Effect of nonsteroidal antiinflammatory drugs on immune function

Effect

of Nonsteroidal Antiinflammatory Immune Function

Drugs on

James S. Goodwin and Jan L. Ceuppens Nonstaroidal antiinflammatory drugs (NSAID) have bean thought to act by inhibiting the production of prostaglandins. Until recently, prostaglandins have bean considered to be mainly proinflammatory compounds, responsible for pain, increased blood flow, and edema. It has more lately bean realized that prostaglandins. particularly PGE,, have important immunoragulatory properties.’ Thus NSAID, by inhibiting PGE, production, modulate immune responses in vitro and in vivo. One or more of these immunomodulating affects of NSAID may play a role in their efficacy as arthritis medications. The purpose of this paper is to review the immunologic role of PGE, and the immunologic affects of NSAID.

PROSTAGLANDINS

AND

IMMUNE

cultures of PBMC. PGE2 is produced in PHAstimulated cultures of PBMC in amounts that, when added exogenously, cause significant inhibition of mitogen stimulation.3*4 Since PGE, is produced in sufficient quantities in PHA cultures to cause an inhibition of ‘H-thymidine incorporation in those cultures, one would expect that the addition of PG synthetase inhibitors to PHA cultures would result in increased ‘H-thymidine incorporation. Indeed, addition of indomethacinenhanced ‘H-thymidine incorporation in PHA cultures, and indomethacin caused a greater percentage increase in ‘H-thymidine incorporation as the dose of PHA was decreased. The percentage increase in 3H-thymidine incorporation ranged from 1,059% f 134% at the lowest dose of PHA (0.2 /*g/mL) to 4 & 3% at the highest dose (20 pg/mL). Similar results were obtained with other cyclooxygenase inhibitors that are chemically unrelated to indomethacin. One possible explanation for this varying response to prostaglandin synthetase inhibitors might be that the PBMC are more sensitive to the inhibiting effects of endogenous PGEz at lower concentrations of mitogen. This indeed appears to be the case. Cultures with the lowest concentrations of PHA show the greatest sensitivity to exogenously added PGE2.4 This would appear to be a general rule for all inhibitors: the lower the mitogen dose is, the more sensitive the culture is to inhibition.’ Physiologic concentrations of PGE, ( 10 ’ M or less) that are comparable to those found in inflammatory sites’ have now been shown by several authors to suppress most in vitro manifestations of T lymphocyte functions: mitogen antigenic stimulation,g Eresponsiveness,4,7,8 lymphokine production rosette formation,““’ (including interleukin II or TCGF),‘2-‘4 the generation of cytotoxic cells in mixed lymphocyte Decreased culture,‘5,‘6 and T cell cytotoxicity.” T cell growth factor (TCGF) or interleukin 11 production contributes to the decreased T cell proliferation caused by PGE2,4 but it is not the only mechanism; the proliferative response of TCGF-dependent cells to added TCGF is also inhibited by small concentrations (1 0m9 M) of

FUNCTION

That inhibition of the immune function by endogenous prostaglandins represents a physiologic feedback mechanism in vivo was first shown by Webb and Osheroff? who found that injection of mice with sheep red blood cells led to an 80-fold increase in intrasplenic PGF,,. Prior administration of a PG synthetase inhibitor prevented the PGF, increase and also resulted in a significantly increased number of plaque-forming spleen cells three to five days after sensitization. Thus antigen challenge initiated prostaglandin production as a negative feedback mechanism. We examined the effects of prostaglandins on mitogen-stimulated 3H-thymidine incorporation in human peripheral blood mononuclear cells (PBMC).3 We found that PGE, or PGE,, but not PGA, PGF,,, or PGF,,, inhibited 3H-thymidine incorporation when added in small concentrations ( 10m9 to 10M6 M) to concanavalin A (Con A)- or phytohemagglutinin (PHA)-stimulated

From the Department of Medicine, University of New Mexico School of Medicine, Albuquerque. NM. James S. Goodwin, M.D.: Department of Medicine, University of New Mexico School of Medicine, Albuquerque, NM; Jan L. Ceuppens, M.D.: Division of Clinical Immunology, Department of Medicine, St. Rafael Hospital, University of Leuven, Leuven, Belgium, Address reprint requests to James S. Goodwin, M.D., Department of Medicine, University of New Mexico School of Medicine, Albuquerque, NM 87131. D I983 by Grune & Stratton, Inc. 0049-0172/83/1301-0009$2.00/0

134

Seminars

in

Arthritis and Rheumatism, Vol. 13, No. 1 (SuppI. 1) (August). 1983

NSAID AND IMMUNE

135

FUNCTION

PGE,.” The cytotoxic activity of non-T lymphocytes, namely, natural killer and antibodydependent cytotoxic activity, are also inhibited by PGE,.‘7m22 THE PGE,-PRODUCING

3)

SUPPRESSOR CELL

In human peripheral blood, the cell that is responsible for PGE, production is a monoalthough there is some evidence in mice cyteV3*23.24 that T cells may produce large amounts of PG.25.26 Monocytes that are exposed to mitozymosan,29 antigen-antigenq3.*’ endotoxin2’ of IgG2’ body complexes,30 or Fc fragments increase their PGE, production. Immune stimulation thus seems to induce a mechanism that tempers the intensity of the response. An increased activity of the prostaglandinproducing suppressor cell, the term we use for these PGE-secreting monocytes, has been shown in several chronic inflammatory conditions, including coccidiomycosis,3’ Brucella abortus infections in cattle,32 multiple sclerosis,33 tuberculosis, and sarcoidosis.34 In addition, many cancers in humans and experimental animals are associated with increased prostaglandin production by host monocytes.35-4’ It appears that at least part of the depressed immune response that is seen in these conditions is due to the immunosuppressive activity of macrophages.

