The production of antibodies to Klebsiella capsular polysaccharides was measured in ... immunosorbent assay specific to each of the 77 Klebsiella serotypes.
INFECTION AND IMMUNITY, Nov. 1994, p. 4838-4843
Vol. 62, No. 11
0019-9567/94/$04.00+0 Copyright © 1994, American Society for Microbiology
Comparison of the Antibody Responses to the 77 Klebsiella Capsular Types in Ankylosing Spondylitis and Various Rheumatic Diseases HANY SAHLY,1* JORN KEKOW,2'3 RAINER PODSCHUN,' MARCO SCHAFF,' WOLFGANG L. GROSS,2'3 AND UWE ULLMANN' Department of Medical Microbiology and Virology, University of Kiel, Kiel,' Department of Rheumatology
of the Medical University at Lubeck, Lubeck2 and the Clinic for Rheumatology at Bad Bramstedt, Bad Bramstedt,3 Germany Received 2 June 1994/Returned for modification 7 July 1994/Accepted 8 August 1994
The production of antibodies to Klebsiella capsular polysaccharides was measured in sera from either HLA-B27-positive (HLA-B27+) or HLA-B27-negative (HLA-B27-) patients with classical ankylosing spondylitis (n = 54). These sera were compared with sera from patients with various rheumatic diseases (n = 82) and HLA-B27+ or HLA-B27- healthy individuals (n = 85). All sera were analyzed by means of an enzyme-linked immunosorbent assay specific to each of the 77 Klebsiella serotypes. The sera from HLA-B27+ patients with ankylosing spondylitis showed a significantly higher antibody frequency to the capsular types K26, K36, and K50 than the sera from HLA-B27- ankylosing spondylitis patients, patients with psoriatic arthritis, systemic lupus erythematosus, rheumatoid arthritis, or reactive arthritis after Yersinia enterocolitica infection, or healthy controls (P < 0.02). The antibodies were of the immunoglobulin G type. No significant antibody response to the other 74 KlebsieUla serotypes, noncapsulated mutants of K26, K36, and K50, or preparations of Citrobacter, Serratia, Hafnia, or MorganeUa spp. or Streptococcus pneumoniae could be detected. The results might suggest a specific association between these capsular types and HLA-B27' ankylosing spondylitis and might imply their predominance in this disease. It has been suggested that the gram-negative microorganism Klebsiella sp. plays an important role in the pathogenesis of ankylosing spondylitis (AS) in HLA-B27-positive (HLA-B27+) individuals (9). However, the association between Klebsiella sp. and AS is still a subject of controversy since no definite evidence of the Klebsiella arthritogenic agent has yet been obtained (26). Klebsiella sp. is a gram-negative bacterium which can cause pneumonia, urinary tract infections, soft tissue infections, and septicemia. On the basis of the structural variability of its capsular polysaccharides, Klebsiella sp. has been classified into 77 serotypes (20). The capsular types differ in their pathogenicity and epidemiological relevance in various infectious diseases, a fact that has been well documented (5, 21, 22, 25). There are, however, no adequate epidemiological studies on the distribution and pathogenicity of Klebsiella serotypes in AS. Previous studies of Klebsiella sp. and AS have dealt with the Klebsiella serotypes arbitrarily, mainly with K21 and K43, rather than considering the entire 77 known Klebsiella serotypes. Hence, it cannot be excluded at this point that other Kiebsiella serotypes play a role in the pathogenesis of AS. Likewise, little is known about the role of Klebsiella capsular polysaccharides in AS, although these antigens have been demonstrated to be an important factor in the pathogenesis of various infectious diseases (7, 8, 11). Our recent enzyme-linked immunosorbent assay (ELISA) studies (27) of antibody responses to capsular polysaccharides
charides of Kiebsiella serotypes K26, K36, and K50 when compared with those of HLA-B27+ or HLA-B27-negative (HIA-B27-) healthy controls. These findings suggest that these capsular types might be involved in AS and might be the predominant capsular types in AS. However, the question remains whether these results are specific to AS. In the present study, the production of antibodies specific to the Klebsiella capsular types in sera from either HLA-B27+ or HLA-B27- patients with AS (AS+B27+ and AS+B27-, respectively) was analyzed by an ELISA. These sera were compared with sera from patients with systemic lupus erythematosus (SLE), psoriatic arthritis (PA), reactive arthritis after Yersinia enterocolitica infection (ReA), or rheumatoid arthritis (RA) and with HLA-B27+ or HLA-B27- healthy controls (AS-B27+ and AS-B27-, respectively).
