final boost, mice
were anesthetized, bled via cardiac puncture, and
sacrificed for tissue harvest. Spleens were removed
and placed into RNAlaterk storage buffer on ice
(Ambion, Austin, TX) prior to processing and RNA
extraction.
2.4. scFv library construction
Spleens were teased apart in ice-cold PBS, transferred to guanidinium denaturation solution, and disrupted using a Dounce homogenizer. Total RNA was
extracted using the ToTALLY RNAk kit (Ambion)
with twice the recommended volume of denaturation
buffer to reduce the viscosity of the extract. Total
RNA (3.7 Ag/50 Al reaction) was subjected to RT-PCR
using the RetroScript kit (Ambion) with oligo-dT18
and M-MLV reverse transcriptase either at 42 jC (VH
samples) or 54 jC (VL samples). PCR and scFv
assembly was performed using standard conditions
(Krebber et al., 1997) to generate a library of 5 105
clones which was subsequently rescued with helper
phage M13KO7.
2.5. Phage display and panning
Gamma-irradiated B. melitensis 16M (6 107 cfu/
ml), in a total volume of 4 ml PBS, was coated
overnight in MaxiSorp Immunotubes (Nunc, Roches-
188
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
ter, NY) at 4 jC. The tube was blocked with 1% BSA/
PBS for 2 h at 37 jC. Phage particles (109) were
added in BSA/PBS, incubated for 1 h at room temperature on an end-over-end turntable, followed by overnight at room temperature without agitation. The
immunotube was washed 15 times with PBS/0.1%
Tween-20, followed by 15 times with PBS alone.
Bound phage were eluted for 10 min in 1 ml 100
mM triethylamine, neutralized with 0.5ml 1M Tris –
HCl pH 7.5, and used to infect exponentially growing
TG1 cells. Infected cells were plated overnight on 2 TY/30 Ag/ml chloramphenicol + 1% glucose. The next
day, cells were scraped from the plates, grown to
exponential phase, and rescued with M13KO7 helper
phage. Infected cells were grown overnight in 2 TY/
30 Ag/ml chloramphenicol + 1% glucose and 30 Ag/
ml kanamycin and 1 mM IPTG at 30 jC. Phage were
purified by polyethylene glycol/NaCl precipitation
and resuspended in a total volume of 2 ml PBS.
One milliliter was used for the next round of panning,
for a total of four rounds.
2.6. Enzyme-linked immunosorbent assay (ELISA)
Wells of MaxiSorp microtitre plates (Nunc) were
coated with 100 Al the same dilution of gammairradiated B. melitensis suspension as used for panning. 100 Al of OD600 = 1.0 suspensions of the other
bacterial antigens were used to coat. Standard
ELISA blocking and washing procedures were used
throughout for both phage and antibody fragments
(Chen et al., 2001) with antibodies used at the
manufacturer’s recommended dilutions. Buffer containing 1 mM calcium chloride was used throughout for FLAG-M1 antibody experiments to satisfy
the requirement for calcium ions. Data points are
the average of duplicate dilutions with each experiment being done at least twice. Antibody fragment
and phAb ELISAs were also carried out with BSAcoated wells as control for each dilution and these
blanks subtracted to counter background-binding
effects.
2.7. Antibody fragment expression vector construction
Primers AHX77 (5V-TATATAGGCCTCGGGGGCCGAATTCGCGGCCGCAGTCGACCATCATC AT C A C C AT C A C G - 3 V) a n d A H X 7 8 ( 5 V-
GCGCAAGCTTGCGATATCACTATGCGGCGCGCCCATTCAGATCCTCTTC-3V) were used for PCR
amplification of the His6 c-myc tag region from
pHEN2 (Griffin library protocol, Medical Research
Council, UK) which in turn was used to replace the
His6 tag in pAK400 (Krebber et al., 1997) via SfiI/
HindIII to create pMoPac1. Subsequently, the chloramphenicol resistance gene (Cm) of pMoPac1 was
replaced with an ampicillin resistance gene. Kunkel
template of pMoPac1 was generated by co-transfection of vector and M13KO7 replicative form into
CJ236Fem, a derivative of E. coli CJ236 cured of
the chloramphenicol resistant FV episome by repeated
growth in the absence of chloramphenicol. Kunkel
mutagenesis using both AHX71 (5V-CCAGTGATTTTTTTCTCCATTTAAATCTTCCTTAGCTCCTGAAAATC-3V) and AHX72 (5V-GGGCACCAATAACTGCCTATTTAAATTTACGCCCCGCCCTGCC-3V) introduced unique SwaI restriction
sites either side of the pMoPac1 Cm gene enabling
facile uni-directional resistance gene exchange employing the vectors own Cm promoter. In this manner,
the h-lactamase gene was amplified from pUC19
using AHX73 (5V-ATATATATTTAAATGAGTATTCAACATTTCCG-3V) and AHX74 (5V-GCGCGCATTTAAATTTACCAATGCTTAATCAGTG-3V),
introduced via SwaI sites and clones selected upon
media containing 200 Ag/ml ampicillin designated
pMoPac10.
