Scand. J. Dent. Res. 1975: 83: 111-119 (Key words: immunotogy; oral mucosa; stomatitis, aphthous)
An immunofluorescence study on the cross-reaction between Strep. 2Aand human oral mucosa O. DONATSKY Department of Oral Surgery, Royal Dental College, Copenhagen, Denmark ABSTRACT - Immunologic cross-reactions between Strep. 2A, fetal human oral mucosa (FHOM), and adult human oral mucosa (AHOM) were investigated by the use of a double layer immunofluorescence staining technique. Rabbit sera were prepared against Strep. 2A and FHOM. Seventeen sera from patients with recurrent aphthous stomatitis (RAS) and with known antibodies against Strep. 2A and AHOM were examined for antibodies against FHOM. The distribution of endpoint titers against FHOM in the sera from RAS differed significantly from the distribution in 17 controls. The immune absorptions in the rabbit sera and in the 17 sera from patients with RAS indicate that some cross-reacting antigenic determinants are shared between Strep. 2A, FHOM, and AHOM. Further, the present results indicate that antibodies are produced against antigenic determinants which are not shared by Strep. 2A, FHOM, and AHOM. The role of cross-reacting antigens in the pathogenesis of RAS is discussed but remains obscure. (Received for publication 20 September, accepted 13 November 1974)
The immunologic aspects of recurrent aphthous stomatitis (RAS) (syn.: recurrent aphthous ulceration), have been dealt with in several studies (Table 1). The possibility of autoimmunity and bacterial hypersensitivity has been suggested (LEHNER 1964, 1967a, b, 1969a; GRAYKowsKi, BARILE, Lee & STANLEY 1966, BRODY & SiLVERMAN 1969, DoLBY 1969, FRANCIS
&
OPPENHEIM
1970,
SALLAY,
HAJOS & BANOCZY 1971C, DONATSKY BENDIXEN
1972,
DONATSKY
&
&
DABEL-
1974a, b). Some of these investiga. ,. ', , ..... ^ , tions mdirectly support the possibility of cross-reactions between fetal human oral STEEN
mucosa (FHOM), adult human oral mucosa (AHOM), and streptococcal antigens present in Strep, sanguis, prototype 2A. The purpose of the present investigation was to use the immunofluorescence (IF)technique to study whether Strep. 2A, FHOM, and AHOM contain crossreacting antigens.
^^l P"""'^"" °J T.TJXlf m Strep. 2 A, FHOM and AHOM was mvesti^^,^^ -^ immunofluorescence experiments as shown in Tables 2-5.
DONATSKY
112 Table 1
Review of the immunologic investigations on recurrent aphthous stomatitis (syn. recurrent aphthous ulceration) Autoimmunity LEHNER (1964, 1967a, b, 1969a, b, c, 1972) BRODY & SiLVERMAN (1969) DOLBY (1969, 1970a, b, 1972) DONATSKY & D.\BELSTEEN (1974b) Bacterial hypersensitivity GRAYKOWSKI et al. (1966) FRANCIS & OPPENHEIM (1970) SALLAY et al. (1971c) DONATSKY & BENDIXEN (1972) DONATSKY & DABELSTEEN (1974a)
Immunity/food proteins TAYLOR et al. (1964) ALP & WRIGHT (1971) THOMAS et aK (1973)
Immunity/virus SALLAY et al. (1971a, b) N.4sz et al. (\91\)
The antigens used were Strep. 2A, fetal human oral mucosa (FHOM), and adult human oral mucosa (AHOM). The Strep. 2A was the strain of Strep, sanguis used by DONATSKY & DABELSTEEN (1974a).
The streptococci were aerobically multiplied on membranes of cellophane placed on horse blood agar for 24-48 h at 37°C. The cultures were washed three times in 0.9% saline. A streptococcal suspension was adjusted at 500 X 10^ organisms per ml, heat killed at 60°C for 30 min and stored at +4°C. Fetal human oral mucosa (FHOM) was extracted in saline-added 2 % Triton® X-100. The extract was adjusted at 4 mg protein per ml and used as FHOM-antigen for immunizing the rabbits. The antisera used were rabbit serum prepared against Strep. 2A (Ra-anti Strep. 2A), rabbit seruin prepared against the saline extract of fetal human oral mucosa (Ra-anti FHOM), and 17 sera from patients with recurrent aphthous stomatitis (RAS) which earlier had shown antibodies against Strep. 2A and against adult human oral mucosa (AHOM). Furthermore, the endpoint titer against FHOM was examined in 17 randomly selected sera from controls.
The immunization of rabbits was carried out by Dakopatts Ltd., Copenhagen. The course of immunization with Strep. 2A lasted 16 weeks. The streptococcal suspension was diluted with equal parts of Freund's incomplete adjuvant. The animal was injected subcutaneously with 0.1 ml of the above Strep. 2A suspension four times at 2-week intervals. After 10 and 14 weeks the rabbit was further injected intravenously with 0.1 ml of the same Strep. 2A suspension. Bleeding took place in week number 16. Antiserum against fetal human oral mucosa (FHOM) was prepared by immunizing the rabbits subcutaneously. The course of this immunization lasted 10 weeks. The saline extract of fetal human oral mucosa was diluted with equal parts of Freund's incomplete adjuvant and 0.1 ml of the diluted suspension was injected four times at 2-week intervals. Bleeding took place in week number 10. The prepared rabbit sera were added 15 niM NaN3 and stored at 4-4°C. The human sera from patients with recurrent aphthous stomatitis were stored at -20°C. The method used for the examination of the sera was a double layer immunofluorescence (IF) staining method. Droplets of streptococcal suspension were placed on shdes and air dried. Biopsies from fetal human oral mucosa (FHOM) and from normal unkeratinized adult human oral mucosa (AHOM) were quick-frozen in cooled pure isopentane and stored in dry ice. The tissues were cut on a cryostat at — 30°C, mounted on slides and air dried for 15 min at room temperature before use. The titrations of antisera, the conjugates, the IF staining, and the microscopic system have been described previously (DABELSTEEN & RYGAARD 1972,
DONATSKY & DABELSTEEN
1974a, b). Heavy chain specific swine antirabbit IgG conjugated with fluorescein isothiocyanate (SwARa IgG/FITC) was included in the study of the rabbit sera prepared against Strep. 2A and FHOM. Like the other conjugates, the SwARa IgG/FITC was prepared and tested by Dakopatts Ltd., Copenhagen (DONATSKY & DABELSTEEN 1974a).
On every day of examination a serum with known activity against the antigens under investigation was included as positive control. Phosphate buffered saline was always included as negative control. Furthermore, in the investigation of the prepared rabbit sera a rabbit serum without activity against Strep. 2A, FHOM or AHOM was further used as a nega-
STREP. 2A AND ORAL MUCOSA tive control. Such a negative control was not available concerning the human sera used, as all our human sera had shown antibodies against Strep. 2A and AHOM (DONATSKY & DABELSTEEN 1974a, b). All the IF stainings were done at least twice, but most were done four or six times. The absorptions of the present human sera were carried out using diluted but strongly positive sera, i.e. if possible, the sera were diluted to a titer two steps below the endpoint titer before absorption, but if the endpoint titers were too small, undiluted sera were used. The prepared rabbit sera were always diluted to the titer 1:2 before absorption. All sera were absorbed with Strep. 2A and AHOM. As it was difficult to obtain fetal human oral mucosa for the present study, the only sera which were absorbed with FHOM were the rabbit sera prepared against Strep. 2A and FHOM. The absorptions were performed with equal parts of bacterial sediments or homogenized oral mucosal tissue for 3-5 h at 21° C followed by 1931 h at -l-4°C.
