Seminars

ARTHRITIS

in

AND RHEUMATISM APRIL 1991

VOL 20, NO 5

Prolactin,

Immunoregulation,

By Luis J. Jara, Carlos Lavalle, Antonio Pindaro

Martinez-Osuna,

and Autoimmune Fraga, Celso Gbmez-Sanchez,

Bernard F. Germain,

Diseases Luis H. Silveira,

and Luis R. Espinoza

Cells of the immune system synthesize prolactin and express mRNA and receptors for that hormone. lnterleukin 1, interleukin 6, y interferon, tumor necrosis factor, platelet activator factor, and substance P participate in the release of prolactin. This hormone is involved in the pathogenesis of adjuvant arthritis and restores immunocompetence in experimental models. In vitro studies suggest that lymphocytes are an important target tissue for circulating prolactin. Prolactin antibodies inhibit lymphocyte proliferation. Prolactin is comitogenic with concanavalin A and induces interleukin 2 receptors on the surface of lymphocytes. Prolactin stimulates ornithine decarboxylase and activates protein kinase C, which are pivotal enzymes in the differentiation, proliferation, and function of lymphocytes. Cyclosporine A interferes with prolactin binding to its receptors on lymphocytes. Hyperprolactinemia has been found in patients with systemic lupus erythematosus. Fibromyal-

gia, rheumatoid arthritis, and low back pain patients present a hyperprolactinemic response to thyrotropin-releasing hormone. Experimental autoimmune uveitis, as well as patients with uveitis whether or not associated with spondyloarthropathies, and patients with psoriatic arthritis may respond to bromocriptine treatment. Suppression of circulating prolactin by bromocriptine appears to improve the immunosuppressive effect of cyclosporine A with significantly less toxicity. Prolactin may also be a new marker of rejection in heart-transplant patients. This body of evidence may have an impact in the study of rheumatic disorders, especially connective tissue diseases. A role for prolactin in autoimmune diseases remains to be demonstrated. Copyright o 1991 by W.B. Saunders Company

From the Division of Rheumatology, Department of lntmwl DiGon of Endocrinology, Department of Internal M&c&, Universir):of South Florida College ofMedicine, Tampa, FL, and the Rheumatic Diseases Unit, Hospital de Especialidades, Centro Medico La Raza, IMSS, Mexico, DF, Mexico. Luis J. Jara, MD: Research Fellow, Division of Rheumatoloa, Department of Internal Medicine, University of South Florida College of Medicine, and Rheumatic Diseases Unit. Hospital de Especialides, Centro Medico La Raza, IMSS, Mexico; Carlos Lavalle, MD, FACP: Rheumatic Diseases Unit. Hospital de Especialidades, Centro Mexico La Raza. IMSS, Mexico; Antonio Fraga, MD, FACP: Rheumatic Diseases Unit, Hospital de Especialidades, Centro Medico La Raza. IMMS, Mexico; Celso Gbmez-SBnchez, MD: Professor of Medicine, Div&ion of Endocrinology, Department of Internal Medicine. Universitvof South Florida College of Medicine:

Luis H. Silveira, MD: Research Fellow, Do&ion of Rheumatology, Department of Internal Medicine, University of South Florida College of Medicine; Pindaro Martinez-Osuna, MD: Research Fellow, Division of Rheumatology, Department of Internal Medicine, University of South Florida College of Medicine; Bernard F. Germain, MD: Professor of Medicine, Division of Rheumatology, Department of Internal Medicine, Unrversity of South Florida CoIlege of Medzcitte: Luis R. Espinoza, MD: Professor ofMedicine, Division of Rheumatology? Department of Internal Medicine. Untver.sity of South Florida College of Medicine. Address reprint requests to Luis R. Espinoza. MD, LSU Medical Center, Section of Rheumatology, 1542 Tulane Ave, New Orleans, LA 70112-2822. Copytight o 1991 by W.B. Saunders Cornpan! 0049.0172/9112005-0001$5.0010

Medicine,

Semw~ars ,n Arthrifjs

and Rheumatism,

INDEX WORDS: Prolactin; autoimmunity; noregulation.

Vol20. No 5 (April), 1991: pp 273.284

immu-

273

274

JARA ET AL

T

HE NEUROENDOCRINE and immune systems are intimately linked and involved in bidirectional communication. A complete regulatory loop between the immune and neuroendocrine systems has been postulated.’ Cells of the immune system express mRNA of adrenocorticotropic hormone, growth hormone (GH), thyroid stimulating hormone, and prolactin (PRL).’ In addition, they also synthesize these hormones3 and have receptors for all of them.4 The central nervous system cells have receptors for cytokines, ie, interleukin 1 (IL-l), and cytokines are synthesized by brain astrocytes and microglial cells.5,6 PRL has an important role in the regulation of immune function.7 It has the ability to promote cell growth and/or differentiation in several tissues and also functions as an immunostimulator.’ Hyperprolactinemia has recently been demonstrated in men and pregnant women with systemic lupus erythematosus (SLE). Clinical observations and experimental studies suggest that PRL may be important in the pathogenesis of SLE and possibly of other autoimmune diseases.‘,” This review discusses the interactions between PRL and the immune system and its association with SLE as well as other autoimmune diseases. PRL: PHYSIOLOGIC

ASPECTS

Prolactin activity was first recognized in 1928 when extracts of hypophysis were shown to induce milk secretion in oophorectomized rabbits.” Over 80 different actions have been ascribed to PRL in vertebrate groups ranging from fish to mammals. These include osmoregulation, growth and development, reproduction, metabolism of carbohydrates and lipids, secretion of mucins, and the more recently demonstrated effects on the immune system.‘* PRL is a single peptide chain of 198 amino acids, with a molecular weight of 23,500 d, and is a globular