A cistron encoding the periplasmic chaperone Skp
was placed downstream of the scFv expression cassette. Primers AHX155 (5V-ATATAAGCTTTAAGGAGATATATATATGAAAAAGTGGTTATTAGCTGCA-3V) and AHX156 (5V- GTGGGGATCCGT G A C G C A G TA G C G G TA A A C G G C A G A C A A A A A A A AT G T C G C A C A AT G T G C G C C AT T T T T C A C T T C A C A G G T C T TAT TAT TTAACCTGTTTCAGTACGTC-3V) were used to
amplify the skp gene from pHELP1 (Hayhurst and
Harris, 1999) and the product inserted into pMoPac10
via HindIII and BamHI to create pMoPac12.
A human kappa constant domain (Ck) (McGregor
et al., 1994) was inserted downstream of the scFv to
enable scAb (scFv-Ck) expression. A pel-specific
primer and AHX88 (5V-CGCGCGTCGACTGACTCTCCGCGGTTG-3V) were used to amplify a scAb
from pIMS100 (Hayhurst, 2000) and the Ck domain
isolated inserted into MoPac10 and MoPac12 via NotI
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
and SalI to create pMoPac15 and 16 respectively. The
constructs are schematically shown in Fig. 1 with
detail of the linker from pMoPac16 from which all
constructs can be deconvoluted.
The super-short g3p gene in plasmid pAK100
(Krebber et al., 1997) was amplified using AHX104
( 5 V- TATA G G C G C G C C G C ATA G A C T G G TGGTGGCTCTGGTTCC-3V) and AHX105 (5VGCGCAAGCTTGCGATATCAAGACTCCTTATTACGC-3V) and mobilized into pMoPac10, 12, 15,
and 16 via AscI and HindIII to form phage display
equivalents pMoPac23, 24, 25 and 26, respectively.
C10 and H1 scFv were mobilized between vectors
using SfiI. All constructs employing PCR amplified
products were sequenced through that region after
being first isolated.
189
2.8. Antibody fragment expression, analysis and
purification
Cell growth, fractionation and analyses were performed essentially as described previously (Chen et
al., 2001) except that the cells were grown in glucosefree TB and induced with 1 mM IPTG. Antibody
fragments were isolated from mid-scale 400 ml cultures with 0.5 ml of nickel charged 6FF resin (Amersham Pharmacia) and eluted in 1 ml of which 0.5 ml
was then applied to a Superdex 200 HR10/30 column
(Amersham Pharmacia) and gel filtration performed
using an Äkta FPLC system (Amersham Pharmacia).
Fractions (0.5 ml) were collected and those representing monomeric scAb pooled and quantified using
micro-BCA (Pierce, Rockford, IL). Proteins used
to calibrate the column were from the molecular
weight gel filtration standard kits from Amersham
Pharmacia.
2.9. MALDI/TOF mass spectrometry of purified scAb
Fig. 1. (a) Schematic diagram of the expression cassettes employed
to rationalize and solve the poor expression characteristics of the
antibody fragments C10 and H1. The parental vector is pAK400
(Krebber et al., 1997) where the antibody fragments are targeted to
the E. coli periplasm using a pelB leader sequence. Abbreviations:
F, FLAG tag; H, His6 tag; M, myc tag; scFv single chain antibody
variable region fragment; Ck, human kappa light chain constant
domain; Skp, seventeen kilo Dalton protein, the periplasmic
chaperone. (b) Detail of the pMoPac16 vector region C-terminal
to scFv inserts with pertinent restriction sites highlighted. S/D
represents a Shine Dalgarno sequence. The inclusion of NotI allows
NcoI/NotI scFv mobilization from older pHEN style libraries as well
as newer unidirectional SfiI based assemblies (Krebber et al., 1997).
The arrow after the His6 tag indicates the likely C terminus of
DscAb which lacks the c-myc tag.