Results Positive reaction of a serum against FHOM or against AHOM appeared as a green fluorescence in the cytoplasm of the stratum spinosum cells. The figures in Table 2 show the endpoint titers (EPT) against Strep. 2A, FHOM, and AHOM in the rabbit sera prepared against Strep. 2A and FHOM, respectively. The immunization with Strep. 2A raised antibodies against Strep. 2A detectable with the present imm.uno-
113
Table 2 Endpoint titers (EPT) against Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM) in rabbit sera prepared against Strep. 2A (Ra-anti Strep. 2A) and against FHOM (Ra-anti FHOM) Endpoint titers (EPT) against Strep. 2A Ra-anti Strep. 2A Ra-anti FHOM
FHOM AHOM
1:64 1:8
1:16 1:16
1:8 1:8
fluorescence technique at the endpoint titer 1 .-64. Some of these antibodies seem to be cross-reacting with FHOM and AHOM, as antibodies against these tissues were detectable with endpoint titers of 1:16 and 1:8, respectively. The immunization with FHOM raised antibodies against FHOM with the endpoint titer 1:16. Some of these antibodies seem to be cross-reacting with Strep. 2A and AHOM. as antibodies against both of these tissues were detectable at an endpoint titer of 1:8. The distribution of the antibody endpoint titers against Strep. 2A, FHOM and AHOM in the sera from patients with recurrent aphthous stomatitis (RAS) is shown in Table 3. Fetal human oral
Table 3 Endpoint titers against Strep. 2A, adult human oral mucosa (AHOM) and fetal human oral mucosa (FHOM) in sera from patients with recurrent aphthous stomatitis (RAS)
Strep 2A AHOM FHOM
1:1
1:2
0 0 0
0 0 0
No. of sera with endpoint titer (EPT) 1:4 1:8 1:16 1:32 1:64 0 7 1
1
5 2 4
1:128
Total no. of sera 17 17 13
DONATSKY
114
with Strep. 2A, all the activity against Strep. 2A, FHOM and AHOM disap24 peared. A similar result was obtained 22 when the rabbit serum prepared against 20 FHOM (Ra-anti FHOM) was absorbed It with FHOM. The absorption of Ra-anti IS Strep. 2A serum with FHOM caused 1 14 1 disappearance of the antibody activity CONTROLS (17)/ RAS 113)/°' • 12 / against all the antigens investigated. 10 When the Ra-anti Strep. 2A serum was / / absorbed with AHOM, the antibody ac/ G tivity against AHOM disappeared, but / 4 / still antibodies against Strep. 2A and 2--• / FHOM were detectable. When the RaLOS ENDPOINT TITER 0 anti FHOM serum was absorbed with l:I6 1:32 1:64 i:12fl • ENDPOINT TITER Strep. 2A, the immunofluorescence stainFig. 1. Cumulative distributions of the end- ings were negative when Strep. 2A or point titers (EPT) against fetal human oral mucosa (FHOM) in 17 sera from controls and AHOM were used as antigens in the 13 sera from patients with recurrent aphthous staining reactions. The absorbed Ra-anti stomatitis (RAS). Sign - • - • - controls, -O-O- FHOM serum still showed antibody acRAS.) tivity against FHOM. When the Raanti FHOM serum was absorbed with mucosa (FHOM) was available for ex- AHOM, similar results were obtained amination of antibodies against FHOM in with negative staining reactions when only 13 sera from RAS. The cumulative AHOM or Strep. 2A were used as antidistributions of the endpoint titers against gens, but a positive staining reaction was FHOM in the 13 sera from RAS and in still seen when FHOM was used as anti17 sera from controls are shown in Fig. 1. gen. The cumulative distributions indicate that The results of the absorption experithe controls and the patients form two ments carried out with Strep. 2A or different populations. The antibody titers AHOM in 17 sera from patients with of the controls and the patients ranged RAS are shown in Table 5. Before abfrom 1:1 to 1:8 and from 1:4 to 1:32, re- sorption all 17 sera were strongly positive spectively. Statistical analysis using the to Strep. 2A and AHOM. The figures in Mann-Whitney U-test (SIEOEL 1956) Table 5 show that when the sera absorbed shows that the mean value of the end- with Strep. 2A or AHOM were used in point titer against FHOM in the group of the immunofluorescence staining reactions patients is significantly higher than the of the homologous antigens, all the sera mean value of endpoint titer in the group revealed negative reactions. When the of controls ( P < 0.001). same sera were prepared on the heterThe results of the absorptions of the ologous antigens, 53 % (9/17) of the sera rabbit sera and the sera from the patients absorbed with Strep. 2A showed negative with RAS are presented in Tables 4-5. reactions against AHOM and 65 % (11/ The results in Table 4 show that when 17) of the sera absorbed with AHOM the rabbit serum prepared against Strep. showed negative reactions against Strep. 2A (Ra-anti Strep. 2A) was absorbed 2A.
NUMBER o r CASES
f
315
STREP. 2A AND QRAL MUCOSA
Table 4 The results of the absorptions of rabbit sera prepared against Strep. 2A (Ra-anti Strep. 2A) and against fetal human oral mucosa (Ra-anti FHOM). These sera were absorbed with Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM), respectively. The effect of the different absorptions- was investigated by a double layer immunofluorescence technique using Strep. 2A, FHOM or AHOM as antigens Antigens Strep. 2A
Sera
Conjugate
Ra-anti Strep. 2A absorbed with Strep. 2A
SwARa IgG/FITC
FHOM Strep. 2A AHOM AHOM Strep. 2A FHOM Strep. 2A FHOM AHOM AHOM FHOM Strep. 2A
Ra-anti FHOM absorbed with FHOM
SwARa IgG/FITC
Ra-anti Strep. 2A absorbed with FHOM
SwARa IgG/FITC
Ra-anti Strep. 2A absorbed with AHOM
SwARa IgG/FITC
Strep. 2A
Negative Negative Negative Negative Negative Negative Negative Positive Positive
Ra-anti FHOM absorbed with Strep.2A
SwARa IgG/FITC
Negative Positive Negative
Ra-anti FHOM absorbed with AHOM
SwARa IgG/FITC
Negative Positive Negative
Table 5 Number of negative sera from 17 patients with recurrent aphthous stomatitis (RAS) after absorption with Strep. 2A or adult human oral mucosa (AHOM). The effect of the absorptions was investigated by a double layer immunofluorescence technique using Strep. 2A or AHOM as antigens Antigens
Negative Negative Negative
FHOM AHOM FHOM Strep. 2A AHOM
Results
Serum absorbed Serum absorbed with Strep. 2A with AHOM 100% (17/17) 5 3 % (9/17)
65% (11/17) 100% (17/17)
Discussion Evidence of immunologic cross-reaction between different microorganisms and tissues has been presented and summarized (KAPLAN 1964). Cross-reaction between Group A streptococci and human heart tissue has been demonstrated (KAPLAN & MEYESERIAN 1962, ZABRISKIE & FREIMER
1966). The possibility of cross-reaction between streptococci and human kidney was proposed by MARKOWITZ, ARMSTRONG & KusHNER (1960). The signifi-
116
DONATSKY
Fig. 2. Graphic illustration of the results of the absorption experiments performed in the rabbit serum prepared - against Strep. 2A (Ra-anti Strep. 2A), using Strep. 2A, fetal human oral mucosa (FHOM), and adult human oral mucosa (AHOM) as antigens. The circles illustrate the different antigens and the hatched area the prepared antibodies.
