Limiting dilution analysis of autoreactive T cells in patients affected by hydatid disease

Journal of Autoimmunity (1992) $733-744

Limiting Dilution Analysis of Autoreactive T Cells in Patients Affected by Hydatid Disease

Francesca Quintieri, Alessandra Siracusano, Rachele Rigan Orsola Pugliese Department

of Immunology, Istituto Superiore

and

di Sanitci, Roma, Italy

(Received 19 March 1992 and accepted ZOJuly 1992)

We evaluated the peripheral autoreactive response in patients with Echitaococcus granubsus compared with healthy individuals. A limiting dilution analysis (LDA) was performed to estimate frequency of autoreactive T cells and to detect the complex mechanism that often underlies the autoreactive response. Different LDA curves were seen in healthy controls and patients. The control group provided a straight line passing through the origin (‘single hit’ kinetics), indicating that only peripheral autoreactive T cells are limiting. On the other hand, autoreactive T cell titration in hydatid patients did not follow ‘single hit’ kinetics, suggesting that two or more cell types are titrated. More interestingly, we observed a very high frequency of autoreactive T cells in hydatld patients (l/l,000 to l/2,500) compared with healthy controls (l/S,000 to l/35,000). These data suggest that chronic infection may provide adequate stimuli for the induction of autoimmune disorders and increase in autoreactive T cells.

Introduction

Autoreactive T cells are distinct from conventional antigen-specific T cells in that their activation does not appear to require the addition of exogenous antigen. They generally recognize self-MHC products and have been described in murine [l-3], guinea pig [4] and human [I-9] systems. Extensive studies on their function have shown that autoreactive T cells can exert helper [2, 3, 5, 6, 81, cytotoxic [7-91 and suppressor [6, 91 activity, suggesting a role in regulating the normal immune Correspondence to: Dr Francesca Quintieri, Immunology Department, Viale Regina Elena 299,00161 Roma, Italy.

Istituto Superiore di Saniti,

733 0896-8411/92/060733+

12 $08.00/O

0 1992 Academic Press Limited

134 Francesca Quintieri et al. response. The autologous mixed lymphocyte reaction (AMLR), either in bulk or in limiting dilution analysis (LDA), has been widely used to study autoreactivity both in healthy individuals and in subjects affected by different autoimmune and inflammatory disorders [lO-141. Pathogenic mycobacteria causing chronic infectious diseases recently have been reported to cause autoimmunity [ 151 and stimulate the generation of putative self-class II MHC-reactive T cells in mice [16] and in man [15]. Previous studies [ 17,181 have shown autoantibody production in patients with hydatid disease due to Echinococcusgranulosus. The present study was carried out to evaluate, by LDA, the peripheral autoreactive T cell response in patients affected by hydatidosis. This technique enables us to estimate frequencies of autoreactive T cells and to detect any difference in the mechanisms regulating the autoreactive response between hydatid patients and health controls.

Materials and methods Blood samples Human blood samples were obtained from nine patients with clinically diagnosed hydatidosis (8 with hepatic and 1 with multiple localization; kindly provided by Prof. F. De Rosa, Institute of Infectious Disease, University ‘La Sapienza’, Policlinico Umberto I, Rome) and from 12 healthy blood donors.

Cell preparation Peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation (Lymphoprep, Nycomed AS, Oslo, Norway). The separated PBMC were washed three times with RPM1 1640 medium (Flow Laboratories, Irvine, UK) supplemented with 10% heat inactivated (56°C 45 min) fetal calf serum (FCS) (Flow Laboratories) and antibiotics (complete medium). To prepare responder T cells, a nylon-wool column method was used as described previously [ 191. Briefly, 4 to 6 x lo6 PBMC in 0.5 ml of complete medium was applied to a nylon-wool column (0.12 g of nylon-wool) and the column was incubated at 37°C for 45 min in a CO, incubator. After incubation, the non-adherent cells were gently eluted with 12 ml of prewarmed complete medium and were used as responder cells. Total irradiated (6,000 R) autologous PBMC were used as stimulator cells.