MATERIALS AND METHODS Patients. Serum samples from patients with AS who were HLA-B27+ (n = 41) or HLA-B27- (n = 13) and from patients with SLE (n = 24), PA (n = 24), ReA (n = 20), or RA (n = 24) were obtained from the Department of Rheumatology at the Medical University of Lubeck, Lubeck, Germany, and the Clinic of Rheumatology in Bad Bramstedt, Germany. The AS patients were defined by the New York criteria (2). A diagnosis of ReA was substantiated by the detection of positive serum IgA titers against Yop protein and detection of Yopl in colon biopsies. Control sera from HLA-B27+ (n = 45) and HLAB27- (n = 40) healthy individuals were obtained from the Department of Transfusion Medicine, University of Kiel, Kiel, Germany. Patients and controls were HLA typed by using the two-stage microlymphocytotoxicity assay described by Terasaki and McClelland (32). Each serum sample was measured for C-reactive protein and total IgA, IgG, and IgM by using the
of the 77 known Kiebsiella serotypes in AS showed increased immunoglobulin G (IgG) responses to the capsular polysac* Corresponding author. Mailing address: Department of Medical Microbiology and Virology, University of Kiel, Brunswiker Str. 4, 24105 Kiel, Germany. Phone: 0049 431 5973316. Fax: 0049 431 5973296.
4838
KLEBSIELLA CAPSULAR POLYSACCHARIDES AND AS
VOL. 62, 1994
4839
TABLE 1. Patients investigated and routine laboratory findings Patient or control group (no. male/no. female)
Mean age (yr [range])
Mean ESRb (mm/h)
AS+B27+ (37/4) AS+B27- (10/3) SLE (4/20)
46 (21-71) 42 (28-64) 42 (21-71) 40 (12-69) 39 (20-62) 55 (29-70) 49 (30-69) 54 (30-69)
18 5 32
PA (8/16) ReA (11/9) RA (10/14) AS-B27+ (40/5) AS-B27- (31/9)
27 17 28
NTd NT
CRPC (mg/dl) 4.1 ± 6.7a
0.3 ± 0.2 3.5 ± 6 2 1.5 2.1 0.27 0.15
± 2.5 ± 5.2 ±2.9 ± 1.2 ± 0.3
Total IgA (mg/dl) 344 ± 152 344 ± 103 252 + 65
Total IgG
Total IgM
1,453 + 370 1,426 + 262 1,357 ± 213
252 ± 105 285 ± 131 252 ± 135
265 + 274 ± 311 ± 340 ± 333 ±
1,461 + 347 1,352 ± 274 1,370 358 1,455 + 245 1,371 ± 211
283 + 74 240 ± 98 225 ±79 243 ± 76 260 ± 68
150 110 95 139 108
(mg/dl)
(mg/dl)
Values are means ± standard deviations. b ESR, erythrocyte sedimentation rate. c CRP, C-reactive protein. d NT, not tested.