Mass spectrometry was performed on a Voyager
MALDI/TOF mass spectrometer (PerSeptive Biosystems, Framingham, MA) in the Protein Microanalysis
Facility of the University of Texas at Austin. Protein
samples were diluted into freshly prepared saturated
sinapinic acid dissolved in 50% acetonitrile, 0.3%
TFA. Sample aliquots of 0.5 or 1.0 ml were spotted
onto stainless steel sample plates and spectra were
collected by averaging 10 –20 laser shots. Samples
were irradiated with a 377-nm nitrogen laser (Laser
Science, Franklin, MA) attenuated and focused on the
sample target using the built-in software. Ions were
accelerated with a deflection voltage of 30 kV. Ions
were differentiated according to their m/z using a
time-of-flight mass analyzer.
3. Results and discussion
3.1. Isolation of Brucella-specific antibody fragments
Mice were immunized with gamma-irradiated B.
melitensis 16M to provide a source of biased immunoglobulin genes for library construction. ELISA
performed with sera collected after 8 weeks indicated
that the immunized mice mounted a strong immune
190
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
response to gamma-irradiated B. melitensis yet no
reactivity to E. coli, with serum from control salineinjected mice showing no reactivity against either
bacterium (data not shown).
A scFv library was constructed from immunoglobulin genes amplified by RT-PCR from the spleens of
two mice. Subsequently, the scFv genes were fused to
M13 g3p and the resulting phage display library was
panned on immobilized gamma-irradiated B. melitensis. Following binding of the phage, the cells were
incubated overnight to reduce potential for selection
of LPS-binders exhibiting high off-rates (Deng et al.,
1994). Monoclonal phage ELISA of 94 clones from
the round 4 population identified eight clones giving
ELISA signals approximately five times above background. These were sequenced to reveal two unique
clones, designated C10 and H1 (sequences submitted
to Genbank).
3.2. Expression of Brucella-specific antibody fragments
The C10 and H1 scFv genes were subcloned from
the phage display vector pAK100 to the periplasmic
expression vector pAK400 (Krebber et al., 1997).
pAK400 was designed for antibody fragment expression in E. coli and consists of a pUC replicon with a
tightly regulated lac promoter (Krebber et al., 1996)
that, upon induction, results in relatively high scFv
expression levels thanks to the presence of a strong
ribosomal-binding site. However, no signal was
obtained in ELISA assays using gamma-irradiated B.
melitensis 16M as capture antigen and osmotic shockates from E. coli BL21 cells containing pAK400 C10
or H1 scFv as the analytes. The capture assay used
anti-FLAG-M1 or anti-His6 as primary antibodies and
anti-mouse-HRP as secondary antibody-conjugate.
Fig. 2. Gamma-irradiated B. melitensis antibody fragment capture ELISA of antibody fragments in osmotic shockates of normalized smallscale expressions for C10 or H1 expressed as scFv from pMoPac10, scFv with Skp from pMoPac12, scAb from pMoPac15 and scAb
with Skp from pMoPac16. Shockates were twofold serially diluted in PBS/BSA. Antibody fragment detection was using anti-FLAG M1
serum, anti-myc 9E10 serum or anti-His6 serum followed by anti-mouse HRP conjugate. Development was 2 h for C10 and overnight for
H1.
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
We reasoned that the absence of ELISA signal could
be due to a combination of two factors: (i) the
occlusion of the N-terminal FLAG peptide epitope
and (ii) the relatively low affinity of anti-His6 sera for
the C-terminally fused hexahistidine tag. To address
these issues, a c-myc peptide epitope that is recognized by the 9E10 monoclonal antibody was placed at
the C terminus of the scFv adjacent to the hexahistidine tag in the style of the commonly used display
vector pHEN2 (Fig. 1). Still, no ELISA signal could
be detected from osmotic shock fractions of cells
expressing C10 or H1 scFv antibodies fused to the
C-terminal c-myc tag. Co-expression of the periplasmic chaperone Skp which is known to improve the
solubility of scFv antibodies (Bothmann and Pluckthun, 1998; Hayhurst and Harris, 1999) did not
remedy the situation. Only the fusion of a human
Ck domain between the scFv and His6-myc tags
resulted in detectable ELISA signals for both C10
and H1 clones using secondary anti-His6 sera or antimyc sera but not anti-FLAG M1 sera (Fig. 2).
Western blot analysis (Fig. 3) of scFv and scAb
fragments from the osmotic shock and spheroplast
fractions (the latter also contains aggregated mem-
191
brane-bound proteins which centrifuge together with
spheroplasts) revealed that:
(a) N-terminal FLAG: The FLAG epitope is detectable in both the scFv and the scAb fragments
indicating that the N terminus of the proteins is
not subject to degradation but is most likely
occluded during the ELISA assay. Co-expression
of Skp increases the production of soluble C10
modestly and that of H1 dramatically in both scFv
and scAb formats. A doublet is visible for each
format with a lower molecular weight species
DscFv predominant and DscAb a relatively minor
species.
