International Immunology, Vol. A. No. 7, pp.
719-727
© 1992 Oxford University Press
Adjuvant arthritis and immunity to the mycobacterial 65 kDa heat shock protein Els J. M. Hogervorst, Jos6e P. A. Wagenaar1, Claire J. P. Boog1, Ruurd van der Zee1, Jan D. A. van Embden2, and Willem van Eden1 Department of Virology, University of Amsterdam, Amsterdam, The Netherlands 'Institute of Infectious Diseases and Immunology, Department of Immunology, Faculty of Veterinary Medicine, University of Utrecht, Utrecht, The Netherlands department of Bacteriology, National Institute of Public Health and Environmental Protection, Bilthoven, The Netherlands
Abstract The mycobacterial 65 kDa heat shock protein (HSP65) is of critical significance in the model of adjuvant arthritis (AA). Arthrltogenlc and protective T cell clones obtained from arthritic rats recognized the 180-188 sequence of HSP65. Previous reports have shown that administration of HSP65 prior to disease induction led to resistance to arthritis In the AA model and in several other models of experimental arthritis. Here, we report the development of Immunity to HSP65 and the critical 180-188 epitope during the course of AA. Following Mycobactertum tuberculosis (MT) immunization both antibodies and T cell responses to HSP65 were detected. Proliferative responses to the 180-188 epitope were seen exclusively in the local draining lymph node cells at day 14 after immunization. The anatomical distribution and course of T cell responses to HSP65 and its 180-188 epitope are compatible with T cell regulated control of the disease. Although lower HSP65 antibody levels were observed in the animals with severe arthritis, in individual animals no evidence was obtained for a relationship between development of HSP65 humoral immunity and arthritis severity. Nevertheless, during disease exacerbation, elicited by HSP65 immunization during disease development, elevated T cell responses against HSP6S and its 180-188 epitope were found. In contrast, we obtained evidence that successful transfer of arthritis resistance to naive recipients depends on the transfer of HSP65 specific T cells. On the basis of these results, it seems that HSP65 plays a crucial role in the T cell regulatory events involved in both the induction of, and protection against, AA. Although, these findings offer perspectives for Immunological intervention of arthritis using HSP65 itself or HSP65 specific T cells, the present findings on the versatile aspects of the immunomodulatory qualities of HSP65 should be taken into account in designing protocols for triggering protective T cell regulation.
Introduction The mycobacterial 65 kDa heat shock protein (HSP65) belongs to the HSP60 family of heat shock or stress proteins. A number of different families of HSPs are known, for example the HSP90, HSP70, and HSP60 family. Besides the mycobacterial HSP65, the HSP60 family comprises the Escherichia coli GroEL, the Rubisco subunit-binding protein in plants, and mammalian HSP60 (1 -3). Synthesis of HSPs is subject to regulatory control dependent on sudden rises of temperature or other stressful events, such as anoxia, the presence of oxygen radicals, nutrient deprivation or infection with microorganisms (4,5). HSPs are present in all living cellular organisms and have retained a
remarkable degree of evolutionary conservation. For the HSP60 family of HSPs the homologies found between bacterial and mammalian counterparts are 50%, in terms of amino acid sequence and an additional 20% when taking into account conserved ammo acid substitutions (2). The strong evolutionary conservation of HSPs indicates that they have similar essential functions for cells from a great diversity of organisms. Some HSPs are also called molecular chaperonins or house-keeping proteins and are, among others, involved in folding, unfolding, and translocation over intracellular membranes of proteins and the assembly or disassembly of complex proteins (4.6,7). The
Correspondence to: E. Hogervorst Transmitting editor: S. H. E. Kaufmann
Received 31 January 1992, accepted 17 March 1992.
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
Key words: autoimmunity, T cell proliferation, immunomodulation, antibody production
720
Mycobacterial HSP65 in adjuvant arthritis
Not only in animal models, but also in patients suffering from rheumatoid arthritis the role of HSP65 has been studied. T cells taken from the synovial fluids of such patients responded to HSP65 (22-25). Children suffering from juvenile rheumatoid arthritis were found to respond to the human HSP6O (26) In adult rheumatoid arthritis patients, rased levels anti-HSP65 antibodies were also found (27,28) and selective expression of HSP60 in the inflamed synovial tissues (29-31) was demonstrated. To understand more about the role and regulatory effect of HSP65, we studied the relation between the humoral immune responses directed to HSP65 and arthritis development. Furthermore, the cellular immune responses against HSP65 and the 180 - 1 8 8 epitope were investigated during disease. In addition, we have examined the effects of the administration of HSP65 itself or HSP65 responsive T cells during arthritis development. Our data confirm the critical significance of HSP65 immunity in AA and the protection against AA found after administration of HSP65 specific T cells may offer possibilities for immunomodulation via T cell vaccination using HSP65- activated T cells.