immunonephelometric method with commercial reagents from Behring. The erythrocyte sedimentation rate was measured by the Westergren method. Further data on patients and controls are shown in Table 1. Preparation of capsular antigens. The capsules were prepared as described previously by our group (27). Klebsiella isolates of capsular types Kl to K82 (NTCC and ATCC capsular reference serotypes) were grown on Worfel-Ferguson agar at 37°C for 24 h and for another 24 h at room temperature to increase capsule production. The bacterial lawn was suspended in 10 ml of 0.9% NaCl, and then the suspension was shaken vigorously for 3 min to detach the capsular polysaccharides. The suspensions were centrifuged at 3,000 x g for 10 min, and the supernatants were filtered (0.2-,urm-pore-size membrane) and lyophilized. The resulting lyophilisates were dissolved in 1 ml of distilled water, and the total carbohydrate content was determined by the phenol-sulfuric acid method described by Chaplin and Kennedy (6). The preparations were diluted in carbonate buffer (1.59 g of Na2CO3, 2.93 g of NaHCO3, and distilled water to 1,000 ml [pH 9.6]), giving a final concentration of 2 ,ug of carbohydrate per ml. Surface antigens of one clinical strain each of Citrobacter freundii, Serratia marcescens, Hafnia alvei, and Morganella morganii, three Streptococcus pneumoniae strains (NTCC K47 and two clinical isolates), and noncapsulated mutants of Kiebsiella capsular types K26, K36, and K50 were prepared as described above and used as negative controls. Noncapsulated mutants. Noncapsulated variants of K26, K36, and K50 were obtained from capsulated parent strains without mutagenic treatment as described by Podschun et al. (23). Briefly, serial dilutions of the parent strains were plated on brain heart infusion agar. Noncapsulated mutants appeared as slightly more translucent colonies or sector of colonies. Noncapsulated mutants were subcultured and tested for the absence of capsule antigen by the capsule swelling method with homologous sera. The API 20E system was used to confirm the biochemical identity of the noncapsulated mutants and parent strains. Identity of lipopolysaccharide (LPS) pattern in parent and mutant strains was demonstrated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of whole-cell lysate preparations. LPS analysis. LPSs of parent strains and mutants were examined by the whole-cell lysate method of Hitchcock (15). Briefly, a loop of bacteria grown on brain heart infusion agar was suspended in 100 ,ul of lysate buffer containing 2% SDS, 4% 2-mercaptoethanol, 10% glycerol, 1 M Tris (pH 6.8), and bromophenol blue. Lysates were heated at 100°C for at least 10 min and subjected to proteinase K digestion (1.25 mg/ml) at
60°C for 60 min. Preparations were separated by SDS-PAGE of the Laemmli system (16) without urea. The stacking gel was 5% polyacrylamide, and the separating gel was 10% polyacrylamide. Gels were stained subsequently by the silver stain method of Tsai and Frasch (35). ELISA. Ninety-six-well plates (MaxiSorp U 16; Nunc, Roskilde, Denmark) were coated overnight at room temperature with 100 ,u1 of antigen dissolved in coating buffer per well. The plates were washed four times with 0.1% Tween-PBS (Tween plus phosphate-buffered saline [pH 7.4]). One hundred microliters of ELISA blocking reagent per well (Boehringer GmbH, Mannheim, Germany) was incubated for 15 -min at room temperature to block nonspecific binding sites. The plates were washed four times, and to each well was added 100 ,ul of test and control sera diluted 1:1,000 in PBS. The plates were incubated at room temperature for 75 min and rinsed four times. One hundred microliters of biotinylated polyvalent goat anti-human serum (IgA, IgG, and IgM; heavy and light chain; Dianova, Hamburg, Germany), at a working dilution of 1:50,000, was added per well, and the plates were incubated at room temperature for 1.5 h. The plates were washed four times, and 100 ,ul of streptavidin peroxidase (Dianova) was added to each well at a dilution of 1:25,000. After 30 min of incubation, the plates were washed four times, and fresh substrate (o-phenylenediamine-hydrochloride) was added. After 25 min, the reaction was stopped with 50 ,ul of 0.5 N sulfuric acid, and the A492 was measured (ELISA processor II; Boehringer). A human anti-K23 serum sample from a patient with Klebsiella K23 septicemia served as a positive control. The specificity of the serum was tested by the capsular swelling reaction with Kiebsiella capsule reference strains. Apart from Klebsiella capsular type K23, this serum did not react with any of the other Kiebsiella capsular types. Pooled normal human serum (NHS) from healthy individuals was used as a negative control. The pooled serum did not react with any Kiebsiella serotype in the capsular swelling reaction. Each positive reaction was then titrated by an ELISA. Increasing dilutions of serum (1:1,000 to 1:5,000) were made in PBS and tested as described above. The end point was taken as the reciprocal of the highest dilution at which a positive reaction could still be clearly detected. Cross-reactions between several capsular types were eliminated by the titration, since the titer of a cross-reaction is lower than that of a homologous reaction (24, 30). All positive sera were subsequently analyzed for classspecific antibodies (IgA, IgG, and IgM) by using anti-human serum IgA (alpha-chain specific), anti-human serum IgG, or
4840
SAHLY ET AL.
