Attempts to demonstrate a polysaccharide capsule in Neisseria gonorrhoeae

APMIS 96: 735-740, 1988

Attempts to demonstrate a polysaccharide capsule in Neisseria gonorrhoeae KARIN REIMANN', HELMUT HEISE' and JENS BLOM' 'Neisseria Department, Statens Seruminstitut, Copenhagen, Denmark, 'Dermatological Clinic, Wilhelm-Pieck-University, Rostock, German Democratic Republic, and 'Department of Biophysics, Statens Seruminstitut, Copenhagen, Denmark

Reimann, K., Heise, H. 8t Blom, J. Attempts to demonstrate a polysaccharide capsule in Neisseria gonorrhoeae. APMIS 96: 135-140, 1988. The presence or absence of a polysaccharide capsule on the human pathogen Neisseria (N.) gonorrhoeae is still a topic of controversy. For this reason we compared the results obtained by light microscopy (dry India ink-Fuchsin stain) and electron microscopy (Alcian blue-lanthanum nitrate stain) of encapsulated strains of N. meningitidis and Streptococcus (S.)pneumoniae and of non-encapsulated strains of S. pneumoniae and Escherichia (E.)coli with those obtained using the same methods on strains of pilliated and non-pilliated N. gonorrhoeae. After staining with India ink-Fuchsin no capsules could be demonstrated on any of the N. gonorrhoeae strains studied. If present the capsules on these cells are too delicate to be identified by light microscopy. After treatment with Alcian blue-lanthanum nitrate sections of cells of N. meningitidis and S. pneumoniae generally showed the presence of a capsular layer. Sections of cells of the non-encapsulated strain of S. pneumoniae which possess C (common)-polysaccharide also showed surface associated capsule-like material. Similarly the surface of the cells of the E. coli strain showed material which appeared to be tufts of pili and/or M (mucoid)-antigen.In experiments where the N. gonorrhoeae cells were harvested as early as after six hours of growth a capsule-like material was demonstrated on cells of all strains studied. Key words: Neisseria gonorrhoeae; polysaccharide; capsule; electron microscopy; Alcian blue-lanthanum nitrate stain. Karin Reimann, Neisseria Department, Statens Seruminstitut, Amager Boulevard 80, DK-2300 Copenhagen S, Denmark.

The presence of capsules on bacteria has for a long time been associated with the virulence of bacteria pathogenic for humans. Neisseriu (N.) rneningitidis has a well-defined polysaccharide capsule; consequently, the possibility that the closely related N. gonorrhoeae might be encapsulated has often been suggested. Several attempts have been made by means of the light microscope (2, 4, 6, 8, 9, 12) and the electron microscope (2, 4, 5 , 8) to determine whether gonococci possess capsulbr material. The existence of a capsule on N. gonbrrhoeae had in

our opinion not been proven since the experiments in which capsule-like material were demonstrated did not comprise bacteria known to be either encapsulated or non-encapsulated. The aim of this study was to apply some of the described methods on gonococci and on appropriate controls in order to confirm or rule out the presence of polysaccharide capsules on N. gonorrhoeae (5,9). Experiments with bacteria grown on media especially developed to promote capsule formation were included. MATERIALS AND METHODS

Received February 2, 1988. Accepted May 12, 1988.

:.

Bacterial strains. Twenty-three random isolates of N. gonorrhoeae from our routine laboratory were studied;