CHANGES IN SENSITIVITY OF LYMPHOCYTES TO PGE,

Changes in the sensitivity of lymphocytes to PGE2 have been observed in several instances. from normal individuals 1) Lymphocytes with the HLA-B12 haplotype have a decreased sensitivity to prostaglandins and histamine, and this could play a role in propensity for autoimmune disorders in these subjects.42 2) Lymphocytes from patients with multiple sclerosis exhibit a defective response to PGE2 in one assay system43 but not in another.44 Further evidence for a PGErelated immune dysfunction in this disease is provided by a study from Zurier’s laboratory that showed that lymphocytes from patients with multiple sclerosis have an increased adherence to measles-infected cells that can be reversed in vitro or in vivo

4)

5)

6)

by the administration of prostaglandin synthetase inhibitors.33*45 Page and coworkers have reported that lymphocytes from patients with juvenile periodontitis are less sensitive to inhibition by PGE, (10m5 M) than are cells from patients with adult periodontitis or normal controls.46 They postulate that a defect in the response of lymphocytes to PGE2 allows for the exuberant immune reactivity that is seen in subjects with juvenile periodontitis. Lymphocytes from healthy subjects over age 70 are much more sensitive to inhibition by exogenous PGE2 than are PBMC from young adults.47A9 This increased sensitivity to PGE would appear to account for some of the depression in the cellular immune function that is found in healthy old people. An increased sensitivity to PGE, was observed after surgery and in women in labor.50 Thus, it would appear that physical stress causes lymphocytes to become more sensitive to PGE2, and the increased sensitivity to inhibition by this endogenous immunomodulator is responsible in part for the depressed cellular immune function that follows physical stress. Lymphocytes from patients with rheumatoid arthritis are also more sensitive than normal subjects to inhibition by exogenous PGE,.” It may be that this increased sensitivity was secondary to the fact that all the patients studied had been on long-term treatment with cyclooxygenase inhibitors for their disease.

EFFECT OF NSAID ON CELLULAR IMMUNE FUNCTIONS IN VIVO

NSAIDs in vivo act as general immunostimulants, increasing cellular immune responses in humans and experimental animals.52-5* Muscoplat et al studied the effect of indomethacin administration on delayed hypersensitivity skin responses in guinea pigs that were previously sensitized to mycobacterium bovis.53 Oral administration of indomethacin that is done simultaneously with intradermal injection of mycobacterium bovis antigen resulted in a more

136

GOODWIN

than twofold increase in skin thickness as compared to the guinea pigs that were not given indomethacin. Delayed skin test responses to picryl guinea pig albumin were also shown to be enhanced by indomethacin in guinea pigs.54 Inhibition of endogenous prostaglandin biosynthesis by indomethacin shortened allograft survival, and this effect was completely abrogated by concurrent injections of a PGE, analogue.” Another in vivo correlate of the immunosuppressive activity of PGE, is that local injection of PGE, can cause reactivation of herpes simplex infection in the mouse ear.56 Natural killer (NK) cell activity is also enhanced in vivo by NSAID. Brunda et al have shown that in vivo administration of indomethacin or aspirin resulted in a marked restoration of NK activity in tumorbearing animals.” Tracey and Adkinson showed that the BCG augmentation of NK activity could be further potentiated by indomethacin in vivo.57 The effect of indomethacin administration in humans on delayed hypersensitivity skin testing has also been studied.S2,5* lndomethacin administration (100 mg/day) for five days did not influence the amount of induration caused by a battery of three common antigens in ten normal subjects.58 When indomethacin was administered to subjects with depressed cellular immunity there was an enhancement of delayed hypersensitivity.52 Two previously anergic patients with common variable immunodeficiency became reactive to skin tests while on indomethacin (100 mg/day) and became anergic again after the medication was stopped. The in vitro response of the patients’ lymphocytes to PHA also rose while they were taking indomethacin (Fig. 1). In summary, several in vivo studies confirm the in vitro experiments that showed that PGE acts as a feedback inhibitor of cellular immune responses and that these immune responses can be enhanced in vivo as well as in vitro by administration of NSAID. CELLULAR ACTIVATION

BY PGE,

Until now we have discussed the suppressive properties of PGEz on immunocyte function. It is clear, however, that under certain experimental conditions, PGE, can deliver activating signals. For example, PGE induces immature thymocytes to differentiate into mature T cells.” Similar data have been obtained with B cell differen-

lndomet

CPM

12000

-

10000

-

8000

-

hacm

AND

CEUPPENS

lndomethacr

lndomethacm

6000 4000 2000 -

Weeks

11

1:

O/4 O/4 214

11 214 l/4

Posltlve

skin

11 014

11 214

I O/4

tests

Fig 1. Effect of indomethacin administration in viva on the response to PHA in vitro and on the response to skin testing in a patient with adult-acquired immunodeficiency. The two shaded areas represent the period of indomethacin administration (26 mg. qid). The in vitro response of the patients’ peripheral blood mononuclear calls to an optimal concentration of PHA and to PHA plus indomethacin (1 Ag/mL) is graphed (Reprinted with permission from the Journal of Clinic& and Labors tory immtmoiogy~~.

tiation6’ PGE also causes activation of suppressor cells in some assay systems. For example, Webb and coworkers showed that PGE, activates a glass wool adherent suppressor T cell to release two peptides,6’ which in turn suppress the response of nonadherent T cells to mitogens.25x26 Fisher et al. have recently demonstrated that PGE2 production is a necessary but not sufficient component of the generation of human Con A-activated suppressor cells that inhibit plasma cell differentiation of B cells as well as T cell responses to mitogens and alloantigens.62 Data from our laboratory suggest an analogous PGE requirement for the generation of another type of suppressor cell, the so-called Tr cell.63 Stobo and associates have shown that PGE, can actually increase the mitogenic reactivity of preincubated low density T cells,64 although it decreases the reactivity of medium and high density T cells. More recently, we have shown that the suppressor-cytotoxic subset of T cells, as defined by monoclonal antibodies, has an increased mitogen-induced proliferation with exposure to PGE if the’ cells are first preincubated.65 Dual effects (inhibition or activation) of PGE have been observed on B cells, natural killer (NK) cells, and macrophages.6675

137

It should be clear from these data that PGE cannot be considered as a universal cell inhibitor and that its physiologic role is very complex. PGE concentration, activation and differentiation state of the target cell, and length of PGEtarget cell interaction all seem to be important factors whose complete roles have not been fully delineated. EFFECT OF PROSTAGLANDINS ON THE HUMORAL IMMUNE RESPONSE

The influence of PGs on the humoral immune response can be mediated through direct effects on B cells, through effects on helper and suppressor T cells, or through effects on other accessory cells. Pharmacologic doses of PGE, (10m6 M or greater) inhibit immunoglobulin production by human B cells in vitro.74.75 Most effects of physiologic concentrations of PGE seem to be mediated through the T cell regulation of antibody production. In one approach to this problem, we first studied the effect of PGE2 on helper and suppressor T cell populations. T-r cells have been shown by Moretta et al to be suppressive for PWMdriven B cell differentiation into plasma cells.76 These T cells bearing a receptor for the Fc portion of IgG (T-y cells) had a 27-fold rise in cyclic AMP after exposure to PGE? whereas the remaining non-T7 cells had a fourfold increase.” Thus, suppressor T cells have a high density of PGE, receptors on their membrane relative to helper cells. This rise in cyclic AMP could be either an inhibition or an activation signal,

Fig. 2. Effect of addition of different concentrations of PGE, on IgM-rheumatoid factor production. In three experiments with pokeweed-mitogen-stimulated lymphocyte cultures, indomethacin (0.5 gg/mLI caused substantial inhibition of IgM-RF production. This was reversed by adding PGE, (0.3 to 3000 nmoll L) to the cultures (Reprinted with permission from the

depending on the cell type and state of activation. Fisher et al. have demonstrated that PGE, in doses of 3 x lo-’ pg/mL can activate human suppressor T cells which subsequently inhibit human B cell differentiation into plasma cells.62 PGE2 has also been shown in vitro to activate mouse suppressor lymphocytes that inhibit primary antibody responses to DNP-LPS.78 In several reports, blocking PG-synthesis enhanced in vitro B cell function, possibly through an effect on helper or suppressor cells.79*80In vivo studies, although the number is restricted, have also shown an enhancement of humoral immune responses with indomethatin. 2.58.81 We have recently used an in vitro culture system of human peripheral blood mononuclear cells that were stimulated with pokeweed mitogen to evaluate the in vitro effect of prostaglandins and prostaglandin synthetase inhibitors on antibody and autoantibody production.75,82 Figure 2 summarizes some in vitro studies in which we measured the effect of NSAID on the production of IgM rheumatoid factor by lymphocytes from patients with rheumatoid arthritis. One million lymphocytes from these patients were cultured with pokeweed mitogen for one week, and the production of IgM-rheumatoid factor in the supernate was measured with a radioimmunoassay. Data from three patients are shown in the figure. In the first patient, 22 ng of rheumatoid factor was produced in the control culture. When a NSAID, indomethacin, was added to the culture, the level was reduced to 8

+ Indomethacln

138

GOODWIN AND CEUPPENS

----.