INFECT. IMMUN.
TABLE 2. Incidence of serum samples with antibodies to Kiebsiella capsular types K26, K36, and K50 in patients and controls Patient or control group
AS+B27+a AS+B27SLE PA ReA RA
AS-B27+a AS-B27-a
% Serum samples positive to Klebsiella capsular polysaccharide(s)
K26
K36
K50
Total
32b
51b 30 12.5 8 20 25 11 7.5
54b
gob
23 25 25 25 12.5 22 27
37 40 32 51 33 33 40
15 8 0 16
0 5 10
0 II-
U)
control
Data reported previously (27). b Significantly higher frequency of antibody to Klebsiella capsular polysaccharides (P < 0.01). a
IgM (Dianova). The detection of the isotypes was performed under the same ELISA conditions as those described above. Statistical analysis. Differences between the groups were evaluated by Pearson's chi-square test. After a Bonferroni correction was performed, the significance of differences was subsequently evaluated by Yate's corrected chi square for 2 x 2 contingency tables. For tables showing a frequency of n of 0.2) (Table 3). Each positive IgG reaction was further titrated by ELISA to 1:8,000. The mean IgG titers to K26, K36, and K50 (5.15 x 103, 4.4 x 103, and 3.9 x 103, respectively) of Klebsiella-reactive sera from AS+B27+ patients were higher than the mean IgG titer of sera from the other control groups. However, these differences did not reach statistical significance in all cases
(Fig. 2). One clinical strain of S. marcescens, C. freundii, and H. alvei, one Morganella morganii strain of clinical origin, three S. pneumoniae strains (one NTCC strain K47 and two clinical isolates), and noncapsulated mutants of the Kiebsiella serotypes K26, K36, and K50 were used as negative controls. Noncapsulated variants from K26, K36, and K50 were obtained without mutagenic treatment. Loss of the K antigen was confirmed by the capsular swelling reaction. While parent strains reacted strongly with homologous antisera, no positive reaction could be observed in the mutant bacteria. No difference in biotype could be detected between mutant and parent strains. LPS patterns of noncapsulated mutants and their parent strains were determined by using SDS-PAGE with silver staining. The LPS patterns of capsulated and noncapsulated mutants were shown to be identical (Fig. 3). Antibodies to S. marcescens, C. freundii, H. alvei, M. morganii, S. pneumoniae, and
KLEBSIELLA CAPSULAR POLYSACCHARIDES AND AS
VOL. 62, 1994
4841
TABLE 3. Anti-capsular IgA, IgG, and IgM in Klebsiella-reactive serum samples from patients and controls K type
K26
Patient
or
AS+B27+ AS+B27SLE (2) PA (0) ReA (3) RA (0) AS-B27+ AS-B27-
(14)C
(2)
K50
a
IgGb
IgM
0.06 ± 0.06 0.03 ± 0.03 0.05 ± 0.002
0.4 ± 0.16
0.18 ± 0.11
0.37 ± 0.11 0.69 ± 0.25
0.16 ± 0.1 0.15 ± 0.03
0.06 ± 0.08
0.53 ± 0.13
0.23 ± 0.1
0.1 ± 0.1 0.02 ± 0.01
0.37 ± 0.18 0.27 ± 0.09
0.17 ± 0.02 0.16 ± 0.19
0.04 ± 0.01
0.1 ± 0.06
0.1 ± 0.06
0.07 ± 0.08 0.07 ± 0.07 0.03 ± 0.016 0.04 ± 0.002 0.06 ± 0.02 0.05 ± 0.03 0.06 ± 0.06 0.07 ± 0.11 0.04 ± 0.01
0.7 ± 0.28 0.87 ± 0.45 0.44 ± 0.2 0.82 ± 0.23 0.9 ± 0.3 0.67 ± 0.23 0.8 ± 0.13 0.94 ± 0.32 0.12 ± 0.06
0.1 ± 0.17 ± 0.1 ± 0.22 ± 0.2 ± 0.2 ± 0.16 ± 0.15 ± 0.13 ±
0.05 ± 0.08 ± 0.06 ± 0.07 ± 0.05 ± 0.06 ± 0.05 ± 0.05 ± 0.04 ±
0.72 ± 0.29 0.69 ± 0.35 0.64 ± 0.29 0.73 ± 0.13 0.46 ± 0.96 0.81 ± 0.41 0.66 ± 0.24 0.54 ± 0.1 0.10 ± 0.07
0.23 ± 0.1 0.18 ± 0.11 0.18 ± 0.03 0.22 ± 0.04 0.27 ± 0.05 0.19 ± 0.13 0.28 ± 0.13 0.23 ± 0.2 0.17 ± 0.1
Cf, 0
I0 0
(2)c (4)c
NHS
K36
AA492a
control
group (n)
AS+B27+ AS+B27SLE (3) PA (2) ReA (4) RA (6) AS-B27+ AS-B27NHS
(24)C (4)