Fig. 3. Graphic illustration of the results of the absorption experiments performed in the rabbit serum prepared against fetal human oral mucosa (Ra-anti FHOM) using Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM) as antigens. The circles illustrate the different antigens and the hatched area the prepared antibodies.
cance of these streptococcal cross-reactions in the pathogenesis of rheumatic carditis and glomerulonephritis has been discussed but is still obscure (KAPLAN 1964, LANCET 1972). Further, cross-reaction between E. coli and colonic mucosa has been shown and discussed in relation to the pathogenesis of ulcerative coli-
against Strep. 2A were all absorbed by Strep. 2A or by FHOM. After the absorption of the Ra-anti Strep. 2A serum with AHOM, the serum still contained antibodies reacting with both Strep. 2A and FHOM. This indicates that all the antigenic determinants of Strep. 2A involved in the present immunization of the rabbit with Strep. 2A or antigenic determinants almost identical with the Strep. 2A determinants are present in FHOM but not in AHOM. That some antigenic determinants or cross-reacting determinants are shared by Strep. 2A and AHOM is indicated by the negative IF staining reaction when AHOM was used as antigen after the absorption with AHOM. All the antibodies prepared against FHOM were only absorbed by FHOM and not by Strep. 2A or AHOM. The absorptions with Strep. 2A or AHOM resulted in negative IF staining reactions when
tis (PERLMANN, HAMMERSTROM, LAGERCRANTZ & GUSTAFSSON 1 9 6 5 ) .
Different studies have previously demonstrated that patients with RAS show immunologic reactions against Strep. 2A, FHOM and AHOM (Table 1). Therefore, these antigens were chosen for the present investigations of the possibility of cross-reaction in RAS. The present results indicate some cross-reactions between Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM). This is graphically illustrated in Figs. 2-3. The antibodies prepared
117
STREP. 2A AND ORAL MUCOSA Strep. 2A or AHOM were used as antigens in the stainings after the absorptions. The IF staining was still positive when FHOM was used. This indicates that only some of the antigenic determinants in FHOM involved in the present immunization of the rabbits with FHOM, or antigenic determinants which are almost identical to those FHOM determinants, are present in Strep. 2A and AHOM. The cross-reacting determinants seem to be identical in Strep. 2A and AHOM. Furthermore, the immunization with FHOM caused antibodies against antigenic determinants not present in Strep. 2A or AHOM. It seems reasonable to suggest that these antibodies are antibodies against pure fetal human oral mucosa antigens. The possibility of cross-reaction between Strep. 2A, FHOM and AHOM was further supported by the findings of antibodies against Strep. 2A, FHOM and AHOM in all the sera from the patients with RAS investigated in the present IF study. Humoral antibodies against Strep. 2A and AHOM have previously been demonstrated by the IF technique (DONATSKY & DABELSTEEN 1974a, b). In the present study, circulating antibodies against FHOM were further detected both in the 13 sera from the patients with RAS and in the 17 sera from the controls. The results showing a statistically significant quantitative difference of the antibody titer to FHOM in the patients with RAS in relation to the controls confirms the results obtained by the agglutination test, the complement fixation test and the precipitation test (LEHNER 1964, 1967a). As the 17 sera from patients with RAS investigated in the present study were selected sera with known activity against Strep. 2A and AHOM, a comparison bethe distribution of endpoint titers
(EPT) against Strep. 2A and AHOM in the sera from the patients and the controls is not statistically reasonable and was therefore not performed.'' The results of the absorptions of the human sera with ^ Strep. 2A or AHOM (Table 5), further support the possibility of cross-reaction between Strep. 2A and AHOM. The figures in Table 5 show that 53 % and 65 % of the sera after absorption with Strep. 2A or AHOM showed negative IF staining reactions against AHOM and Strep. 2A, respectively. These results indicate that some antigenic determinants or cross-reacting antigenic determinants may be shared by Strep. 2A and AHOM, but in some sera antibodies against different non cross-reacting antigenic determinants in Strep. 2A and AHOM are present. The human sera from patients with RAS were not absorbed with FHOM. However, in the light of the present experimental IF results a similar effect of absorption with FHOM is likely. The present results indicating cross-reactions between Strep. 2A, FHOM, and AHOM are in agreement with the different studies showing humoral or cellular immunity to Strep. 2A, FHOM, and AHOM in patients with RAS (LEHNER 1964, 1967a, b, 1969a; GRAYKOWSKI et al. 1966, BRODY & SiLVERMAN 1969, DOLBY 1969, SALLAY et al. 1971c, DONATSKY & BENDIXEN 1972, STEEN 1974a, b).
DONATSKY
&
DABEL-
As far as a cross-reaction between Strep. 2A and FHOM is concerned, the conclusions on the present results are at variance with the conclusions presented by WILTON & LEHNER (1968) and LEHNER (1972). These author did not succeed in reverse absorption with Strep. 2A and FHOM using six sera from patients with RAS and guinea pig serum prepared against Strep. 2A. Nevertheless, these
118
DONATSKY
findings may show the same results as demonstrated and discussed in the present paper, i.e. that antigenic determinants which are not shared by Strep. 2A and FHOM may cause antibodies still reacting with the homologous antigenic determinants after absorption of the serum with the heterologous antigenic determinants. A similar phenomenon has been demonstrated and discussed in relation to rheumatic carditis (KAPLAN 1964). This author suggests that the presence of antibodies in the sera other than those which react with the cross-reacting antigenic determinants may prevent detection of the cross-reacting antibodies (KAPLAN 1964). The present results indicate cross-reactions between -Strep. 2A, FHOM, and AHOM. Previous studies have demonstrated humoral or cellular immunity to Strep. 2A, FHOM, and AHOM in patients with RAS (LEHNER 1964, 1967a, b, 1969a; GRAYKOWSKI et al. 1966, BRODY & SiLVERMAN 1969, DOLBY 1969, SALLAY et al. 1971c, DONATSKY & BENDIXEN 1972, DONATSKY & DABELSTEEN 1974a, b), indirectly indicating the possibility of crossreactions. The role of such cross-reacting antigens in the pathogenesis of RAS remains obscure. Some possibilities are (1) sensitization to streptococcal antigens with formation of antigen-antibody complexes in the oral mucosa, (2) antistreptococcal humoral and/or cellular immunity reacting directly against oral mucosa, (3) secondary release of oral mucosa cross-reacting with streptococcal antigens, and (4) secondary release of streptococcal antigens cross-reacting with oral mucosa. In conclusion the present IF study indicates that Strep. 2A FHOM, and AHOM share some cross-reacting antigens. The role of these antigens in the pathogenesis of recurrent aphthous stomatitis (RAS) is, however, obscure and needs further investigation.
