Modulatory effect of hydrocortisone on T-lymphocyte regulatory activity in pokeweed mitogen-driven Ig production

CLINICAL

IMMUNOLOGY

AND

26, 240-248

IMMUNOPATHOLOGY

(1983)

Modulatory Effect of Hydrocortisone on T-Lymphocyte Regulatory Activity in Pokeweed Mitogen-Driven Ig Production Jo& A. BRIEVA,' EMILIO Srrvicio

de Inn~u~~~iogrn,

Centro

G. DE LA CONCHA, DORA PASCUAL-SALCEDO,AND ALFREDO BOOTELLO Esprcial

Rumtin Madrid-34.

y Cujul. Spain

Curreteru

de Colmenur

Km

9.100.

The effect of hydrocortisone (HCO) on pokeweed mitogen (PWM)-driven lg production from human tonsil B- plus T-lymphocyte cocultures was analyzed. IgM and IgG production decreased in cultures with a B/T ratio of 10/l or higher tP i 0.001) and increased in cultures with a B/T ratio of l/I or lower (P < 0.01) when 10m5 M HCO was added. Similar effects were observed on Ig production from peripheral blood lymphocyte cultures. These effects were only obtained at the HCG doses of 10-j and 10m6 M. and no modifications by monocyte depletion were observed. The function of HCO-pretreated B cells did not differ from that of control B cells when cocultured with fresh T cells. However, the HCO pretreatment of T cells before coculturing with fresh B cells affected the Ig production in the same way as the addition of the drug to fresh cocultures. These results suggest that HCO modulates the PWM-driven Ig production by affecting helper and suppressor T-cell activity.

INTRODUCTION

It is well established that glucocorticoids (GC) exert a powerful and multifaceted influence on the human immune response (1, 2). In rive, the administration of GC produces a decrease in the serum immunoglobulin (Ig) levels (3). Such a treatment does not, however, affect the primary and secondary responses to several antigens (4) and under certain circumstances increased specific antibody production has been reported (5). In addition, the administration of GC can reduce the elevated serum Ig levels present in autoimmune diseases (6, 7), and, in some cases of immunodeficiency, GC can enhance the patients’ serum Ig levels (8, 9). It is becoming quite clear that Ig production requires a complex series of cellular interactions between regulator cells (helper and suppressor T lymphocytes and monocytes) and B lymphocytes. The pokeweed mitogen (PWM)-driven Ig production from human lymphocyte cultures has been shown to be an adequate system by which to analyze these interactions (10-13). The addition of GC at pharmacological doses to PWM-stimulated peripheral blood lymphocyte (PBL) cultures results in an enhancement of Ig production (14, 15). This phenomenon has been explained as either a modulation of the B-cell triggering signal (14) or a T-independent stimulation of B cells ( 16). This work shows that the addition of hydrocortisone (HCO) to human lympho’ To whom correspondence UCLA School of Medicine,

should be addressed: Department of Microbiology 43-239 Center for the Health Sciences, Los Angeles. 240

0090-1229/83/010240-09$01.50/0 Copyright All rights

0 1983 by Academic Press, Inc of reproduction in any form reserved

and Immunology. Calif. 90024.

EFFECT

OF

HYDROCORTISONE

ON

REGULATORY

T CELLS

241

cyte cultures stimulated by PWM can result in a decrease or an increase of the Ig production depending on the B/T ratio present in the culture. Further analyses indicated that HCO could act by altering T-cell function. MATERIALS