AS+B27+ AS+B27SLE (6) PA (6) ReA (5) RA (3) AS-B27+ AS-B27NHS
(24)C (3)
(5)C (3)c
(10)C (11)C
0.07 0.05 0.025 0.03 0.016 0.03 0.09 0.07 0.02
0.07 0.12 0.009 0.12 0.12 0.1 0.04 0.03 0.02
Values are means ± standard deviations of sera with antibodies to Kiebsiella
capsular antigens. A4492 equalsA492 of specific reaction minus backgroundA492. P < 0.02 versus control NHS. c Data reported previously (27).
noncapsulated mutants of the Kiebsiella serotypes K26, K36, and K50 were absent or detectable in fewer than 10% of all serum samples. DISCUSSION
Capsular polysaccharides are an important factor in the virulence of bacteria. There is a well-documented association between certain capsular serotypes and various infectious diseases. In contrast, it has been proposed that capsular antigens are not involved in the arthritogenic activity of Kiebsiella sp., which causes AS in HLA-B27+ individuals (12). This suggestion is based on the observation that only some of the K21 and K43 strains that have been examined displayed interaction with HLA-B27+ cells from AS patients. However, apart from the lack of reproducibility (3), the previous studies of Klebsiella sp. and AS have not taken into account the whole spectrum of the 77 known Kiebsiella serotypes. Hence, it cannot be excluded at this point that there is an association between the capsular type and arthritogenic activity or that certain Kiebsiella capsular polysaccharides play a role in the pathogenesis of AS. In the search for antibodies to Kiebsiella sp. in AS, our recent serological study (27) was the first to consider the entire 77 Kiebsiella serotypes. Among the 77 Kiebsiella serotypes, a significantly elevated incidence of antibody response to K26, K36, and K50 was recorded for sera from HLA-B27+ patients with AS when compared with that recorded for sera from healthy controls. Thus, we suggested
NHS K26
K 36
K 5U
K Type FIG. 2. Mean titers of specific IgG to Kiebsiella capsular types K26, K36, and K50 in Klebsiella-reactive sera from AS+B27+ and AS+B27patients, sera from patients with SLE, PA, RA, or ReA, and sera from AS-B27+ and AS-B27- healthy individuals. An asterisk indicates that mean IgG titers to Kiebsiella capsular polysaccharides were significantly lower than those of AS+B27+ sera (P < 0.05).
that these serotypes predominate in AS. The aim of the present study was to verify whether these previous results are specific for HLA-B27+ AS. The antibody response to the entire 77 Kiebsiella capsular types was analyzed in sera of HLA-B27+ and HLA-B27- AS patients and compared with that in sera from patients with different rheumatic diseases. Our results show significantly higher frequencies of antibodies to capsular polysaccharides K26, K36, and K50 in sera from HLA-B27+ patients with AS than in sera from HLA-B27- AS patients, sera from patients with PA, SLE, RA, or ReA, and sera from healthy individuals. Antibodies to the other 74 Kebsiella capsular polysaccharides, to preparations of C. freundii, S. marcescens, H. alvei, M. morganii, and S. pneumoniae, and to noncapsulated mutants of K26, K36, and K50 were absent or detectable in fewer than 10% of all serum samples. These results might indicate a specific predominance of these
j~
~
~
0)
core
region -i A
B
C
D
E
F
FIG. 3. SDS-PAGE of LPS showing the core oligosaccharide region and the 0 polysaccharide repeating units from capsulated Kiebsiella reference strains K26 (A), K36 (C), and K50 (E) and from the corresponding noncapsulated mutants (B, D, and F).