Abbreviations: AA, adjuvant arthritis; AS, ankylosing spondylitis; BRC, bromocriptine; Con-A, concanavalin-A; G-A, cyclosporine A, FCA, Freund’s complete adjuvant; GH, growth hormone; NK, natural killer; ODC, ornithine decarboxylase; PAF, platelet activator factor; PKC, protein kinase C; PRL, prolactin; RA, rheumatoid arthritis; SLE, systemic lupus erythematosus.

protein consisting of about 25% a-helix, 33% p turn, and 8% random coil.‘” Immunoreactive PRL has been found in a variety of mammalian tissues and its synthesis has been demonstrated in human chorion,” endometrium,” a few tumor cell lines,16T and B lymphocytes,‘7,‘8 and the central nervous system.19 The rate of PRL synthesis and release is modulated by a variety of hormones and stimulator-y and inhibitory factors produced in the hypothalamus. PRL release is stimulated by serotonin, thyrotropin releasing hormone, and vasoactive intestinal peptide and inhibited by dopamine, opiates, and neuroleptics.2” IL-l”,” and probably other lymphokines participate in this regulation2’ affecting brain, autonomical, and endocrine functions (Table 1). Recently it has been shown that platelet activator factor (PAF) and substance P participate in the release of PRL.24,2”PAF is now known to have a wide spectrum of biological activities, including actions on the immune system,26 and substance P activates rheumatoid synovial cells to produce prostaglandins and metalloproteinases.27 On the other hand, PAF also participates in SLE pathogenesis.” PRL AND THE IMMUNE

SYSTEM

In 1930, Smith” noted that thymus glands of rats ceased to grow immediately after hypophysectomy and regressed to less than one half their weight as compared with controls. This study is regarded as the first to indicate the role of PRL in thymic physiology. However, only recently the relation between PRL and the immune system has been studied both in vivo

Table

1:

Regulation

of Prolactin

the lmmunomodulator lmmunomodulator

Secretion

Action

Reference

IL-I

?I

21, 22, 23

IL-6

+

23

IFN-y

_

40

TNF-(Y

_

39

PAF

+

24

*

25

Substance Abbreviations:

P +, enhancement:

at

Level

-, suppression; IL-l, interleu-

kin 1; IL-6, interleukin 6; IFN-y, interferon-y, necrosis factor-a; PAF, platelet activating factor.

TNF-u, tumor

PROLACTIN, IMMUNOREGULATION,

(experimental and in vitro.

AND AUTOIMMUNITY

models and clinical observation)

PRL. and Adjuvant Arthritis

In rats, Freund’s complete adjuvant (FCA) injected at the base of the tail induces an arthritic disease (adjuvant arthritis; AA) and there is a great amount of evidence suggesting that it is the result of an aberrant immune response.” Other abnormalities that may develop include balanitis, mild conjunctivitis, and uveitis. This picture is very similar to Reiter syndrome in humans.31 Nagy and Berczi reported that PRL, GH, and placental lactogen are potent in restoring the immune reactivity of hypophysectomized animals,32-34and the treatment of rats with the dopaminergic ergot alkaloid bromocriptine (BRC) inhibited the following immune reactions: contact sensitivity skin reaction to dinitrochlorobenzene, AA, experimental allergic encephalitis, and antibody formation to sheep red blood cells and bacterial lipopolysaccharide. These results suggest that BRC suppresses immunity by inhibition of PRL.35,36Neidhart and Larson studied the neuroendocrine response during FCA-induced immune stimulation, particularly the release of endogenous pituitary hormones. The activity of ornithine decarbowlase (ODC), the first enzyme involved in polyamine biosynthesis, was used in this study as a biological marker of the stimulation of different parts of the central nervous system, the pituitary gland, and lymphoid tissues following immune stimulation. The circadian rhythms of pituitary ODC activity and plasma PRL level, 3 to 4 days

Fig

1:

Relationships

among immune system, neuroendocrine system and adjuvant arthritis.

275

after FCA, showed that enhancement of enzymatic activity during the dark phase correlated with a marked release of PRL. These in vivo studies showed that FCA influences central nervous system pathways and lymphoid tissue (bone marrow, thymus, spleen, and lymph nodes), and supports the notion that endogenous PRL is involved in some immunologic early events that lead to the development of AA.37.38Following the injection of FCA the thymus and activated leucocytes might emit signals (ie, IL-l, IL-6, interferon, and tumor necrosis factor) to the neuroendocrine system, which are able to modify PRL secretion and other neuroregulators.39,4” These substances might modulate immune system activation and participate in the pathogenesis of AA (Fig 1). Recently, Wolar et al have shown that methylacetylenic putrescine, and inhibitor of ODC, prevents the development of collagen-induced arthritiq4’ an animal model which has many similarities to human rheumatoid arthritis (RA).42 Furthermore, this inhibition correlates with a reduction in anti-collagen II antibody. On the other hand Flescher et al presented evidence to suggest that excessive polyamines can contribute to the IL-2 deficiency found in RA. Blocking of polyamine production with inhibitors of ODC results in increased IL-2 production by RA mononuclear cells4’ Whyte and Williams administered BRC to arthritic mice immediately postpartum, and found that the drug suppressed the clinical exacerbation of collagen-induced arthritis.44 T lymphocytes are known to be involved in the pathogenesis of collagen-induced arthritis.45 It is possible that

276

the ability of T lymphocytes to respond to antigenic stimulation may be mediated by PRL through the interaction with ODC activity, leading to an excessive polyamine production and IL-2 deficiency. In this regard, recently, YuLee* has shown that ODC mRNA levels and various growth-related genes in cultured T-cells (Nb2 T lymphoma cells) are stimulated by PRL. PRL and Lymphocyte Function

