Demonstration of antigen-specific immune response against Streptococcus sanguis

182 0

Journal of Dermatological Science, 5 ( 1993) 182- 189 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. 0923-181l/93/$06.00

DESC 00215

Demonstration

of antigen-specific immune response against Streptococcus sanguis

Norihisa

Ishii”, Emiko Isogaib, Yuko Yamakawaa, Hiroshi Nakajima”, Shigeaki Ohno”, Hiroshi Isogaid, Shunji Hayashi”, Kenji Yokotaf and Keiji Oguma’

Departments of ‘Dermatology and cOphthalmology, Yokohama City University School of Medicine, Yokohama, ‘Department of Preventive Dentistry, School of Dentistry, Higashi Nippon Gakuen University, Hokkaido, dDivision of Animal Experimentation and eDepartment of Microbiology. Sapporo Medical College, Sapporo and JDepartment of Microbiology, Okayama University School of Medicine, Okayama, Japan

(Received 21 January 1993; accepted 9 March 1993)

Key words: Antibody production; Footpad swelling; Genetic control; Helper T cells; Streptococcus sanguis; T cell proliferation

Abstract The genetic control of Streptococcus sanguis antigen response was studied. Mice sensitized with inactivated S. sanguis organisms antigen-injected at the base of the tail developed footpad swelling. Those with an I-Ak,q.’ region of H-2 showed a strong footpad response, whereas those with an I-AbvdgSregion showed a weak response to S. sanguis cell wall antigen. Footpad response was mediated by CD4+, 8- T cells by using in vitro monoclonal antibody treatment. Similar evidence of genetic control was obtained with an in vitro T cell proliferation assay. However, quantitation of antibodies against S. sanguis showed that antibody production was not controlled by H-2. These results indicated that both in vivo footpad swelling and in vitro T cell proliferation responses were functions of helper (CD3+, 4+, 87 T cells and controlled by the I-A region of H-2.

Introduction Behcet’s disease (BD) is a multisystemic disorder, mainly observed in Mediterranean areas and Japan. The etiology of BD is unknown, but Streptococcus sanguis has been implicated [ 1- 51. One possibility is that the mucosal manifestations of BD are triggered by strepCorrespondence to: Norihisa Ishii, Department of Dermatology, Yokohama City University School of Medicine, 3-9 Fukuura, Kanazawa-ku, Yokohama 236, Japan.

tococci, especially S. sanguis [2]. Arthritis and ocular manifestations may require a second trigger via the T cell network. Peptidoglycans of streptococci have a number of biological activities in mammalian systems, including functioning as an immunological adjuvant and activating macrophages. Associated cell wall polymers serve to protect the peptidoglycans to a variable extent from degradation by lytic enzymes. The cell wall complex of streptococci is a convenient agent for provoking chronic inflammation in experimental models. Injection of cell wall complexes can

183

induce arthritis [6] and it has been shown that T cells play a crucial role in chronic erosive streptococcal cell wall induced arthritis [7]. The present study investigated the antigenspecific immune response induced by cell wall and membrane antigens from S. sanguis isolated from patients with BD. Congenic and recombinant mice with different H-2 haplotypes were used for this study. We demonstrated that inoculation of mice with these antigens resulted in the induction of antigen-specific T cells and antibody response.

BlO.A(3R) and A.TL mice were kindly donated by Dr. Jan Klein, Max Planck Institute for Biology, Tubingen, Germany and BlO.A, A.BY. BlO.D2, BlO.A(2R), BlO.A(4R), BlO.BR, BlO.DA@ONS), BlO.RIII, A.SW, A.TH, BlO.HTT and BlO.T(6R) mice were kindly provided by the National Institute of Genetics, Mishima, Japan. These and other mice were maintained under specific pathogen-free conditions in our animal center at the Yokohama City University School of Medicine. Mice aged 8-15 weeks were used in the experiments.