4842
SAHLY ET AL.
Klebsiella capsular types in AS. The specific IgG titers to K26, K36, and K50 were markedly higher in AS+B27+ patients than controls. However, the differences did not reach statistical significance in all cases. The increased frequency of antibody response in AS+B27+ sera was not due to a nonspecific immune response since there was no difference in the total immune globulin levels between sera from AS patients and sera from the other groups. Morever, the lack of responsiveness to noncapsulated variants indicated that this antibody response is not due to contaminating LPS in the antigen preparation since no differences in the LPS and protein pattern could be detected between mutant and parent strains. Furthermore, it is remarkable that the antibody response to these capsular types did not differ significantly between patients suffering from other rheumatic diseases with characteristic polyclonal B-cell activation, such as SLE, and healthy controls. Thus, we suggest that this increased antibody response is specific to HLA-B27+ AS patients. Our findings that the antibodies to polysaccharides of K26, K36, and K50 in sera of AS+B27+ patients are of the IgG type contradict previous antibody studies by various groups which showed elevated anti-Kiebsiella antibodies mainly of the IgA type (10, 19, 29, 33, 34). For that reason, an intestinal site of antigen contact was assumed. However, the elevated IgG titers to capsular polysaccharides do not necessarily exclude a mucosal site of antigen contacts. Stodell et al. (31) found the number of IgG-containing cells in the rectal lamina propria to be higher in patients with AS than in controls. Therefore, it is conceivable that local antigens stimulate a local immunocyte reaction in the large bowel, as is postulated for chronic inflammatory diseases. At this stage, little is known about the distribution of these capsular types in the bowel of AS patients. Further isolation and serotyping of Klebsiella sp. from the bowel of AS patients are therefore necessary. HLA-B27, which occurs in 96% of patients with AS (4, 28), has been suggested to be the arthritogenic gene predisposing for AS (1). Two main theories exist as to how Kiebsiella sp. contributes to the development of AS in HLA-B27+ individuals (9). Ebringer's molecular mimicry theory suggests a crossreactivity between Kiebsiella sp. and some epitopes of the B27 molecule. By contrast, the receptor theory postulates that class I major histocompatibility complex acts as a receptor cavity for Klebsiella antigens, making it vulnerable to immunological attacks. Interestingly, in contrast to the HLA-B27+ sera from AS patients, HLA-B27- sera from AS patients did not show significant elevation of specific antibodies to Klebsiella capsular polysaccharides K26, K36, and K50 when compared with sera from healthy controls. These results might imply an HLA-B27restricted role of the capsular polysaccharides in AS. However, it is important to emphasize that these elevated antibody responses do not necessarily mean that the capsule is involved in the pathogenesis of AS. The arthritogenic activity of Kiebsiella sp. might be caused by other antigens such as peptides, which would be consistent with the fact that the molecular mimicry and the receptor theory have been suggested only for peptides (18). Nevertheless, the question remains whether Klebsiella capsular polysaccharides carry B-cell epitopes that cross-react with epitopes on the HLA-B27 molecule or whether they interact with the B27 molecule, rendering it susceptible to immunological processes. Recently, it was reported that the antibody response to pneumococcal capsular polysaccharides is regulated by T cells in a major histocompatibility complex-restricted manner (13). Since pneumococcal capsular polysaccharides cross-react with Kiebsiella capsular polysaccharides (14), a similar manner of immune response to
INFECT. IMMUN.