Bernton et al in 1988 examined the T celldependent induction of activated tumoricidal macrophages in mice infected with either Mycobacterium bovis (strain BCG) or Listeria monocytogenes (LM) or inoculated with killed Propionibactetium acnes (PA). In this study, mice were treated with BRC and such treatment prevented T-cell dependent induction of macrophage tumoricidal activity after the intraperitoneal injection with either BCG, LM, or PA. Concomitant treatment with bovine PRL reversed this effect. Of the multiple events leading to macrophage activation in vivo, the production by T lymphocytes of y-interferon was the most impaired in BRC-treated mice. The critical influence of pituitary PRL release on maintenance of lymphocyte function and on lymphokine-dependent macrophage activation suggests that in mice lymphocytes are an important target tissue for circulating PRL.47 In studies designed to evaluate the in vitro effects of exogenous PRL on lymphocyte proliferation with serum-free and serum supplemented media, Hartmann et a14”were unable to find conditions where the addition of PRL to either culture system significantly affected proliferative responses. Surprisingly, when antisera to PRL were added to neutralize PRL (as a negative control) a profound inhibition of cell proliferation resulted. Antibodies to other pituitary hormones did not have any effect. Such inhibition appears specific for both murine and human growth factor-dependent cells and involved failure of anti-PRL treated lymphoid cells to respond to growth factors, as opposed to altered production of autocrine growth factors, namely IL-2 and IL-4. Thus, antibodies to PRL appear to block an event occurring in the Gl to S phase transition of these cell lines that constitutively express growth factor receptors. Collectively, these data indicate that lymphocytes

JARA ET AL

synthesize a cross-immunoreactive PRL-like protein (ir-PRL) and that ir-PRL is required for normal proliferation in response to mitogen or growth factors. Recent evidence supports a role for PRL in the mitogenesis of T lymphocytes. At suboptima1 concentrations of concanavalin-A (Con-A), the addition of rat PRL to murine splenocytes resulted in a dose-dependent increase in mitogenesis.17 Therefore, PRL is comitogenic with Con-A. In this sense, Mukherjee et a149have shown recently that PRL induces IL-2 receptors on the surface of lymphocytes in vitro. Splenocytes from rats incubated in the presence of PRL responded to synthetic IL-2 by dividing. Preincubation of PRL with PRL antiserum markedly reduced the subsequent incorporation levels of radiolabeled thymidine, whereas preincubation with GH antiserum was without effect. The IL-Zinduced proliferative response of PRL-treated lymphocytes implied that PRL induced IL-2 receptors on the cell surface. This was confirmed directly by flow cytometric analysis of PRL-treated lymphocytes stained with an antirat IL-2 receptor (anti-TAC) antibody. While these data offer little insight into an understanding of the mechanism(s) by which PRL might exert its IL-2 effect, this finding shows that PRL by itself is mitogenic. PRL,: A New Cytokine (?)

PRL has been shown to be a mitogenic agent for Nb2 T lymphoma cells. Shiu et a150demonstrated that these cells possess receptors that bind only PRL. In 1983, Russell et al showed that cyclosporine, an immunosuppresssive agent used in human organ transplantation5’ and recently in autoimmune diseases,” blocks PRLstimulated ODC activity in lymphocyte-containing tissues of the rat such as the thymus and spleen.53 The induction of increased ODC activity is an integral event regulating lymphocyte differentiation, proliferation, and function.” Subsequently, Russell et al reported the presence of PRL receptors on T, B, and monocytic cells of human peripheral blood and spIeen,5i.56 the estimated receptor number was 360 per cell. The specific binding of PRL to lymphocytes can be enhanced or blocked by cyclosporine depending on the concentration of the drug. Administration of cyclosporine A (Cs-A) also rapidly

PROLACTIN, IMMUNOREGULATION,

277

AND AUTOIMMUNITY

increased the serum PRL level in rats fourfold as compared with the control values within 1 hour of injection; such elevation could be blocked by BRC. This study suggests a role for PRL in the mechanism(s) of the immunosuppressive action of Cs-A. The discovery of PRL receptors on T and B lymphocytes suggested that immunomodulation by PRL was a primary response to the hormone and opened the door to exploration of the ability of PRL to directly affect cells of the immune system. It is known that these cells as well as many others have specific receptors for PRL (Table 2). In 1986, Hiestand et a15’provided evidence for the involvement of PRL in the maintenance of T cell immune competence. Their data showed that a reduction of serum PRL levels by BRC led to a decrease of lymphocyte responsiveness toward antigenic stimulation in rats. This phenomenon was observed both in vitro (mixed lymphocyte reaction) and, more importantly, in vivo (graft v host reaction). Dot blot hybridization showed that cytoplasmic RNA (mRNA) from lymphocytes hybridized with both PRL and GH cDNA probes. This study suggests that the ability of T lymphocytes to respond fully to antigenic stimulation may depend on the presence of PRL of pituitary origin bound to receptors on their outer membrane. On presentation of antigen these lymphocytes then produce a secondary signal in the form of a PRL/GH-

related polypeptide that can amplify either the response of the same cell to produce lymphokines (autocrine action) or the mitogenic response of neighboring lymphocytes (exocrine action), or both. Binding of antigens or mitogens to their cell surface receptors induces a secondary signal. In the same way as other immunomodulators, PRL can amplify the response of immune cells.5x In 1987, Montgomery et al” showed that lymphocytes may produce a PRL-like protein, In this study the cell subset(s) responsible for the production of this PRL-like molecule was not identified. However, Con-A is a T cellspecific mitogen and the production of PRL bioactivity paralells the course of Con-Ainduced proliferation, suggesting that this PRLlike factor is produced by this lymphocyte subset. The PRL-like material has a molecular mass of 46 kd, different from that of the anterior pituitary PRL entity in mice, which is 24 kd, and it has a peptide map homology similar to murine pituitary PRL. Di Mattia et al and Gellersen et al’x.sshave shown that PRL is synthesized by a human B-lymphoblastoid cell line (IM-9-P) and is indistinguishable from pituitary human PRL. These authors also demonstrate that PRL secretion in this line is regulated by dexamethasone but not by other hormones known to modulate PRL secretion in the pituitary or decidua. The proliferative action of PRL on T lymphocytes may be a complex interaction between two