Polyclonal activation Peripheral blood mononuclear cells and enriched T cells from two normal subjects and two hydatid patients were stimulated with an anti-CD3 monoclonal antibody (mAb) or with phytohaemagglutinin (PHA). The anti-CD3 mAb, a culture supernatant purified on a protein A column, was kindly provided by Dr G. Di Felice. Flat-bottomed 96-well microtiter plates were precoated with 200 ul/well of the mAb (1 l.tg/ml) by overnight incubation. Responder cells were plated at different concentrations and kept in culture for 24 h. Phytohaemagglutinin (Famitogen,

Autoreactive T cells in hydatidosis

735

Burroughs Wellcome & Co) was added to responder cells at a concentration of 1 ug/ml and kept in culture for 72 h. In both cases the proliferation was assayed by the uptake of [3H]-thymidine (Amersham) during the last 20 h of culture. The proliferative activity was expressed as the mean counts per minute (cpm) of triplicate cultures + standard deviation (SD). Establishment of autoreactive T cell lines Enriched T cells (2.5 x 106) were stimulated with an equal number of irradiated autologous PBMC in 5 ml of complete medium using plastic culture flasks. To maintain growth, the T cells were restimulated with irradiated autologous PBMC once a week, and after 2 weeks of culture 20 U/ml of recombinant interleukin-2 (rIL-2, Janssen Chimica, Beerse, Belgium) were added to the cultures twice a week. Cell lines were maintained in culture for at least 3 weeks. Assay of specijkity of autoreactive T cell lines Five days after the last stimulation, 1 x lo5 responding T cells were plated in each well of a 96-well flat-bottomed microtiter culture plate and cultured in 200 ~1 of complete medium with an equal number of irradiated stimulator cells. Culturing was performed for 4 days and proliferation activity was assayed as previously described. In order to examine the HLA specificity of autoreactive T cells, Epstein-Barr virus-transformed B cell lines (EBV-BCLs), homozygous for HLA class I and class II antigens (from the 10th Histocompatibility Workshop), were used as stimulators. For the same purpose, anti-HLA-DR (monomorphic) mAb L243 (gift of Dr G. Lombardi) and anti-HLA-A,B,C mAb (TEC-HLA-A,B,C, Techno Genetics, Recordati, Milan, Italy) were used to block autoreactive proliferation. Cultures were set up as described for limiting dilution analysis (see below). Responder cells were seeded only at a concentration of 100,000 cells/well. mAbs were added at the beginning of the cultures at concentrations of 5.6 and 2.8 ug/ml. Limiting dilution analysis Enriched T lymphocytes, isolated from PBMC, were cultured in flat-bottomed 96-well microtiter plates at graded concentrations ranging from 390 to 50,000 cells/well with 50,000 autologous irradiated (6,000 R) PBMC as stimulators in each well. Cultures were set up in 24 replicate wells in a total volume of 200 ~1 of complete medium. IL-2 was added at days 7 and 10 of cultures. At day 12 of incubation, 20 h before termination, 0.5 @/well of [3H]-thymidine were added. The proliferative activity of each well was considered positive when the cpm exceeded the mean of the negative control (T cells alone) by at least three standard deviations. Analysis of the distribution of precursor cells was performed by plotting the negative logarithm of the fraction of the non-responding cultures ( - In Fo) on the y axis and the cell input on the x axis. The semilog plot was used to estimate the frequency of autoreactive precursor cells because, by interpolating the non-responding cultures at the level of 0.37, the size

736 Francesca Quintieri et al. of the sample containing an average of one precursor cell can be estimated 120,211. The precursor frequency of cultures exhibiting non-‘single-hit’ kinetics was roughly calculated as described by Lefkovits and Waldmann [22]. Briefly, Fo =0.37 was plotted from line 0.1 and AN cells were read directly on the titration curve; frequency = 1/AN.

Statistical analysis Linear portions of LDA data were analysed by a linear least-square regression analysis of the semilogarithmic plot of the number of cells per well versus the fraction of negative cultures. Correlation values ranged from 0.964 to 0.995. Differences between frequencies of autoreactive T cell precursors were analysed using the nonparametric Wilcoxon rank sum test.

Results Proliferative response to polyclonal activators To compare proliferative responses to polyclonal activators between normal subjects and hydatid patients, an anti-CD3 mAb and PHA were used. Table 1 shows that both PBMC and enriched T cells from controls and patients give a similar response to anti-CD3 and PHA, even at low cell concentrations. Therefore, it is reasonable to believe that the total responder T cell frequencies in hydatid patients are comparable to those found in normal subjects.