Acknowledgments — The study was supported by Danish State Research Foundation, Copenhagen, Denmark. The author is grateful to Dr. K. STOLTZE, Institute for Graduate Studies, Royal Dental College, Copenhagen, for statistical advice. The fluorescence microscope was kindly provided by Dr. E. DABELSTEEN, Department of Oral Pathology, Royal Dental College, Copenhagen.
References ALP, M . H . & WRIGHT, R . : Autoantibodies to
reticulin in patients with idiopathic steatorrhoea, coeliac disease, and Crohn's disease, and their relation to immunoglobulins and dietary antibodies. Lancet 1971: 2: 682-685. BRODY, H . A. & SILVERMAN, S.: Studies on re-
current oral aphthae. I. Clinical and laboratory comparisons. Oral Surg. 1969: 27: 2734. DABELSTEEN, E . & RYGAARD, J.: A sensitive
immunofluorescence technique for detecting blood group substances A and B. Findings in oral epithelium. Acta Pathol. Microbiol. Scand. 1972: 80: 433-439. DOLBY, A. E.: Recurrent aphthous ulceration. Effect of sera and peripheral blood lymphocytes upon oral epithelial tissue culture cells. Immunology 1969: 17: 709-714. DOLBY, A. E.: Mikulicz's recurrent oral aphthae. The effect of hydrocortisone succinate sodium upon the in vitro lymphocyte cytotoxicity. Br. Dent. J. 1970a: 128: 579-580. DOLBY, A. E.: Mikulicz's recurrent oral aphthae. The effect of anti-lymphocyte serum upon the in vitro cytotoxicity of lymphocytes from. patients for oral epithelial cells. Clin. Exp. Immunol. 1970b: 7: 681-686. DOLBY, A. E.: The effect of lymphocytes from sufferers from recurrent aphthous ulceration upon colon cells in tissue culture. Gut 1972: 13: 387-389. DONATSKY, O . & BENDIXEN, G.: In vitro demonstration of cellular hypersensitivity to Strep. 2A in recurrent aphthous stomatitis by means of the leucocyte migration test. Acta Allergol. 1972: 27: 137-144. DONATSKY, O . & DABELSTEEN, E . : An immunofluorescence study on the humoral immunity to Strep. 2A in recurrent aphthous stomatitis. Acta Pathol. Microbiol. Scand. B 1974a: 82: 107-112. DONATSKY, O . & DABELSTEEN, E . : An immunofluorescence study on the humoral immu-
STREP. 2A AND ORAL MUCOSA nity to adult human oral mucosa in recurrent aphthous stomatitis. Acta AllergoL: in press 1974b. FRANCIS, T . C . & OPPENHEIM, J. J.: Impaired
lymphocyte stimulation by some streptococcal antigens in patients with recurrent aphthous stomatitis and rheumatic heart disease. Clin. Exp. Immunol. 1970: 6: 573-586. GRAYKOWSKI, E . A., BARILE, M . F., LEE, W .
B. & STANLEY, H . R . : Recurrent aphthous stomatitis. Clinical, therapeutic, histopathologic, and hypersensitivity aspects. /. Am. Med. Assoc. 1966: 196: 637-644. KAPLAN, M . H . : Immunologic cross-reaction between group A streptococcal cells and mammalian tissue. Possible relationship to induction of autoimmunity in rheumatic fever. In: UHR, J . W . (ed.): The streptococcus, rheumatic fever and glomerulonephritis. Williams & Wilkins Company, Baltimore 1964, p. 169-184.
lit
phritogenic streptococci. Nature 1960: 187: 1095-1097.
(Lond.)
NASZ, I., KuLCSAR, G., DAN, P. & SALLAY, K . :
A possible pathogenic role for virus-carrier lymphocytes. / . Infect. Dis. 1971: 124: 214216. PERLMANN,
P.,
HAMMERSTROM,
S.,
LAGER-
CRANTZ, R. & GusTAFSsoN, B. E.: Antigen from colon of germfree rats and antibodies in human ulcerative colitis. Ann. N. Y. Acad. Sci. 1965: 124: 377-394. SALLAY, K . , DAN, P., GECK, P., KULCSAR, G.
& NASZ, I.: Immunofluorescent studies on circulating lymphocytes in oral mucosal diseases. Arch. Derm. Forsch. 1971a: 241: 1521. SALLAY, K . , DAN, P., KULCSAR, G. & NASZ, L :
Transformation of lymphocytes from patients with recurrent aphthae. Rev. Immunol. 1971b: 35: 17-21. SALLAY, K . , HAJOS, M . K . & BANOCZY, J.: Un-
tersuchungen iiber den allergischen Pathomunological cross-reaction between group A mechanismus der chronisch rezidivierenden streptococcal cells and human heart tissue. Aphten. Allerg. Immunol. 1971c: 17: 17-23. Lancet 1962: 1: 706-710. SiEGEL, S.: Non-parametric statistics for the beLANCET (editorial): Streptococcal disease and havioral sciences. McGraw-Hill Book Comnephritis. Lancet 1972: 1: 129-130. pany, New York 1956, p. 116-127. LEHNER, T . : Recurrent aphthous ulceration TAYLOR, K . B., TRUELOVE, S. C. & WRIGHT, and autoimmunity. Lancet 1964: 2: 1154R.: Serologic reactions to gluten and cow's 1155. milk proteins in gastrointestinal disease. GasLEHNER, T . : Autoimmunity and management troenterol. 1964: 46: 99-108. of recurrent oral ulceration. Br. Dent. J. THOMAS, H . C , FERGUSON, A., MCLENNAN, J. 1967a: 122: 15-20. G. & MASON, D . K . : Food antibodies in oral LEHNER, T . : Stimulation of lymphocyte transdisease: a study of serum antibodies to food formation by tissue homogenates in recurproteins in aphthous ulceration and other rent oral ulceration. Immunology 1967b: 13: oral diseases. / . Clin. Pathol. 1973: 26: 371159-166. 374. LEHNER, T . : Pathology of recurrent oral ulceraWILTON, J. M. A. & LEHNER, T . : An investigation and ulceration in Behcet's syndrome: tion into the antigenic relationship between light, electron and fluorescence microscopy. oral bacteria and oral mucosa. / . Dent. Res. /. Pathol. 1969a: 97: 481-494. 1968: 47: 1001 (abstract). LEHNER, T . : Immunoglobulin estimation of ZABRISKIE, J. B. & FREIMER, E. H . : An immublood and saliva in human recurrent oral ulnological relationship between the group A ceration. Arch. Oral Biol. 1969b: 14: 357streptococcus and mammalian muscle. / . 364. Exp. Med. 1966: 124: 661-683. LEHNER, T . : Characterization of mucosal antibodies in recurrent aphthous ulceration and Behcet's syndrome. Arch. Oral Biol. 1969c: Address: 14: 843-853. LEHNER, T . : Immunologic aspects of recurrent Department of Oral Surgery oral ulcers. Oral Surg. 1972: 33: 80-85. Royal Dental College MARKOWITZ, A. S., ARMSTRONG, S. H . & 4 Universitetsparken KusHNER, D. S.: Immunological relation- DK-2100 Copenhagen 0 ships between the rat glomerulus and ne- Denmark KAPLAN, M . H . & MEYESERIAN, M . : An im-
An immunofluorescence study on the cross-reaction between Strep. 2Aand human oral mucosa O. DONATSKY Department of Oral Surgery, Royal Dental College, Copenhagen, Denmark ABSTRACT - Immunologic cross-reactions between Strep. 2A, fetal human oral mucosa (FHOM), and adult human oral mucosa (AHOM) were investigated by the use of a double layer immunofluorescence staining technique. Rabbit sera were prepared against Strep. 2A and FHOM. Seventeen sera from patients with recurrent aphthous stomatitis (RAS) and with known antibodies against Strep. 2A and AHOM were examined for antibodies against FHOM. The distribution of endpoint titers against FHOM in the sera from RAS differed significantly from the distribution in 17 controls. The immune absorptions in the rabbit sera and in the 17 sera from patients with RAS indicate that some cross-reacting antigenic determinants are shared between Strep. 2A, FHOM, and AHOM. Further, the present results indicate that antibodies are produced against antigenic determinants which are not shared by Strep. 2A, FHOM, and AHOM. The role of cross-reacting antigens in the pathogenesis of RAS is discussed but remains obscure. (Received for publication 20 September, accepted 13 November 1974)