Table 2: Prolactin Receptors in the Cells of the Immune System and in Some Tissues I. Immune

IV. Others:

system:

macrophages

prostate

T-lymphocytes

uterus

B-lymphocytes

kidney

natural killer cells

liver skeletal muscle

II. Endocrine

system:

anterior

pituitary

adrenal

lung erythrocyte aorta

pancreas (islets of Langerhans)

lymphoma

ovary

mastocytoma

(T-cell derived) (mast cells?)

testis mammary Ill. Centralnervous

glands system:

brain Adapted from Russell DH. New aspects of prolactin and immunity: A lymphocyte-derived kinase C activation. Trends Pharmacol Sci 1989;10:40-44

prolactin-like product and nuclear protein

278

JARA ET AL

factors: the level of circulating PRL from the anterior pituitary (endocrine action), and the ability of activated T and B cells to divide and produce a PRL-like factor with possible autocrine (on the same cell that produces them) and paracrine (on neighboring cells) actions. So far, the lymphocyte PRL-like substance appears to be a novel PRL variant and a new cytokine.

nucleus suggests that unoccupied PRL receptors are present in isolated nuclei. The implications of this study are important for our understanding of the nuclear function of polypeptide hormones such as PRL and its interaction with the immune system, ie, it is possible that PRL induction of IL-2-R on rat splenic lymphocytes might be explained by PKC activation.

PRL and Protein Kinase C

PRL. and Natural Killer Cells

The mitogenic activation of resting T lymphocytes involves two distinct cellular events: the synthesis of IL-2 and the synthesis and expression of IL-2 receptors. It is therefore significant that an immediate consequence of T cell activation is the phosphorylation of a wide range of proteins on serine (Ser), threonine (Thr), and tyrosine (Tyr) residues.60 Briefly, when antigen interacts with T-cell receptors (or when mitogens such as phytohemagglutinin or Con-A interact with T cell plasma membrane receptors) the following cascade of molecular events is thought to occur. Phospholipase C in the endoplasmic reticulum is activated, which in turn increases phosphoinositol turnover to produce diacylglycerol and inositol triphosphate. The latter stimulates the release of bound calcium from the endoplasmic reticulum to produce an increase in the concentration of cytosolic-free calcium. The increased free calcium activates calmodulin which, together with diacylglycerol, activates protein kinase C (PKC), a crucial enzyme in the regulation of cellular function and growth including the cells of the immune system.61 Evidence from a number of cell types and tissues, such as vascular smooth muscle,6’ suggests that the intracellular action of PRL involves the activation of PKC. Also, Buckley et aVj3have found that addition of PRL to purified rat liver nuclei and splenocyte nuclei results in a rapid activation by several 100-fold of PKC. This dose-dependent effect was blocked by a rat anti-PRL-receptor monoclonal antibody and by anti-PRL antiserum. The time course (3 minutes) for PRL activation of PKC in splenic nuclei is similar to that reported for antibodies directed against Ia and immunoglobulin antigens and for CAMP on nuclear PKC activity in intact murine B lymphocytes.64 The observation of receptor-mediated activation of PKC in the

Forni et al”’ in 1983 demonstrated that destruction of the tubero-infundibular region of the hypothalamus led to a persistently abrogated natural killer (NK) activity in mice. These observations suggest that neuroendocrine mechanisms play a relevant role in the control of NK activity. Gerli et alhhin 1986 studied human NK cell activity in 11 BRC-treated and 23 untreated hyperprolactinemic women and in 63 agematched healthy women using a “Cr-release assay. NK cell activity was significantly reduced in untreated hyperprolactinemic patients with respect to normal subjects, but therapy with BRC restored NK cell function of patients to the levels of normal controls, suggesting that hyperprolactinemia is also immunosuppressive. However, Matera et al” studied the functional state of native NK cells in a large group of patients with hyperprolactinemia of functional or tumoral origins with no radiological demonstration of hypothalamic damage and their results showed that the NK activity from patients with hyperprolactinemia is not different from that of normal subjects. Alarc&-Segovia et al have found decreased NK function in hyperprolactinemia (personal communication, 1986). Recently, Matera et aY have shown that highly purified NK cells express binding sites for PRL (660 receptors per cell). At concentrations corresponding to the therapeutic range, Cs-A induced a complete inhibition of PRL binding. By contrast, very low concentrations increased PRL binding to more than 100% of control levels. These data strongly support the idea of a close relation between PRL and NK cell lineage in humans. PROLACTIN,

IMMUNE

SYSTEM, AND DISEASE

There are a number of observations in humans that may indicate that PRL does indeed influence the immune system in both health and