Methods
a M. tuberculosis HSP70 expressing recombinant E. coli stran (33). The 180-188 peptide was prepared by solid-phase techniques previously described by Steward and Young (34). Arthritis induction with MT Rats were immunized intradermally at the base of the tail with 100 /J of 10 mg/ml MT in mineral oil (Incomplete Freund's Adjuvant, Difco Laboratories, Detroit, Ml). The rats were observed for clinical arthritis and were scored by grading each paw from 0 to 4 based on erythema, swelling and deformity of the joint. 0 = no erythema or swelling, 1 = slight erythema or swelling of one of the toes, 2 = erythema and swelling of more than one toe, 3 = erythema and swelling of the ankle, and 4 = complete erythema and swelling of toes and ankle and the incapacity to bend the ankle. All four legs were scored, so the highest score achievable was 16. ELISA (or the detection of HSP specific antibodies HSP65 or HSP70 (0.2 pg) was used to coat wells of (Greiner) ELISA plates. The plates were incubated for 2 h at 37°C. The wells were washed with phosphate-buffered saline (pH 7.4) containing 0.05% (v/v) Tween 20 (PBS-Tween). The sera, obtaned from MT immunized rats, were added in dilution of 1:100 and 1 200 in PBS - Tween and incubated for 2 h at 37°C. After washes with PBS-Tween, goat-anti-rat Ig conjugate (Nordic, Tilburg, The Netherlands) was added in a dilution of 1:4000 in PBS-Tween and incubated for 2 h. After washing, the ABTS substrate (2,2-azinodi-bis(3-ethylbenzthiazoline-6-sulphonic acid) (Sigma, no. A-1888) in 0.1 M citrate phosphate buffer and H2O^) was added and after 0.5 h the absorbance at 405 nm was measured in a Titertek multiscan reader. Proliferative responses At consecutive points of time after MT immunization, lymph nodes and spleens were removed and cell suspensions were made. The erythrccytes present in the spleen were removed by a FicollIsopaque gradient. Proliferative responses were measured in flatbottom microtiter plates (Costar) in triplicate wells. Each well contained 2 x i O 5 cells suspended in 0.2 ml DMEM (Gibco), supplemented with 10% FCS (Gibco), 5 x 10~5 M /3-mercaptoethanol, glutamine, penicillin, and streptomycin, in the presence or absence of antigen. The cells were cultured for 3 days and for the last 16 h pulsed with [3H]thymidine. The cultures were harvested on fibreglass filters and thymidine incorporation was measured in a Beta-plate (LKB, Wallace Oy, Turku, Finland) in counts per minute (c.p.m.).
Animals Inbred male Lewis rats were obtained from the University of Ljmburg, Maastricht, The Netherlands. All rats were used at 6 - 8 weeks of age.
Exacerbation of the disease by administration of HSP65 At day 0 rats were immunized with MT for arthritis induction. At day 7, 9, and 11, 250 ng HSP65 in 250 /»l PBS was injected i.v. and arthritis development was scored.
Antigens The following antigens were used for immunization of the animals or in proliferation assays: heat-killed MT H37 RA (Difco Laboratories, Detroit, Ml), the 65 kDa HSP was derived from E. coli K12 carrying plasmid pRIB1300 expressing the M. bovis BCG HSP65 protein gene (9). The protein was purified from a temperature-induced culture according to the procedure described by Thole ef al. (32). The 70 kDa HSP was derived from
Protection induction by spleen cells obtained from rats recovered from AA Rats were immunized with MT and at day 35, when remission of the disease was seen, the spleens were isolated. Erythrccytes were removed by a Ficoll-lsopaque gradient and after washing the cells, 5 x 108 cells were injected intravenously in naive rats. At the same time 1 mg MT was injected at the base of the tail.
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
mycobacterial HSP65 has been shown to be a strongly immunogenic protein (8). In addition, it has been demonstrated to be involved in experimentally induced adjuvant arthritis (AA) (9). AA is an animal model for rheumatoid arthritis and is induced in Lewis rats by immunization with heat-killed Mycobacterium tuberculosis (MT) in Freund's incomplete adjuvant (10). The first indications that HSP65 played a role in AA were obtained by the characterization of arthritogenic and protective T cell clones generated from arthritic rats (11-13). The epitope recognized by these clones was identified as a nine amino acid sequence (180-188) of mycobacterial HSP65 (9). Later, the epitope was further characterized as amino acid sequence 180-186 (14). Furthermore, vaccination with HSP65 also has been shown to induce protective immunity in various bacterial induced experimental arthritis models, like AA, streptococcal cell wallinduced arthritis (9,15,16) and in non-bacterial associated pristane-induced arthritis (17). Partial protective effects were seen in collagen type ll-induced arthritis (16,18). Previously, we have shown that responsiveness to the 180 - 1 8 8 epitope was related to susceptibility to AA. In addition, a role for HSP65 specific T cells in controlling the T cell responses to the 180 - 1 8 8 epitope was described (19). Recently, it was shown that the 180-188 epitope by itself could induce protection against AA (20,21).
Mycobacterial HSP65 in adjuvant arthritis 721 (Table 2). No significant levels of antibodies with specificity for HSP70 were found (Table 1).
Treatment of AA by HSP65 specific T cells Rats were immunized with MT for arthritis induction. At day 14 post-immunization the inguinal lymph nodes were removed and a cell suspension was made. The cells were tested for proliferation to HSP65 and 180 - 1 8 8 and also cultured for 3 days in the presence of HSP65. The cells were cultured at a concentration of 5 x 106 cells/ml in the presence of 10 ^g/ml HSP65. After 4 days dead cells were removed by a Ficoll-lsopaque gradient and after washing the cells 3 times, 1 x 1 0 ' cells were injected i.v. into rats who had received MT for arthritis induction 14 days earlier. The remaining cells were cultured at a concentration of 5 x 106 ceils in the presence of 5% EL-4 supernatant as source of IL-2, and 3 and 7 days later the animals received another injection of 1 x 107 cells and arthritis scores were followed.
Cellular immune responses in AA
Statistical analysis For data concerning arthritis scores the Mann-Whitney IMest was used. For analysis of the antibody data, Student's r-test was performed.