AND METHODS

Cell preparation. Tonsils were obtained from subjects undergoing routine tonsillectomy for chronic tonsillitis. Single-cell suspensions were obtained from tonsillar stroma by teasing with forceps in Hanks’ balanced salt solution (HBSS) as previously described (11). Peripheral blood lymphocytes were obtained from normal adult volunteers by centrifugation of heparinized venous blood on a sodium metrizoate/Ficoll gradient (Lymphoprep, Hyegaard and Co. As., Oslo). The cells were washed three times in HBSS before culturing or further processing. Tonsil lymphocytes were used except where stated otherwise. Reactives. Pokeweed mitogen (lot A393903, Grand Island Biological Co., Grand Island, N.Y.) was used at a final dilution of 4 pi/ml which in preliminary experiments was shown to be optimal for Ig production. Hydrocortisone (hydrocortisone-21-hemisuccinate, Sigma Chemical CO., St. Louis, MO.) was used at the indicated doses. Culture medium. All cultures were carried out in RPM1 1640 medium (Flow Laboratories, Irvine, Scotland) supplemented with penicillin 100 units/ml, streptomycin sulfate 100 ,&ml, 200 n&Z L-glutamine, and 10% fetal calf serum (Flow Laboratories). Cell separation. Mononuclear cells were separated into T-cell-enriched (T-cell fraction) and T-cell-depleted (B-cell fraction) populations by previously described techniques (11). Briefly, T cells were rosetted with neuraminidase-treated sheep erythrocytes. To achieve maximal rosetting, pelleted ceils were gently resuspended after 15 min and centrifugation was repeated. Rosettes were then allowed to form for 30 min at room temperature, placed on a sodium metrizoate/Ficoll gradient, and centrifuged at 400s for 30 min. The B-cell preparation at the top of the gradient contained < 1% E-rosetting cells. The sheep erythrocytes in the T-cell fraction at the bottom of the gradient were lysed with 0.91% ammonium chloride for 10 min at room temperature. This population was usually contaminated by <2% surface Ig-positive cells, ~1% monocytes, and when cultured with PWM alone failed to produce detectable Ig. Afterward the cells were washed three times and suspended in culture medium for use. Preincuhation with hydrocortisone. Unfractionated tonsil lymphocytes at a density of 2 x lo6 cells/ml were incubated in culture medium in polystyrene petri dishes (Corning Glassware, Corning, N.Y.) with PWM (4 PI/ml) in the presence or absence of 10-j A4 HCO. After 6 hr the cells were recovered, washed three times in HBSS, and separated into T- and B-cell fractions. These cells will be referred to as HCO-treated and control T and B cells. Monocyte depletion. Tonsil B-cell fractions at a density of 2 x lo6 cells/ml in culture medium were incubated in polystyrene petri dishes for 2 hr. The nonadherent cells were decanted and reincubated under the same conditions for another period of 2 hr. At the end, the nonadherent cells were recovered and washed three times in HBSS. The number of monocytes as defined by positive nonspecific

242

BRIEVA

ET AL.

esterase staining was reduced from 14.8 * 5.2 to 2.6 + 1.3% (P < 0.005) after depletion. Culture conditions. Cultures were performed in flat-bottom microplates (Nunc, Komstrup, Denmark) in triplicate (0.25 ml per well) by adding graded numbers of T cells (up to 1 X 10” cells/ml) to a constant number of B cells (1 x lo6 cells/ml). In other experiments graded numbers of B cells (up to 0.5 x lo6 cells/ml) were added to a constant number of T cells (1 x lo6 cells/ml). All cultures were stimulated by PWM (4 yl/ml) and HCO was added or not added as indicated. The cultures were incubated for a period of 8 days. All incubations were made at 37°C in humidified air with 5% CO,. Differences in cell recovery or viability between cultures with or without HCO were not found, as determined by trypan blue staining. Analysis afIg synthesis. After 8 days the supernatant IgM was measured by double-antibody inhibition radioimmunoassay as previously described ( 17). In some experiments IgG was also measured by the same method. As it was observed that the effect of HCO on both immunoglobulin classes was similar, IgG was not measured in all the experiments and usually only IgM data are presented. Statistical analysis. Values are expressed as the mean F SEM. Significance was established by using Student’s t test for paired samples. In all cases P < 0.05 was considered significant. RESULTS Ejjkct of HCO OII the Ig productionjiwm PWM-stimulated cultures of dgferent mixtures of B and T cells. Figure 1A shows the effect of 10-j M HCO on the

PWM-driven IgM production from tonsil B lymphocytes in the presence of graded numbers of tonsil T lymphocytes. When B lymphocytes were cultured alone the IgM production did not change significantly with HCO addition. When small

.ll, 0

02

1 w:, ; ,, 0 02 1 51 T LYMPHOCYTES (x106)

,

1

. 51

ADDED

FIG. I, Effect of 10-sM HCO on PWM-driven IgM (A) and IgG (B) production from tonsil B plus T cocultures. Graded amounts of T cells were cultured with a constant number of B cells (1 X 10” cells/ml) and HCO was added (0) or not added (0) along with PWM. The results are expressed as means 2 SEM of 19 different experiments.