The immunologic aspects of recurrent aphthous stomatitis (RAS) (syn.: recurrent aphthous ulceration), have been dealt with in several studies (Table 1). The possibility of autoimmunity and bacterial hypersensitivity has been suggested (LEHNER 1964, 1967a, b, 1969a; GRAYKowsKi, BARILE, Lee & STANLEY 1966, BRODY & SiLVERMAN 1969, DoLBY 1969, FRANCIS
&
OPPENHEIM
1970,
SALLAY,
HAJOS & BANOCZY 1971C, DONATSKY BENDIXEN
1972,
DONATSKY
&
&
DABEL-
1974a, b). Some of these investiga. ,. ', , ..... ^ , tions mdirectly support the possibility of cross-reactions between fetal human oral STEEN
mucosa (FHOM), adult human oral mucosa (AHOM), and streptococcal antigens present in Strep, sanguis, prototype 2A. The purpose of the present investigation was to use the immunofluorescence (IF)technique to study whether Strep. 2A, FHOM, and AHOM contain crossreacting antigens.
^^l P"""'^"" °J T.TJXlf m Strep. 2 A, FHOM and AHOM was mvesti^^,^^ -^ immunofluorescence experiments as shown in Tables 2-5.
DONATSKY
112 Table 1
Review of the immunologic investigations on recurrent aphthous stomatitis (syn. recurrent aphthous ulceration) Autoimmunity LEHNER (1964, 1967a, b, 1969a, b, c, 1972) BRODY & SiLVERMAN (1969) DOLBY (1969, 1970a, b, 1972) DONATSKY & D.\BELSTEEN (1974b) Bacterial hypersensitivity GRAYKOWSKI et al. (1966) FRANCIS & OPPENHEIM (1970) SALLAY et al. (1971c) DONATSKY & BENDIXEN (1972) DONATSKY & DABELSTEEN (1974a)
Immunity/food proteins TAYLOR et al. (1964) ALP & WRIGHT (1971) THOMAS et aK (1973)
Immunity/virus SALLAY et al. (1971a, b) N.4sz et al. (\91\)
The antigens used were Strep. 2A, fetal human oral mucosa (FHOM), and adult human oral mucosa (AHOM). The Strep. 2A was the strain of Strep, sanguis used by DONATSKY & DABELSTEEN (1974a).
The streptococci were aerobically multiplied on membranes of cellophane placed on horse blood agar for 24-48 h at 37°C. The cultures were washed three times in 0.9% saline. A streptococcal suspension was adjusted at 500 X 10^ organisms per ml, heat killed at 60°C for 30 min and stored at +4°C. Fetal human oral mucosa (FHOM) was extracted in saline-added 2 % Triton® X-100. The extract was adjusted at 4 mg protein per ml and used as FHOM-antigen for immunizing the rabbits. The antisera used were rabbit serum prepared against Strep. 2A (Ra-anti Strep. 2A), rabbit seruin prepared against the saline extract of fetal human oral mucosa (Ra-anti FHOM), and 17 sera from patients with recurrent aphthous stomatitis (RAS) which earlier had shown antibodies against Strep. 2A and against adult human oral mucosa (AHOM). Furthermore, the endpoint titer against FHOM was examined in 17 randomly selected sera from controls.
The immunization of rabbits was carried out by Dakopatts Ltd., Copenhagen. The course of immunization with Strep. 2A lasted 16 weeks. The streptococcal suspension was diluted with equal parts of Freund's incomplete adjuvant. The animal was injected subcutaneously with 0.1 ml of the above Strep. 2A suspension four times at 2-week intervals. After 10 and 14 weeks the rabbit was further injected intravenously with 0.1 ml of the same Strep. 2A suspension. Bleeding took place in week number 16. Antiserum against fetal human oral mucosa (FHOM) was prepared by immunizing the rabbits subcutaneously. The course of this immunization lasted 10 weeks. The saline extract of fetal human oral mucosa was diluted with equal parts of Freund's incomplete adjuvant and 0.1 ml of the diluted suspension was injected four times at 2-week intervals. Bleeding took place in week number 10. The prepared rabbit sera were added 15 niM NaN3 and stored at 4-4°C. The human sera from patients with recurrent aphthous stomatitis were stored at -20°C. The method used for the examination of the sera was a double layer immunofluorescence (IF) staining method. Droplets of streptococcal suspension were placed on shdes and air dried. Biopsies from fetal human oral mucosa (FHOM) and from normal unkeratinized adult human oral mucosa (AHOM) were quick-frozen in cooled pure isopentane and stored in dry ice. The tissues were cut on a cryostat at — 30°C, mounted on slides and air dried for 15 min at room temperature before use. The titrations of antisera, the conjugates, the IF staining, and the microscopic system have been described previously (DABELSTEEN & RYGAARD 1972,
DONATSKY & DABELSTEEN
1974a, b). Heavy chain specific swine antirabbit IgG conjugated with fluorescein isothiocyanate (SwARa IgG/FITC) was included in the study of the rabbit sera prepared against Strep. 2A and FHOM. Like the other conjugates, the SwARa IgG/FITC was prepared and tested by Dakopatts Ltd., Copenhagen (DONATSKY & DABELSTEEN 1974a).
On every day of examination a serum with known activity against the antigens under investigation was included as positive control. Phosphate buffered saline was always included as negative control. Furthermore, in the investigation of the prepared rabbit sera a rabbit serum without activity against Strep. 2A, FHOM or AHOM was further used as a nega-
STREP. 2A AND ORAL MUCOSA tive control. Such a negative control was not available concerning the human sera used, as all our human sera had shown antibodies against Strep. 2A and AHOM (DONATSKY & DABELSTEEN 1974a, b). All the IF stainings were done at least twice, but most were done four or six times. The absorptions of the present human sera were carried out using diluted but strongly positive sera, i.e. if possible, the sera were diluted to a titer two steps below the endpoint titer before absorption, but if the endpoint titers were too small, undiluted sera were used. The prepared rabbit sera were always diluted to the titer 1:2 before absorption. All sera were absorbed with Strep. 2A and AHOM. As it was difficult to obtain fetal human oral mucosa for the present study, the only sera which were absorbed with FHOM were the rabbit sera prepared against Strep. 2A and FHOM. The absorptions were performed with equal parts of bacterial sediments or homogenized oral mucosal tissue for 3-5 h at 21° C followed by 1931 h at -l-4°C.