In order to monitor cellular responses to HSP65 during AA, lymphocyte proliferation assays were performed with cells obtained from different lymphoid organs at various stages of AA. The development and incidence of the AA in our stock of Lewis rats is shown in Fig. 1. On average, disease started at day 14.9 ± 0.3 (mean day ± SEM). The mean maximum arthritis score measured was 10.8 ± 0.4 and the mean day of the maximum arthritis score was found at day 21.7 ± 0.4. At different stages of the disease spleen and lymph node cells were isolated and proliferate responses directed to MT, HSP65, and the 180-188 epitope of HSP65 were determined. We tested a concentration range from 1 to 25 /ig/ml and in Fig. 2 the results obtained with an optimal concentration of 10 ^g/ml MT, HSP65, and 180 -188 peptde are shown. Every point of time represents 5 - 2 5 tested animals. We also tested spleen cells obtained from naive animals for their responses to the antigens. The responses to MT, HSP65, and the 180-188 epitope, given as stimulation indexes (±SEM), were 4.5 ± 0.4, 2.4 ± 0.4, and 1.4 ± 0.2 respectively. Seven days after immunization low responses against MT were found in the draining inguinal lymph node cells (Fig. 2A) and no responses to HSP65 or peptide 180 - 1 8 8 were
Results Humoral immune responses in AA To study the humoral immune responses to HSP65 in AA, Lewis rats were immunized intradermally at the base of the tail with MT. At various times, blood samples were taken and tested for the presence of antibodies directed to HSP65 and HSP70 using an ELISA. The results are shown in Table 1. No HSP65 specific antibodies were detected prior to MT immunization or 9 days after arthritis induction. However, at day 14 high levels of antibodies were found. The antibody levels started to decline at days 28 and 50, but still exceeded the levels measured preceding MT immunization. We also compared the severity of the arthritis in relation to the amount of HSP65 specific antibodies present (measured at days 28 and 50). In animals with low arthritis scores (maximum score of 0 - 4 ) , the levels of antibodies slightly exceeded those of animals suffering from more severe arthritis
Table 2. Correlation between HSP65 antibody levels and severity of AA
14 32
Arthritis score3
HSP65 antibodies'5
0-4 5-16
0.96 ± O.Otf 0.75 ± 0.05
a
Animals were grouped according to their maximal arthritis score measured during disease. 'The ELISA was performed with sera obtained at day 28 or 50 after MT administration. c Significantly different (P < 0 05)
1OO
Table 1. Development of antibodies8 against HSP65 and not to HSP70 in MT-immunized animals Days after immunization
OD value of sera tested for antibodies directed to HSP65 HSP70
0 g 14 28 50
0.12 0.13 1.05 o.g2 0.71
± 0.01
b
0.01 b ± 0.03= ± 0.05 ± 0.07
0.13 0.13 0.15 019 0.20
± ± ± ±
0.01 0.01 0.02 0.01 0.01
a The sera were tested in a dilution erf 1:100 and shown is the optical density ±SEM measured at 405 nm (see Methods). "Significantly different from day 14 (P < 0.005), day 28 (P < 0.005), and day 50 (P < 0.05). c Significantly different from day 0 (P < 0.005), day 9 (P < 0.005), day 28 (P < 0.05), and day 50 (P < 0.005).
20
3O
4O
5O
DAYS AFTER MT ADMINISTRATION Fig. 1. The mean couse of arthritis ( • ) and arthritis incidence (O) of a group of 70 Lewis rats studied in 15 different experiments.
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
Also spleen cells from naive animals were isolated and injected as control ceils. After 15 days arthritis development was scored.
722
Mycobacterial HSP65 in adjuvant arthritis
B
7
9
14
21
38
SO
DAYS AFTER MT ADMINISTRATION
7
9
14
21
38
OO
DAYS AFTER MT ADMINISTRATION
D
9
14
21
36
50
7
9
14
21
38
00
DAYS AFTER MT ADMINISTRATION
7
9
14
21
36
00
DAYS AFTER MT ADMINISTRATION
7
9
14
21
30
SO
DAYS AFTER MT ADMINISTRATION
Fig. 2. The proliferative responses to MT, HSP65, and the 180-188 epitope of inguinal lymph node cells (A and B), popliteal lymph node cells (C and D), and spleen cells (E and F). The results are indicated as stimulation indices (SI) ±SEM in bars. The black bars represent the responses to MT, the dark hatched bars the responses to HSP65, the light hatched bars the responses to the 180 - 188 peptide. The results shown were obtained using 2 x 105 cells in the presence of an optimal concentration (10 ^g/ml) antigen.
found (Fig. 2B). However, marked proliferative responses to HSP65 were found at day 9, which declined gradually at days 14, 21, and 35. The courses of the response to MT and HSP65 were similar. A response to the 180-188 epitope was only detected at day 14 after MT administration, which correlated with the moment at which the first signs of arthritis were noticed (see Fig. 1).
In the popliteal lymph node cells (Fig. 2C and D), responses to MT and HSP65 were not found until day 14. At that moment the first signs of arthritis could be seen. The responses to MT further increased at days 21 and 50, which differed from the results obtained with HSP65. Maximal proliferative responses to HSP65 were found at day 21, and declined at days 35 and 50. These results show that the day of maximal proliferative responses
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
7
DAYS AFTER MT ADMINISTRATION
Mycobacterial HSP65 in adjuvant arthritis 723 correlates with the day at which the maximum arthritis score is found. No responses to the 180 - 1 8 8 epitope were found in the popliteal lymph nodes after immunization with MT. We also tested the proliferate responses of spleen cells. Responses to MT and HSP65 were found 14 days after MT immunization (Fig. 2E and F) and increased during the course of disease to a maximum at day 50. No prdiferative responses of the spleen cells were found to the 180-188 epitope. In individual animals, no evidence was found for the correlation between the arthritis severity and the level of cellular immune responses to HSP65 in inguinal lymph nodes, popliteal lymph nodes or spleens (Fig. 3). Even in animals that failed to develop AA after MT immunization (score 0), responses to HSP65 were seen.
