Download Variations in Surface Protein Composition Associated

Journal of General Microbiology (1979), 114,305-312. Printed in Great Britain
305
Variations in Surface Protein Composition Associated with
Virulence Properties in Opacity Types of Neisseria gonorrhoeae
By P. R. L A M B D E N , J. E. HECKELS, L. T. JAMES A N D P. J. W A T T
Department of Microbiology, Southampton University Medical School,
Southampton General Hospital, Southampton SO9 4 X Y
(Received 3 1 January 1979)
The biological properties of a series of opacity variants of Neisseria gonorrhoeae P9 have
been examined. A novel protein, designated protein IId* (mol. wt 28850), was identified
within the set of heat-modifiable surface proteins previously reported. All variants producing
extra outer membrane proteins were less sensitive to the bactericidal action of serum than the
prototype transparent strain, with protein Ira* (mol. wt 28 500) being associated with
increased resistance. The production of a different protein, protein 11* (mol. wt 29000),
was correlated with resistance to low molecular weight antimicrobial agents (penicillin,
fusidic acid, Cu2+,Zn2+). Increased adhesion to human buccal epithelial cells was demonstrated in all variants that produced extra surface proteins. These variants did not show
increased binding to hexyl- and phenyl-substituted Sepharose gels suggesting that hydrophobic interaction was not responsible for their cohesive properties. The prototype strain
lacking additional proteins demonstrated the greatest binding to erythrocytes, indicating
that adhesion to buccal cells and red blood cells is mediated by different mechanisms. One
variant producing protein IIa* showed increased association with leukocytes, whereas
another producing protein IIb * showed decreased association with leukocytes. These
results show that the heat-modifiable surface proteins are important virulence attributes of
the gonococcus : this must be considered in the selection of strains for vaccine trials.
INTRODUCTION
The characteristic symptoms of gonorrhoea are a result of the ability of the gonococcus
to attach to and subsequently invade non-squamous mucosal surfaces. This tissue penetration invokes local inflammation but dissemination of the infection is determined by the
susceptibility of the invading gonococcus to antibody and complement in the host’s serum.
Studies are now in progress to explain these virulence attributes in terms of the surface
properties of the gonococcus.
Pili are responsible for an increased attachment of gonoccoci to mucosal cells (Swanson,
1973) and the recently reported capsule appears to enhance resistance to phagocytosis
(Richardson & Sadoff, 1977)and to the bactericidal effect of normal human serum (Ward et al.,
1978). However, the role of outer membrane components in virulence has been less easy
to establish since variants lacking specific components have not been readily available.
Nevertheless, lipopolysaccharide has been shown to be the major target for the bactericidal
effect of natural antibody and complement (Ward et al., 1978). The other major antigens of
the gonococcal surface are the outer membrane proteins, two of which have been isolated
and purified from Neisseria gonorrhoeae strain P9 (Heckels, 1977). Recent studies have
shown that further outer membrane proteins are present in strains which have been typed
according to colonial opacity (Swanson, 1978). A similar pattern was obtained with
0022-1287/79/0000-8583$02.00 @ 1979 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
306
P. R. L A M B D E N A N D OTHERS
variants of strain P9 (Lambden & Heckels, 1979). Knowledge of the role that the extra
surface proteins may play in the potential virulence of gonococci is scarce but their variation
in clinical isolates obtained throughout the menstrual cycle has been correlated with their
increased susceptibility to the elevated levels of proteolytic enzymes present in secretions
during the luteal phase of the cycle (James & Swanson, 1978). Little information, however,
is available on what, if any, selective advantage is conferred on the gonococcus by the
acquisition of the additional outer membrane proteins. In this study, we report the use of a
series of variants of strain P9 with defined differences in outer membrane protein composition
which can be related to significant differences in gonococcal attachment, antibiotic sensitivity
and susceptibility to serum killing.
METHODS
Bacterial strains and growth conditions. Neisseria gonorrhoeae strain P9 was grown on clear typing medium
(Swanson, 1978) enriched with a supplement similar in composition to commercial Isovitalex (BBL) except
that L-cystine was omitted. Opacity variants were selected using the illumination system of Swanson (1978)
as described by Lambden & Heckels (1979). Colonies of defined opacity were purified by single colony
isolation and stored in liquid nitrogen.
In experiments which required SH-labelled gonococci, D- [ l-3H]glucose (500 mCi mm01-~; The Radiochemical Centre, Amersham) was added to the medium to a final concentration of 25 pCi ml-l.
Sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis. Outer membrane preparations obtained
from the variants by extraction with 0.2 M-lithium acetate (Heckels, 1977) were subjected to SDS-polyacrylamide gel electrophoresis essentially as described by Lambden & Heckels (1979), but using a linear
gradient of 10 to 20% (w/v) acrylamide in the separating gel.
Standardization of gonococcal suspensions. Bacteria were harvested from solid medium into complete
Dulbecco phosphate buffered saline pH 7.4 (PBS ; Oxoid). To remove gonococcal aggregates, the suspension
was homogenized by vortex mixing and centrifuged at 500 g for 10 min for serum killing experiments or at
150 g for 1 min for attachment experiments. A sample of the suspension obtained was diluted 10-fold with
PBS and the absorbance was read in a Pye Unicam SP1800 spectrophotometer at 560nm. Under these
conditions an A 5 6 0 of 0.1 corresponded to an original concentration of 2-8x lo8 colony forming units
(c.f.u.) ml-l. The original suspension was diluted in the appropriate suspension medium to the required
concentrat ion.
Determination of susceptibility to killing by normal human serum. Suspensions of gonococcal variants were
added to a fourfold dilution of normal human serum in PBS to give a final cell density of 1 x lo4c.f.u. ml-l.
The suspensions were gently shaken at 37 "C and samples were withdrawn at 15 min intervals for determination of the number of survivors by viable counting. Results were plotted as the logarithm of the percentage
of survivors against sampling time and the best straight line fit was obtained by linear regression analysis.
Determination of susceptibility to antimicrobial agents. Standard suspensions of the variant gonococci were
prepared at lo6 c.f.u. ml-l. Using a phage-typing machine (P.H.L.S., Colindale, London) loopfuls of each
suspension were transferred to agar plates of typing medium containing a range of concentrations of the
appropriate antimicrobial agent. Plates were incubated for 24 h at 36 "C in air containing 5 9; (v/v) CO,
and then scored for growth. The minimum inhibitory concentration (m.i.c.) was taken as the lowest concentration of agent which abolished visible growth.
Determination of gonococcal attachment. Buccal epithelial cells were scraped with wooden spatulas from
several volunteers, pooled, suspended in PBS and centrifuged at 500 g for 5 min. The pellet was washed
three times in PBS and finally suspended at a packed cell volume of 10 yo in Basal Minimum Eagle (Modified)
medium with Hanks' salts (Flow Laboratories) buffered with 0.05 ~-N-2-hydroxyethylpiperazine-N'-2
ethanesulphonic acid (Hepes) to pH 7.4 (TCM). 3H-Labelled gonococci suspended in the same medium
(1 ml) at a concentration of 2.8 x lo7c.f.u. ml-1 were added to an equal volume of buccal cell suspension
and incubated with gentle mixing at 37 "C for 1 h. Samples of the suspension (4x 0.4 ml) were then layered
on to cushions (7 ml) of 6 % (w/v) dextran in 0.9% (w/v) NaCl (Dextraven; Fisons) in polystyrene centrifuge tubes. The tubes were centrifuged at 500 g for 2 min to deposit cells and adherent gonococci. The
contents of the top half of each tube were aspirated to remove most of the non-attached gonococci and the
remainder was frozen at -70 "C. The bottom 1 cm portion was then cut from the tube. After thawing, the
cells and adherent bacteria were transferred to scintillation vials and counted in a Packard Tri-Carb scintillation counter using PCS scintillation cocktail (Amersham-Searle). The percentage of gonococci adhering was
determined as: [lo0 x (mean c.p.m. in pellet)- (c.p.m. pelleted in control without buccals)]/[total c.p.m.
applied to tube].
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
307
Surface proteins and virulence of gonococci
Attachment of gonococci to erythrocytes and to hydrophobic Sepharose gels was determined in a similar
mannx using 10% (v/v) and 20 % (v/v) suspensions, respectively. Before scintillation counting, erythrocytes
were lysed by the addition of 1 M-NaOH (100 pl) and decolourized by incubation with 30% (w/v) hydrogen
peroxide (100 4) at 20 "C overnight.
Determination of gonococcal-leukocyte association. The ability of variant gonococci to attach to human
polymorphonuclear (PMN) leukocytes was determined using the method of King & Swanson (1978).