Klebsiella capsular polysaccharides is possible. This hypothetical model of arthritogenic activity of Kebsiella sp. in connection with HLA-B27 remains speculative and needs further verification. Proceeding from the assumption that Klebsiella sp. causes AS in connection with HLA-B27, two questions need to be answered. How do HLA-B27- individuals develop the clinical syndrome of AS, and how does this theory harmonize with the fact that 10% of the serum samples from AS+B27+ patients did not react with Klebsiella sp.? It has been suggested that HLA-B27+ AS and HLA-B27- AS are different entities with different pathogenic properties (17). However, this suggestion is speculative at this stage since no clear evidence has been obtained for a different pathomechanism in HLA-B27+ or HLA-B27- disease. Morever, doubt has been raised on the role of Kiebsiella sp. in the pathogenesis of AS in HLA-B27+ individuals (26). Clearly, to further clarify the role of Kiebsiella sp. in HLA-B27+ AS, intensive research taking K26, K36, and K50 into particular account is required.
1.
2.
3.
4. 5.
6. 7. 8.
9.
10.
11.
12.
13.
14. 15.
REFERENCES Baines, M., and A. Ebringer. 1992. HLA and disease. Mol. Aspects Med. 13:263-378. Bennet, P. H., and P. H. N. Wood. 1966. Population studies of the rheumatic diseases, p. 456. In P. H. Bennet and P. H. N. Wood (ed.), Proceedings of the Third International Symposium. Excerpta Medica Foundation, New York. Beukelmann, C., J. Quarles, H. C. Van Ufferd, and F. 0. M. J. Van Bree. 1990. Trial and error in producing ankylosing spondylitisselective antisera according to Andrew Geczy. Scand. J. Rheumatol. Suppl. 87:74-80. Caifrey, M. F. P., and D. C. 0. James. 1973. Human lymphocyte antigen association in ankylosing spondylitis. Nature (London) 242:121. Casewell, M., and H. G. Talsania. 1979. Predominance of certain Klebsiella capsular types in hospital in the United Kingdom. J. Infection. 1:77-79. Chaplin, M. F., and J. F. Kennedy (ed.). 1986. Carbohydrate analysis. IRL Press, Oxford. Cryz, S. J., E. Furer, and R. Germanier. 1984. Experimental Kiebsiella pneumoniae burn wound sepsis: role of capsular polysaccharide. Infect. Immun. 43:440-441. Domenico, P., W. G. Johanson, Jr., and D. C. Strauss. 1982. Lobar pneumonia in rats produced by clinical isolates of Klebsiella pneumoniae. Infect. Immun. 37:327-335. Ebringer, A. 1992. Ankylosing spondylitis is caused by Klebsiella. Evidence from immunogenetic, microbiologic, and serologic studies. Rheum. Dis. Clin. N. Am. 18:105-121. Ebringer, A., N. L. Cox, I. Abuljadayel, M. Shuloom, S. Khalafpour, T. Pfaszynska, F. Shodjai-Moradi, and C. Wilson. 1988. Klebsiella antibodies in ankylosing spondylitis and Proteus antibodies in rheumatoid arthritis. Br. J. Rheumatol. 27(Suppl. II): 72-85. Ehrenworth, L., and H. Baer. 1956. The pathogenicity of Klebsiella pneumoniae for mice: the relationship to the quantity and rate of production of type-specific polysaccharide. J. Bacteriol. 72:713717. Geczy, A. F., A. Kerri, H. V. Bashir, and J. Edmonds. 1980. A factor(s) in Klebsiella culture filtrates specifically modifies an HLA-B27-associated cell-surface component. Nature (London) 283:782-784. Griflioen, A. W., G. T. Rijkers, E. A. H. Toebes, and B. J. M. Zegers. 1991. The human in vitro anti-type 4 pneumococcal polysaccharide antibody response is regulated by suppressor T cells. Scand. J. Immunol. 34:229-236. Heidelberger, M., W. Nimmich, J. Eriksen, and S. Stirm. 1978. More on cross-reaction of pneumococci and Klebsiella. Acta. Pathol. Microbiol. Scand. Sect. B 86:313-320. Hitchcock, P. J. 1984. Analyses of gonococcal lipopolysaccharide in whole-cell lysates by sodium dodecyl sulfate-polyacrylamide gel
VOL. 62, 1994
16. 17. 18. 19. 20. 21. 22.