x
Y~
1
1-4
5-8
13-16
9-12
ARTT-RITIS SCORE
10T
20
30
Exacerbation of AA after HSP65 administration Previous experiments had shown that immunization with HSP65 prior to MT administration resulted in protection against AA. Now, the influence of HSP65 immunization during development of the disease was studied. Therefore, animals were immunized with MT and injected with 250 HQ HSP65 7,9, and 11 days later. Thus, HSP65 was injected before signs of arthritis occurred. Figure 4 shows the course of AA in control animals, which were immunized with PBS and of animals injected with HSP65. The control animals showed a disease development comparable to the course of disease shown in Fig. 1. The HSP65 immunized animals, however, demonstrated not only a significantly increased severity of the disease (maximum score in control animals of 10.0 ± 1.0, in HSP65 injected group 13.1 ± 0.9), but also an accelerated onset of the disease (arthritis score at day 14 of 4.5 ± 1.2 and 10 4 ± 4.0 respectively). The day at which the maximum score was reached was earlier in the HSP65 immunized animals (20.7 ± 1.3 in control group versus 17.6 ± 1 . 1 in HSP65 treated group). The same effect was seen when a higher dose of HSP65 (1 mg) was given or when HSP65 was administered when first signs of arthritis occurred (data not shown). Spleen cells collected from animals suffering from AA exacerbations after HSP65 administration were tested for prdiferative responses to MT, HSP65, and the 180-188 epitope. As shown in Table 3, these animals had higher proliferate responses to MT, HSP65 and to some extent to the 180-188 epitope than the contrd rats.
40
DAYS AFTER MT ADMINISTRATION Fig. 4. Effect of administration of HSP65 at day 7, 9, and 11 after arthritis induction. Rats were immunized mtradermally at day 0 with 1 mg MT in IFA at the base of the tail At day 7, 9, and 11, 250 ^g HSP65 was injected i.v. and arthritis was scored (' P < 0.05). ( • ) HSP65 inoculated animals (n = 8), (O) PBS control animals (n = 6)
Table 3. Increased prdiferative responses3 of spleen cells to HSP65 are found in rats suffering HSP65 induced AA exacerbations Maximum scoreb
Immunization
Booster
Antigen 0»g/ml)
c.p.m.
MT
PBS PBS PBS PBS
none MT HSP65 180-188
543 21593 2688 730
± ± ± ±
102 4553 429 122
10.8 ± 1.1
(10) (25) (25)
HSP65 HSP65 HSP65 HSP65
none MT HSP65 180-188
975 47217 24045 1931
± ± ± ±
132 6190 8402 186
13.6 ± 0.9°
(10) (25) (25)
MT
"Rats (n = 7/group) were immunized with MT for arthritis induction. At days, 7, 9, and 11 a booster was given of 250 j»g HSP65 or PBS as control. Prdiferative responses were tested 19 days after MT administration. •The mean maximum arthritis score is shown. c Significantly different from PBS contrd group (P < 0.05).
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
Fig. 3. Correlation between the severity of the disease and the proliferate responses to mycobacterial HSP65 of lymphocytes obtained from MT-immunized animals. The prdiferative responses (SI) measured at day 14 and 22 are shown as pooled data.
724
Mycobacterial HSP65 in adjuvant arthritis
Spleen cells responsive to HSP65 obtained from rats after remission of AA transfer AA resistance
Effect of administration of HSP65 specific T lymphocytes on arthritis seventy To see whether disease remitting T cells could be selected by in vitro HSP65 re-stimulation of MT primed T cells, the following experiment was done. HSP65 specific T cells were obtained from the inguinal lymph nodes of arthritic rats, 14 days after MT immunization. The cells were activated in vitro by culturing the cells in the presence of 10 ^g/ml HSP65. Following this in vitro stimulation raised proliferative responses were demonstrated for MT and HSP65 but not for the 180 - 1 8 8 epitope. The responses to these antigens of in vitro HSP65 restimulated cells were 13.6, 28.0, and 1.1, respectively. In non-restimulated cells these responses were 10.1, 4.4, and 2 1 respectively. The cells were
O
10
20
3O
4O
Discussion Previous investigations in the model of AA have demonstrated the critical role of HSP65 in this disease. Firstly, T cell clones with specificity for this protein have been shown to elicit AA and to induce protection against AA (9,12,13). Further analyses have defined the epitope recognized by the clones to be the 180 -188 amino acid sequence of HSP65 (9). Secondly, immunization with HSP65 prior to arthritis induction by MT immunization leads to disease resistance (9). Recently, in other models of experimentally induced arthritis, including models without any mjcrobnal induction of disease, the significance of HSP65 immunity has also been demonstrated. Induction of disease has been shown to coincide with the development of responses to HSP65 and, in addition, pre-immunization has led to the development of arthritis resistance (15-18). In order to further understand the role of HSP65 in experimental arthritis, in this study we investigated the humoral response to HSP65 and the cellular immune response to HSP65 and its 180 -188 epitope during the course of AA. In addition, we have tested the immunomodulatory activities of HSP65 specific polyclonal T cell populations in the model. Not earlier than 14 days after immunization with MT, antibodies to HSP65 were detected. No significant levels of mycobacterial HSP70 specific antibodies were found. This finding indicated that
5O
DAYS AFTER MT ADMINISTRATION
25
3O
35
40
DAYS AFTER MT ADMINISTRATION Fig. 5. Protective effect of spleen cells isolated from rats after remission of AA. Spleen cefls were isolated from rats 35 days after arthritis induction; 5 x 108 cells were injected i.v., simultaneously with 1 mg MT for arthritis induction and arthritis was scored ('P < 0.005): ( • ) rats immunized with T cells from rats with AA in remission (n = 7), (O) rats immunized with T cells from healthy control rats (n