Normalcontrols (>70 years of age)

No mdomethacin lndomethacin0.5

Normalcontrols (age 20-40)

1

PGE Concentration (ng/ml)

0

24 46 72 96 120 144 166 192 Time after beginningof the culture(hours)

Fig. 3. PGE, production by peripheral blood mononuclear cells (1 x ld/mL) stimulated with PWM (11200) and cultured without or with indomethacin (0.5 gg/mL). Two experiments were performed. Approximately 6 to 7 ng/mL PGE (2 x lo-‘Ml is produced after 46 hours.

ng, and then as prostaglandin Ez was readded to these cultures with indomethacin, the IgM rheumatoid factor gradually increased to control levels. The same effect was seen in the other two patients. In all three cases, the amount of PGE that was required to restore the rheumatoid factor levels to control values was approximately 30 nmol/L. Figure 3 shows that prostaglandin E, is normally produced in these cultures. This shows data from two experiments with prostaglandin production quantitated on the vertical axis. Approximately 10m8 M prostaglandin E is produced in the first 24 hours of the culture. Addition of indomethacin to these cultures eliminates 90% of this production. Figure 4 provides a summary of experiments on patients with rheumatoid arthritis and also on young and old normal donors whose lymphocytes produce the rheumatoid factor. It can be seen that in every case the addition of indomethacin lowers the production of the rheumatoid factor, and the readdition of PGE at 3 to 30 nmol/L restores rheumatoid factor production to the control values. In summary, prostaglandin E is normally produced in these cultures. If PGE production is blocked by the addition of indomethacin, then IgM-rheumatoid factor production is decreased. If PGE is added back to the cultures, then IgM rheumatoid factor production is restored back to normal. In other words, endogenous prostaglandin E stimulates IgMrheumatoid factor production. When this PGE is

0

No

IND

arugs

IND PG;z

No IND arugs

INC PdE2

No IND orugs

IND P;E;E~

Fig. 4. Effects of indomethacin (0.5 pg/mL) and PGE, (30 nmol/L) on IgM-rheumatoid factor production. Lymphocytes from ten patients with rheumatoid arthritis, 15 healthy controls age >70 years, and 14 healthy controls age 20 to 40 years were cultured with pokeweed mitogen. Open circles = rheumatoid arthritis patients who received no medication for seven days before blood was taken. In all three groups, indomethacin caused a significant inhibition of IgM-RF production IP = 0.002 for rheumatoid arthritis patients; P = 0.02 for controls over 70 years; and P = 0.0001 for controls age 20 to 40 years, by a paired t test). Addition of 30 nmol/L PGE, caused a significant enhancement in IgM-RF production over the level with indomethatin (P = 0.003 for rheumatoid arthritis patients; P = 0.01 for controls over 70 years; and P = 0.02 for controls aged 20 to 40 years) (Reprinted with permission from the Lancetnl.

removed, the rheumatoid factor falls. When this PGE is added back, the rheumatoid factor rises. We then looked at the comparative ability of three different NSAID to inhibit IgM rheumatoid factor production in vitro. All drugs were tested in vitro in concentrations that were found in the sera of patients who were taking these agents. As shown in Table 1, all three NSAID inhibited rheumatoid factor production. There are several possible mechanisms to Table 1. Effect of Nonsteroidal Antiinflammatory Agents on IgM-Rheumatoid Factor Production in Vitro by Lymphocytes From Patients With Rheumatoid Arthritis Drug (&mL)

PercentInhibition (%)

lndomethacin (0.5)

55

lndomethacin (5.0)

44

Carprofen ( 10)

60

Piroxicam ( 10)

80

*Results are expressed as percent inhibition of Ighll-rheumatoid factor production in comparison to cultures with no drugs. Drug concentrations that ware used ware comparable to serum concentrations in patients who ware taking the drugs (Adapted from Ceuppens et al*‘).

NSAID AND IMMUNE

FUNCTION

139

Table 2. Inhibition by lndomethacin PGE, of IgM-Rheumatoid

and Stimulation

Factor Production

by

Requires

the Presence of a Radiosensitive T Cell IgM-RFProduction (n&L)

No Drugs

lndomethacin (0.5 PgImLl

lndomethacin + PGE, (30 nnwl/L)

8fT

48 + 22

13 f 5

47 + 21

8 + T (1000 rad)

45k

37 k 16

34 + 8

17

*Results are the mean + SEM of six experiments. In 8 + T cultures, indomethacin caused a 73% inhibition of IgM-RF [email protected] duction which was completely reversed by PGE, (comparing cultures with no drugs with cultures plus indomethacin. P < 0.01; comparing cultures plus indomethacin with cultures plus indomethacin and PGE,. P -c 0.01

by a paired t-test). In the 8 +

irradiated T cultures indomethacin caused a slight (18%). but no significant, inhibition of IgM-RF production, and addition of PGE, had no effect (Reproduced from Ceuppens et aI” with permis-

-h E s G B 8 IL o 0

explain stimulation of rheumatoid factor production by endogenous PGE. The PGE could directly stimulate B cells. It could also stimulate helper T cells that are required for the B cell to produce the rheumatoid factor. Alternatively, PGE could inhibit suppressor T cells. The suppressor T cells are tonically suppressing the helper T cells. In preliminary experiments we found that PGE had no effect on isolated B cells or on B cells + helper T cells. Thus, it appeared that prostaglandin E was working by inhibiting suppressor T cells. We tested this directly in two ways. Suppressor cells are often radiosensitive; so we first asked whether NSAID inhibited and prostaglandin E stimulated rheumatoid factor production if the T cells were first irradiated prior to the culture. Table 2 shows the effects of radiation. Irradiating T cells prior to culture eliminated the inhibitory effect of NSAID and the stimulatory effect of PGE,. This would sug-

120-

No drug

110 -

lndomethacin (0.5 pg/ml)

100 -

lndomethacin plus PGE2 (30 nM)

go80 7060-

g

5ow

2 g

40-

$

3o 20 10 o-

.