PROLACTIN, IMMUNOREGULATION,

AND AUTOIMMUNITY

disease. Thyroid autoimmunity was evaluated in 92 hyperprolactinemic patients, 4 of whom had acromegaly. High titers of thyroglobulin and microsomal antibodies were found in 1 acromegalic and in 12 women with either adenomatous or idiopathic hyperprolactinemia.h4 Ishibashi and Yamaji”’ examined antithyroid antibodies in sera from 73 erythroid patients with prolactinoma and 40 patients with acromegaly. Statistical analysis showed that the occurrence of both positive antimicrosomal antibody and antithyroglobulin antibody was significantly higher in prolactinoma women than in normal and in acromegalic women. In five women, both antibodies disappeared after BRC therapy and six of nine prolactinoma patients had a higher B cell population. These results suggest that hyperprolactinemia is associated with a higher incidence of positive antithyroid antibodies, which may be consequent to nonspecific stimulation by PRL of B lymphocyte function. Vidaller et al, ” in 1986 studied the function of peripheral blood lymphocytes from four patients with tumoral hyperprolactinemia and normal ovarian function before and after PRL levels were normalized by treatment with BRC. Lymphocytc responses were suppressed to Con-A or pokeweed mitogen and to a lesser extent to phytohemagglutinin. The spontaneous Con-A induced suppression and production of IL-2 was low in two patients and was increased after drug treatment. All these findings give insight on the immunomodulatory role of PRL in vivo. Some of these abnormalities, such as B lymphocyte stimulation, with overproduction of antibodies, decreased response to mitogens and lowered suppression and production of IL-2, have been demonstrated in SLE patients.7’.71 PRL also was measured in 13 untreated and 42 treated patients suffering from celiac disease. Increased PRL values were found in four of the five untreated men, whereas the eight untreated women had normal levels. PRL was elevated in all 35 patients responding favorably to a glutenfree diet. In contrast. significantly lower PRL levels were found in seven patients not responding to the diet.” In 1987, Lavalle et al” described increased basal levels of serum PRL in men with SLE. Serum PRL levels were above normal limits in seven of eight patients studied (P < .Ol). It is of

279

interest that chloroquine, a drug used in the treatment of SLE, inhibits PRL secretion from cultured anterior pituitary cells,‘5 was used in the one patient with normal serum PRL levels. Recently, the same group, studied nine women (five active and four inactive) with SLE during pregnancy. Patients with SLE showed hyperprolactinemia in comparison with normal pregnancy (Table 3) and the five active SLE patients had the highest levels of PRL.“’ Both studies open the possibility that PRL can participate in SLE immunopathogenesis and offer a link between basic studies that support a direct role of PRL in the humoral and cell mediated immune response and clinical observations of SLE patients. In this regard, the recent observations about a patient with refractory neuropsychiatric SLE treated successfully with BRC and intravenous immunoglobulin,7h and the finding that BRC decreases early mortality in female NZB/B mice, are of note.77 Primary fibromyalgia and RA have been studied in relation with diurnal hormone variation of plasma cortisol, GH, PRL. and thyroidstimulating hormone.” Patients with fibromyalgia syndrome had loss of diurnal variation in plasma cortisol compared with RA patients. Serum PRL, GH, and thyroid-stimulating hormone levels were within normal limits and demonstrated expected diurnal variations. However, Ferraccioli ct al reported that a subset of fibromyalgia, RA and low-back pain patients, entities characterized by persistent pain, present several endocrinological abnormalities, namely cortisol nonsuppression to dexamethasone and a hyperprolactinemic response to thyrotropinreleasing hormone.“’ These studies support the hypothesis that chronic pain and stress are important determinants of neuroendocrinologic abnormalities. In regard to PRL and RA, these

Table

3:

Prolactin

in Pregnant

SLE Patients

Prolactin (ng/mL) Weeks

Control

10-19

127 2 41

20-25

119k35

26-30

119 2 26

191 t 48*

31-37

166 k 12

262 rt 55*

Adaptedfrom *P < .Ol

ref.10.

Patients -___ 81 244 168 2 41*

280

results also support the notion of a close relation between the neuroendocrine and immune systems. In 1985, Hedner and Bynke reported four patients with endogenous iridocyclitis treated with BRC for hyperprolactinemia, galactorrhea, and Parkinson disease and found alleviation of the iridocyclitis.8” Of interest three of these patients had spondyloarthopathies: two with ankylosing spondylitis (AS) and one with Reiter syndrome. Recently, ovarian function was studied in women with AS and PRL level was found normal in all 17 patients.81 The levels of PRL in Reiter syndrome, psoriatic arthritis, and other reactive arthritides are unknown. Determination of PRL values in these disorders is of interest because in AA, an excellent model of reactive arthritis3’ several studies have shown hyperprolactinemia.” Palestine et al demonstrated the effect of PRL suppression by BRC in combination with lower doses of Cs-A on experimental autoimmune uveitis.82 BRC plus low-dose Cs-A led to marked decreases in the incidence of experimental autoimmune uveitis and anti-S antigen antibody titers as well as in the lymphocyte proliferative assay. This study suggests that BRC can enhance the immunosuppression of low-dose Cs-A and supports the idea that PRL has a role in the immunosuppresive action of CS-A.‘~ Recently, the same group” demonstrated that the administration of Cs-A, Cs-A plus BRC, or BRC, results in down regulation of antinuclear autoantibody levels of PRL in patients with uveitis. The levels of PRL were determined in these patients and the mean serum prolactin level was 5.4 ngimL. After therapy with BRC, the mean serum prolactin level was less than 2.0 ng/mL in all patients.84 Dougados et alX5investigated the effect of BRC in 6 of the 12 Cs-Atreated RA patients previously reported. The addition of BRC did not potentiate Cs-A efficacy since no significant clinical improvement was observed and there was no reduction in Cs-A daily dosage, except in one patient. The small number of patients studied, however, and the low BRC dosage used preclude any definite conclusion. In contrast, of 35 psoriatic arthritis patients who were systematically treated with increasing doses of BRC, starting at 2.5 mg up to 30 mg per day, significant remission was observed in 77%.8h Larson et al, and Carrier et a18’-“”proposed