Results Positive reaction of a serum against FHOM or against AHOM appeared as a green fluorescence in the cytoplasm of the stratum spinosum cells. The figures in Table 2 show the endpoint titers (EPT) against Strep. 2A, FHOM, and AHOM in the rabbit sera prepared against Strep. 2A and FHOM, respectively. The immunization with Strep. 2A raised antibodies against Strep. 2A detectable with the present imm.uno-
113
Table 2 Endpoint titers (EPT) against Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM) in rabbit sera prepared against Strep. 2A (Ra-anti Strep. 2A) and against FHOM (Ra-anti FHOM) Endpoint titers (EPT) against Strep. 2A Ra-anti Strep. 2A Ra-anti FHOM
FHOM AHOM
1:64 1:8
1:16 1:16
1:8 1:8
fluorescence technique at the endpoint titer 1 .-64. Some of these antibodies seem to be cross-reacting with FHOM and AHOM, as antibodies against these tissues were detectable with endpoint titers of 1:16 and 1:8, respectively. The immunization with FHOM raised antibodies against FHOM with the endpoint titer 1:16. Some of these antibodies seem to be cross-reacting with Strep. 2A and AHOM. as antibodies against both of these tissues were detectable at an endpoint titer of 1:8. The distribution of the antibody endpoint titers against Strep. 2A, FHOM and AHOM in the sera from patients with recurrent aphthous stomatitis (RAS) is shown in Table 3. Fetal human oral
Table 3 Endpoint titers against Strep. 2A, adult human oral mucosa (AHOM) and fetal human oral mucosa (FHOM) in sera from patients with recurrent aphthous stomatitis (RAS)
Strep 2A AHOM FHOM
1:1
1:2
0 0 0
0 0 0
No. of sera with endpoint titer (EPT) 1:4 1:8 1:16 1:32 1:64 0 7 1
1
5 2 4
1:128
Total no. of sera 17 17 13
DONATSKY
114
with Strep. 2A, all the activity against Strep. 2A, FHOM and AHOM disap24 peared. A similar result was obtained 22 when the rabbit serum prepared against 20 FHOM (Ra-anti FHOM) was absorbed It with FHOM. The absorption of Ra-anti IS Strep. 2A serum with FHOM caused 1 14 1 disappearance of the antibody activity CONTROLS (17)/ RAS 113)/°' • 12 / against all the antigens investigated. 10 When the Ra-anti Strep. 2A serum was / / absorbed with AHOM, the antibody ac/ G tivity against AHOM disappeared, but / 4 / still antibodies against Strep. 2A and 2--• / FHOM were detectable. When the RaLOS ENDPOINT TITER 0 anti FHOM serum was absorbed with l:I6 1:32 1:64 i:12fl • ENDPOINT TITER Strep. 2A, the immunofluorescence stainFig. 1. Cumulative distributions of the end- ings were negative when Strep. 2A or point titers (EPT) against fetal human oral mucosa (FHOM) in 17 sera from controls and AHOM were used as antigens in the 13 sera from patients with recurrent aphthous staining reactions. The absorbed Ra-anti stomatitis (RAS). Sign - • - • - controls, -O-O- FHOM serum still showed antibody acRAS.) tivity against FHOM. When the Raanti FHOM serum was absorbed with mucosa (FHOM) was available for ex- AHOM, similar results were obtained amination of antibodies against FHOM in with negative staining reactions when only 13 sera from RAS. The cumulative AHOM or Strep. 2A were used as antidistributions of the endpoint titers against gens, but a positive staining reaction was FHOM in the 13 sera from RAS and in still seen when FHOM was used as anti17 sera from controls are shown in Fig. 1. gen. The cumulative distributions indicate that The results of the absorption experithe controls and the patients form two ments carried out with Strep. 2A or different populations. The antibody titers AHOM in 17 sera from patients with of the controls and the patients ranged RAS are shown in Table 5. Before abfrom 1:1 to 1:8 and from 1:4 to 1:32, re- sorption all 17 sera were strongly positive spectively. Statistical analysis using the to Strep. 2A and AHOM. The figures in Mann-Whitney U-test (SIEOEL 1956) Table 5 show that when the sera absorbed shows that the mean value of the end- with Strep. 2A or AHOM were used in point titer against FHOM in the group of the immunofluorescence staining reactions patients is significantly higher than the of the homologous antigens, all the sera mean value of endpoint titer in the group revealed negative reactions. When the of controls ( P < 0.001). same sera were prepared on the heterThe results of the absorptions of the ologous antigens, 53 % (9/17) of the sera rabbit sera and the sera from the patients absorbed with Strep. 2A showed negative with RAS are presented in Tables 4-5. reactions against AHOM and 65 % (11/ The results in Table 4 show that when 17) of the sera absorbed with AHOM the rabbit serum prepared against Strep. showed negative reactions against Strep. 2A (Ra-anti Strep. 2A) was absorbed 2A.
NUMBER o r CASES
f
315
STREP. 2A AND QRAL MUCOSA
Table 4 The results of the absorptions of rabbit sera prepared against Strep. 2A (Ra-anti Strep. 2A) and against fetal human oral mucosa (Ra-anti FHOM). These sera were absorbed with Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM), respectively. The effect of the different absorptions- was investigated by a double layer immunofluorescence technique using Strep. 2A, FHOM or AHOM as antigens Antigens Strep. 2A
Sera
Conjugate
Ra-anti Strep. 2A absorbed with Strep. 2A
SwARa IgG/FITC
FHOM Strep. 2A AHOM AHOM Strep. 2A FHOM Strep. 2A FHOM AHOM AHOM FHOM Strep. 2A
Ra-anti FHOM absorbed with FHOM
SwARa IgG/FITC
Ra-anti Strep. 2A absorbed with FHOM
SwARa IgG/FITC
Ra-anti Strep. 2A absorbed with AHOM
SwARa IgG/FITC
Strep. 2A
Negative Negative Negative Negative Negative Negative Negative Positive Positive
Ra-anti FHOM absorbed with Strep.2A
SwARa IgG/FITC
Negative Positive Negative
Ra-anti FHOM absorbed with AHOM
SwARa IgG/FITC
Negative Positive Negative
Table 5 Number of negative sera from 17 patients with recurrent aphthous stomatitis (RAS) after absorption with Strep. 2A or adult human oral mucosa (AHOM). The effect of the absorptions was investigated by a double layer immunofluorescence technique using Strep. 2A or AHOM as antigens Antigens
Negative Negative Negative
FHOM AHOM FHOM Strep. 2A AHOM
Results
Serum absorbed Serum absorbed with Strep. 2A with AHOM 100% (17/17) 5 3 % (9/17)
65% (11/17) 100% (17/17)
Discussion Evidence of immunologic cross-reaction between different microorganisms and tissues has been presented and summarized (KAPLAN 1964). Cross-reaction between Group A streptococci and human heart tissue has been demonstrated (KAPLAN & MEYESERIAN 1962, ZABRISKIE & FREIMER
1966). The possibility of cross-reaction between streptococci and human kidney was proposed by MARKOWITZ, ARMSTRONG & KusHNER (1960). The signifi-
116
DONATSKY
Fig. 2. Graphic illustration of the results of the absorption experiments performed in the rabbit serum prepared - against Strep. 2A (Ra-anti Strep. 2A), using Strep. 2A, fetal human oral mucosa (FHOM), and adult human oral mucosa (AHOM) as antigens. The circles illustrate the different antigens and the hatched area the prepared antibodies.