EFFECT

OF

HYDROCORTISONE

ON

REGULATORY

T CELLS

243

numbers of T cells were cocultured with B cells (B/T ratios of 5011 and 1011) HCO provoked a reduction of IgM production to approximately 30% of controls (P < 0.001). At a B/T ratio of 2/l no effect was found, but greater IgM production was observed in cultures with HCO when higher numbers of T cells (B/T ratio of l/l) were added (P < 0.01). The maximal IgM production obtained in cultures with HCO was higher than that obtained without the drug (P < 0.05). Therefore, HCO treatment appears to give rise to a parallel curve of IgM production but displaced to the right. Similar effects were seen on IgG production (Fig. IB). Identical experiments were also carried out using PBL and very similar results were obtained (Figs. 2A and B). Subsequently, experiments were performed to see the effect of HCO on cultures with a B/T ratio lower than l/l. Since the addition of more T cells on these cultures was not possible without affecting viability conditions by overcrowding, cultures were made by adding graded numbers of B cells to a constant number of T cells (Fig. 3). A consistent enhancing effect on IgM production (P < 0.005) was found at all B/T ratios tested. Effect of different doses of HCO. The HCO dose dependency for the previously observed effects was investigated (Table 1). HCO at lo-” and 10e6 M had similar activity for increasing and decreasing Ig production. Greater (1O-4 M) or lower ( 10e7 M) concentrations were less effective. Efect of rnonocyte depletion. In Fig. 4 the effects of HCO on IgM production by cultures of B and T cells before and after monocyte depletion are compared. Although the monocyte number was reduced seven- to eightfold, HCG addition affected both populations in a similar manner. Activity qf B- and T-cell fractions after preincubation u+th HCO. In order to determine which cell was the target for the HCO effects, the activity of HCGtreated and control T and B cells was assessed by subsequent coculturing with

11 cl0

02

1

51

T LYMPHOCYTES

0 (~10~)

.02

1

5 1.

ADDED

FIG. 2. Effect of 10m5 M HCO on PWM-driven IgM (A) and IgG (B) production from peripheral blood B plus T cocultures. Graded amounts of T cells were cultured with a constant number of B cells (1 x IO6 cells/ml), and HCO was added (0) or not added (0) along with PWM. The results are expressed as means k SEM of 7 different experiments.

244

BRIEVA

5

ET AL.

.25

325

B LYMPHOCYTES

06

(~10~)

03

ADDED

FIG. 3. Effect of 10e5 M HCO on PWM-driven IgM production from cultures with B/T ratios lower than l/l. Graded numbers of B cells were cultured with a constant number of T cells (1 x lo6 cells/ml) and HCO was added (0) or not added (0) along with PWM. The results are expressed as means tSEM of 9 different experiments.

fresh B and T cells, respectively. In Fig. 5 the IgM production from fresh B cells plus HCO-treated and control T cells are compared. The function of control T cells was similar to that of fresh T cells (data not shown). The HCO-treated T cells when added in low numbers (B/T ratio of 50/l) provoked a reduction of the IgM production (P < 0.01). At B/T ratios of 10/l and 2/l no difference was observed. However, when a B/T ratio of l/l was reached, IgM production was higher in those cultures containing HCO-treated T cells (P < 0.01). The maximal IgM

EFFECT

TABLE 1 OF SEVERAL DOSES OF HCO ON PWM-DRIVEN PRODUCTION FROM BIT COCULTURES T lymphocytes

HCO

( x lOYm1)

added

IgM

to B lymphocytes

(1 x 10Yml)

dose 00

0

0.02

0.1

0.6" (20.2)