Development and characterization of autoreactive T cell cultures Two CD4+ autoreactive T cell lines were obtained from one healthy control and one hydatid patient. T cells from the two lines were able to proliferate when stimulated with autologous irradiated non-T cells. In order to establish whether the proliferative response observed was indeed specific for self-MHC antigens, restriction experiments were set up using EBV-BCLs as stimulator cells. Table 2 shows that only EBV-BCLs that shared DR antigens with autologous stimulating cells were able to induce a good proliferative response of T cell lines. For the same purpose, inhibition experiments with an anti-MHC class I and II molecule mAb were also set up. Results are reported in Figure 1. The proliferative response of T cell lines was blocked by the anti-class II mAb, in agreement with results of restriction experiments. No inhibition was observed using anti-MHC class I or an unrelated (anti-cholera toxin) mAb.

Analysis of autoreactive T cell precursors Different LDA curves were seen in healthy controls (Figure 2) and patients affected by E. granulosus (Figure 3). The control group provided a straight line passing through the origin. This reflects a ‘single hit’ curve, indicating that in our experimental system only peripheral autoreactive T cells are limiting. In fact, deviation of

1

enriched

EnTC

Patient PBMC

EnTC*

Control PBMC

EnTC

Patient PBMC

EnTC*

Control PBMC

Responder cells

*EnTC: nylon-wool

Expt.

T cells

105 5x 104 104 105 5x 104 104 105 5x 104 104 105 5x 104 104 5x 104 104 103 5x 104 104 103 5x 104 104 103 5x 104 104 103

Cells/well

PHA

anti-CD3

Polyclonal activator 2

Expt.

EnTC

Patient PBMC

EnTC

Control PBMC

EnTC

Patient PBMC

EnTC*

Control PBMC

Responder cells 5x 104 2.5 x lo4 1.2 x 10 5x 104 2.5 x lo4 1.2 x 104 5 x 104 2.5 x IO4 1.2 x 104 5x 104 2.5 x lo4 1.2 x 104 5x 104 104 5x 103 5x 104 104 5x 103 5x 104 104 5x 103 5x 104 104 5 x 103

Cells/well

response to polyclonal activators

12,112+ 1,938 4,817+732 342+ 141 21,694+4,710 2,251+ 482 88+16 12,279 + 2,089 6,335 + 1,529 419rt: 162 10,953 + 3,757 1,167+209 63+ 18 86,005 + 15,380 10,585 + 2,480 179+80 45,708 + 3,272 1,358 + 247 104+37 100,751+8,150 13,317* 1,291 309+ 122 67,177+7,885 2,072 & 622 232k98

cpm + SD

Table 1. Proliferative

PHA

anti-CD3

Polyclonal activator

9,260 f 1,855 956 f 207 86+44 1,178+407 114k24 55_f 12 6,052 + 1,057 558 f 64 72+21 3,568 f 842 363 f 67 70f46 129,943 f 6,692 18,162+ 1,146 2,138 f 1,565 38,325 + 3,549 152_f43 95&-24 81,122*3,193 8,963f 1,16 2885 153 28,534 k 1,460 199+ 139 103 * 33

cpm f SD

738

Francesca Quintieri et al. Table 2. Assay of specificity of an autoreactive

Responder T cell line

MP1MP MP MP MP MP MP

Irradiated EBV-BCL

MP OP ws7 ws31 ws 15 WS 29

T cell line Proliferative response

DR antigens

(cpm + SD)*

-

123&35 4,802 f 1,092 3,409 k 859 4,088 f 1,300 2,085 + 354 441 f79 87&-212

5,7 2,7 7,7 5,5 w6, w6 1, 1

*cpm values were obtained as the mean of triplicate cultures f standard deviation. tMP autoreactive T cell line was obtained as described in materials and methods.