(b) C-terminal proximal His6 tag: The His6 tag is not
subject to degradation and is present on scFv and
scAb. The His6 tag is also present on DscFv and
DscAb.
(c) C-terminal myc tag: The absence of a band
corresponding to soluble scFv indicates that the
c-myc epitope is subject to extensive degradation.
Conversion of the scFv into the scAb format
partially alleviated this problem and co-expression
of Skp resulted in a further increase in the
Fig. 3. Western blot of osmotic shockates (o) as used in ELISA in Fig. 2 and the remaining spheroplast fractions (s). + indicates Skp
coexpression and indicates no Skp supplement. Electrophoresis was on a 12% Laemmli gel and the Western blots were probed sequentially
with three different antisera to establish a correlate between ELISA signal and immuno-reactive protein content. Following each probing was a
2 10-min room temperature stripping step (Kaufmann et al., 1987) with overnight reblocking with milk. Exposures represent approx. 5s.
Approximate positions of molecular weight markers are shown in kDa and were deduced from the positions of Rainbow markers (Amersham
Pharmacia) on the Immobilon-P membrane. Anti-h-lactamase serum was used to confirm the fractionation had succeeded on a separate blot
(data not shown).
192
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
production of soluble material. c-myc-specific
signal is not evident at locations expected for
DscFv or DscAb indicating these species have lost
the c-myc epitope.
The data in Fig. 3 indicate that the fusion of the Ck
domain greatly enhanced the ability to detect the
antibody protein as well as the overall expression
level especially when combined with Skp co-expression. Indeed, following IMAC and FPLC purification
scAb yields were far in excess of scFv yields (Fig. 4)
allowing us to generate several hundred micrograms
of scAb fragment per liter of pMoPac16 expression
cultures. Yields of the corresponding scFv were consistently at least 10-fold lower.
3.3. Mass spectral analysis of IMAC-FPLC purified
scAb
The monomeric scAb isolated by FPLC was found
to migrate as a doublet in Coomassie stained SDSPAGE gels (Fig. 5a). The abundance of the lower
molecular weight species increases over time suggesting copurification of a contaminating protease. Both
proteins possessed the FLAG tag and His6 tag yet
only the higher molecular weight species possessed
the c-myc tag as determined by Western blotting (data
Fig. 4. FPLC chromatography of IMAC eluates resulting from midscale expression culture of C10 scAb from pMoPac16 (solid line)
and C10 scFv from pMoPac12 (dotted line). Molecular weight
markers: (1) aldolase—158 kDa; (2) bovine serum albumin—67
kDa; (3) ovalbumin—43 kDa; (4) chymotrypsin—25 kDa. The
expression and purification experiments were carried out three times
and the same pattern observed in that scAb material by far exceeded
scFv material.
Fig. 5. (a) SDS-PAGE of IMAC/FPLC purified monomeric C10
scAb protein revealing the two species scAb and DscAb following
Coomassie blue staining. (b) MALDI/TOF analysis of the scAb
sample to elucidate the site at which the c-myc epitope is lost.
not shown). As the entire c-myc sequence is not
required for recognition by anti-c-myc monoclonal
9E10 (Hilpert et al., 2001), it was pertinent to determine precisely where the myc tag was being ‘‘lost’’.
Therefore, the scAb was analysed by MALDI/TOF
spectrometry yielding the masses of the doublet components (Fig. 5b). Since the N-terminal DYKD
sequence is present as determined by anti-FLAG M1
reactivity, BioEdit (Hall, 1999) was used to deduce
the masses of C-terminal truncations of scAb indicating that DscAb most likely ended immediately after
the last histidine residue of the His6 as shown in Fig.
1b. The predicted mass of 39,689.58 Da is within
experimental limits of the MALDI/TOF spectrometry
determination of 39,667 F 30 Da. We have since
found several scFv isolated from a semi-synthetic
library (the Griffin library) and expressed from
pHEN2 can be subject to the same degradation pattern
indicating that the cleavage is not a vector-specific
artifact but rather a global problem with early tag
design (data not shown).
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
193
(data not shown). Unlike purified antigens, the target
to which C10 and H1 are specific may be relatively
rare and/or occluded and so the one-to-one stoichiometry of secondary antibody HRP conjugate to
detection tag on the scAb may be limiting sensitivity.
However, when scFv are displayed as g3p fusions, the
secondary antibody– HRP conjugate specific for the
major coat protein of M13, g8p, is able to amplify the
scFv/antigen binding event by virtue of each phage
possessing several thousand copies of the g8p protein.