Materials and Methods

Bacteria and antigen preparation S. sanguis 113-20 (a strain isolated from

Mice

the oral cavity of a patient with BD) was grown in BHI medium in anaerobic conditions at 37°C for 24 h. Sedimented bacteria were washed three times in phosphatebuffered saline (PBS) and stored at -20°C before use. Bacterial cell walls were prepared by a method described previously [8]. Briefly, S. sanguis organisms were sonicated in an ice bath with a cell disrupter Branson model 250 sonilier (Branson, CT) for 30 min and unbroken cells were removed by centrifugation (6000 x g, 4”C, 20 min). The supernatants were again centrifuged at 20 000 x g (4”C, 30 min) and the resultant supernatants were further purified by centrifugation at 100 000 x g (4”C, 3 h). The precipitate was applied to a Sephadex G- 1000 (Pharmacia, Uppsala, Sweden) column (2.2 x 45 cm) equilibrated with 10 mM phosphate buffer (pH 7.4) and designated as membrane antigen. The supernatant after centrifugation at 100 000 x g was designated as soluble membrane antigen. The precipitate of the first 6 000 x g centrifugation was treated with 0.2 mg/ml of proteinase K (Wako Pure Chemical, Osaka, Japan) at 37°C for 18 h. After being washed, the preparation was lyophilized and used as cell wall antigen.

The congenic strains used in this study and their H-2 haplotypes are listed in Table I.

TABLE I Genetic mapping of DTH to S. sanguis cell wall antigena Strain

H-2 region ~ KAESD

Footpad swelling (x10_2mm)

Response status

Mean f S.E. Bl0.A k k k d d C57BL/lO b b o b b b b o b b A.BY BlO.MBR b k k k q BlO.D2 d d d d d BIO.A(ZR) k k k d b BlO.A(4R) k k o b b BlO.A(3R) b b b d d BlO.BR k k k k k BlO.DA(80NS) q q o q s r r r r r BlO.RIII ASW s s 0 s s A.TL s k k k d A.TH ssosd BlO.HTT ssskd BlO.T(BR) q q o q d

43.3 zt 4.7 17.0 f 2.1 14.3 zt 1.8 42.7 f 3.6 14.8 f 2.7 49.7 zt 3.8 46.3 f 4.1 13.4 zt 1.7 53.2 zt 4.9 64.4 zt 5.8 53.8 + 2.2 19.7 zt 2.4 57.0 f 2.5 14.9 f 2.9 10.1 zt 2.8 60.0 f 5.6

High Low Low High Low High High Low High High High Low High Low Low High

aFootpad swelling was gauged 24 h after challenge. Each group of experiments involved six to ten mice. Footpad swelling of naive mice challenged without sensitization was 7-14 x 10-z mm.

184

Antibodies

Monoclonal Thy-l .2-specific, L3T4 (CD4)specific and Lyt-2.2(CD8)-specific antibodies were purchased from Cedarlane Laboratories, Ltd. (Hornby, Ontario, Canada). Sensitization and elicitation offootpad swelling S. sanguis antigen was diluted in Hanks’

balanced salt solution (HBSS, Gibco) and emulsified with an equal volume of incomplete Freund’s adjuvant (Difco, Hedingen, Stuttgart, Germany). Fifty million S. sanguis in a 50 ~1 volume were injected subcutaneously at the base of the tail. Ten days after sensitization, the mice were challenged in both hind footpads with 25 ~1 of a solution containing 1 pg of antigen. The thickness of the footpads was measured before and after challenge and the response was calculated as the difference (in units of 10m2mm). The results were expressed as mean * standard error (SE.) of the mean. Differences between experimental and control groups were determined using the Mann-Whitney test and P < 0.05 was taken as the level of significance. Adoptive transfer of DTH

Cells were collected from para-aortic and inguinal lymph nodes 10 days after mice were sensitized with S. sanguis. One volume of cell suspension (lO’/ml) was incubated with the appropriate concentration of each antibody at 4°C for 45 min. Then, after two washes of the suspension, one volume (lO’/ml) of low tox rabbit complement (Cedarlane) (diluted 1:4) was added and incubation was continued at 37°C for 1 h. The cells then were washed three times at 4°C with HBSS. A suspension of 4 x 10’ lymph node cells was injected intravenously into naive syngeneic mice via the tail vein in a 0.5 ml volume of HBSS. Lymph node cell proliferative assay