735

REIMANN

the only known clinical data were sex and age of the patients from whom we isolated the strains and the anatomical site from which the isolate was obtained. These strains were studied together with two piliated variants of laboratory strains (F 62 and 6650) and four non-piliated variants of laboratory strains (F 62, 6650, WHO 3 and WHO 5) (7). The encapsulated strains were two laboratory strains of N. meningitidis (M 1027 serogroup A and C 1 I serogroup C (originally provided by C. E. Frasch, Bethesda, USA)) and one strain of Streptococcus (S.) pneumoniae, type 3 (ATCC 10813). One non-encapsulated strain of S. pneumoniae, R 36a (ATCC 27336) and one non-encapsulated strain of E. coli K I2 W300 I were also included. (The two strains of S. pneumoniae were provided by J. Henrichsen, WHO Collaborating Centre for Reference and Research on Pneumococci and the E. coli was provided by F. 0rskov, WHO Collaborating Centre for Reference and Research on Escherichia and Klebsiella, Statens Seruminstitut, Denmark). Media. GC medium. 36 g/L Gc medium base (Difco) supplemented with I % isovitalex (BBL),glucose content 0.1%. I/2 GPH medium: (gonococcus peptone haemoglobin) as described by Hendley et al. (9,except that we used 0.7%and 0.45% agarose (Sigma and BRL) instead of 1.2% agarose (Grand Island Biological Co.). Even though we had reduced the agarosecontent the agar strength was still high at 0.7%, but at 0.45% the agar strength was similar to that recommended for our routine GC medium; the glucose content was 0.05% which is half the amount usually used in media for the growth of gonococci; Hendley et al. ( 5 ) had found a low glucose content supportive for capsule production. 10% bloodagar: made with defibrinated horse-blood, glucose content 0.04%. Trypsin broth: liquid medium which contains 0.3% maltose. Growth conditions. All cultures were incubated at 36 “C (the Neisseria in a C0,-incubator). Gonococci and meningococci were incubated for 17, 12 or 6 hours, if inoculated on GC medium, and for 17 or 6 hours if inoculated on 112 GPH medium. The pneumococci were not able to grow on either of the above mentioned media and were therefore inoculated either on 10% bloodagar plates (incubation time 17 hours) or in trypsin broth (incubation time five hours). The inoculum for the trypsin broth was prepared from pneumococci propagated overnight in serum broth and then diluted 1:30 in the trypsin broth. The E. coli strain was inoculated on the GC medium and incubated for 4 hours. Light microscopy. Dry India ink stain (9): The bacteria were suspended in a small drop of India ink (Rothring tusch) on a microscope slide. By means of a cover glass the suspension was smeared into a thin film which was allowed to dry. The bacteria were stained for 7-8 minutes with 1% Fuchsin. The slides were thoroughly rinsed with tap-water, dried and examined under a light microscope at 1000 x magnification. Electron microscopy. Alcian blue-lanthanum nitrate method. The cells were treated with Alcian blue-lanthanum nitrate using the method described by Shea ( 13)

736

el

al

with minor modifications. Colonies on the surface of agar plates were covered with fixative containing 3% glutaraldehyde, 0.01 M CaCI, and 0.5% (w/v) Alcian blue 8GS (E. Gurr) in 0.1 M cacodylate buffer, pH 7.2. After fixation at room temperature for 2 hours, the agar surface was gently rinsed three times with cacodylate buffer. The fixed colonies were then punched out or removed carefully with a loop and washed once in cacodylate buffer. Cells grown in liquid media (not illustrated) were fixed overnight at 4 “Cin glutaraldehyde and Alcian blue and then further treated in the same way as the cells removed from the plates. In one experiment MgC1, was added to the Alcian blue solution to a concentration of 0.03 M in order to optimize the Alcian blue capsular polysaccharide reaction ( 1 1). Cells removed with a loop were pelleted (8,800 g for 2 minutes), enrobed in melted (45 “C) 1.5% Noble agar (Difco) and after hardening the agar was cut into small cubes. All agar blocks were postfixed for 2 hours or overnight in cacodylate buffer containing 1% OsO, and IYo lanthanum nitrate at room temperature. This was followed by dehydration in increasing concentrations of ethanol alone or followed by propylene oxide and embedding in Spurr’s low viscosity resin (Polysciences Inc. Warrington, Pa, USA) or Vestopal-W (Martin Jaeger, Geneva, Switzerland), respectively. Further preparation of the embedded material as well as electron microscopy of the sections produced was carried out as described by Blom et al. (1).