The close packing of g8p (Makowski, 1993; Kneissel
et al., 1999) and bivalency of the IgG based anti-M13
HRP conjugate will most likely prevent each and
every copy of g8p being bound by conjugate. Yet
even if 1 in 10 g8p were bound it would result in a
100-fold signal amplification.
Having established that phage-borne scFv were the
superior detection reagent and having elevated periplasmic expression levels of scFv by conversion to
Fig. 6. ELISA detection of gamma-irradiated B. melitensis 16M
using (a) IMAC purified scAb protein or (b) phage displayed scFv
in pAK100 for the two clones C10 and H1. scAb binding employed
anti-His6-HRP for detection while phage binding employed anti
g8p-HRP. Color development was allowed to proceed for 15 min.
3.4. Detection of B. melitensis with the recombinant
antibodies
Purified scAb protein was used in ELISA detection
of B. melitensis alongside phage displayed scFv from
pAK100. scAb protein capture detection employed
anti-His6-HRP conjugate and phAb capture detection
employed anti-M13 HRP. We consistently noticed
improved signals for phage displayed antibodies compared to their soluble counterparts even though the
latter were present in the assay at a significant molar
excess (Fig. 6). The use of various anti-His sera and
polyclonal anti-Ck sera failed to improve upon the
sensitivity of scAb mediated B. melitensis detection
Fig. 7. Clone C10 phage ELISA detection of gamma-irradiated B.
melitensis using phage display vectors pMoPac23 displaying scFv,
pMoPac24 displaying scFv with Skp coexpression, pMoPac25
displaying scAb and pMoPac26 displaying scAb with Skp
coexpression. pAK100 displaying scFv was grown fresh and used
as a positive control each time. The monoclonal phage stocks were
prepared by M13KO7 superinfection of TG-1 bearing the relevant
vector and display allowed to proceed for 18 h at 25 jC with 1 mM
IPTG in TB plus 2% glucose (plus 30 Ag/ml chloramphenicol for
pAK100 and 200 Ag/ml ampicillin for pMoPac vectors). Each
vector was rescued, titered and subjected to ELISA on three
separate occasions with the same trend in evidence, i.e.
(phAbscFv + skp)>(phAbscFv)=(phAbscAb + skp)>(phAbscAb)>
pAK100.
194
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
scAb, we sought to combine the two approaches as
periplasmic solubility of antibody fragments can be
proportional to the efficiency of their display (Coia et
al., 1997; Bothmann and Pluckthun, 1998, 2000).
However, we consistently observed that phage displaying scFv were almost an order of magnitude more
effective in detecting B. melitensis than phage displaying scAb (Fig. 7). Skp co-expression improved
the capacity for detection even further. It appears that
the larger size of the scAb (42 kDa as opposed to 27
kDa for an scFv) and its propensity to dimerise (Fig. 4
and McGregor et al., 1994) may be decreasing the
frequency with which scAb-g3p fusions can assemble
onto virions.
The specificity of the phAb-C10 antibody towards
B. melitensis was evaluated. Reactivity was observed
only with gamma-irradiated B. melitensis sample and
not with any of the Yersiniae tested (Fig. 8a). This is
in contrast to an existing diagnostic monoclonal antibody YST9.2.1 which reacts not only with B. melitensis but Y. pseudotuberculosis (NADC 5561) by
virtue of the shared O:9 LPS epitope for which YST
9.2.1 is specific (Fig. 8b). Thus, phAb C10 is clearly
able to distinguish species sharing the same major
LPS antigen. This is most useful in the diagnostic
setting as bacteria sharing common LPS epitopes are
often co-isolated and difficult to distinguish.
4. Conclusions
Fig. 8. Cross-reactivity ELISA of (a) phAbscFv (C10) and (b)
mouse monoclonal YST 0.2.1 for strains of Yersinia. Detection of
phAb capture utilized anti-M13-HRP and detection of YST 0.2.1
was revealed with anti-mouse HRP conjugate. Solid circles, Y.
pseudotuberculosis (ATCC 6905); open circles, Y. pseudotuberculosis (NADC 5560); solid triangles, Y. pseudotuberculosis (ATCC
29833); open triangles, Y. pseudotuberculosis (NADC 5561); solid
squares, Y. enterocolitica (ATCC 23715); Y. rhodei (ATCC 43380);
solid diamonds, Y. ruckeri (ATCC 29473); open diamonds, gammairradiated B. melitensis.