JARA ET AL

PRL as a new marker of rejection in hearttransplant patients. This marker seems a particularly useful predictor because its plasma concentration increased between 2 and 8 days before rejection episodes. In support of this idea, Wilner et al” studied the effect of hypoprolactinemia induced by a new drug, CQP 201403, a dopamine agonist, on the survival of heterotopic cardiac allografts and the ability of peripheral blood lymphocytes to respond in vitro to plant mitogens. CQP treatment greatly enhanced the immunosuppressive effect of Cs-A on graft survival and on in vitro lymphocyte function. The authors concluded that modulation by PRL may be a useful adjunct to Cs-A immunosuppression. Suppression of circulating PRL levels with BRC also improved the immunosuppressive effect of Cs-A, thereby lowering the frequency of early acute rejection as well as infection after cardiac transplantation.” CONCLUSIONS

AND PROSPECTS

The studies reviewed here support a link between PRL, the immune system, and autoimmune diseases. The implications of this link are numerous, ranging from understanding pathophysiologic mechanisms to developing new clinical therapies. The main findings can be summarized as follows. (1) Cells of the immune system can synthesize PRL and other biologically active neuroendocrine peptide hormones. (2) PRL receptors are present on normal T, B, NK-cells and on monocytes. (3) PRL play a role in the development of adjuvant arthritis and restores immunocompetence in experimental models. (4) In vitro lymphocyte functions are decreased in hypoprolactinemia and are restored by PRL. (5) PRL stimulates ODC and thereby polyamine synthesis in many tissues, including tissues of the immune system. (6) PRL can activate PKC, a pivotal enzyme in many cellular functions and cell growth regulation, including the immune system. (7) Cs-A interferes with PRL binding to its receptors on lymphocytes. (8) Hyperprolactinemia has been shown in a subset of SLE patients. (9) Experimental autoimmune uveitis and patients with uveitis, whether or not associated with spondyloarthropathies, may respond to BRC. This drug has been used successfully in psoriatic arthritis patients and in a single patient with SLE. (10) Suppression of circulating PRL by BRC appears to improve the

PROLACTIN, IMMUNOREGULATION,

281

AND AUTOIMMUNITY

immunosuppressive effect of Cs-A with significantly less toxicity. The evidence accumulated may have an impact in the study of rheumatic disorders, especially in connective tissue diseases. In this regard, a number of pertinent questions emerge. (1) What is the prevalence of hyperprolactinemia in patients with connective tissue diseases? (2) What is the relation among altered PRL levels, clinical activity, and expression of immune response, ie, antinuclear antibodies? (3) Is there an abnormality in immune cells PRLreceptors, as well as in the synthesis and secretion of PRL by T and B lymphocytes, of these patients? Does PRL influence receptors and synthesis of lymphokines of SLE patients? (4) Does PRL alter proliferation of SLE patients lymphocytes? (5) Is the response to PRL modified in patients with reactive arthritis and uveitis. in a similar way to that seen in AA? (6) Is the combination of low-dose Cs-A plus BRC useful in the treatment of these diseases? (7) What is the role of hyperprolactinemia in clinical reactivation of SLE-pregnancy? We believe the interaction between the immune and neuroendocrine systems is altered in patients with autoimmune diseases. The immune system fails to discriminate appropriately between self- and non-self-antigens, consequently generating both specific and nonspecific responses against the antigens. Lymphokines and other stimulators are necessary to activate self-reactive T cells. In this context, immunologic signals (ie, IL-1 and other immunomodulators) can directly affect the neuroendocrine

system, by their receptors, and produce the release of hormones (ie, PRL, neurotransmitters, or growth factors). These neuroregulators exert direct influences over immunological functions also by their receptors on T and B lymphocytes; therefore, hyperactive cells of the immune system produce more neuro and immunomodulators and complete a two-way humoral interaction between the immune and neuroendocrine systems. A combination of cytokines and neuromodulators may function in an additive, synergistic, or antagonistic manner on the immune-neuroendocrine system. This in-vivo activation signal leads to hyperactivity of B lymphocytes and exhausts the T cells, so they appear activated by cell surface markers although they are functionally depressed, and as a consequence they have an abnormal production of lymphokines. On the other hand, the neuroendocrine axis may have a defective response to inflammatory stimuli and contribute to the susceptibility to autoimmune diseases. In support of this concept, Sternberg et al have shown that Lewis strain rats have a defective hypothalamic response to inflammatory stimuli, resulting in a blunted corticosterone response from the adrenal gland.“’ The defect appears to lie in the regulation of synthesis of mRNA for corticotropin-releasing hormone.q3 New ideas and investigations for further confirmation of these studies are necessary, but regardless of their results, interaction between the neuroendocrine and immune systems will have a major impact on our understanding of the physiology, diagnosis, and therapy of connective tissue diseases.

REFERENCES I. Blalock JE. Smith EM: A complete regulatory loop between the immune and neuroendocrine systems. Fed Proc 44: 108-111. 1985 2. Weigent DA. Blalock JE: Interactions between the neuroendocrine and immune systems: Common hormones and receptors. Immunol Rev 100:79-108, 1987 3. Khansari D, Mm-go A, Faith R: Effects of stress on the immune system. Immunol Today 11:170-175, 1990 4. Bhalla AK: Hormones and the immune response. Ann Rheum Dis 48: l-6, 1989 5. Plats-Salaman C: Immunomodulators and feeding regulation: A humoral link between the immune and nervous system. Brain Behav lmmun 3:193-213, 1989 h. Fontana A, Frei K, Bodmer S, et al: Immune-mediated encephalitis on the role of antigen-presenting cells in brain tissue. Immunol Rev 100:185-201, 1987 7. Berczi I, Nagy E: The effect of prolactin and growth hormone on hemolymphopoietic tissue and immune func-

1, Kovacs K (eds): Hormones MTP Press. 1987. pp 145-171

tion. in Berczi

Lancaster

and Immunity.