23. 24. 25. 26.
KLEBSIELLA CAPSULAR POLYSACCHARIDES AND AS
electrophoresis: stable association of lipopolysaccharide with the major outer membrane protein (protein I) of Neisseria gonorrhoeae. Infect. Immun. 46:202-212. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature (London) 227:680-685. Linssen, A. 1990. B27-positive versus B27-negative disease. Scand. J. Rheumatol. Suppl. 87:111-119. Maclean, L. 1992. HLA-B27 subtypes: implications for the spondylarthropathies. Ann. Rheum. Dis. 51:929-931. Maki-Ikola, O., K. Lehtinen, R. Vainionpaa, R., and P. Toivanen. 1991. Bacterial antibodies in ankylosing spondylitis. Clin. Exp. Immunol. 84:472-475. 0rskov, I., and M. A. Fife-Asbury. 1979. New Kiebsiella capsular antigen, K82, and the deletion of five of those previously assigned. Int. J. Syst. Bacteriol. 27:386-387. Podschun, R. 1990. Phenotypic properties of Klebsiella pneumoniae and oxytoca isolated from different sources. Zentralbl. Hyg. Umweltmed. 189:527-535. Podschun, R., P. Heineken, U. Ullmann, and H. G. Sonntag. 1986. Comparative investigations of Klebsiella species of clinical origin: plasmid pattern, biochemical reactions, antibiotic resistances and serotypes. Zentralbl. Bakteriol. Hyg. 262:335-345. Podschun, R., I. Penner, and U. Ullmann. 1992. Interaction of Klebsiella capsule type 7 with human polymorphonuclear leucoytes. Microb. Pathog. 13:371-379. Rennie, R. P., and I. B. R. Duncan. 1974. Combined biochemical and serological typing of clinical isolates of Klebsiella. Appl. Microbiol. 28:534-539. Riser, E., and P. Noone. 1981. Klebsiella capsular type versus site of isolation. J. Clin. Pathol. 34:552-555. Russel, A. S., and M. E. Suarez Almazor. 1992. Ankylosing
27.
28. 29. 30.
31.
32. 33.
34.
35.
4843
spondylitis is not caused by Klebsiella. Rheum. Dis. Clin. N. Am. 18:95-104. Sahly, H., R. Podschun, R. Sass, B. Broker, W. L. Gross, J. Kekow, and U. Ullmann. 1994. Serum antibodies to Klebsiella capsular polysaccharides in ankylosing spondylitis. Arthritis Rheum. 37: 754-759. Schlosstein, L., P. I. Terasaki, R. Bluestone, and C. M. Pearson. 1973. High association of an HLA antigen, B27, with ankylosing spondylitis. N. Engl. J. Med. 288:704-706. Shodjai-Moradi, F., A. Ebringer, and I. Abuljadayel. 1992. IgA antibody response to Klebsiella in ankylosing spondylitis measured by immunoblotting. Ann. Rheum. Dis. 51:233-237. Simoons-Smit, A. M., A. M. Verwel-van Vught, I. Y. R. Kanis, and D. M. MacLaren. 1985. Biochemical and serological investigations on clinical isolates of Klebsiella. J. Hyg. 95:265-276. Stodell, M. A., R C. Butler, V. A. Zemelman, K. Henry, and D. A. Brewerton. 1984. Increased number of IgG-containing cells in the rectal lamina propria of patients with ankylosing spondylitis. Ann. Rheum. Dis. 43:172-176. Terasaki, P. I., and J. D. McClelland. 1964. Microdroplet assay of human serum cytotoxins. Nature (London) 204:998-1000. Trull, A., A. Ebringer, G. S. Panayi, R Ebringer, and D. C. 0. James. 1984. HLA-B27 and the immune response to enterobacterial antigens in ankylosing spondylitis. Clin. Exp. Immunol. 55:74-80. Trmll, A., R Ebringer, G. S. Panayi, D. Colthorpe, and A. Ebringer. 1983. IgA antibodies in ankylosing spondylitis. Scand. J. Rheumatol. 12:249-253. Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal. Biochem. 119:115-119.