Figure 1. Effect of anti-class I and II MHC mAbs on the proliferative response of autoreactive T cells. A clear inhibition was obtained with anti-DR mAb q, while both an anti-MHC class I KZiand unrelated m mAbs were ineffective ( n responder cells alone, 0 autoreactive response).

linearity would occur as a consequence of an impairment in the accessory cell network. In addition, the straight line allows us to estimate precisely the frequency of autoreactive T cells by interception of regression lines at 0.37 negative cultures. In

\

\

__‘_ ,________

‘\

. ______ _____.. F

t

._______ -----_____ _____ . K F 1

1

Figure 2. Limiting dilution assay of peripheral autoreactive T cell precursors of 12 controls. Number of cells added per culture is plotted on the x axis. Fraction of negative cultures is plotted on the y axis. The intercept of regression lines at 0.37 negative cultures extrapolates to the cell dose that contains an average of one autoreactive T cell precursor well.

0.01i:

0.1

0.37

0.1‘Al:::

0.37

0 9 9 9

OPkS83sI

0

88888

Cells/well

740

Francesca Quintieri et al.

/’

000‘0~ ooo‘ot7 0oo‘OE 000‘02 Ocn’Ol

0:

i : j

-

LliLI I’

k

000‘0s ooo‘oi7 ooo‘os 000’02 000‘0l

a, <

0

; : : ;

l

I

.

;

‘”

5

-

. ,

”

I

000’0s

E

000‘0P OOO‘E ooo‘oz 000’0l 0

\

: : *; :

,

ooo‘oc¶

j

000‘0~ 000‘0s

: I

omoz

;

000‘0l 0 t-

.*

: r.

i

si d

6

6 d

I’

-E

SaJniln:,

k d

6

BulpuodsaJ-uou

)o uolrm~j

Autoreactive T cells in hydatidosis

141

Table 3. Frequency of peripheral autoreactive T cell precursors

HC (12) Hydatid patients (9) HC vs hydatid patients

Range

Median

1/5,00&1/35,000 l/1,000-l/2,500 Wilcoxon rank sum test P < 0.005

l/14,750 l/1,400

healthy controls, precursor frequencies of autoreactive T cells ranged from 1/S,OOO to l/35,000 (Table 3). In contrast, autoreactive T cell titration in hydatid patients did not follow ‘single hit’ kinetics. However, at higher cell concentration, the curves exhibited a definite linear portion; the extrapolated straight line intercepted the y axis at level two or more, indicating that two or more cell types are titrated. Precursor frequencies in these patients, roughly estimated as described in Materials and methods, ranged from l/ 1,000 to l/2,500 (Table 3). Statistical analysis clearly showed a significant difference in terms of autoreactive T cell precursors between normal subjects and hydatid patients (P
In the present study we have obtained the first evidence that peripheral autoreactive T cells are greatly enhanced in patients affected by hydatid disease. Moreover, LDA experiments indicated that peripheral autoreactive T cell proliferation was brought about by different mechanisms in hydatid patients and healthy controls. Indeed, the ‘single hit’ kinetics found in LDA of the control group demonstrated that only one cell population is limiting; Onethe other hand, the non-‘single hit’ kinetics of the LDA curves found in hydatid patients suggest that at least a second cell population is limiting for the induction of autoreactive T cells. Many studies of autoreactive T cells have been carried out in several pathologies, mostly of autoimmune etiology. The principal approach was autologous mixed lymphocyte reaction (AMLR), either in bulk cultures or in LDA. A decreased AMLR has been demonstrated in multiple sclerosis [13], primary biliary cirrhosis [23], systemic lupus erythematosus [24] and Crohn’s disease (manuscript in preparation). However, other recent reports have stressed that autoimmune diseases may result from enhanced autoreactivity of T lymphocytes. These activated autoreactive cells may damage tissues directly or indirectly by recruiting other cell types. Lindor et al. [14] have used AMLR to investigate the frequencies and the phenotypes of autoreactive T lymphocytes in primary sclerosing cholangitis, often associated with autoimmune disorders. In a large group of patients they found evidence of enhanced activation of autoreactive T cells of the suppressor/cytotoxic phenotype. A limiting dilution system was established by Schleiser et al. [lo] to analyse the frequency and regulation of autoreactive T cells in patients with rheumatoid arthritis

742

Francesca Quintieri et al. Cells/well l0.000

15.000

20.000

25.000

Figure 4. Limiting dilution assay of peripheral autoreactive T cell precursors of one hydatid patient. The assay was carried out in both human serum (- n -) and fetal calf serum (- 0 -_).