RESULTS

Light Microscopy All the bacterial strains mentioned under Material and Methods except s.pneurnoniae type 3 and the E. coli strain were studied with the quick and easy dry India ink method (9). All strains presented the same picture after staining: a dark India ink background in which “pink bodies” (the bacteria) were surrounded by white rings. The width of the rings did not vary with the bacterial species but with the thickness of the ink smear. A thick smear resulted in nothing or a narrow white ring around the “pink body”, and a very thin smear resulted in a wide ring round the bacteria. Electron Microscopy All the bacterial strains studied were treated according to the Alcian blue-lanthanum nitrate method ( 13). In cases where the strains were grown for only 6 hours, MgC I was added to the Alcian blue stain ( 1 I). Thin sections of meningococci grown on the GC medium revealed a homogeneous capsule of varying thickness (Fig. 1). A

POLYSACCHARIDE CAPSULE IN N. GONNORRHOEA

Fig. 1. Cells of strain N. meningitidis Cl 1 grown for 12 hours on GC-medium. The cells present layers of capsular (C) material. x 45,000. Fig. 2. Cells of strain N. meningitidis C11 grown for 6 hours on 1/2 GPH medium. Capsular material (C) is only visible on the cells situated at the outer periphery of the microcolony. Bar = 0.5 pm. x 15,000. Fig. 3. A cell of strain S. pneumoniae type 3 grown for 17 hours on 10% blood agar. A thick layer of homogeneous capsular material (C) is present. x 45,000.

The figures show sections of cells obtained from cultures of the studied bacteria treated with the Alcian blue-lanthanum nitrate. All sections are counter stained with magnesium uranyl acetate and lead nitrate. The bar on each electron micrograph represents 0.1 pm, unless otherwise stated.

Fig. 4. Cells of the non-encapsulated strain S. pneumoniae R36a grown for 17 hours on 10% blood agar. A thin and uneven layer of C-polysaccharide (CP) has reacted with Alcian blue-lanthanum nitrate at the surface of the cell wall. x 45,000. Fig. 5. Cells of E. coli K12 W300 1 grown for 4 hours on GC medium. An uneven layer of electron-dense material is found on the cells. Material which resemblesbundles of pili (P) is present between the cells. x 45,000.

737

REIMANN el al.

Fig. 6 . Piliated cells of strain N. gonorrhoeae F62 grown for 6 hours on 1/2 GPH medium. The cells present obvious layers of capsule-like material. Material which may be small bundles of pili (P)are found between the cells. x 45.000. Fig. 7. Cells from a microcolony of the piliated N . gonorrhoeae strain F62 grown for 17 hours on GC medium. Note the varying reaction with Alcian blue; only two of the cells show capsule-like material (CL) on the surface. Bar = 0.5 pm. x 15,000, Fig. 8. Higher magnification of cells from the same preparation as seen in Fig. 7. A thin capsule-like layer (CL) is found on some of the cells. x 45,000.

positive capsular reaction was also seen when the same strain was grown on the glucose-poor medium 1/2 GPH, but the capsular layer was only revealed in regions at the periphery of the microcolonies. The capsules on the cells lying further inside the colony showed no reaction (Fig. 2). Thin sectionsof S. pneumoniae type 3 grown on 1OYo blood agar showed cells with very thick and homogeneous capsules (Fig. 3). In between the cells scattered amounts of capsular material could also be found. The non-encapsulated S. pneumoniae R36a (Fig. 4) and the non-encapsulated E. coli K12 W3001 (Fig. 5 ) both showed an uneven and loose layer of electron-dense material adhering to their cell walls. These layers strongly resembled capsular material in appearance. In between the cells some electron-dense material was also present. After growth for 6 hours on GC medium and 1/2 GPH medium both laboratory and wild strains of N. gonorrhoeae showed capsule-like material (Fig. 6). Bundles of pili could also be 738

distinguished. Gonococcal cultures grown for 17 hours on the same media only occasionally showed cells surrounded by capsule-like material (Figs. 7, 8). DISCUSSION A bacterial capsule has been defined as a covering layer outside the cell wall that has a definite external surface (4, 15). Human bacterial diseasescharacterized by invasion of the blood by the infectious organism are caused, almost entirely, by encapsulated bacteria, as for example the meningococcus; the closely related gonococcus may also dissiminate but only in about 1% of the cases; the main target for the gonococcus is the mucous membranes and encapsulation is not mandatory for bacterial colonization of these sites. Since 1977, several investigators have tried by