6.5 (kl.1)

21.9 (23.7)

18.6 (24.0)

11.6 (24.2)

10 -I

0.4 (20.1)

5.1 (-tO.8)

17.6 (23.2)

15.1 (24.1)

8.7 (21.9)

10-S

0.5 (L0.2)

I .8 (i-0.4) P -c 0.001

6.9 (kl.5) P i 0.001

21.9 (k3.3)

25.9 (24.8) P c; 0.01

, o-”

0.7 (20.3)

2.3 (~~0.6) P < 0.005

8.1 (k2.0) P < 0.005

20.7 (-1-4.1)

23.8 (k5.3) P -: 0.01

IO- 7

0.6 (k0.1)

3.7 (kl.8)

19.2 (25.6)

23.1 (k7.4)

0

U Data

are expressed

as means

t SEM

of IgM

production

0.5

(&ml)

of 11 different

1

14.4 (i-5.8) experiments

EFFECT

OF

HYDROCORTISONE 50/j

B/T

ON

‘W,

*/, l/1

REGULATORY v/1

T CELLS

245

24 ‘I’

‘O/,

i

0

.02 T

1 5 1 LYMPHOCYTES

0

.02 (~10~)

1 5 1 ADDED

FIG. 4. Effect of 10mS M HCO on PWM-driven IgM production from monocyte-containing (A) or monocyte-depleted (B) B plus T cocultures. Graded amounts of T cells were cultured with a constant number of B cells (1 x IO6 cells/ml) and HCO was added (0) or not added (0) along with PWM. The results are expressed as means k SEM of 7 different experiments.

production obtained in cultures with HCO-treated T cells was higher than that observed with control T cells (P < 0.05). These effects were qualitatively similar to those demonstrated on fresh populations (Figs. 1 and 2). Thus, the effect of HCO on T cells resulted in a displacement to the right of the IgM production curve, so that severalfold greater numbers of HCO-treated T lymphocytes were required to produce similar amounts of IgM as in control cultures. When T lymphocytes alone instead of unfractionated populations were used in the preincubation period, the effects were not consistent (data not shown). Significant differences were not detected in the IgM production between HCO-treated and control B cells when cocultured with fresh T cells (Fig. 6).

+

c’------?

--

0 T

02 LYMPHOCYTES

1 (~10~)

5

1

ADDED

FIG. 5. Function of HCO-pretreated T cells. Unfractionated populations were preincubated with PWM in the presence or absence of IO-” M HCO for 6 hr. After that. HCO-treated (0) or control (0) T cells were separated and tested by coculturing graded numbers of these cells with a constant number of fresh B cells (1 x lO”cells/ml). The results are expressed as means r+_ SEM of 6 different experiments.

246

BRIEVA

11 t+ 0

ET

.02 T LYMPHOCYTES

AL.

1 (x10%,,)

.5

;

ADDED

FIG. 6. Function of HCO-pretreated B lymphocytes. Unfractionated populations were preincubated with PWM in the presence or absence of lo-“M HCO for 6 hr. After that HCO-treated (0) and control (0) B cells were separated and tested by coculturing a constant number of these cells (1 X lo6 cells/ml) with graded numbers of fresh T cells. The results are expressed as means k SEM of 6 different experiments.