Leukocytes were prepared from peripheral blood by gelatin precipitation of erythrocytes and washed, and
~
6.5 x lo5 cells were pipetted on to clean glass coverslips. After allowing 1 h for cell adhesion, 1 . 4 lo'
gonococci were added to the coverslips which were then incubated for 20 min at 36 "Con an orbital shaker
rotating at 140 rev. min-l. The coverslips were washed, air-dried and stained with acridine orange (Smith &
Romnel, 1977). Coded coverslipswere counted using an Ortholux I1 microscope (Leitz, Wetzlar, W. Germany)
fitted with Ploem I incident ultraviolet illumination (barrier filter K570; excitation filter VP 425 3 mm
BG3 and a TK5lO dichroic mirror with a K515 suppression filter).
+
RESULTS
Identijication oj' a novel outer membrane protein in N . gonorrhoeae strain P9
The major surface proteins in colonial variants of strain P9 of the gonococcus are shown
in Table 1. A major outer membrane protein designated protein I (mol. wt 36000) was
present in each, together with the variations in the four proteins in the molecular weight
range 27500 to 29000 as previously reported (Lambden & Heckels, 1979). Using SDSpolyacrylamide gel electrophoresis with a linear gradient of 10 to 20 yo (w/v) acrylamide, a
fifth outer membrane protein, protein IId*, was identified in the variant P9-11. This particular variant was originally described as containing proteins 11* and IIa* but the greater
resolution obtained with the gradient polyacrylamide system established that the protein
11" in P9-11 consistently showed greater mobility than protein 11" from variants P9-6 and
P9-19 (Fig. 1). The apparent molecular weight of the new protein (IId*) was very close to
the molecular weight of protein II* (mol. wt 29000) and was estimated to be 28850 on the
10 to 20% gradient gel system.
Sensitivity to normal lzuman serum
The susceptibility of each of the variants to the bactericidal action of normal human
serum is represented graphically in Fig. 2. The prototype, non-pilated variant P9-1 (containing only lipopolysaccharide and protein I as major components of the outer membrane)
gave 21 % survivors after 45 min and was the most sensitive to serum. The pilated variant
P9-2 and variants P9-6 (II*), P9-16 (IIb*) and P9-19 (II*, IIc*) were more resistant to
serum killing although less than 50% survived after exposure to serum for 45 rnin. A
second set of variants P9-9 (IIa*, IIb*), P9-11 (IId*, Ha*) and P9-13 (Ira*) showed marked
resistance to serum killing with more than 70% survivors after 45 min. The unique feature
of these variants is the presence of protein IIa* in the outer envelope.
Table 1. Outer membrane proteins in variants of N . gonorrlioeae P9
Outer
membrane
protein
Protein I
Protein 11"
II'd*
I1a*
IIb*
llc"
Pilin
Molecular
weight
36000
29 000
28 850
28 500
28 000
27500
19500
+ , Present; -,not detected.
M I C 114
20
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
308
P. R. L A M B D E N A N D O T H E R S
1
2
3
4
5
6
7
8
9
I
II*
IId *
Ila"
IIb*
Ik*
Fig. 1. SDS-polyacrylamide gel electrophoresisof outer membrane preparations from each of the
gonococcal variants: 1, P9-1; 2, P9-6; 3, P9-9; 4, P9-11; 5, P9-13; 6,P9-16; 7, P9-19; 8, P9-11:
9, P9-6. The separating gel contained a linear gradient of 10 to 20% (w/v) acrylamide.
Sensitivity to antimicrobial agents
Clear differences in the pattern of sensitivity to penicillin, fusidic acid, Cu2+and Zn2+
were seen amongst the variants (Table 2). Two variants P9-6 (II*) and P9-19 (11" IIc*)
consistently showed a two-fold greater m.i.c. on plates containing penicillin, fusidic acid
and Cu2+ and an increased resistance to Zn2+ compared with other isolates. The outer
membrane profiles showed that protein 11* was unique to the two resistant variants. Isolates
lacking protein 11* showed an equal sensitivity to the agents tested, although the two transparent variants P9-1 and P9-2 (protein I only) were slightly more resistant to Zn2+ than
other protein II*-deficient variants. All isolates were equally susceptible to tetracycline
(m.i.c. 0.5 pg ml-l).
A t tachmen t of gonococcal variants
Attachment to buccal epithelial cells showed clear differences between the variants
(Table 3) with the greatest enhancement over the prototype P9-1 shown by P9-2 which
differed only in the possession of pili. However, the variants which lacked pili but contained
additional outer membrane proteins also had a considerable advantage in attachment,
especially variants P9-6 (II*), P9-11 (IId*, IIa*) and P9-13 (IIa*) which showed a t least a
twofold increase.