The proliferative assay previously described

by Ishii et al. [9] was employed. After sensitization, the draining lymph nodes (para-aortic and inguinal) were removed aseptically, and single-cell suspensions were prepared from the pooled lymph nodes of two to three mice. After two washes, the cells were resuspended in culture medium consisting of RPMI-1640 supplemented with 10% heat-inactivated fetal calf serum (Gibco), penicillin (100 units/ml), streptomycin (100 &ml) and L-glutamine (2 mM final concentration). The cells were cultured in 0.2 ml total volumes on microculture plates. Optimal responses were obtained with 4 x lo5 viable cells per well in the presence of antigen. Peak proliferative responses occurred on day 4 and were measured by thymidine incorporation, after 1 PCi of [3H]methyl thymidine was added for the last 15 h of culture. Proliferation was mediated by T cells, as shown by the fact that pretreatment of the lymph node cells with anti-Thy I .2 plus rabbit complement resulted in a markedly diminished response (data not shown). Each determination was performed in tive wells and data are expressed as counts/ min f standard deviation (SD.). Responses significantly above background (i.e. cells without antigen) were calculated using the unprimed t-test. Quantitation of antibodies specijk for S. sanguis

Antibodies were quantified by a modified ELISA as described by Ota et al. [lo]. Briefly, antigens were diluted in distilled water to a reaction volume of 50 ~1, then dispensed into immunoplate wells and allowed to dry completely by incubation at 37°C overnight. After blocking, the immunoplate wells were reacted at 37°C for 1 h with 100 ~1 of serum from mice of 10 days after sensitization and from nonsensitized mice. The wells were washed five times with Tween-PBS and reacted at 37°C for 1 h with anti-mouse IgG, IgGl, IgG2a,

18.5

IgG2b and IgG3 labelled with horseradish peroxidase (Zymed Co., Ltd., PA). The wells were finally washed with Tween-PBS and reacted with a substrate mixture of ophenylenediamine and H202 at 37°C for 10 min. Absorbance was measured at 492 nm with an immunoreader (EIA reader, Model MTP-22 Corona Electric Co., Ltd., Ibaraki, Japan).

into the hind footpads (Fig. 1). Footpad swelling was monitored for 96 h. Maximal footpad swelling was observed between 24 and 48 h in mice inoculated with cell wall or cell membrane. Soluble membrane antigen did not induce the DTH reaction. On the basis of these results, footpad swelling was measured 24 h after challenge with cell wall antigen in the following experiments.

Results

Genetic control of DTH response to S. sanguis antigen

DTH response using several fractions of S. sanguis

The next experiment was designed to study the genetic control of DTH response to S. sanguis cell wall antigen in various congenic and recombinant mice. As shown in Table I, BlO.A, BIO.MBR, BlO.A(ZR), BlO.A(4R), BlO.BR, BlO.DA(80NS), BlO.RIII, A.TL and BlO.T(6R) showed high responses, whereas C57BL/lO, A.BY, BlO.D2, BlO.A(3R),A.SW,

The DTH reaction was used to study T cell response to S. sanguis antigens in vivo. Bl0.A mice were injected subcutaneously at the base of the tail with inactivated S. sanguis organisms. Ten days later DTH reactions were induced by injecting S. sanguis antigens

l=&Q_

~

Fig. I. DTH response group of experiments

in Bl0.A involved

mice to cell wall (0). membrane five mice. Footpad

(r)

.u

48

72

and soluble

membrane

24 Hours

0

swelling of naive mice challenged

without

96

(a)

fractions

sensitization

from S. sanguis. Each was

1 to 8 x IO-* mm.

186 TABLE II Treatment of DTH effector lymph node cells with various monoclonal antibodies plus complementa Recipient Number of mice mice

A.TL A.TL A.TL A.TL A.TL A.TL

5 5 5 5 5 5

A.TL

5

Monoclonal antibody treatment C alone Anti-Thy- I .2 + C Anti-CD4 + C Anti-CD8 + C Serumb No transfer (positive control) No transfer (negative control)