DISCUSSION

The present work shows that HCO can markedly decrease as well as increase the PWM-driven Ig production from human lymphocyte cultures. These effects depend on the B/T ratio present in the culture: the HCO decreasing effect is detected at a B/T ratio of 10/l or higher whereas the increasing effect requires B/T ratios of l/l or lower. No differential effects were found at a B/T ratio of 2/l, most likely because the equilibrium between the suppressing and enhancing properties of HCO is established at this point. The increasing effect of pharmacological doses of GC on Ig production from PWM-stimulated PBL cultures has been repeatedly reported (14, 15). The cultures in our experiments with a B/T ratio similar to that of PBL also showed an enhancing effect on Ig synthesis. On the other hand, diminution of IgG production from bone marrow cell cultures after GC administration has been shown (18). Since the bone marrow B-lymphocyte number is severalfold higher than that of T lymphocytes, the explanation for this finding can be related to our results. It has been shown that the HCO effect is exerted on the early events after PWM activation (14, 15). We have investigated the activity of T- and B-cell fractions briefly precultured in the presence of PWM and HCO. B-cell function seemed to be largely insensitive to HCO since neither the addition of the drug to cultures of B cells alone nor the pretreatment of these cells modified their subsequent activity. A lack of effect of HCO on B-cell cultures stimulated by PWM has also been found by Fauci ef al. (14). This would be in agreement with the finding that B lymphocytes are more resistant than T lymphocytes to the inhibitory effect of GC on the lymphoproliferative response to PWM (19). However, Cooper et al. (16) have reported that the addition of 10e6 M prednisolone enhanced the PWM-driven IgG production from cultures of B cells alone or pretreated B cells.

EFFECT

OF

HYDROCORTISONE

ON

REGULATORY

T CELLS

247

The effect of HCO on Ig production from B plus T cocultures was mimicked by HCO-pretreated T lymphocytes when cocultured with fresh B lymphocytes. This suggests that, although several mechanisms are probably involved in the effects produced by HCO, the T lymphocytes play a role in these phenomena. The fact that HCO pretreatment of T cells alone does not consistently affect the activity of those cells indicates that a non-T cell may be required for the HCO-induced effect on T lymphocytes. This inducer role does not seem to be exerted by monocytes since a significant reduction of those cells does not prevent the HCO effects. Nevertheless, since the depletion obtained was only partial, the possibility cannot be ruled out that a small number or a nonadherent-to-plastic subset of monocytes could mediate the effects of HCO. It is well established that GC at the doses used depresses a number of T-cell functions in ~~i~~ as well as in vitro (1, 2). In our experiments a severalfold greater number of HCO-treated than control or fresh T cells was required to reach similar IgM production. This modified activity of T lymphocytes was not due to a nonspecific cytotoxic effect of 10-j M HCO since the cell viability and recovery of cultures with or without the hormone were similar, and the addition of a higher dose of HCO (lo-’ M) was shown to be ineffective. Therefore, HCO could act by diminishing T-cell activity. Thus, the addition of HCO to cultures with high B/T ratios or the coculture of B cells with small amounts of HCO-treated T cells would induce a decrease of Ig production by reducing the capability of T lymphocytes to help B lymphocytes: likewise the increasing effect of HCO when added to cultures with a B/T ratio of l/l or lower or the addition of HCO-treated T lymphocytes in greater amounts would also be explained as a diminution of the well-known suppressor ability that high numbers of T lymphocytes exert in this system (11, 12). So HCO would act on helper as well as suppressor T-cell activities since otherwise an increase or decrease of Ig production should be expected at all B/T ratios. The action of HCO could be exerted either on inducer T cells required for the expression of helper and suppressor activities, or on both functional subsets of T lymphocytes themselves. In this regard a radiosensitive helper T-cell subset which predominantly exerts its activity on cultures with a high B/T ratio has recently been identified (20). The effect of HCO on helper T-cell activity in cultures with such a B/T ratio might be exerted on this same subset. On the other hand, GC has been reported to inhibit several suppressor T-lymphocyte functions. These include the inhibition of those suppressor T cells occurring naturally (21-231, stimulated by Con A (21), present in cord blood (24), generated in the autologous mixed-lymphocyte reaction (25), and exhibited by PBL in some pathological states (9. 26, 27). In summary, HCO modulates Ig production in PWM-stimulated cultures by affecting the predominant (helper or suppressor) T-lymphocyte activity. Since these effects are reached at pharmacologically attainable doses of the drug, the possibility arises that such effects have an in \ri\~ role. It could be related to the apparently opposite effects that GC administration exerts on serum Ig levels and antibody production in different pathological states, by modulating the excessive positive or negative influences from T cells detected in these diseases.

248

BRIEVA

ET AL.