POLYSACCHARIDE CAPSULE IN N . GONNORRHOEA

means of the light and the electron microscope to elucidate whether or not the gonococcus is encapsulated like the closely related pathogen, the meningococcus. Hendley et al. (4), James & Swanson (6) and Richardson & Sadoff( 12)used the wet India ink technique (3) and the light microscope; they concluded that most gonococci had capsules, but no one included both encapsulated and nonencapsulated control strains in their material. Later Melly et al. (8)claimed that the wet India ink technique is liable to produce artifacts which may be misinterpreted as capsules. Demarco de Horrnaeche et al. (2) also used light microscopy but they stained their specimens with the Leishman stain. They investigated two gonococcal strains and found capsules. Miiller (9)used a dry India ink technique with a Fuchsin counterstain to visualize capsules on the gonococci by light microscopy. He observed that about 50% of 142 strains possessed capsules and claimed that encapsulation correlated with chronic gonococcal infection in the patients. These last two studies comprised neither positive nor negative controls. We used the same method as Muller (9) to investigate piliated and non-piliated gonococci, two encapsulated meningococcal strains and one non-encapsulated pneumococcus for the presence of a capsule. We found the method useless for the purpose as all bacteria presented the same type of picture i.e. a “pink body” surrounded by a ring, the width of which was correlated to the thickness of the India ink smear. In our opinion the ring does not represent a capsule, but an empty gap in the ink layer produced by shrinkage of the bacterial body during the drying process. Melly et al. (8) used the India ink technique adapted to electron microscopy. They included positive as well as negative controls and found that less than 1% of the gonococci studied had structures that could possibly be interpreted as capsules. Electron microscopy on Alcian blue-lanthanum nitrate treated cells ( 13) was used by Demarco de Hormaeche et al. (2) and Hendley et al. ( 5 ) to demonstrate capsular material. Hendley et al. stressed the importance of the growth medium for the production of capsules and the lability of the capsular material when manipulated. Hendley et al. ( 5 ) designed a growth medium, 1/2 GPH, with a low glucose content which appeared to facilitate capsule production. When they fixed and stained bacteria, grown to log-phase, directly on the agar surface before the organisms were manipulated,

they found that three gonococcal strains were fully encapsulated and that two were only partly encapsulated, strain F 62 being one of the latter. For our electron microscopy studies we tried to follow the method of Hendley et al. ( 5 ) as closely as possible (see Material and Methods). The positive controls, the meningococcusand the encapsulated pneumococcus clearly showed capsular material on their surface. The meningococcus produced the greatest amounts of capsular material when it had the best conditions for growth, i.e. on the GC medium. When grown on 1/2 GPH medium, the meningococci displayed a dense capsular layer only on the cells belonging to the surface region of the microcolonies. This phenomenon probably indicatesthat the Alcian blue-lanthanum nitrate is unable to penetrate into the microcolony and react with the underlying cells. Another explanation could be that only cells in direct contact with the medium at the periphery of the colonies are able to produce capsules. The gonococci only produced significant amounts of capsule-like material after 6 hours of growth and in this case, as for the meningococci, predominantly on peripheral cells. This observation may account for some of the disagreementsas to whether or not the gonococci are encapsulated. Our experiments did not support the theory of Hendley et al. ( 5 ) that a glucose-poor medium promotes capsule production. We found that the meningococci and especially the gonococci grew very poorly on the 1/2 GPH medium. Much to our surprise the non-encapsulated S. pneumoniae R 36a and the non-encapsulated E. coli K 12 W300 1 also showed capsule-like material on the surface of the cells. Recent experiments have shown that the positively stained material on the pneumococci consists of so-called C-polysaccharide (14). The loose tufty layers seen on the E. coli cells could represent bundles of pili emanating at their surface and/or other material reacting with Alcian blue-lanthanum nitrate, e.g. the M-antigen. The observed staining of the C-polysaccharide and surface structures of E. coli suggests that a morphological indication of possible encapsulation should be confirmed by a chemical identification. Noegel & Gotschlich (10) found in their attempts to isolate capsular material from the gonococcus, that the capsulated cells produced considerableamounts of polyphosphate, and consequently they suggested that this might have a capsular role. 739

REIMANN ef al.