Fig. 3. Graphic illustration of the results of the absorption experiments performed in the rabbit serum prepared against fetal human oral mucosa (Ra-anti FHOM) using Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM) as antigens. The circles illustrate the different antigens and the hatched area the prepared antibodies.
cance of these streptococcal cross-reactions in the pathogenesis of rheumatic carditis and glomerulonephritis has been discussed but is still obscure (KAPLAN 1964, LANCET 1972). Further, cross-reaction between E. coli and colonic mucosa has been shown and discussed in relation to the pathogenesis of ulcerative coli-
against Strep. 2A were all absorbed by Strep. 2A or by FHOM. After the absorption of the Ra-anti Strep. 2A serum with AHOM, the serum still contained antibodies reacting with both Strep. 2A and FHOM. This indicates that all the antigenic determinants of Strep. 2A involved in the present immunization of the rabbit with Strep. 2A or antigenic determinants almost identical with the Strep. 2A determinants are present in FHOM but not in AHOM. That some antigenic determinants or cross-reacting determinants are shared by Strep. 2A and AHOM is indicated by the negative IF staining reaction when AHOM was used as antigen after the absorption with AHOM. All the antibodies prepared against FHOM were only absorbed by FHOM and not by Strep. 2A or AHOM. The absorptions with Strep. 2A or AHOM resulted in negative IF staining reactions when
tis (PERLMANN, HAMMERSTROM, LAGERCRANTZ & GUSTAFSSON 1 9 6 5 ) .
Different studies have previously demonstrated that patients with RAS show immunologic reactions against Strep. 2A, FHOM and AHOM (Table 1). Therefore, these antigens were chosen for the present investigations of the possibility of cross-reaction in RAS. The present results indicate some cross-reactions between Strep. 2A, fetal human oral mucosa (FHOM) and adult human oral mucosa (AHOM). This is graphically illustrated in Figs. 2-3. The antibodies prepared
117
STREP. 2A AND ORAL MUCOSA Strep. 2A or AHOM were used as antigens in the stainings after the absorptions. The IF staining was still positive when FHOM was used. This indicates that only some of the antigenic determinants in FHOM involved in the present immunization of the rabbits with FHOM, or antigenic determinants which are almost identical to those FHOM determinants, are present in Strep. 2A and AHOM. The cross-reacting determinants seem to be identical in Strep. 2A and AHOM. Furthermore, the immunization with FHOM caused antibodies against antigenic determinants not present in Strep. 2A or AHOM. It seems reasonable to suggest that these antibodies are antibodies against pure fetal human oral mucosa antigens. The possibility of cross-reaction between Strep. 2A, FHOM and AHOM was further supported by the findings of antibodies against Strep. 2A, FHOM and AHOM in all the sera from the patients with RAS investigated in the present IF study. Humoral antibodies against Strep. 2A and AHOM have previously been demonstrated by the IF technique (DONATSKY & DABELSTEEN 1974a, b). In the present study, circulating antibodies against FHOM were further detected both in the 13 sera from the patients with RAS and in the 17 sera from the controls. The results showing a statistically significant quantitative difference of the antibody titer to FHOM in the patients with RAS in relation to the controls confirms the results obtained by the agglutination test, the complement fixation test and the precipitation test (LEHNER 1964, 1967a). As the 17 sera from patients with RAS investigated in the present study were selected sera with known activity against Strep. 2A and AHOM, a comparison bethe distribution of endpoint titers
(EPT) against Strep. 2A and AHOM in the sera from the patients and the controls is not statistically reasonable and was therefore not performed.'' The results of the absorptions of the human sera with ^ Strep. 2A or AHOM (Table 5), further support the possibility of cross-reaction between Strep. 2A and AHOM. The figures in Table 5 show that 53 % and 65 % of the sera after absorption with Strep. 2A or AHOM showed negative IF staining reactions against AHOM and Strep. 2A, respectively. These results indicate that some antigenic determinants or cross-reacting antigenic determinants may be shared by Strep. 2A and AHOM, but in some sera antibodies against different non cross-reacting antigenic determinants in Strep. 2A and AHOM are present. The human sera from patients with RAS were not absorbed with FHOM. However, in the light of the present experimental IF results a similar effect of absorption with FHOM is likely. The present results indicating cross-reactions between Strep. 2A, FHOM, and AHOM are in agreement with the different studies showing humoral or cellular immunity to Strep. 2A, FHOM, and AHOM in patients with RAS (LEHNER 1964, 1967a, b, 1969a; GRAYKOWSKI et al. 1966, BRODY & SiLVERMAN 1969, DOLBY 1969, SALLAY et al. 1971c, DONATSKY & BENDIXEN 1972, STEEN 1974a, b).
DONATSKY
&
DABEL-
As far as a cross-reaction between Strep. 2A and FHOM is concerned, the conclusions on the present results are at variance with the conclusions presented by WILTON & LEHNER (1968) and LEHNER (1972). These author did not succeed in reverse absorption with Strep. 2A and FHOM using six sera from patients with RAS and guinea pig serum prepared against Strep. 2A. Nevertheless, these
118
DONATSKY
findings may show the same results as demonstrated and discussed in the present paper, i.e. that antigenic determinants which are not shared by Strep. 2A and FHOM may cause antibodies still reacting with the homologous antigenic determinants after absorption of the serum with the heterologous antigenic determinants. A similar phenomenon has been demonstrated and discussed in relation to rheumatic carditis (KAPLAN 1964). This author suggests that the presence of antibodies in the sera other than those which react with the cross-reacting antigenic determinants may prevent detection of the cross-reacting antibodies (KAPLAN 1964). The present results indicate cross-reactions between -Strep. 2A, FHOM, and AHOM. Previous studies have demonstrated humoral or cellular immunity to Strep. 2A, FHOM, and AHOM in patients with RAS (LEHNER 1964, 1967a, b, 1969a; GRAYKOWSKI et al. 1966, BRODY & SiLVERMAN 1969, DOLBY 1969, SALLAY et al. 1971c, DONATSKY & BENDIXEN 1972, DONATSKY & DABELSTEEN 1974a, b), indirectly indicating the possibility of crossreactions. The role of such cross-reacting antigens in the pathogenesis of RAS remains obscure. Some possibilities are (1) sensitization to streptococcal antigens with formation of antigen-antibody complexes in the oral mucosa, (2) antistreptococcal humoral and/or cellular immunity reacting directly against oral mucosa, (3) secondary release of oral mucosa cross-reacting with streptococcal antigens, and (4) secondary release of streptococcal antigens cross-reacting with oral mucosa. In conclusion the present IF study indicates that Strep. 2A FHOM, and AHOM share some cross-reacting antigens. The role of these antigens in the pathogenesis of recurrent aphthous stomatitis (RAS) is, however, obscure and needs further investigation.