8. Russell DH: New aspects of prolactin and immunity: A lymphocyte-derived prolactin-like product and nuclear protein kinase C activation. Trends Pharmacol Sci 10:40-44. 1989 9. Lavalle C, Loyo E, Paniagua between systemic 1987

R et al: Correlation

study

prolactin and androgens in male patients with lupus erythematosus. J Rheumatol l-4:268-272,

10. Fraga A, Jara L, Graef A, et al: Prolactin and gonadal hormones in pregnant SLE patients. Arthritis Rheum. 32:S24, 1989 (abstr) 11. Stricker P, Grueter F: Action del’hyphophysesur la montee laiteuse. 99: 1978-1980, 1928 12. Lamberts

du lobe anterieur CR Sot Biol (Par)

SWJ. Mac Leod RM: Regulation

of prolac-

282

tin secretion at the level of the lactotroph. Physiol Rev 70:279-318, 1990 13. Nicoll CS, Mayer GL, Russell SM: Structural features of prolactin and growth hormones that can be related to their biological properties. Endocr Rev 7:169-203, 1986 14. Golander A, Hurley T, Barret J, et al: Prolactin synthesis by human chorion decidual tissue: a possible source of amniotic fluid prolactin. Science 202:311-313, 1978 15. Maslar IA, Riddick DH: Prolactin production by human endometrium during the normal menstrual cycle. Am J Obstet Gynecol135:751-754,1979 16. Rosen SW, Weintraub BD, Aaronson SA: Nonrandom ectopic protein production by malignant cells: Direct evidence in vitro. J Clin Endocrinol Metab 50:834841,198O 17. Montgomery D, Zukoski C, Shah G, et al: Concanavalin A-stimulated murine splenocytes produce a factor with prolactin-like bioactivity and immunoreactivity. Biochem Biophys Res Commun 145:692-698,1987 18. DiMattia G, Gellersen B, Bohnet H, et al: A human B-lymphoblastoid cell line produces prolactin. Endocrinology 122:2508-2517, 1988 19. De Vito WJ: Distribution of immunoreactive prolactin in the male and female rat brain: Effects of hypophysectomy and intraventricular administration of colchicine. Neuroendocrinology 47:87-99, 1989 20. Fluckiger E, Del Pozo E, van Werder K: Control of prolactin secretion, in Fluckiger E, Del Pozo E, von Werder K (eds): Prolactin, Physiology, Pharmacology and Clinical Findings. Berlin, Germany, Springer-Verlag, 1982, pp 24-41 21. Bernton EW, Beach JE, Holaday JW, et al: Release of multiple hormones by a direct action of interleukin-1 on pituitary cells. Science 238:519-521, 1987 22. Schettini G, Lorio T, Meucci 0, et al: Interleukin-1-B modulation of prolactin secretion from rat anterior pituitary cells: involvement of adenylate cyclase activity and calcium mobilization. Endocrinology 126:1435-1441, 1990 23. Yamaguchi M, Matsuzaki N, Hirota K, et al: Interleukin 6 possibly induced by interleukin ll3 in the pituitary gland stimulates the release of gonadotropins and prolactin. Acta Endocrinol (Copenh) 122:201-205, 1990 24. Camoratto AM, Grandison L: Platelet-activating factor stimulates prolactin release from dispersed rat anterior pituitary cells in vitro. Endocrinology 124:1502-1506, 1989 25. Arisawa M, Snyder G, Yu WH, et al: Physiologically significant inhibitory hypothalamic action of substance P on Prolactin release in the male rat. Neuroendocrinology 52122-27, 1990 26. Braquet P, Touqui L, Shen TY, et al: Perspectives in platelet-activating factor research. Pharmacol Rev 39:97145,1987 27. Lotz M, Carson DA, Vaughan JH: Substance P activation of rheumatoid synoviocytes: Neural pathway in pathogenesis of arthritis. Science 235:893-895, 1987 28. Tetta C, Bussolino F, Modena V, et al: Release of platelet-activating factor in systemic iupus erythematosus. Int Arch Allergy Appl Immunol91:244-256, 1990 29. Smith PE: The effect of hypophysectomy upon the involution of the thymus in the rat. Anat Ret 47:119-129, 1930

JARA ET AL

30. Van Eden W, Holoshitz J, Cohen I: Antigenic mimicry between mycobacteria and cartilage proteoglycans: the model of adjuvant arhtritis. Concepts Immunopathol4:144170,1987 31. Wilder RL: Animal models of reactive arthritis, in Espinoza L, Goldenberg D, Arnett F, et al (eds): Infections in the rheumatic diseases. Philadelphia, PA, GruneStratton, 1988, pp 311-316 32. Nagy E, Berczi I: Immunodeficiency in hypophysectomized rats. Acta Endocrinol (Copenh) 89:530-537. 1978 33. Berczi I, Nagy E, Kovacs K, et al: Regulation of humoral in rats by pituitary hormones. Acta Endocrinol (Copenh) 98:506-513,198l 34. Nagy E, Berczi I, Friesen HG: Regulation of immunity in rats by lactogenic and growth hormones. Acta Endocrinol (Copenh) 102:351-357, 1983 35. Nagy E, Berczi I, Wren G, et al: Immunomodulation by bromocriptine. Immunopharmacology 6:231-243, 1983 36. Spangelo BL, Hall NR, Ross PC, et al: Stimulation of in vivo antibody production and concanavalin-A-induced mouse spleen cell mitogenesis by prolactin. Immunopharmacology 14:11-20, 1987 37. Neidhart M: Bromocriptine microcapsules inhibit ornithine decarboxylase activity induced by Freund’s complete adjuvant in lymphoid tissues of male rats. Endocrinology 125:2846-2852, 1989 38. Neidhart M, Larson DF: Freund’s complete adjuvant induces ornithine decarboxylase activity in the central nervous system of male rats and triggers the release of pituitary hormones. J Neuroimmunol26:97-105, 1990 39. Gaillard RC, Turnill D, Sappino P, et al: Tumor necrosis factor inhibits the hormonal response of the pituitary gland to hypothalamic releasing factors. Endocrinology 127:101-106, 1990 40. Vankelecom H, Carmeliet P, Heremans H. et al: Interferon-I inhibits stimulated adrenocorticotropin, prolactin, and growth hormone secretion in normal rat anterior pituitary cell cultures. Endocrinology 126:2919-2926, 1990 41. Wolar JA, Logan D, Bowlin T: Methylacetylenic putrescine (MAP), an inhibitor of polyamine biosynthesis. prevents the development of collagen-induced arthritis. Cell Immunol 125:498-507, 1990 42. Courtenay JS, Dallman MJ, Dayan AD, et al: Immunization against heterologous type II collagen induces arthritis in mice. Nature 283:666-668, 1980 43. Flescher E, Bowlin T, Ballester A, et al: Increased polyamines may downregulate interleukin 2 production in rheumatoid arthritis. J Clin Invest 83:1356-1352,1989 44. Whyte A, Williams RO: Bromocriptine suppresses postpartum exacerbation of collagen-induced arthritis. Arthritis Rheum 31:927-928, 1988 45. Holmdahl R, Klareskog L, Rubin K, et al: T lymphocytes in collagen II-induced arthritis in mice: Characterization of arthritogenic collagen II-specific T-cell lines and clones. Stand J Immunol22:295-306, 1985 46. Yu-lee L: Prolactin stimulates transcription ofgrowthrelated genes in Nb2 T Lymphoma cells. Mol Cell Endocrinol68:21-28, 1990 47. Bernton EW, Meltzer MS, Holaday JW: Suppression of macrophage activation and T-lymphocyte function in hypoprolactinemic. Science 239:401-404,1988