719-727
© 1992 Oxford University Press
Adjuvant arthritis and immunity to the mycobacterial 65 kDa heat shock protein Els J. M. Hogervorst, Jos6e P. A. Wagenaar1, Claire J. P. Boog1, Ruurd van der Zee1, Jan D. A. van Embden2, and Willem van Eden1 Department of Virology, University of Amsterdam, Amsterdam, The Netherlands 'Institute of Infectious Diseases and Immunology, Department of Immunology, Faculty of Veterinary Medicine, University of Utrecht, Utrecht, The Netherlands department of Bacteriology, National Institute of Public Health and Environmental Protection, Bilthoven, The Netherlands
Abstract The mycobacterial 65 kDa heat shock protein (HSP65) is of critical significance in the model of adjuvant arthritis (AA). Arthrltogenlc and protective T cell clones obtained from arthritic rats recognized the 180-188 sequence of HSP65. Previous reports have shown that administration of HSP65 prior to disease induction led to resistance to arthritis In the AA model and in several other models of experimental arthritis. Here, we report the development of Immunity to HSP65 and the critical 180-188 epitope during the course of AA. Following Mycobactertum tuberculosis (MT) immunization both antibodies and T cell responses to HSP65 were detected. Proliferative responses to the 180-188 epitope were seen exclusively in the local draining lymph node cells at day 14 after immunization. The anatomical distribution and course of T cell responses to HSP65 and its 180-188 epitope are compatible with T cell regulated control of the disease. Although lower HSP65 antibody levels were observed in the animals with severe arthritis, in individual animals no evidence was obtained for a relationship between development of HSP65 humoral immunity and arthritis severity. Nevertheless, during disease exacerbation, elicited by HSP65 immunization during disease development, elevated T cell responses against HSP6S and its 180-188 epitope were found. In contrast, we obtained evidence that successful transfer of arthritis resistance to naive recipients depends on the transfer of HSP65 specific T cells. On the basis of these results, it seems that HSP65 plays a crucial role in the T cell regulatory events involved in both the induction of, and protection against, AA. Although, these findings offer perspectives for Immunological intervention of arthritis using HSP65 itself or HSP65 specific T cells, the present findings on the versatile aspects of the immunomodulatory qualities of HSP65 should be taken into account in designing protocols for triggering protective T cell regulation.
Introduction The mycobacterial 65 kDa heat shock protein (HSP65) belongs to the HSP60 family of heat shock or stress proteins. A number of different families of HSPs are known, for example the HSP90, HSP70, and HSP60 family. Besides the mycobacterial HSP65, the HSP60 family comprises the Escherichia coli GroEL, the Rubisco subunit-binding protein in plants, and mammalian HSP60 (1 -3). Synthesis of HSPs is subject to regulatory control dependent on sudden rises of temperature or other stressful events, such as anoxia, the presence of oxygen radicals, nutrient deprivation or infection with microorganisms (4,5). HSPs are present in all living cellular organisms and have retained a
remarkable degree of evolutionary conservation. For the HSP60 family of HSPs the homologies found between bacterial and mammalian counterparts are 50%, in terms of amino acid sequence and an additional 20% when taking into account conserved ammo acid substitutions (2). The strong evolutionary conservation of HSPs indicates that they have similar essential functions for cells from a great diversity of organisms. Some HSPs are also called molecular chaperonins or house-keeping proteins and are, among others, involved in folding, unfolding, and translocation over intracellular membranes of proteins and the assembly or disassembly of complex proteins (4.6,7). The
Correspondence to: E. Hogervorst Transmitting editor: S. H. E. Kaufmann
Received 31 January 1992, accepted 17 March 1992.
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
Key words: autoimmunity, T cell proliferation, immunomodulation, antibody production
720
Mycobacterial HSP65 in adjuvant arthritis
Not only in animal models, but also in patients suffering from rheumatoid arthritis the role of HSP65 has been studied. T cells taken from the synovial fluids of such patients responded to HSP65 (22-25). Children suffering from juvenile rheumatoid arthritis were found to respond to the human HSP6O (26) In adult rheumatoid arthritis patients, rased levels anti-HSP65 antibodies were also found (27,28) and selective expression of HSP60 in the inflamed synovial tissues (29-31) was demonstrated. To understand more about the role and regulatory effect of HSP65, we studied the relation between the humoral immune responses directed to HSP65 and arthritis development. Furthermore, the cellular immune responses against HSP65 and the 180 - 1 8 8 epitope were investigated during disease. In addition, we have examined the effects of the administration of HSP65 itself or HSP65 responsive T cells during arthritis development. Our data confirm the critical significance of HSP65 immunity in AA and the protection against AA found after administration of HSP65 specific T cells may offer possibilities for immunomodulation via T cell vaccination using HSP65- activated T cells.