/

B+T

B+OKT4(+)

T

B+OKT4(+)T OKT8( + )T

+

Fig. 5. Production of the IgM-rheumatoid factor by B + T cells. by B + helper T cells (OKT4( + 1).and by B + helper T cells (OKT4( + 1) + suppressor T cells (OKTB( + 1).

GOODWIN

140

gest that endogenous prostaglandin was inhibiting a radiosensitive suppressor T cell. We tested this more directly by separating helper and suppressor T cells with monoclonal antibodies. Figure 5 shows data from two experiments. Once again, when B+ T cells were cultured, indomethacin inhibited and prostaglandin stimulated rheumatoid factor production. However, when B cells were cultured with isolated helper T cells, indomethacin no longer inhibited and PGE, no longer stimulated. However, when suppressor cells, identified by the 0KT8 monoclonal antibody, were added back to this system, then once again indomethacin inhibited and PGEz stimulated rheumatoid factor production. Thus, we think that the mechanism of inhibition of rheumatoid factor production by NSAID is as outlined in Fig. 6. The B cells are making the rheumatoid factor with the help of T helper cells. However, in this system T suppressor cells would normally shut off the T helper cells, resulting in no autoantibody production. However,

0

-

B

Rheumatoid Factors

t 0

PGE

NSAID Fig. 6. rheumatoid

Outline factor

of mechanism production.

whereby

by NSAID

inhibit

AND

CEIJPPENS

PGEz produced by the monocytes in the cell culture tonically inhibits suppressor cell activity. Thus, the suppressor cells are shut off, allowing the T helper cells to help the B cells make the rheumatoid factor. When NSAID is added to the system, it shuts off PGE production, allowing the suppressor cells to grow up and shut off the helper cell function, resulting in a decreased rheumatoid factor production. Although endogenous PGE, stimulates rheumatoid factor production in vitro, it was not immediately clear that prostaglandin has a comparable role in vivo. The levels of PGE that are produced in the test tube (approximately lOma M) are 100 to 1,000 times higher than those found in the serum of patients with rheumatoid arthritis or normal controls. However, it is interesting to note that high concentrations of prostaglandin are found in the synovial fluid of rheumatoid j0ints.j PGE, in levels of 5 x lo-” M are commonly found in the synovial fluid in patients with rheumatoid arthritis. These high local concentrations of prostaglandin may explain why the rheumatoid factor is preferentially produced in the rheumatoid joint. Thus, the prostaglandin can selectively shut off the suppressor ceil function in the joint, resulting in rheumatoid factor production. However, in peripheral tissues with normal concentrations of prostaglandin, the suppressor cell function remains normal and little or no rheumatoid factor is produced. We then tested this hypothesis by placing 20 patients with rheumatoid arthritis on piroxicam (Feldene), a new NSAID.83 All 20 patients had been on one or another NSAID prior to the study. They were then placed on placebo medication and the NSAID were stopped for two weeks. In all cases, the serum rheumatoid factor levels rose. The patients were then placed on piroxicam 20 mg once a day for 10 weeks. The rheumatoid factor levels quickly came back to control. Surprisingly, they continued to decrease throughout the 1O-week length of the study; so that after 10 weeks the mean rheumatoid factor levels of the 20 patients was 60% of previous levels, when they were on the other nonsteroidals prior to initiation of the study. In every patient the rheumatoid factor level was lower after 10 weeks of piroxicam than it was on the other nonsteroidal drugs. This lowering of the rheumatoid factor level on piroxicam parallels the increased potency of

NSAID AND IMMUNE

141

FUNCTION

piroxicam in decreasing levels in vitro.

the rheumatoid

factor

CONCLUSION

Prostaglandin synthetase inhibitors are now currently used as antiinflammatory agents in many immunologic diseases such as rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE). Although their antiinflammatory activity

is obvious, it is less clear what their effect is on the basic disease process. In SLE and probably also in RA, suppressor cell function is decreased.84 If our data that endogenous PGEz tonically inhibits a suppressor cell for immunoglobulin and autoantibody formation have releVance to the in vivo situation, then PG-synthesis inhibitors would be expected to improve the basic disease process.