Attachment to erythrocytes showed a different pattern (Table 3). Again the pilated
variant attached most avidly with an approximately twofold advantage over the prototype
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
309
Sairface proteins and virulence of gonococci
1
1
1
15
45
30
Time
(iiiin)
Fig. 2. Sensitivity of gonococcal variants to the bactericidal effect of normal human serum. Points
represent the logarithm of the percentage of survivors at each time and the straight lines were
drawn from linear regression analysis of the data. Variants used were: 1 (0),P9-1; 2 (O), P9-2;
3 (O), P9-6; 4 (m), P9-9; 5 (A), P9-11; 6 (A),P9-13; 7 (V),P9-16; 8 (v),P9-19.
Table 2. Sensitivity of N. gonorrhoeae P9 variants to antimicrobial agents
Results are expressed as the minimum inhibitory concentration in ,ug ml-l.
Tetracycline
Penicillin
Fusidic acid
CUS04.5H20
ZnS01.7H20
P9-1
P9-2
P9-6
P9-9
P9-11
P9-13
P9-16
P9-19
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.16
0.10
10
80
0.16
0.10
30
80
0.32
0.20
20
90
0.16
0.10
10
70
0.16
0.10
10
70
0.16
0.10
10
70
0.16
0.32
0.10
0.20
10
20
70
100
Table 3. Attachment of N. gonorrhoeae P9 variants to buccal epithelial cells,
erythrocytes and hydrophobic Sepharose gets
Results are expressed as the percentage of 3H-labelled gonococci sedimentingwith the buccal cells,
erythrocytes or gels, as described in Methods.
Buccal epithelial cells*
Erythrocytes*
Phenyl-Sepharose*
Hexyl-Sepharose
PPl
12.0
24.5
8.9
6.2
P9-2
38.0
45.4
10.0
5.1
P9-6
24.0
19.0
3.9
0.5
P9-9
22-2
14.7
7.1
4.1
P9-11
27.3
15.1
8.2
4.4
P9-13
26.1
15.5
8.5
6.1
P9-16
17.7
11.0
8-6
5.6
P9-19
20.6
11.4
8.1
6.2
*Average values from three separate experiments.
Table 4. Association of variants of N . gonorrhoeae P9 with PMN leukocytes
Results are expressed as the percentage of polymorphs with associated gonococci and are the
mean values from three separate experiments.
Association
P9-1
10.0
P9-2
8.0
P9-6
21-5
P9-9
20.5
P9-11
20.2
P9-13
32.0
P9-16
4.6
P9-19
21.0
P9-1. However, those variants possessing additional outer membrane proteins all attached
to erythrocytes with decreased efficiency compared with P9- 1.
Since bacterial adhesion may be profoundly affected by surface hydrophobicity, the
attachment of the variants to hydrophobic gels was studied (Table 3). Little difference was
seen with the exception of P9-6 (protein 11* only) which showed decreased binding to both
phenyl- and hexyl-Sepharose gels.
20-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
310
P. R . L A M B D E N A N D O T H E R S
Leukocyte association
Differences in gonococcal-leukocyte association were seen between the variants (Table 4).
The presence of pili had little influence on leukocyte association but alterations in outer
membrane composition had a significant effect. Variant P9-16 (I1 b*) showed decreased
association whereas the other variants showed two- to threefold enhancement over the
prototype P9- 1.
DISCUSSION
Some caution is necessary in comparisons between the variants of N . gonorrhoeae P9
since acquisition of a new outer membrane protein may result in a concomitant decrease
in one previously present (Heckels, 1978) and perhaps other as yet unknown alterations in
membrane composition. Nevertheless, differences occur between the variants in properties
which directly affect potential virulence and appear to be correlated with particular outer
membrane proteins.
Gonococci present in urethral exudates (Ward et al., 1970) and infected subcutaneous
chambers in guinea pigs (Penn et al., 1977) exhibit greater resistance to serum killing than
after subculture on laboratory media. A putative gonococcal vaccine should retain the
factor(s) responsible for this critical aspect of gonococcal virulence. Data presented in this
paper show that the production of outer membrane proteins in addition to protein I
conferred on the variants an increased resistance to serum killing. The most resistant variants
were distinguished from the others in that they each produced protein Ira*. Clearly any
understanding of the mechanism of serum resistance will require knowledge of the relative
amounts and spatial arrangement of the components present in the outer membrane. It is
currently unknown whether protein I (Hildebrandt et al., 1978) or the lipopolysaccharide
(Guymon et al., 1978a) is responsible for serum resistance in gonococci isolated from
patients with disseminated infections. Another possible resistance mechanism is for an
external structure to coat the vulnerable outer membrane, for example, the recently described capsule (Richardson & Sadoff, 1977). Presumably pili wrapped around the gonococcal
surface account for the enhanced resistance of variant P9-2 (Fig. 2). Despite the evidence
that gonococcal resistance to serum killing is multifactorial, the use of isogenic variants has
shown that one outer membrane protein (Ha*) is associated with considerable resistance.