Acknowledgments — The study was supported by Danish State Research Foundation, Copenhagen, Denmark. The author is grateful to Dr. K. STOLTZE, Institute for Graduate Studies, Royal Dental College, Copenhagen, for statistical advice. The fluorescence microscope was kindly provided by Dr. E. DABELSTEEN, Department of Oral Pathology, Royal Dental College, Copenhagen.
References ALP, M . H . & WRIGHT, R . : Autoantibodies to
reticulin in patients with idiopathic steatorrhoea, coeliac disease, and Crohn's disease, and their relation to immunoglobulins and dietary antibodies. Lancet 1971: 2: 682-685. BRODY, H . A. & SILVERMAN, S.: Studies on re-
current oral aphthae. I. Clinical and laboratory comparisons. Oral Surg. 1969: 27: 2734. DABELSTEEN, E . & RYGAARD, J.: A sensitive
immunofluorescence technique for detecting blood group substances A and B. Findings in oral epithelium. Acta Pathol. Microbiol. Scand. 1972: 80: 433-439. DOLBY, A. E.: Recurrent aphthous ulceration. Effect of sera and peripheral blood lymphocytes upon oral epithelial tissue culture cells. Immunology 1969: 17: 709-714. DOLBY, A. E.: Mikulicz's recurrent oral aphthae. The effect of hydrocortisone succinate sodium upon the in vitro lymphocyte cytotoxicity. Br. Dent. J. 1970a: 128: 579-580. DOLBY, A. E.: Mikulicz's recurrent oral aphthae. The effect of anti-lymphocyte serum upon the in vitro cytotoxicity of lymphocytes from. patients for oral epithelial cells. Clin. Exp. Immunol. 1970b: 7: 681-686. DOLBY, A. E.: The effect of lymphocytes from sufferers from recurrent aphthous ulceration upon colon cells in tissue culture. Gut 1972: 13: 387-389. DONATSKY, O . & BENDIXEN, G.: In vitro demonstration of cellular hypersensitivity to Strep. 2A in recurrent aphthous stomatitis by means of the leucocyte migration test. Acta Allergol. 1972: 27: 137-144. DONATSKY, O . & DABELSTEEN, E . : An immunofluorescence study on the humoral immunity to Strep. 2A in recurrent aphthous stomatitis. Acta Pathol. Microbiol. Scand. B 1974a: 82: 107-112. DONATSKY, O . & DABELSTEEN, E . : An immunofluorescence study on the humoral immu-
STREP. 2A AND ORAL MUCOSA nity to adult human oral mucosa in recurrent aphthous stomatitis. Acta AllergoL: in press 1974b. FRANCIS, T . C . & OPPENHEIM, J. J.: Impaired
lymphocyte stimulation by some streptococcal antigens in patients with recurrent aphthous stomatitis and rheumatic heart disease. Clin. Exp. Immunol. 1970: 6: 573-586. GRAYKOWSKI, E . A., BARILE, M . F., LEE, W .
B. & STANLEY, H . R . : Recurrent aphthous stomatitis. Clinical, therapeutic, histopathologic, and hypersensitivity aspects. /. Am. Med. Assoc. 1966: 196: 637-644. KAPLAN, M . H . : Immunologic cross-reaction between group A streptococcal cells and mammalian tissue. Possible relationship to induction of autoimmunity in rheumatic fever. In: UHR, J . W . (ed.): The streptococcus, rheumatic fever and glomerulonephritis. Williams & Wilkins Company, Baltimore 1964, p. 169-184.
lit
phritogenic streptococci. Nature 1960: 187: 1095-1097.
(Lond.)
NASZ, I., KuLCSAR, G., DAN, P. & SALLAY, K . :
A possible pathogenic role for virus-carrier lymphocytes. / . Infect. Dis. 1971: 124: 214216. PERLMANN,
P.,
HAMMERSTROM,
S.,
LAGER-
CRANTZ, R. & GusTAFSsoN, B. E.: Antigen from colon of germfree rats and antibodies in human ulcerative colitis. Ann. N. Y. Acad. Sci. 1965: 124: 377-394. SALLAY, K . , DAN, P., GECK, P., KULCSAR, G.
& NASZ, I.: Immunofluorescent studies on circulating lymphocytes in oral mucosal diseases. Arch. Derm. Forsch. 1971a: 241: 1521. SALLAY, K . , DAN, P., KULCSAR, G. & NASZ, L :
Transformation of lymphocytes from patients with recurrent aphthae. Rev. Immunol. 1971b: 35: 17-21. SALLAY, K . , HAJOS, M . K . & BANOCZY, J.: Un-
tersuchungen iiber den allergischen Pathomunological cross-reaction between group A mechanismus der chronisch rezidivierenden streptococcal cells and human heart tissue. Aphten. Allerg. Immunol. 1971c: 17: 17-23. Lancet 1962: 1: 706-710. SiEGEL, S.: Non-parametric statistics for the beLANCET (editorial): Streptococcal disease and havioral sciences. McGraw-Hill Book Comnephritis. Lancet 1972: 1: 129-130. pany, New York 1956, p. 116-127. LEHNER, T . : Recurrent aphthous ulceration TAYLOR, K . B., TRUELOVE, S. C. & WRIGHT, and autoimmunity. Lancet 1964: 2: 1154R.: Serologic reactions to gluten and cow's 1155. milk proteins in gastrointestinal disease. GasLEHNER, T . : Autoimmunity and management troenterol. 1964: 46: 99-108. of recurrent oral ulceration. Br. Dent. J. THOMAS, H . C , FERGUSON, A., MCLENNAN, J. 1967a: 122: 15-20. G. & MASON, D . K . : Food antibodies in oral LEHNER, T . : Stimulation of lymphocyte transdisease: a study of serum antibodies to food formation by tissue homogenates in recurproteins in aphthous ulceration and other rent oral ulceration. Immunology 1967b: 13: oral diseases. / . Clin. Pathol. 1973: 26: 371159-166. 374. LEHNER, T . : Pathology of recurrent oral ulceraWILTON, J. M. A. & LEHNER, T . : An investigation and ulceration in Behcet's syndrome: tion into the antigenic relationship between light, electron and fluorescence microscopy. oral bacteria and oral mucosa. / . Dent. Res. /. Pathol. 1969a: 97: 481-494. 1968: 47: 1001 (abstract). LEHNER, T . : Immunoglobulin estimation of ZABRISKIE, J. B. & FREIMER, E. H . : An immublood and saliva in human recurrent oral ulnological relationship between the group A ceration. Arch. Oral Biol. 1969b: 14: 357streptococcus and mammalian muscle. / . 364. Exp. Med. 1966: 124: 661-683. LEHNER, T . : Characterization of mucosal antibodies in recurrent aphthous ulceration and Behcet's syndrome. Arch. Oral Biol. 1969c: Address: 14: 843-853. LEHNER, T . : Immunologic aspects of recurrent Department of Oral Surgery oral ulcers. Oral Surg. 1972: 33: 80-85. Royal Dental College MARKOWITZ, A. S., ARMSTRONG, S. H . & 4 Universitetsparken KusHNER, D. S.: Immunological relation- DK-2100 Copenhagen 0 ships between the rat glomerulus and ne- Denmark KAPLAN, M . H . & MEYESERIAN, M . : An im-