or Lyme disease and healthy controls. The frequency of autoreactive T cells in blood and synovial fluid was significantly higher in the patients. Furthermore, their data suggested that defective T suppressor function could be a necessary condition for the development of autoimmune diseases. In our study, depletion of CD8+ cells in the responder population did not change the profile of the LDA curve of our patients (data not shown), indicating that the CD8+ cell population has no influence on the imbalance of the regulatory network of autoreactivity. More interestingly, we observed a very high frequency of autoreactive T cells in hydatid patients. We can exclude that our findings could be a consequence of either a non-specific hyperactivity of circulating T lymphocytes or a generalized increase in responder cell number, both due to hydatidosis. Indeed, we did not find any difference in proliferative responses to polyclonal activators between patient and control lymphocytes. Similar findings have been reported in tuberculous patients, and evidence suggests that chronic infections may provide adequate stimuli for the induction of autoimmune disorders and increase in autoreactive T cells. The enhanced frequency of autoreactive T cells in hydatid patients could be due to the persistent presence of circulating parasite antigens: an elevated production of

Autoreactive T cells in hydatidosis

743

IFN-), increases HLA class II antigen expression on stimulator cells, potentiating the antigenic stimulus that triggers the autoreactive response and, structural similarity between components of the infectious agent and ‘self molecules could lead to the breakdown of immunological ‘self tolerance and the generation of harmful autoantibodies and/or autoreactive effector T cells [25-271. Ameglio et al. [ 181 have previously hypothesized similar mechanisms to explain the presence of anti-HLA reactivity in sera of hydatid patients. Such antigenic structures could also account for the increase in peripheral autoreactivity here described. In conclusion, the results presented here clearly indicate that the frequency of autoreactive T cells is elevated in hydatid patients. Moreover, the shape of the LDA curves suggests that different cell interactions characterize the autoreactive phenomenon. Whether self-MHC-reactive T lymphocytes play an active role in the protection or in the pathogenesis of hydatid disease is still an open question. References 1. Clayberger,

2.

3. 4.

5. 6.

7.

8.

9.

10.

11.

12.

13.

C., A. H. Dekruyff, and H. Cantor. 1984. Immunoregulatory activities of autoreactive T cells: an I-A-specific T cell clone mediates both help and suppression of antibody responses. 3. Immunol. 132: 2237-2243 Finnegan, A., B. Needleman, and R. J. Hodes. 1984. Activation of B cells by autoreactive T ceils: Cloned autoreactive T cells activate B cells by two distinct pathways. 3. Zmmunol. 33: 78-85 Saito, T. and K. Rajewsky. 1985. A self-Ia reactive T cell clone directly stimulates every hundredth B cell and helps antigen-specific B cell responses. Eur.3. Immunol. 15: 927-934 Dosreis, G. A. and E. M. Shevach. 1985. Analysis of autoreactive I region-restricted T cell colonies isolated from the guinea pig syngeneic mixed lymphocyte reaction and from immune responses to conventional foreign antigens. Eur.3. Immunol. 15: 466-472 Leung, D. Y. M., M. C. Young, and A. S. Geha. 1986. Induction of IgG and IgE synthesis in normal B cells by autoreactive T cell clones. 3. Zmmunol. 136: 2851-2865 Kotani, H., H. Mitsuya, A. F. Jarrett, G. G. Yenokida, S. P. James, and W. Strober. 1986. An autoreactive T cell clone that can be activated to provide both helper and suppressor function.3. Immunol. 136: 1951-1959 Tilkin, A. F., M. Bagot, M. Kayibanda, J. P. Vernant, and J. P. Levy. 1986. A human autoreactive T cell line specific for minor histocompatibility antigen(s) isolated from a bone-marrow grafted patient. 3, Immunol. 137: 3772-3776 Tilkin, A. F., J. Michon, D. Juy, M. Kayibanda, Y. Henin, G. Sterkers, H. Betuel, and J. P. Levy. 1987. Autoreactive T cell clones of MHC class II specificities are produced during responses against foreign antigens in man. 3. Immunol. 138: 674-679 Rosenkrantz, K., B. DuPont, D. Williams, and N. Flomenberg. 1987. Autocytotoxic and autosuppressor T-cells generated from autologous lymphocyte cultures. Human Immunol. 19: 189-203 Schleisier, M. G., G. Haas, G. Wolff-Vorbeck, I. Melchers, and I?. Hans-Hartmut. 1989. Autoreactive T cells in rheumatic disease. (1) Analysis of growth frequencies and autoreactivity of T cells in patients with rheumatoid arthritis and Lyme disease. 3. Autoimmunity 2: 3 l-49 Del Gallo, F., G. Lombardi, E. Piccolella, M. S. Gilardini Montani, P. Del Porto, 0. Pugliese, G. Antonelli, and V. Colizzi. 1990. Increased Autoreactive T cell Frequency in Tubercolous Patients. Int. Arch. Allergy Appl. Immunol. 91: 36-&2 Alcocer-Varela, J., A. Laffon, D. Alarcon-Segovia, and G. Ibanez de Kasep. 1984. Early proliferative response in human autologous mixed lymphocyte reaction in scleroderma.3. Rheumatology 1 l( 1): 48-5 1 Hirsch, A. L. 1986. Defective autologous mixed lymphocyte reactivity in multiple sclerosis. Clin. Exp. Immunol. 64: 107-l 13