ACKNOWLEDGMENTS The authors wish to thank Francisca Jaime, Nieves Menendez, and Reyes Urbiola for excellent technical assistance, Stephen Bendoski for the preparation of the manuscript, and Dr. R. H. Stevens for critical advice. This work was supported by Grant TR180920181 of Fondo de Investigaciones Sanitarias, Instituto National de la Salud of Spain.

REFERENCES I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1 I. 12. 13. 14. IS. 16. 17.

Claman. H. N., N. Eng/. J. Med. 287, 388, 1972. Fauci. A. S., .I. Immunopharmucol. 1, 1, 1978. Butler, W. T.. and Rossen, R. T., .J. C’lirl. Invest. 52, 2629, 1973. Butler, W. T.. Couch, R. B.. Rossen, R. D., and Hersh, E. M., J. C&. Irrvest. 53, 14a, 1974. Tuchinda. M.. Newcomb, R. W., and De Vald, B. L., Int. Arch. Allergy Appl. Immunoj. 42, 533, 1972. Ragan. C., Drokdest, A. W., and Boots, R. H.. Amer. 1. Med. 7, 741. 1949. Hahn. B. H., Kantor, 0. S., and Osterlandd, C. K.. AH~I. Intern. Med. 83, 597, 1975. Soothill, J. F., Hill, L. E., and Rowe, D. S., In “Immunologic Deficiency Diseases in Man” (D. Bergsma and R. A. Good, Eds.), pp. 71-79. National Foundation Press, New York, 1968. Waldmann, T. A., Blease, R. M., Broder, S., and Krakauer, R. S., Ann. Intern, Med. 88, 226, 1978. Keightley, R. G., Cooper, M. D., and Lawton. A. R.. J. Immunol. 117, 1538, 1976. Janossy. G.. Gomez de la Concha, E., Luquetti, A., Snajdr, M. J., Waxdal, M. J., and PlattsMills, T. A. E.. Stand. J. Immunol. 6, 109, 1977. Saxon, A. R., Stevens, R. H., and Ashman, R. F., J. Immlrnol. 118, 1972, 1977. Gmelig-Meyling, F., and Waldmann. T. A.. J. Immunol. 126, 529, 1981. Fauci, A. S.. Pratt, K. R., and Whalen, G.. J. Immunol. 119, 598, 1977. Cooper, D. A., Duckett, M.. Petts, V., and Penny. R.. C/in. Exp. Immunol. 37, 145, 1979. Cooper, D. A., Duckett. M., Hansen. P., Petts, V., and Penny, R., C/in. E.up. Immtrnol. 44, 129. 1981. de la Concha. E. G., Oldham, G., Webster, A. D. B., Asherson. G. L., and Platts-Mills, T. A. E., Clin.

Exp.

Immunol.

27,

208,

1977.

18. McMillan, R.. Longmire, R., and Yelenosky, R., J. Immurrol. 116, 1592, 1976. 19. Blomgren. H.. and Andersson, B.. Exp. Cell Res. 97, 233, 1976. 20. Uchiyama, T.. Nelson, D. L., Fleisher, T. A., and Waldmann. T. A.. J. Immunol. 126, 1398, 1981. 21. Haynes, B. F., and Fauci, A. S.. Cell. Immrnol. 44, 157, 1979. 22. Haynes. B. F., Katz, P., and Fauci, A. S., Cell. Immunol. 44, 169, 1979. 23. Saxon, A.. Stevens, R. H., Ramer. S. J., Clements. P. J.. and Yu, D. T. Y.. J. Clin. fn\vst. 61, 992, 1978. 24. Morito, T., Bankhurst, A. D.. and Williams, R. C.. J. Clin. In\vst. 64, 990, 1979. 25. James, S. P.. Yenokida, G. G.. Graeff. A. S.. Elson, C. D., and Strober. W.. J. Immur~ol. 127, 2605, 1981. 26. Litwin. S. D., Rubinstein. A.. Atassi, B., and Sicklick, M.. J. Clin. Immunol. 1, 94. 1981. 27. Litwin, S. D.. J. Imm(c&. 122, 728, 1979. Received

May

13. 1982: accepted

with

revisions

August

11, 1982.