Footpad swelling

(x 10-t mm) Mean f SE. 41.6 f 17.0 zt 20.3 zt 45.5 + 11.1 f 50.5 zt

3.2 2.7 3.0 3.2 2.8 2.8

side of the I-A region, while results obtained from C57BL/lO and BlO.MBR suggest that an Ir control gene maps to the right side of the IA region. These results therefore show that the S. sanguis cell wall specific DTH reaction is controlled by the genes within the I-A region. Treatment of DTH effector lymph node cells with various monoclonal antibodies

13.7 zt 2.4

aLymph node cells from sensitized A.TL mice were incubated with the indicated monoclonal antibodies at 4°C for 45 min, then with complement (C) for another 60 min. After three washes, 4 x lo7 lymph node cells were injected intravenously into naive syngeneic A.TL mice. After challenge with S. sang& footpad swelling was measured with a microdial thickness gauge. bSerum (0.5 ml) from A.TL mice sensitized with S. sanguis cell wall was transferred intravenously to naive A.TL mice.

A.TH and BlO.HTT showed low responses. These results suggest that the responses to S. sanguis cell wall were controlled by the H-2 complex. Furthermore, BlO recombinant studies revealed that the I-A region is important for these responses. Thus, the results obtained from C57BL/lO, BlO.A(QR) and BlO.BR indicate that an Ir control gene maps to the left

The next experiment was designed to investigate the component responsible for the reaction induced by S. sanguis cell wall antigen, Lymph node cells from sensitized A.TL mice were treated with various monoclonal antibodies plus complement. The treated cells were injected intravenously into naive syngenic A.TL mice. After challenge with S. sanguis, footpad swelling was measured. As shown in Table II, treatment of sensitized A.TL lymph node cells with Thy-l.Zspecific or CDCspecific antibody plus complement abolished the ability to transfer the footpad swelling response. However, treatment of these lymph node cells with CD8-specific antibody plus complement failed to modify this response. These results indicate that CD4+, 8- T cells are responsible for transfer of the footpad swelling response. T cell response to S. sanguis antigens in vitro

T cell proliferative

response was studied

TABLE III Specific T cell response to S. sanguis cell wall antigena Mouse strain

BIO.A C57BL/lO BlO.D2 BlO.BR

[3H]thymidine incorporation after stimulation with cell wall fraction antigen (counts/min * SD. (S.I.)b) 25 &ml

2.5 &ml

0.25 &ml

0 @ml

1 896 zt 451 (22.6) 951 zt 161 (3.1) 553 zt 255 (1.7) 1490 f 152 (14.5)

1053 zt 717 (12.5) 486 zt 65 (1.6) 413 f 231 (1.3) 870 zt 584 (8.4)

634 l 376 f 318 f 356 zt

84 zt 12 (1.0) 311 f 156 (1.0) 310 f 200 (1.0) 103 l 5 (1.0)

69 (7.5) 9 (1.2) 85 (1.0) 57 (3.5)

aLymph node cells from sensitized mice were incubated with the indicated concentration of S. sanguis cell wall antigen for 4 days. bStimulation index.

187

Cell wall

0.3

&I% r-l-r

0.2 8 P 0.1 n

IgGI

IgGeb

IgGza

*

Cell

,

I

lgGs membrane

*

0.3

8 0.2 P 0.1

t Soluble *

mzmbrane n

0.3

0.2 8 V 0.1

0

I&

IgG

I

IgG

2a

IgG

2b

IgG

3

Fig. 2. Antibody response to S. sanguis antigens in different inbred mouse strains. W Bl0.A. (U) BIO.BR, E3C57BL/lO, 0 BIO.D2. AOD, the difference between antibody response in sensitized mice and that in unsensitized mice (AOD + SD.). Each group of experiments involved five mice. * P < 0.01(two-sample r-test).

using several BlO recombinant mice. Mice were sensitized with cell wall antigen, and proliferative response was evaluated on day 4. As assessed by [ 3H]thymidine incorporation

in lymph node cells, the response was dependent on the dose of antigen (Table III). High responsiveness was observed in B 1O.A and BlO.BR with all concentrations of anti-

188

gen. However, low responsiveness was observed in C57BWlO and BlO.D2. These results indicated that the degree of responsiveness observed with footpad swelling was equivalent to that seen with T cell proliferation. Levels of IgG antibody to S. sanguis