REFERENCES 1. Goodwin JS, Webb DR: Regulation of the immune response by prostaglandins. Clin Immunol Immunopathol 15:106-122,198O 2. Webb DR, Osheroff PL: Antigen stimulation of prostaglandin synthesis and control of immune responses. Proc Nat1 Acad Sci USA 43: 130&l 304,1976 3. Goodwin JS, Bankhurst AD, Messner RP: Suppression of human T-cell mitogenesis by prostaglandin. Existence of a prostaglandin-producing suppressor cell. J Exp Med 146:1719-1734, 1977 4. Goodwin JS, Messner RP, Peake CT: Prostaglandin suppression of mitogen-stimulated lymphocytes in vitro. J Clin Invest 62:753-760, 1978 5. Goodwin JS, Messner RP, Williams RC: Inhibitors of T-cell mitogenesis: Effect of mitogen dose. Cell Immunol 45303-308, 1979 6. Robinson DR, Levine L: Prostaglandin concentrations in synovial fluid in rheumatic diseases: Action of indomethatin and aspirin, in Robinson HJ, Vane JR, (eds): Prostaglandin Synthetase Inhibitors. New York, Raven Press, 1974; pp 223-228 7. Ferraris VA, DeRubertis FR: Release of prostaglandin by mitogen- and antigen-stimulated leukocytes in vitro. J CIin Invest 54378-389, 1974 8. Novogrcdsky A, Rubin AL, Stenzel KH: Selective suppression by adherent cells, prostaglandin and cyclic AMP analogues of blastogenesis induced by different mitogens. J ImmunoI 122:1-7, 1979 9. Muscoplat CC, Klausner DJ, Brunner CJ, et al: Regulation of mitogen- and antigen-stimulated lymphocyte blastogenesis by prostaglandins. Infect Immun 26:31 l-315, 1979 IO. Erten U, Emre T, Cavdar AO, et al: An in vitro study on the effect of prostaglandins E, and F, alpha on E-rosette forming activity of normal lymphocytes. Prostaglandins Med 5:255-258; 1980 II. Venza-Teti, Misefari A, Sofo V, et al: Interaction between prostaglandins and human t lymphocytes: Effect of PGE, on E-receptor expression. Immunopharmacology 2:165-171.1980 12. Gordon D, Bray M, Morley J: Control of lymphokine secretion by prostaglandins. Nature 262:401402, 1976 13. Baker PE, Fahey JV, Munck A: Prostaglandin inhibition of T-cell proliferation is mediated at two levels. Cell Immunol61:52-61, 1981 14. Rappaport RS, Dodge GR: Prostaglandin E inhibits the production of human interleukin 2. J Exp Med 155:943948, 1982

15. Darrow TL, Tomar RH: Prostaglandin-mediated regulation of the mixed lymphocyte culture and generation of cytotoxic cells. Cell Immunol 56:177-183, 1980 16. Leung KH, Mihich E: Prostaglandin modulation of development of cell-mediated immunity in culture. Nature 288:597-600, 1980 17. Schultz RM, Stoychkov JN, Pavlidis N, et al: Role of E-type prostaglandins in the regulation of interferon-treated macrophage cytotoxic activity. J Reticuloendothel Sot 26:93-l 02, 1979 18. Brunda MJ, Herberman RB, Holden HT: Inhibition of murine natural killer cell activity by prostaglandins. J lmmunol 124:2682-2687,198O 19. Roder J, Klein M: Target effector natural killer cell system. IV. Modulation tides. J Immunol 123:2785-2790, 1979

interaction in the by cyclic nucleo-

20. Garovoy MR, Strom TB, Kaliner M, et al: Antibodydependent lymphocyte-mediated cytotoxicity mechanism and modulation by cyclic nucleotides. Cell Immunol20: 197204, 1975 21. Trofatter KF, Daniels, CA: Interaction of human cells with prostaglandins and cyclic AMP modulators. I. Effects of complement-mediated lysis and antibody-dependent cellmediated cytolysis of herpes symplex virous-infected human fibroblasts. J Immunol 122: 1363-I 370, 1979 22. Droller MJ, Schneider MU, Perlman P: A possible role of prostaglandin: The inhibition of natural and antibcdydependent cell-mediated cytotoxicity against tumor cells. Cell Immunol39: 165-l 77.1978 23. Kennedy MS, Stobo JD, Goldyne ME: In vitro synthesis of prostaglandins and related lipids by populations of human peripheral blood mononuclear cells. Prostaglandins 20:135-145.1980 24. Kurland JI, Bockman R: Prostaglandin E production by human blood monocytes and mouse peritoneal macrophages. J Exp Med 147:952-957, 1978 25. Webb DR, Nowowiejski I: Mitogen-induced changes in lymphocyte prostaglandin levels: A signal for the induction of suppressor cell activity. Cell Immuno! 41:72-85, I978 26. Webb DR, Rogers RJ, Nowowiejski I: Endogenous prostaglandin synthesis and the control of lymphocyte function Proc NY Acad Sci 332:262-370, 1980 27. Passwell JH, Dayer JM, Merler E: Increased prostaglandin production by human monocytes after membrane receptor activation. J Immunol 123:115-120, 1979 28. Humes JL, Bonney RJ, Pebes L, et al: Macrophage