Recombinant antibody methodology was used to
generate antibody fragments specific for the potential
bioweapon B. melitensis. Specific washing steps were
included in the phage panning procedure to eliminate
phage that might be cross-reactive with strains
expressing the same dominant LPS epitope. The isolated Brucella-specific scFvs were poorly expressed in
E. coli, and they displayed a propensity to lose the
commonly used C-terminal c-myc peptide tag due to
proteolytic degradation. Gene fusion with a human
immunoglobulin kappa constant domain to produce
scAb constructs as well as chaperone co-expression
using Skp were employed to elevate significantly the
production of purified antibody fragment.
Despite these improvements, we consistently observed that phAb were two orders of magnitude more
sensitive in capture ELISA formats compared to the
homologous scAb constructs. Indeed, similar phAb
signal amplification has recently been shown to be of
use in detecting low levels of Bacillus spores (Zhou et
al., 2002) and can doubtless be further improved by
polyvalent display (Rondot et al., 2001). Another
technical advantage of the phAb format is that production of phage particles is significantly more convenient than recombinant scFv or even scAb
expression and isolation from E. coli.
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
The isolated antibody fragment referred to as C10
was able to discriminate between B. melitensis and
other bacteria that share the same major surface LPS
antigen. C10 did not appear to be LPS specific and we
are currently elucidating the nature of the antigen.
This same strategy of avoiding LPS-specific antibodies while using phAbs to detect relatively rare but
unique epitopes should allow rapid definition of
bacteria within mixed populations and be of application to environmental sensing.
Acknowledgements
We thank the following for generously providing
reagents: Kevin O’Connell ECBC, Maryland for
antisera and bacteria; Greg Winter laboratory, Cambridge, England for pHEN2; Andreas Plückthun,
Zurich, Switzerland for the pAK series of vectors.
We are indebted to Mitch Magee of the Texas Center
for Infectious Diseases, San Antonio, TX, for the
animal procedures. We thank Klaus Linse and staff of
the Core Facility, Institute for Cellular and Molecular
Biology, University of Texas at Austin for the protein
microanalysis work.
This work was supported by a Beckman Research
Technology Initiative Award (Project No. 26750695),
Department of Defense Measurement and Signal
Intelligence DOD DAAD17-01-D-001 with the
United States Army Research Laboratory and the
National Institutes of Health HL66880-02 Subcontract
GMO-010207 (to GG).
References
Berger, F.M., Fukui, G.M., Ludwig, B.J., Rosselet, J.P., 1969. Increased host resistance to infection elicited by lipopolysaccharides from Brucella abortus. Proc. Soc. Exp. Biol. Med. 131, 1376.
Boel, E., Bootsma, H., de Kruif, J., Jansze, M., Klingman, K.L.,
van Dijk, H., Logtenberg, T., 1998. Phage antibodies obtained
by competitive selection on complement-resistant Moraxella
(Branhamella) catarrhalis recognize the high-molecular-weight
outer membrane protein. Infect. Immun. 66, 83.
Boschiroli, M.L., Foulongne, V., O’Callaghan, D., 2001. Brucellosis: a worldwide zoonosis. Curr. Opin. Microbiol. 4, 58.
Bothmann, H., Pluckthun, A., 1998. Selection for a periplasmic
factor improving phage display and functional periplasmic expression. Nat. Biotechnol. 16, 376.
Bothmann, H., Pluckthun, A., 2000. The periplasmic Escherichia
195
coli peptidylprolyl cis,trans-isomerase FkpA: I. Increased functional expression of antibody fragments with and without cisprolines. J. Biol. Chem. 275, 17100.
Bundle, D.R., Gidney, M.A., Perry, M.B., Duncan, J.R., Cherwonogrodzky, J.W., 1984. Serological confirmation of Brucella
abortus and Yersinia enterocolitica O:9 O-antigens by monoclonal antibodies. Infect. Immun. 46, 389.
Caroff, M., Bundle, D.R., Perry, M.B., 1984. Structure of the Ochain of the phenol-phase soluble cellular lipopolysaccharide of
Yersinia enterocolitica serotype O:9. Eur. J. Biochem. 139, 195.
Casey, J.L., Coley, A.M., Tilley, L.M., Foley, M., 2000. Green
fluorescent antibodies: novel in vitro tools. Protein Eng. 13, 445.
Chen, G., Dubrawsky, I., Mendez, P., Georgiou, G., Iverson, B.L.,
1999. In vitro scanning saturation mutagenesis of all the specificity determining residues in an antibody binding site. Protein
Eng. 12, 349.
Chen, G., Hayhurst, A., Thomas, J.G., Harvey, B.R., Iverson, B.L.,
Georgiou, G., 2001. Isolation of high-affinity ligand-binding
proteins by periplasmic expression with cytometric screening
(PECS). Nat. Biotechnol. 19, 537.