The levels of serum IgG antibody to S. sanguis were evaluated by an ELISA in various strains of mice sensitized to S. sanguis. The differences between sensitized and unsensitized mice in the levels of each IgG and its subclass antibodies to S. sanguis are shown in Fig. 2. IgG2a antibodies showed no significant difference between sensitized and non-sensitized mice. The differences were seen among BlO congenic mice in responses to IgG and its subclass antibodies, but no responsive tendency was seen to antigens. Moreover, levels of antibodies in BIO.A (I-Ak) and BlO.BR (I-Ak) were not similar. These results indicated that antibody production against S. sanguis was not controlled by H-2. Discussion The genetic control of S. sanguis antigen response was studied by the methods of DTH, T cell proliferation assay and antibody titration. The I-A region of H-2 plays an important role in DTH. This in vivo reaction showed results similar to those observed in the T cell proliferation assay. These findings support the concept that identical T cells are responsible for footpad swelling and the T cell proliferative response to S. sanguis antigen. However, genetic control was not clear in the antibody titration experiment in the present study. The finding that mice with an I-Ak9q,’ region showed a strong reaction, whereas those with an I-Ab*d*sregion had only a weak response, indicates that responsiveness to S. sanguis antigens in mice is controlled by a gene of the H-2 class II. Genetic control

within the I-A region of class II has been observed for many antigens [9,1 l-151. Immunogenetic factors are critical in the susceptibility of an individual to BD, i.e. HLA-B51 and HLA-DRw52 were reported to be signiticantly increased in patients with BD [16]. Our results indicated that class II (I-A) genes influence the S. sanguis immune response, but that class I (K and D) genes do not. The difference in class I restriction between humans and mice suggests that the etiology of BD is multi-factorial, including S. sanguis and that several epitopes reactive to T cells exist with class I and class II genes. In the mouse, only a class II restricted response against S. sanguis cell wall antigen was shown, using the DTH assay and T cell proliferation. Response to S. sanguis cell membrane antigen also showed the same type of genetic restriction manner as that seen with cell wall antigen (data not shown). It might be that HLADRw52 was highly responsive to S. sanguis antigens and that HLA-B5 1 was influenced by other antigens. In the mouse experimental autoimmune uveoretinitis model, dual regulation of susceptibility by both MHC class II and background genes has been demonstrated using retinal antigen and Bordetella pertussix toxin [ 171.However, the S. sanguis induced immune response was controlled only by the class II region. Further research is necessary to understand the complex genetic mechanism in autoimmune diseases induced by bacterial infection. T cells play a crucial role in chronic erosive streptococcal cell wall induced arthritis [7]. These results indicated that CD4+ T cells are critical in the DTH response to streptococcal cell wall antigen. However, CD8+ T cells did not suppress the DTH response to the antigens. Antibody production against S. sanguis was not controlled by H-2, because antibody titration did not differ among mice with different

189

H-2 regions. These results seem to indicate that different kinds of immune-competent cells work in each response: T cells in the DTH and T cell proliferative responses and T and B cells in antibody production. It is conceivable that certain streptococcal infections may induce augmented expression of superantigen molecules such as heat shock proteins (HSP). Lehner et al. reported that some streptococcal antigens are associated with HSP and suggested their association with BD [18]. T cells recognize processed HSPs in the context of MHC molecules. Bacterial HSPs could be preferentially processed through the MHC class II pathway [19,20]. Our results with S. sanguis antigen showed class II restricted T cell regulation. Therefore, it would be interesting to analyze HSP in a model of S. sanguis infection. Further immunogenetic studies are needed to elucidate the etiopathogenesis of S. sanguis infection and BD. References The Behcet’s Disease Research Committee of Japan: Skin hypersensitivity to streptococcal antigens and the induction of systemic symptoms by the antigens in Behcet’s disease - A multiple study. J Rheumatol 16: 506-511, 1989. Isogai E, Isogai H, Fujii N, et al.: Adhesive properties of Streptococcus sanguis isolated from patients with Behcet’s disease. Microb Ecol Health Dis 3: 321-328, 1990. Isogai E, Ohno S, Kotake S, et al.: Chemiluminescence of neutrophils from patients with Behcet’s disease and its correlation with an increased proportion of uncommon serotypes of Streptococcus sanguis in the oral flora. Arch Oral Biol 35: 43-48, 1990. Isogai E, Ohno S, Takeshi K, et al.: Close association of Sfreprococcus sunguis uncommon serotypes with Bechcet’s disease. Biftdobact Microflora 9: 27-41, 1990. Isogai E, Isogai H, Yokota K, et al.: Platelet aggregation induced by uncommon serotpyes of Srreprococcus sanguis isolated from patients with Behcet’s disease. Arch Oral Biol 36: 425-429, 1991. Cromartie WJ, Craddock JG, Schwab JH, Anderle SK, Yang CH: Arthritis in rats after systemic injection of streptococcal cells or cell walls. J Exp Med 146: 1585-1602, 1977.