142

synthesis and release prostaglandins in response to inflammatory stimuli. Nature 269: 149-15 1, I977 29. Ellner JJ, Spagnuolo PJ: Suppression of antigen and mitogen-induced human T-lymphocyte DNA synthesis by bacterial lipopolysaccharide: Mediation by monocyte activation and production of prostaglandins. J Immunol I23:26892695, 1979 30. Bonney RJ, Naruns PP. Davies P, et al: Antigenantibody complexes stimulate the synthesis of prostaglandins by mouse peritoneal macrophages. Prostaglandins 18:605616,1979 31. Cantanzaro A: Suppression of T-cell function in coccidioidomycosis. Clin Res 27:36A, 1979 32. Muscoplat CC, Rakich PM, Thoen CO, et al: Enhancement of the lymphocyte blastogenic and delayed hypersensitivity skin responses by indomethacin. Infect Immun 20:627-63 1, 1978 33. Dore-Duffy P, Zurier RB: Lymphocyte adherence in multiple sclerosis. Role of monocytes and increased sensitivity of MS lymphocytes to prostaglandin E. Clin Immuno Immunopathol 19:303-3 13, 198 1 34. Goodwin JS, DeHoratius R, Israel H, et al: Suppressor cell function in sarcoidosis. Ann Intern Med 90:169-173, 1979 35. Goodwin JS, Messner RP, Bankhurst AD, et al: Prostaglandin-producing suppressor cells in Hodgkin’s disease. N Engl J Med 297~963-964. 1977 36. Pelus LM. Bockman RS: Increased prostaglandin synthesis by macrophages from tumor-bearing mice. J Immunol 123:2 118-2125, 1979 37. Bockman RS: Stage-dependent reduction in T colony formation in Hodgkin’s disease. Coincidence with monocyte synthesis of prostaglandin. J Clin Invest 66:523-53 I, 1980 38. Deshazo RD: Indomethacin-responsive mononuclear cell dysfunction in Hodgkin’s disease. Clin Immunol Immunopathol 17:66-75, 1980 39. Bankhurst AD, Kauffman G, Goodwin JS, et al: Suppressor leukocytes in anergic patients with solid tumors. J Lab Clin Immunol3:159-163. 1980 40. Vosixa G, Thies J: Effect of indomethacin on lymphocytes from cancer patients: Differentiation of patient types. Clin Immunol Immunopathol I3:[email protected], 1979 41. Balch CM, Tildem AB: Prostaglandin-mediated suppression of immuno competence in patients with melanoma. Surg Forum 3 1:407-411,1980 42. Staszak CS, Goodwin JS, Troup GM, et al: Decreased sensitivity to prostaglandin and histamine in lymphocytes from normal HLA-BI2 individuals: A possible role in autoimmunity. J Immunol 125:181-185, 1980 43. Kirby PJ, Morley J, Ponsford JR, et al: Defective PGE reactivity in leukocytes of multiple sclerosis patients. Prostaglandins 11:621-630, 1976 44. Goodwin JS, Messner RP: Prostaglandin E inhibition of mitogen stimulation in patients with multiple sclerosis. Prostaglandins 15:28 l-286, 1978 45. Dore-Duffy P, Zurier RB: Lymphocyte adherence in multiple sclerosis: ElTect of aspirin. J Clin Invest 63: 154-I 57, 1979 46. Page RC, Clagett JA, Engel LD, et al: Effects of prostaglandin on the antigen- and mitogen-driven responses of peripheral blood lymphocytes from patients with adult and

GOODWIN AND CEUPPENS

juvenile periodontitis. Clin Immunol Immunopathol 11:7787, 1978 47. Goodwin JS, Messner RP: Sensitivity of lymphocytes to prostaglandin E2 increases in subjects over age 70. J Clin Invest 64:434-439, 1979 48. Goodwin JS: Changes in lymphocyte sensitivity to prostaglandin E, histamine, hydrocortisone and X-irradiation with age. Clin Immunol Immunopathol25:243-25 1, I982 49. Goodwin JS: Increased sensitivity to prostaglandin E, in old people. Prostaglandins Med 3:395-400, 1979 50. Goodwin JS, Bromberg S, Staszak C, et al: Effect of physical stress on sensitivity of lymphocytes to inhibition by prostaglandin E,. J Immunol 127:5 18-522, 198 I 51. Wolinsky S, Goodwin JS, Messner RP, et al: Role of prostaglandins in the depressed cell-mediated immune response in rheumatoid arthritis. Clin Immunol Immunopathol 17:31-37, 1980 52. Goodwin JS, Murphy S, Bankhurst AD, et al: Partial reversal of the cellular immune defect in common variable immunodeficiency with indomethacin. J Clin Lab Immunol 1:197~199, 1978 53. Muscoplat CC, Rakich PM, Thoen CO, et al: Enhancement of lymphocyte blastogenic and delayed hypersensitivity with skin responses by indomethacin. Infect Immun 20:627-63 1, 1978 54. Lipsmeyer E: Effect of cimetidine and indomethacin on delayed hypersensitivity. Transplantation 33:107-109, 1982 55. Anderson CB, Jaffee BM, Graff RJ: Prolongation of murine skin allografts by prostaglandin E,. Transplantation 231444-447, I977 56. Blyth WA, Hill TJ, Field HJ, et al: Reactivation of herpes simplex virus infection by ultraviolet light and possible involvement of prostaglandins. J Gen Viral 33:547-550, 1976 57. Tracey DE, Adkinson NF: Prostaglandin synthesis inhibitors potentiate the BCG-induced augmentation of natural killer cell activity. J Immunol 125:136-141, 1980 58. Goodwin JS, Selinger DS, Messner RP, et al: Effect of indomethacin in vivo on humoral and cellular immunity in humans. Infect Immun 19:43&433, 1978 59. Bach MA, Fournier C, Bach JF: Regulation of thetaantigen expression by agents altering cyclic AMP level and by thymic factor. Ann NY Acad Sci 249:316-327, 1975 60. Burchiel SW, Warner NL: Cyclic AMP modulation of Fc receptor expression on a pre-B cell lymphoma. J Immunol 124:1016-1021, 1980 61. Rogers TJ, Nowowiejski I. Webb DR: Partial characterization of a prostaglandin-induced suppressor factor. Cell Immunol50:82-93, I980 62. Fischer A, Durandy A, Griscelli C: Role of prostaglandin E2 in the induction of nonspecific T lymphocyte suppressor activity. J Immunol 126:1452Zl455, 1981 63. Kaszubowski P, Goodwin JS: Monocyte-produced prostaglandin induces Fc gamma receptor expression on human T-cells. Cell Immunol68:343-348, 1982 64. Stobo JD, Kennedy MS, Goldyne ME: Prostaglandin E modulation of the mitogenic response of human T cells. Different response of T-cell subpopulations. J Clin Invest 64:1188-1203, 1979 65. Gualde N, Goodwin JS: Etfects on prostaglandin E,