Coia, G., Ayres, A., Lilley, G.G., Hudson, P.J., Irving, R.A., 1997.
Use of mutator cells as a means for increasing production levels
of a recombinant antibody directed against Hepatitis B. Gene
201, 203.
Corbel, M.J., 1997. Brucellosis: an overview. Emerg. Infect. Dis.
3, 213.
de Greeff, A., van Alphen, L., Smith, H.E., 2000. Selection of
recombinant antibodies specific for pathogenic Streptococcus
suis by subtractive phage display. Infect. Immun. 68, 3949.
Deng, S.J., MacKenzie, C.R., Sadowska, J., Michniewicz, J.,
Young, N.M., Bundle, D.R., Narang, S.A., 1994. Selection of
antibody single-chain variable fragments with improved carbohydrate binding by phage display. J. Biol. Chem. 269, 9533.
Doern, G.V., 2000. Detection of selected fastidious bacteria. Clin.
Infect. Dis. 30, 166.
Emanuel, P.A., Dang, J., Gebhardt, J.S., Aldrich, J., Garber, E.A.,
Kulaga, H., Stopa, P., Valdes, J.J., Dion-Schultz, A., 2000. Recombinant antibodies: a new reagent for biological agent detection. Biosens. Bioelectron. 14, 751.
Fernandez-Lago, L., Vallejo, F.J., Trujillano, I., Vizcaino, N., 2000.
Fluorescent whole-cell hybridization with 16S rRNA-targeted
oligonucleotide probes to identify Brucella spp. by flow cytometry. J. Clin. Microbiol. 38, 2768.
Franz, D.R., Jahrling, P.B., Friedlander, A.M., McClain, D.J., Hoover, D.L., Bryne, W.R., Pavlin, J.A., Christopher, G.W., Eitzen
Jr., E.M. 1997. Clinical recognition and management of patients
exposed to biological warfare agents. JAMA 278, 399.
Griffiths, A.D., Malmqvist, M., Marks, J.D., Bye, J.M., Embleton,
M.J., McCafferty, J., Baier, M., Holliger, K.P., Gorick, B.D.,
Hughes-Jones, N.C., et al., 1993. Human anti-self antibodies
with high specificity from phage display libraries. EMBO J.
12, 725.
Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95.
Hayhurst, A., 2000. Improved expression characteristics of singlechain Fv fragments when fused downstream of the Escherichia
196
A. Hayhurst et al. / Journal of Immunological Methods 276 (2003) 185–196
coli maltose-binding protein or upstream of a single immunoglobulin-constant domain. Protein Expr. Purif. 18, 1.
Hayhurst, A., Georgiou, G., 2001. High-throughput antibody isolation. Curr. Opin. Chem. Biol. 5, 683.
Hayhurst, A., Harris, W.J., 1999. Escherichia coli skp chaperone
coexpression improves solubility and phage display of singlechain antibody fragments. Protein Expr. Purif. 15, 336.
Hilpert, K., Hansen, G., Wessner, H., Kuttner, G., Welfle, K., Seifert, M., Hohne, W., 2001. Anti-c-myc antibody 9E10: epitope
key positions and variability characterized using peptide spot
synthesis on cellulose. Protein Eng. 14, 803.
Hoogenboom, H.R., de Bruine, A.P., Hufton, S.E., Hoet, R.M.,
Arends, J.W., Roovers, R.C., 1998. Antibody phage display
technology and its applications. Immunotechnology 4, 1.
Hoogenboom, H.R., Lutgerink, J.T., Pelsers, M.M., Rousch, M.J.,
Coote, J., Van Neer, N., De Bruine, A., Van Nieuwenhoven,
F.A., Glatz, J.F., Arends, J.W., 1999. Selection-dominant and
nonaccessible epitopes on cell-surface receptors revealed by
cell-panning with a large phage antibody library. Eur. J. Biochem. 260, 774.
Iliades, P., Kortt, A.A., Hudson, P.J., 1997. Triabodies: single chain
Fv fragments without a linker form trivalent trimers. FEBS Lett.
409, 437.
Kaufmann, S.H., Ewing, C.M., Shaper, J.H., 1987. The erasable
Western blot. Anal. Biochem. 161, 89.
Kerschbaumer, R.J., Hirschl, S., Kaufmann, A., Ibl, M., Koenig, R.,
Himmler, G., 1997. Single-chain Fv fusion proteins suitable as
coating and detecting reagents in a double antibody sandwich
enzyme-linked immunosorbent assay. Anal. Biochem. 249, 219.