Methods
a M. tuberculosis HSP70 expressing recombinant E. coli stran (33). The 180-188 peptide was prepared by solid-phase techniques previously described by Steward and Young (34). Arthritis induction with MT Rats were immunized intradermally at the base of the tail with 100 /J of 10 mg/ml MT in mineral oil (Incomplete Freund's Adjuvant, Difco Laboratories, Detroit, Ml). The rats were observed for clinical arthritis and were scored by grading each paw from 0 to 4 based on erythema, swelling and deformity of the joint. 0 = no erythema or swelling, 1 = slight erythema or swelling of one of the toes, 2 = erythema and swelling of more than one toe, 3 = erythema and swelling of the ankle, and 4 = complete erythema and swelling of toes and ankle and the incapacity to bend the ankle. All four legs were scored, so the highest score achievable was 16. ELISA (or the detection of HSP specific antibodies HSP65 or HSP70 (0.2 pg) was used to coat wells of (Greiner) ELISA plates. The plates were incubated for 2 h at 37°C. The wells were washed with phosphate-buffered saline (pH 7.4) containing 0.05% (v/v) Tween 20 (PBS-Tween). The sera, obtaned from MT immunized rats, were added in dilution of 1:100 and 1 200 in PBS - Tween and incubated for 2 h at 37°C. After washes with PBS-Tween, goat-anti-rat Ig conjugate (Nordic, Tilburg, The Netherlands) was added in a dilution of 1:4000 in PBS-Tween and incubated for 2 h. After washing, the ABTS substrate (2,2-azinodi-bis(3-ethylbenzthiazoline-6-sulphonic acid) (Sigma, no. A-1888) in 0.1 M citrate phosphate buffer and H2O^) was added and after 0.5 h the absorbance at 405 nm was measured in a Titertek multiscan reader. Proliferative responses At consecutive points of time after MT immunization, lymph nodes and spleens were removed and cell suspensions were made. The erythrccytes present in the spleen were removed by a FicollIsopaque gradient. Proliferative responses were measured in flatbottom microtiter plates (Costar) in triplicate wells. Each well contained 2 x i O 5 cells suspended in 0.2 ml DMEM (Gibco), supplemented with 10% FCS (Gibco), 5 x 10~5 M /3-mercaptoethanol, glutamine, penicillin, and streptomycin, in the presence or absence of antigen. The cells were cultured for 3 days and for the last 16 h pulsed with [3H]thymidine. The cultures were harvested on fibreglass filters and thymidine incorporation was measured in a Beta-plate (LKB, Wallace Oy, Turku, Finland) in counts per minute (c.p.m.).
Animals Inbred male Lewis rats were obtained from the University of Ljmburg, Maastricht, The Netherlands. All rats were used at 6 - 8 weeks of age.
Exacerbation of the disease by administration of HSP65 At day 0 rats were immunized with MT for arthritis induction. At day 7, 9, and 11, 250 ng HSP65 in 250 /»l PBS was injected i.v. and arthritis development was scored.
Antigens The following antigens were used for immunization of the animals or in proliferation assays: heat-killed MT H37 RA (Difco Laboratories, Detroit, Ml), the 65 kDa HSP was derived from E. coli K12 carrying plasmid pRIB1300 expressing the M. bovis BCG HSP65 protein gene (9). The protein was purified from a temperature-induced culture according to the procedure described by Thole ef al. (32). The 70 kDa HSP was derived from
Protection induction by spleen cells obtained from rats recovered from AA Rats were immunized with MT and at day 35, when remission of the disease was seen, the spleens were isolated. Erythrccytes were removed by a Ficoll-lsopaque gradient and after washing the cells, 5 x 108 cells were injected intravenously in naive rats. At the same time 1 mg MT was injected at the base of the tail.
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
mycobacterial HSP65 has been shown to be a strongly immunogenic protein (8). In addition, it has been demonstrated to be involved in experimentally induced adjuvant arthritis (AA) (9). AA is an animal model for rheumatoid arthritis and is induced in Lewis rats by immunization with heat-killed Mycobacterium tuberculosis (MT) in Freund's incomplete adjuvant (10). The first indications that HSP65 played a role in AA were obtained by the characterization of arthritogenic and protective T cell clones generated from arthritic rats (11-13). The epitope recognized by these clones was identified as a nine amino acid sequence (180-188) of mycobacterial HSP65 (9). Later, the epitope was further characterized as amino acid sequence 180-186 (14). Furthermore, vaccination with HSP65 also has been shown to induce protective immunity in various bacterial induced experimental arthritis models, like AA, streptococcal cell wallinduced arthritis (9,15,16) and in non-bacterial associated pristane-induced arthritis (17). Partial protective effects were seen in collagen type ll-induced arthritis (16,18). Previously, we have shown that responsiveness to the 180 - 1 8 8 epitope was related to susceptibility to AA. In addition, a role for HSP65 specific T cells in controlling the T cell responses to the 180 - 1 8 8 epitope was described (19). Recently, it was shown that the 180-188 epitope by itself could induce protection against AA (20,21).
Mycobacterial HSP65 in adjuvant arthritis 721 (Table 2). No significant levels of antibodies with specificity for HSP70 were found (Table 1).
Treatment of AA by HSP65 specific T cells Rats were immunized with MT for arthritis induction. At day 14 post-immunization the inguinal lymph nodes were removed and a cell suspension was made. The cells were tested for proliferation to HSP65 and 180 - 1 8 8 and also cultured for 3 days in the presence of HSP65. The cells were cultured at a concentration of 5 x 106 cells/ml in the presence of 10 ^g/ml HSP65. After 4 days dead cells were removed by a Ficoll-lsopaque gradient and after washing the cells 3 times, 1 x 1 0 ' cells were injected i.v. into rats who had received MT for arthritis induction 14 days earlier. The remaining cells were cultured at a concentration of 5 x 106 ceils in the presence of 5% EL-4 supernatant as source of IL-2, and 3 and 7 days later the animals received another injection of 1 x 107 cells and arthritis scores were followed.
Cellular immune responses in AA
Statistical analysis For data concerning arthritis scores the Mann-Whitney IMest was used. For analysis of the antibody data, Student's r-test was performed.