The data presented in Table 2 indicate that protein IIa* offers no protection against
water-soluble, low molecular weight antimicrobial agents such as antibiotics, Cu2f, which
is released from intrauterine contraceptive devices (Cohen & Thomas, 1974), or Zn2+,which
approaches a level of 1 mg ml-I in prostatic secretions (Fair & Wehner, 1976). The results
suggest that protein 11" may be responsible for enhanced resistance possibly by affecting
envelope permeability in such a way that penicillin, fusidic acid, Cu2+and Zn2+are excluded
from their target sites. This variant is clearly distinct from the permeability mutants described by Guymon et al. (1978b) which were characterized by an increase in the subunit
molecular weight of protein I or the appearance of a minor outer membrane protein of
molecular weight 52 000.
The interaction of gonococci with host-cell surfaces is a critical early stage in the pathogenesis of gonorrhoea but the mechanisms involved are complex and ill understood. Our
results indicate that the presence of the additional outer membrane proteins enhances the
binding of gonococci t o buccal cells. Conversely, the presence of extra outer membrane
proteins decreased attachment to erythrocytes indicating that the two cohesive processes are
mediated by different components. However the pilated variant P9-2 showed the greatest
attachment to both buccal cells and erythrocytes. The increased attachment of enteropathogenic Escherichia coli mediated by pili has been correlated with increased hydrophobicity of pilated strains as revealed by binding to hydrophobic Sepharose gels (Smyth
et al., 1978). Binding of the gonococcal variants to hydrophobic gels did not reveal such
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
Surface proteins and virulence of gonococci
311
substantial differences although one variant, P9-6, showed less hydrophobic character than
the others. This variant showed good adhesion to epithelial cells suggesting that the hydrophobic interactions with the host-cell surface may play only a minor role in gonococcal
attachment. Nevertheless, the correlation of variation in epithelial cell attachment with
alterations in surface proteins is likely to be of considerable significance in gonococcal
virulence and makes possible the investigation of the molecular mechanism responsible for
adhesion.
Another cell-cell interaction which has been associated with potential virulence is that
between gonococci and PMN leukocytes. Increased leukocyte association has been correlated with the presence of an additional protein (LA) factor, in the molecular weight
range 28000 to 29000, in the gonococcal outer membrane (King & Swanson, 1978). The
variants of strain P9 also had altered leukocyte associations but, as reported by Swanson
et al. (1975), pili had little effect. Unlike the results present by these workers, the current
study demonstrates that one protein (IIb*) of molecular weight 28000 is associated with
decreased leukocyte association whereas all the other variants show increased association.
Thus, we were not able to identify a single LA factor, although it may be relevant that the
variant showing greatest leukocyte association (P9-13) contains a single extra protein (I1a*)
with a molecular weight (28 500) which is close to that reported for the LA factor of King &
Swanson (1 978).
Several proteins in the outer membranes of Gram-negative bacteria have been associated
with specific functions including structural integrity of the cell, uptake of specific nutrients
and formation of diffusion pores (Di Rienzo et al., 1978). While it would appear unlikely
that the major proteins expressed at the gonococcal surface function purely as virulence
factors, nevertheless, the variations in protein composition directly affect the potential virullence of the organism. These factors must be carefully considered in producing gonococcal
components for serological diagnosis and vaccine trials since it would be unlikely that a
random mixture of colony forms selected in vitro would necessarily contain those factors
critical for virulence in the natural infection.
This work was supported by a Medical Research Council Programme Grant.
REFERENCES
'
COHEN,L. & THOMAS,
G. (1974). Copper versus the HECKELS,
J. E. (1977). The surface properties of
gonococcus in vivo. British Journal of Venereal
NeisAeria gonorrhoeae : isolation of the major
Diseases 50, 364-366.
components of the outer membrane. Journal of
DIRIENZO,
J. M., NAKAMURA,
K. & INOUYE,
M.
General Microbiology 99, 333-341.