The results of our experiments did not give the final answer to the question of whether or not gonococci are encapsulated but under certain conditions gonococci seem to be able to produce a capsule-like material. Our experiments lead us to conclude that Alcian blue-lanthanum nitrate indeed reacts with capsular polysaccharide, as shown by others, but as it also reacts with other materials on the bacterial surface, e.g. C-polysaccharide of the pneumococci, pili bundles and/or M-antigen on E. coli, a positive result becomes inconclusive. Unfortunately, a negative result may also be inconclusive as we have demonstrated that the Alcian blue-lanthanum nitrate stain only seems to react with material situated on cells at the periphery of the microcolonies and not with cells from deeper regions. The authors wish to thank Inga Lind and Aksel BirchAndersen for their valuable help and advice throughout this study. We thank Helene Ravn, Anna Grethe Overguard and Finn Laursen for their technical assistance, and Sandra Hyrnan and Grethe Seidenfaden for their secretarial assistance.

REFERENCES Blorn, J., Mama, B. & Wiik, A,: A study of Russell bodies in human monoclonal plasma cells by means of immunofluorescence and electron microscopy. Acta path. microbiol. scand. Sect. A, 84: 335-349, 1976.

Dernarco de Horrnaeche, R., Thornley, M . J. & Glauert, A . M.: Demonstration by light and electron microscopy of capsules on gonococci recently grown in vivo. J. Gen. Microbiol. 106: 81-91, 1978. Duguid, J. P. The demonstration of bacterial capsules and slime. J. Path. Bact. 63: 673-685, 1951. Hendley, J. O., Powell, K. R.. Rodewald, R., Holz-

740

grefe, H . H. & Lyles, R.: Demonstration of a capsule on Neisseria gonorrhoeae. N. Engl. J. Med. 296: 608-61 1, 1977.

5 . Hendley, J. O., Powell, K. R., Salornonsky, N. L. & Rodewald, R. R.: Electron microscopy of the gonococcal capsule. J. Infect. Dis. 143: 796-802, 198 1. 6. James, J. F. & Swanson, J.: The capsule of the gonococcus. J. Exp. Med. 145: 1082-1086, 1977. 7. Jephcott, A . E., Reyn, A. & Birch-Andersen, A.: Neisseria gonorrhoeae 111. Demonstration of presumed appendages to cells from different colony types. Acta path. microbiol. scand. Sect. B, 79: 437-439, 197 1.

8 . Melly, M . A . , McGee, Z . A., Horn, R. G.,Morris, F. & Glick, A . D.: An electron microscopic India ink technique for demonstrating capsules on microorganisms: Studies with Streptococcus pneurnoniae, Staphylococcus aureus, and Neisseria gonorrhoeae. J. Infect. Dis. 140: 605-609, 1979. 9. Muller, G. v.: Kapselnachweis bei Neisseria-gonorrhoeae-Isolaten von asymptomatischer, akuter und chronischer Gonorrhoe. Dermatol. Monatsschr. 166: 235-237, 1980. 10. Noegel, A . & Gotschlich, E. C.: Isolation of a high molecular weight polyphosphate from Neisseria gonorrhoeae. J. Exp. Med. 157: 2049-2060, 1983. 11. Powell, K. R., Hendley, J. O.,Pohl, K. E., Freidberg, A. & Volk, W . A , : Quantitation of acidic capsular polysaccharides by Alcian blue binding. Analyt. Biochem. 119: 31-37, 1982. 12. Richardson, W . P. & Sadox J. C.: Production of a capsule of Neisseria gonorrhoeae. Infect. Immun. 15: 663-664, 1977. 13. Shea, S. M.: Lanthanum staining of the surface coat of cells: Its enhancement by the use of fixatives containing Alcian blue or cetylpyridinium chloride. J. Cell Biol. 51: 6 1 1-620, 197 1. 14. Surensen, U. B. S. & Henrichsen, J.: Cross-reations between pneumococci and other streptococci due to C polysaccharide and F antigen. J. Clin. Microbiol. 25: 1854-1859, 1987. 15. Wilkinson, J. F.: The extracellular polysaccharides of bacteria. Bact. Rev. 22: 46-73, 1958.