Kittelberger, R., Hilbink, F., Hansen, M.F., Ross, G.P., Joyce, M.A.,
Fenwick, S., Heesemann, J., Wolf-Watz, H., Nielsen, K., 1995.
Serological cross-reactivity between Brucella abortus and Yersinia enterocolitica 0:9 II the use of Yersinia outer proteins for
the specific detection of Yersinia enterocolitica infections in
ruminants. Vet. Microbiol. 47, 271.
Kneissel, S., Queitsch, I., Petersen, G., Behrsing, O., Micheel, B.,
Dubel, S., 1999. Epitope structures recognised by antibodies
against the major coat protein (g8p) of filamentous bacteriophage fd (Inoviridae). J. Mol. Biol. 288, 21.
Krebber, A., Burmester, J., Pluckthun, A., 1996. Inclusion of an
upstream transcriptional terminator in phage display vectors
abolishes background expression of toxic fusions with coat protein g3p. Gene 178, 71.
Krebber, A., Bornhauser, S., Burmester, J., Honegger, A., Willuda,
J., Bosshard, H.R., Pluckthun, A., 1997. Reliable cloning of
functional antibody variable domains from hybridomas and
spleen cell repertoires employing a reengineered phage display
system. J. Immunol. Methods 201, 35.
Lindquist, E.A., Marks, J.D., Kleba, B.J., Stephens, R.S., 2002.
Phage-display antibody detection of Chlamydia trachomatis-associated antigens. Microbiology 148, 443.
Makowski, L., 1993. Structural constraints on the display of foreign
peptides on filamentous bacteriophages. Gene 128, 5.
Maynard, J.A., Maassen, C.B., Leppla, S.H., Brasky, K., Patterson,
J.L., Iverson, B.L., Georgiou, G., 2002. Protection against anthrax toxin by recombinant antibody fragments correlates with
antigen affinity. Nat. Biotechnol. 20, 597.
McGregor, D.P., Molloy, P.E., Cunningham, C., Harris, W.J., 1994.
Spontaneous assembly of bivalent single chain antibody fragments in Escherichia coli. Mol. Immunol. 31, 219.
Moreno, E., Moriyon, I., 2002. Brucella melitensis: a nasty bug
with hidden credentials for virulence. Proc. Natl. Acad. Sci.
U. S. A. 99, 1.
Pearce, L.A., Oddie, G.W., Coia, G., Kortt, A.A., Hudson, P.J.,
Lilley, G.G., 1997. Linear gene fusions of antibody fragments
with streptavidin can be linked to biotin labelled secondary
molecules to form bispecific reagents. Biochem. Mol. Biol.
Int. 42, 1179.
Redkar, R., Rose, S., Bricker, B., DelVecchio, V., 2001. Real-time
detection of Brucella abortus, Brucella melitensis and Brucella
suis. Mol. Cell. Probes 15, 43.
Reiter, Y., Brinkmann, U., Jung, S.H., Lee, B., Kasprzyk, P.G.,
King, C.R., Pastan, I., 1994. Improved binding and antitumor
activity of a recombinant anti-erbB2 immunotoxin by disulfide
stabilization of the Fv fragment. J. Biol. Chem. 269, 18327.
Rondot, S., Koch, J., Breitling, F., Dubel, S., 2001. A helper phage
to improve single-chain antibody presentation in phage display.
Nat. Biotechnol. 19, 75.
Short, M.K., Jeffrey, P.D., Demirjian, A., Margolies, M.N., 2001. A
single H:CDR3 residue in the anti-digoxin antibody 26-10 modulates specificity for C16-substituted digoxin analogs. Protein
Eng. 14, 287.
Sreevatsan, S., Bookout, J.B., Ringpis, F., Perumaalla, V.S., Ficht,
T.A., Adams, L.G., Hagius, S.D., Elzer, P.H., Bricker, B.J.,
Kumar, G.K., Rajasekhar, M., Isloor, S., Barathur, R.R., 2000.
A multiplex approach to molecular detection of Brucella abortus and/or Mycobacterium bovis infection in cattle. J. Clin. Microbiol. 38, 2602.
Sumerkan, B., Gokahmetoglu, S., Esel, D., 2001. Brucella detection in blood: comparison of the BacT/Alert standard aerobic
bottle, BacT/Alert FAN aerobic bottle and BacT/Alert enhanced
FAN aerobic bottle in simulated blood culture. Clin. Microbiol.
Infect. 7, 369.
Zhou, B., Wirsching, P., Janda, K.D., 2002. Human antibodies
against spores of the genus Bacillus: a model study for detection
of and protection against anthrax and the bioterrorist threat.
Proc. Natl. Acad. Sci. U. S. A. 99, 5241.