744

Francesca Quintieri et al.

14. Lindor, K. D., R. H. Wiesner, N. F. LaRusso, and H. A. Homburger. 1987. Enhanced autoreactivity of T-lymphocytes in primary sclerosing cholangitis. Hepatology 7(5): 884-888 15. Shoenfeld, Y. and D. A. Isenberg. 1988. Mycobacteria and autoimmunity. Immunology Today 9: 178-182 16. Muller, I., B. Maier, V. Brinkmann et al. 1986. Autoreactive T cell clones from mice infected with Mycobacterium bovis, strain bacillus Calmette-Guerin (BCG) I. Phenotype, specificity and in vitro function. Immunobiology 171: 366-380 17. Mori, H., B. Wernli, N. Weiss, and R. Franklin. 1986. Autoantibodies in humans with cystic or alveolar echinococcosis. Trans. R. Sot. Trap. Med. Hyg. 81: 978-980 18. Ameglio, F., F. Saba, A. Bitti, A. Aceti, N. Tanigaki, R. Sorrentino, A. Doley, and R. Tosi. 1987. Antibody reactivity to HLA class I and II in sera from patients with hydatidosis. J. Inject. Dis. 156: 673-675 19. Handwerger, B. S. and R. H. Schwartz. 1974. Separation of murine lymphoid cells using nylon wool columns. Transplantation 18: 544-547 20. Lefkovits, I. and H. Waldmann. 1984. Limiting dilution analysis of cells of immune system. I. The clonal basis of the immune response. Zmmunol. Today 5: 265-268 21. Waldmann, H. and I. Lefkovits. 1984. Limiting dilution analysis of cells of immune system. II. What can be learned? Immunol. Today 5: 295-298 22. Lefkovits, I. and H. Waldmann. 1979. Limiting Dilution Analysis of Cells in the Immune System. Cambridge University Press, Cambridge. pp. 81-82 23. Lindor, K. D., R. H. Wiesner, E. A. Dickson, and H. A. Homburger. 1988. The autologous mixed lymphocyte reaction in primary biliary cirrhosis: analysis of activation and blastogenesis of autoreactive T lymphocytes. Hepatology 8: 1555-1559 24. Takada, S., Y. Ueda, N. Suzuki, Y. Murakawa, T. Hoshino, I. Green, A. D. Steinberg, D. A. Horwitz, and T. Sakane. 1985. Abnormalities in autologous mixed lymphocyte reaction-activated immunologic processes in systemic lupus erythematosus and their possible correction by interleukin 2. Eur. J. Immunol. 15: 262-267 25. Oldstone, M. B. A. 1987. Molecular mimicry and autoimmune disease. Cell 50: 819-820 26. Fujinami, R. S. and M. B. A. Oldstone. 1985. Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism of autoimmunity. Science 230: 1043-1045 27. Ziegler, J. L. and Stites. 1986. Hypothesis: AIDS in an autoimmune disease directed at the immune system and triggered by a lymphotropic retrovirus. Clin. Zmmunol. Immunopathol. 41: 305-313