(1978). The outer membrane proteins of Gram- HECKELS,
J. E. (1978). The surface properties of
negative bacteria: biosynthesis, asszmbly and
Neisseria gonorrhoeae : topographical distribution
functions. Annual Review of Biochemistry 47,
of the outer membrane protein antigens. Journal
481-532.
of General Microbiology 108, 21 3-219.
FAIR,
W. R. & WEHNER,
N. (1976). The prostatic HILDEBRANDT,
J. F., MAYER,L. W., WANG,S. P. &
antibacterial factor: identity and significance.
BUCHANAN,
T. M. (1 978). Neisseria gonorrhoeae
Progress in Clinical and Biological Research 6,
acquire a new principal outer membrane protein
when transformed to resistance to serum bacteri383-403.
GUYMON,
L. F., LEE,T. J., WALSTAD,
D., SCHMON- cidal activity. Infection and Immunity 20,267-272.
YER, A. & SPARLING,
P. F. (1978a). Altered outer JAMES,
J. F. & SWANSON,
J. (1978). Studies on gonomembrane components in serum-sensitive and
coccus infection XIII. Occurrence of color/
serum-resistant strains of Neisseria gonorrhoeae.
opacity colonial variants in clinical cultures.
In Immunobiology of Neisseria gonorrhoeae, pp.
Infection and Immunity 19, 332-340.
139-141. Edited by G. F. Brooks, E. C. Gotschlich, KING,G. J. & SWANSON,
J. (1978). Studies on gonoK.K.Holmes, W.D.Sawyer & F.E.Young.
coccus infection XV. Identification of surface
Washington, D.C. : American Society for Microproteins of Neisseria gonorrhoeae correlated with
biology.
leukocyte association. Infection and Immunity 21,
GUYMON,
L. F., WALSTAD,
D. L. & SPARLING,
P. F.
575-584.
(1978b). Cell envelope alterations in antibiotic- LAMBDEN,
P. R. & HECKELS,
J. E. (1979). The insensitive and -resistant strains of Neisseria gonorfluence of outer membrane protein composition
rhoeae. Journal of Bacteriology 136, 391-401.
on the colonial morphology of Neisseria gonor-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00
3 12
P. R. L A M B D E N A N D O T H E R S
rhoeae strain P9. FEMS Microbiology Letters SWANSON,
J. (1973). Studies on gonococcus infection
5, 263-265.
IV. Pili: their role in attachment of gonococci to
PENN,C. W., VEALE, D. R. & SMITH,H. (1977).
tissue culture cells. Journal of Experimental
Selection from gonococci grown in vitro of a
MediciKe 137, 571-589.
colony type with some virulence properties of SWANSON,
J. (1978). Studies on gonococcus infection
organisms adapted in vivo. Journal of General
XII. Colony color and opacity variants of gonoMicrobiology 100, 147-158.
cocci. Infection and Immunity 19, 320-331.
N C H A R C S ~W.
N ,P. & SADOFF,
J. C. (1977). Produc- SWANSON,
J., SPARKS,E., YOUNG,D. & KING,G.
tion of a capsule by Neisseria gonorrhoeae. Infec(1975). Studies on gonococcus infection X. Pili
tion and Immunity 15, 663-664.
and leukocyte association factor as mediators of
SMITH,D. L. & ROMNEL,
F. (1977). A rapid microinteractions between gonococci and eukaryotic
method for the simultaneous detection of phagocells in vitro. Infection and Immunity 11, 1352cytic and microbicidal activity of human peri1361.
pheral blood leucocytes in vitro. Journal of WARD,M. E., GLYNN,
A. A. &WATT,P. J. (1970).
Immunological Methods 17, 241-247.
Gonococci in urethral exudates possess a virulence
SMYTH,
C. J., JONSSON,
P., OLSSON,
E., SONDERLIND, factor lost on subculture. Nature, London 277,
O., ROSENGREN,
J., HJERTEN,
S. & WADSTROM,
T.
382-384.
(1978). Differences in hydrophobic surface charac- WARD,M. E., LAMBDEN,
P. R., HECKELS,
J. E. &
WATT, P. J. (1978). The surface properties of
teristics of porcine enteropathogenic Escherichiu
coli with or without K88 antigen as revealed by
Neisseria gonorrhoeae : determinants of susceptibihydrophobic interaction chromatography. Infeclity to antibody complement killing. Journal of
tion and Immunity 22, 462-472.
General Microbiology 108, 205-212.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 12 Aug 2017 19:04:00