143

and preincubation on lectin-stimulated proliferation human T-cell subsets. Cell Immunol 70:373, 1982

of

66. Schultz RM, Pavlidis NA, Stylos WA, et al: Regulation of macrophage tumoricidal function: A role for prostaglandins of the E series. Science 202:320-321, 1978 67. TatTet S, Russell SW: Macrophage-mediated tumor cell killing: Regulation of expression of cytolytic activity by prostaglandin E. J Immunol 126:424427, I98 I 68. Drysdale B, Shin HS: Activation of macrophages for tumor cell cytotoxicity: Identification of indomethacin sensitive and insensitive pathways. J Immunol 127:76&765, 1981 69. McCarthy ME, Zwilling BS: Differential effects of prostaglandins on the antitumor activity of normal and BCG-activated macrophages. Cell Immunol60:91-99, 1981 70. Schnyder J, Dewald B, Baggiolini M: Effects of cyclooxygenase inhibitors and prostaglandin E, on macrophage activation in vitro. Prostaglandins 18:411421, 1979 71. Razin E, Bauminger S, Globerson A: Effect of prostaglandins on phagocytosis of sheep erythrocytes by mouse peritoneal macrophages. J Reticuloendothel Sot 23:237-242, 1978 72. Cantarow WD, Cheung HT, Sundharadas G: Effects of prostaglandins on the spreading, adhesion, and migration of mouse peritoneal macrophages. Prostaglandins 16:39-46, 1978 73. Wahl LM, Olsen CE, Wahl SM, et al: Prostaglandin and cyclic AMP regulation of macrophage involvement in connection tissue destruction. Ann NY Acad Sci 332:271278, 1979 74. Morito T, Bankhurst AD, Williams RC: Studies on the modulation of immunoglobulin production by prostaglandins. Prostaglandins 20:383-390, 1980

75. Ceuppens JL, Goodwin JS: Endogenous prostaglandin E2 enhances polyclonal immunoglobulin production by tonically inhibiting T suppressor cell activity. Cell Immunol 70:41, 1982 76. Moretta L, Webb SR, Grossi CE, et al: Functional analysis of two human T-cell subpopulations: Help and suppression of B-cell responses by T cells bearing receptors for IgM or IgG. J Exp Med 146:184-200, 1977 77. Goodwin JS, Kaszubowski PA, Williams RC, Jr: Cyclic adenosine monophosphate response to prostaglandin E2 on subpopulations of human lymphocytes. J Exp Med 150:1260-1264,1979 78. Fulton AM, Levy JG: The induction of nonspecific T suppressor lymphocytes by prostaglandin E,. Cell Immunol 595460, 1981 79. Mattingly JA, Kemp JD: Suppression of secondary plaque-forming cell responses by rat splenic adherent cells. Cell Immunol48:195-200, 1979 80. Webb DR, Nowowiejski I: The role of prostaglandins in the control of primary 19s immune response to sRBC. Cell Immunol33:1-lo,1977 81. Kazimiera D, Grinwich KD, Plescia OJ: Tumormediated immuno-suppression: Prevention by inhibitors of prostaglandin synthesis. Prostaglandins 14:1175-l 182, 1977 82. Ceuppens JL, Rodriguez M, Goodwin JS: Non-steroidal anti-inflammatory agents inhibit the synthesis of IgM rheumatoid factor in vitro. Lancet l:528-530, 1982 83. Goodwin JS, Ceuppens JL, Rodriguez MA: Effects of administration of nonsteroidal antiinflammatory agents and indices of cellular immune status and serum rheumatoid factor levels in patients with rheumatoid arthritis. JAMA, 1983 (in press) 84. Talal N: Disordered immunologic regulation and autoimmunity. Transplant Rev 31:24&263, 1976