In order to monitor cellular responses to HSP65 during AA, lymphocyte proliferation assays were performed with cells obtained from different lymphoid organs at various stages of AA. The development and incidence of the AA in our stock of Lewis rats is shown in Fig. 1. On average, disease started at day 14.9 ± 0.3 (mean day ± SEM). The mean maximum arthritis score measured was 10.8 ± 0.4 and the mean day of the maximum arthritis score was found at day 21.7 ± 0.4. At different stages of the disease spleen and lymph node cells were isolated and proliferate responses directed to MT, HSP65, and the 180-188 epitope of HSP65 were determined. We tested a concentration range from 1 to 25 /ig/ml and in Fig. 2 the results obtained with an optimal concentration of 10 ^g/ml MT, HSP65, and 180 -188 peptde are shown. Every point of time represents 5 - 2 5 tested animals. We also tested spleen cells obtained from naive animals for their responses to the antigens. The responses to MT, HSP65, and the 180-188 epitope, given as stimulation indexes (±SEM), were 4.5 ± 0.4, 2.4 ± 0.4, and 1.4 ± 0.2 respectively. Seven days after immunization low responses against MT were found in the draining inguinal lymph node cells (Fig. 2A) and no responses to HSP65 or peptide 180 - 1 8 8 were
Results Humoral immune responses in AA To study the humoral immune responses to HSP65 in AA, Lewis rats were immunized intradermally at the base of the tail with MT. At various times, blood samples were taken and tested for the presence of antibodies directed to HSP65 and HSP70 using an ELISA. The results are shown in Table 1. No HSP65 specific antibodies were detected prior to MT immunization or 9 days after arthritis induction. However, at day 14 high levels of antibodies were found. The antibody levels started to decline at days 28 and 50, but still exceeded the levels measured preceding MT immunization. We also compared the severity of the arthritis in relation to the amount of HSP65 specific antibodies present (measured at days 28 and 50). In animals with low arthritis scores (maximum score of 0 - 4 ) , the levels of antibodies slightly exceeded those of animals suffering from more severe arthritis
Table 2. Correlation between HSP65 antibody levels and severity of AA
14 32
Arthritis score3
HSP65 antibodies'5
0-4 5-16
0.96 ± O.Otf 0.75 ± 0.05
a
Animals were grouped according to their maximal arthritis score measured during disease. 'The ELISA was performed with sera obtained at day 28 or 50 after MT administration. c Significantly different (P < 0 05)
1OO
Table 1. Development of antibodies8 against HSP65 and not to HSP70 in MT-immunized animals Days after immunization
OD value of sera tested for antibodies directed to HSP65 HSP70
0 g 14 28 50
0.12 0.13 1.05 o.g2 0.71
± 0.01
b
0.01 b ± 0.03= ± 0.05 ± 0.07
0.13 0.13 0.15 019 0.20
± ± ± ±
0.01 0.01 0.02 0.01 0.01
a The sera were tested in a dilution erf 1:100 and shown is the optical density ±SEM measured at 405 nm (see Methods). "Significantly different from day 14 (P < 0.005), day 28 (P < 0.005), and day 50 (P < 0.05). c Significantly different from day 0 (P < 0.005), day 9 (P < 0.005), day 28 (P < 0.05), and day 50 (P < 0.005).
20
3O
4O
5O
DAYS AFTER MT ADMINISTRATION Fig. 1. The mean couse of arthritis ( • ) and arthritis incidence (O) of a group of 70 Lewis rats studied in 15 different experiments.
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
Also spleen cells from naive animals were isolated and injected as control ceils. After 15 days arthritis development was scored.
722
Mycobacterial HSP65 in adjuvant arthritis
B
7
9
14
21
38
SO
DAYS AFTER MT ADMINISTRATION
7
9
14
21
38
OO
DAYS AFTER MT ADMINISTRATION
D
9
14
21
36
50
7
9
14
21
38
00
DAYS AFTER MT ADMINISTRATION
7
9
14
21
36
00
DAYS AFTER MT ADMINISTRATION
7
9
14
21
30
SO
DAYS AFTER MT ADMINISTRATION
Fig. 2. The proliferative responses to MT, HSP65, and the 180-188 epitope of inguinal lymph node cells (A and B), popliteal lymph node cells (C and D), and spleen cells (E and F). The results are indicated as stimulation indices (SI) ±SEM in bars. The black bars represent the responses to MT, the dark hatched bars the responses to HSP65, the light hatched bars the responses to the 180 - 188 peptide. The results shown were obtained using 2 x 105 cells in the presence of an optimal concentration (10 ^g/ml) antigen.
found (Fig. 2B). However, marked proliferative responses to HSP65 were found at day 9, which declined gradually at days 14, 21, and 35. The courses of the response to MT and HSP65 were similar. A response to the 180-188 epitope was only detected at day 14 after MT administration, which correlated with the moment at which the first signs of arthritis were noticed (see Fig. 1).
In the popliteal lymph node cells (Fig. 2C and D), responses to MT and HSP65 were not found until day 14. At that moment the first signs of arthritis could be seen. The responses to MT further increased at days 21 and 50, which differed from the results obtained with HSP65. Maximal proliferative responses to HSP65 were found at day 21, and declined at days 35 and 50. These results show that the day of maximal proliferative responses
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
7
DAYS AFTER MT ADMINISTRATION
Mycobacterial HSP65 in adjuvant arthritis 723 correlates with the day at which the maximum arthritis score is found. No responses to the 180 - 1 8 8 epitope were found in the popliteal lymph nodes after immunization with MT. We also tested the proliferate responses of spleen cells. Responses to MT and HSP65 were found 14 days after MT immunization (Fig. 2E and F) and increased during the course of disease to a maximum at day 50. No prdiferative responses of the spleen cells were found to the 180-188 epitope. In individual animals, no evidence was found for the correlation between the arthritis severity and the level of cellular immune responses to HSP65 in inguinal lymph nodes, popliteal lymph nodes or spleens (Fig. 3). Even in animals that failed to develop AA after MT immunization (score 0), responses to HSP65 were seen.