7

8

9

IO

II

12

13

I4

15

I6

17

I8

19

20

Yoshino S, Cleland LG, Mayrhofer G, Brown RR, Schwab JH: Prevention of chronic erosive streptococcal cell wall-induced arthritis in rats by treatment with a monoclonal antibody against the T cell antigen receptor ofi. J Immunol 146: 4187-4189, 1991. Yokota K, Hayashi S, Fujii N, et al.: Antibody response to oral streptococci in Behcet’s disease. Microbial Immunol 36: 815-822, 1992. Ishii N, Baxevanis CN, Nagy ZA, Klein J: Selection of H2 molecules for the context of antigen recognition by T lymphocytes. Immunogenetics 14: 283-292, 198I. Ota F, Kiso M, Akiyama Y, Satomi Y, Yasuoka M, Fukui K: Serological typing of reference strains and clinical isolates of streptococcus mutants by agglutination reactions and/or radioimmunoassay. Zentralbl Bakteriol Hyg A265: 330-339, 1987. Ishii N, Aoki I, Aihara Y, Hikawa N. Takahashi T, Okuda K: Analysis of hen egg white lysozyme (HEL)specific delayed type hypersensitivity hybridomas. Hybridoma 4: 311-318, 1985. Ishii N, Nakajima H, Aoki I, Ishii T, Takahashi T, Okuda K: Delayed-type hypersensitivity to allogeneic mouse epidermal cell antigens. I. Lyt-1+2- T cells are important for DTH. Immunogenetics 23: 351-356, 1986. Ishii N, Ichiyama S, Ishii T, Aoki I, Okuda K, Nakajima H: Genetic control of delayed-type hypersensitivity to Dpenicillamine antigen. Int Arch Allergy Appl Immunol85: 150-153, 1988. Ishii N, Ishii H, Ono H, Horiuchi Y, Nakajima H, Aoki I: Genetic control of nickel sulfate delayed-type hypersensitivity. J Invest Dermatol 94: 673-676, 1990. Ishii N, Ishii H, Nakajima H, Aoki I, Shimoda N: Contact sensitivity to mercuric chloride is associated with I-A region in mice. J Dermatol Sci 2: 268-273, 1991. Ohno S, Matsuda H: Studies of HLA antigens in Behcet’s disease in Japan, in Recenr Advances in BehGer S Disease. Edited by T Lehner, CG Barnes. Royal Society of Medicine Services, London, New York, 1986, pp 1I- 15. Caspi RR, Grubbs BG, Chan CC, Chader GJ, Wiggert B: Genetic control of susceptibility to experimental autoimmune uveoretinitis in the mouse model. Concomitant regulation by MHC and non-MHC genes. J Immunol 148: 2384-2389, 1992. Lehner T, Lavery E, Smith R, van der Zee R, Mizushima Y, Shinnick T: Association between the 65-kilodalton heat shock protein, Streptococcus sanguis, and the corresponding antibodies in Behcet’s syndrome. Infect Immun 59: 1431-1441, 1991. Koga T, Wand-Wiirttenberger A, DeBruyn J, Munk ME, Schoel B, Kaufmann SHE: T cells against a bacterial heat shock protein recognize stressed macrophages. Science 245: 1112-1115, 1989. Born W, Happ MP, Dallas A, et al.: Recognition of heat shock proteins and yS cell function. Immunol Today 11: 40-43. 1990.