x
Y~
1
1-4
5-8
13-16
9-12
ARTT-RITIS SCORE
10T
20
30
Exacerbation of AA after HSP65 administration Previous experiments had shown that immunization with HSP65 prior to MT administration resulted in protection against AA. Now, the influence of HSP65 immunization during development of the disease was studied. Therefore, animals were immunized with MT and injected with 250 HQ HSP65 7,9, and 11 days later. Thus, HSP65 was injected before signs of arthritis occurred. Figure 4 shows the course of AA in control animals, which were immunized with PBS and of animals injected with HSP65. The control animals showed a disease development comparable to the course of disease shown in Fig. 1. The HSP65 immunized animals, however, demonstrated not only a significantly increased severity of the disease (maximum score in control animals of 10.0 ± 1.0, in HSP65 injected group 13.1 ± 0.9), but also an accelerated onset of the disease (arthritis score at day 14 of 4.5 ± 1.2 and 10 4 ± 4.0 respectively). The day at which the maximum score was reached was earlier in the HSP65 immunized animals (20.7 ± 1.3 in control group versus 17.6 ± 1 . 1 in HSP65 treated group). The same effect was seen when a higher dose of HSP65 (1 mg) was given or when HSP65 was administered when first signs of arthritis occurred (data not shown). Spleen cells collected from animals suffering from AA exacerbations after HSP65 administration were tested for prdiferative responses to MT, HSP65, and the 180-188 epitope. As shown in Table 3, these animals had higher proliferate responses to MT, HSP65 and to some extent to the 180-188 epitope than the contrd rats.
40
DAYS AFTER MT ADMINISTRATION Fig. 4. Effect of administration of HSP65 at day 7, 9, and 11 after arthritis induction. Rats were immunized mtradermally at day 0 with 1 mg MT in IFA at the base of the tail At day 7, 9, and 11, 250 ^g HSP65 was injected i.v. and arthritis was scored (' P < 0.05). ( • ) HSP65 inoculated animals (n = 8), (O) PBS control animals (n = 6)
Table 3. Increased prdiferative responses3 of spleen cells to HSP65 are found in rats suffering HSP65 induced AA exacerbations Maximum scoreb
Immunization
Booster
Antigen 0»g/ml)
c.p.m.
MT
PBS PBS PBS PBS
none MT HSP65 180-188
543 21593 2688 730
± ± ± ±
102 4553 429 122
10.8 ± 1.1
(10) (25) (25)
HSP65 HSP65 HSP65 HSP65
none MT HSP65 180-188
975 47217 24045 1931
± ± ± ±
132 6190 8402 186
13.6 ± 0.9°
(10) (25) (25)
MT
"Rats (n = 7/group) were immunized with MT for arthritis induction. At days, 7, 9, and 11 a booster was given of 250 j»g HSP65 or PBS as control. Prdiferative responses were tested 19 days after MT administration. •The mean maximum arthritis score is shown. c Significantly different from PBS contrd group (P < 0.05).
Downloaded from http://intimm.oxfordjournals.org/ at Carleton University on July 18, 2015
Fig. 3. Correlation between the severity of the disease and the proliferate responses to mycobacterial HSP65 of lymphocytes obtained from MT-immunized animals. The prdiferative responses (SI) measured at day 14 and 22 are shown as pooled data.
724
Mycobacterial HSP65 in adjuvant arthritis
Spleen cells responsive to HSP65 obtained from rats after remission of AA transfer AA resistance
Effect of administration of HSP65 specific T lymphocytes on arthritis seventy To see whether disease remitting T cells could be selected by in vitro HSP65 re-stimulation of MT primed T cells, the following experiment was done. HSP65 specific T cells were obtained from the inguinal lymph nodes of arthritic rats, 14 days after MT immunization. The cells were activated in vitro by culturing the cells in the presence of 10 ^g/ml HSP65. Following this in vitro stimulation raised proliferative responses were demonstrated for MT and HSP65 but not for the 180 - 1 8 8 epitope. The responses to these antigens of in vitro HSP65 restimulated cells were 13.6, 28.0, and 1.1, respectively. In non-restimulated cells these responses were 10.1, 4.4, and 2 1 respectively. The cells were
O
10
20
3O
4O
Discussion Previous investigations in the model of AA have demonstrated the critical role of HSP65 in this disease. Firstly, T cell clones with specificity for this protein have been shown to elicit AA and to induce protection against AA (9,12,13). Further analyses have defined the epitope recognized by the clones to be the 180 -188 amino acid sequence of HSP65 (9). Secondly, immunization with HSP65 prior to arthritis induction by MT immunization leads to disease resistance (9). Recently, in other models of experimentally induced arthritis, including models without any mjcrobnal induction of disease, the significance of HSP65 immunity has also been demonstrated. Induction of disease has been shown to coincide with the development of responses to HSP65 and, in addition, pre-immunization has led to the development of arthritis resistance (15-18). In order to further understand the role of HSP65 in experimental arthritis, in this study we investigated the humoral response to HSP65 and the cellular immune response to HSP65 and its 180 -188 epitope during the course of AA. In addition, we have tested the immunomodulatory activities of HSP65 specific polyclonal T cell populations in the model. Not earlier than 14 days after immunization with MT, antibodies to HSP65 were detected. No significant levels of mycobacterial HSP70 specific antibodies were found. This finding indicated that
5O
DAYS AFTER MT ADMINISTRATION
25
3O
35
40
DAYS AFTER MT ADMINISTRATION Fig. 5. Protective effect of spleen cells isolated from rats after remission of AA. Spleen cefls were isolated from rats 35 days after arthritis induction; 5 x 108 cells were injected i.v., simultaneously with 1 mg MT for arthritis induction and arthritis was scored ('P < 0.005): ( • ) rats immunized with T cells from rats with AA in remission (n = 7), (O) rats immunized with T cells from healthy control rats (n