PROLACTIN, IMMUNOREGULATION,

4X. Hartmann DP, Holaday of lymphocyte proliferation FASEB J 3:2194-2202, 1989

AND AUTOIMMUNITY

JW, Bernton EW: Inhibition by antibodies to prolactin.

4Y. Mukherjee P, Mastro A, Hymer WC: Prolactin induction of interleukin-2 receptors on rat splenic lymphocytes. Endocrinology 126:88-94, 1990 50. Shiu RPC. Elsholtz HP, Tanaka T, et al: Receptormediated mitogenic action of prolactin in a rat lymphoma cell line. Endocrinology 113:159-165, 1983 5 I, Cohen DJ. Loertscher R, Rubin MF, et al: Cyclosporine: A new immunosuppressive agent for organ transplantation. Ann Intern Med 101:667-682, 1984 52. Schnider R (ed): Cyclosporin in autoimmune disease. First International Symposium, Basle, March 18-20. Springer Verlag, Berlin, New York, Tokyo, Sept 1985 5.3. Russell DH, Larson DF, Womble JR, et al: Cyclosporin A inhibits ornithine decarboxylase induction in kidney in response to prolactin, growth hormone and insulin. Physiologist 26:A-13, 1983 53. Klimpel CR, Byus CV, Russell DH, et al: Cyclic AMP dependent protein kinase activation and the induction of ornithine decarboxylase during lymphocyte mitogenesis. J lmmunol 123:817-824, 1979 55. Russell DH, Matrisian L, Kibler R, et al: Prolactin receptors on human lymphocytes and their modulation by cyclosporine. Biochem Biophys Res Commun 121:899-906. 1084 5f). Russell DH, Kibler R, Matrisian L et al. Prolactin receptors on human T and B lymphocytes: Antagonism of prolactin binding by cyclosporine. J Immunol 134:30273031.1985

57. Hiestand PC, Mekler P, Nordann R, et al: Prolactin as a modulator of lymphocyte responsiveness provides a possible mechanism of action for cyclosporine. Proc Nat1 Acad Sci USA 83:2599-2603, 1986 5X. Dinarello CA. Mier JW: Lymphokines. 3 17:940-945, I987

N Engl J Med

5Y. Gellersen B. DiMattia, Friensen HG, et al: Regulation of prolactin secretion in the human B-lymphoblastoid cell line IM-9-P3 by dexamethasone but not other regulators of pituitary prolactin secretion. Endocrinology 125:28532861. 198’) 60. Wedner HJ, Bass G: Tyrosine.phosphorylation of a 66,000 Mr soluble protein in lectin-activated human peripheral blood T lymphocytes. J Immunol 136:4226-4231, 1986 61. Shaw LM: Advances in cyclosporine measurement. and therapeutic monitoring. 129’). 1308. 1989

pharmacology, Clin Chem 35:

62. Sauro MD. Buckley AR, Russell DH, et al: Prolactin stimulation of protein kinase C activity in rat aortic smooth muscle. Life Sci 44:1787-1792, 1989 63. Buckley A. Crowe P. Russell DH: Rapid activation of protein kinase C in isolated rat liver nuclei by prolactin, a known hepatic mitogen. Proc Nat1 Acad Sci USA X5:86498653. 198X 64. Russell DH, Zorn NE, Buckley AR, et al: Prolactin and known modulators of rat splenocytes activate nuclear protein kinase C. European J Pharmacol 188:139-152. 1990 65. Forni CT, Bindoni M, Santoni A, et al: Radiofrequency destruction of the tuberoinfundibolar region of

283

hypotalamus permanently abrogates NK cell activity in mice. Nature 306:181-184, 1983 66. Gerli R, Rambotti P, Nicoletti I, et al: Reduced number of natural killer cells in patients with pathological hyperprolactinemia. Clin Exp Immunol64:399-406. 1986 67. Matera L, Ciccarelli E, Cesano A, et al: Natural killer activity in hyperprolactinemic patients. Immunopharmacology 18:143-146, 1989 68. Mater-a L, Muccioli G, Cesano A, et al: Prolactin receptors on large granular lymphocytes: dual regulation by cyclosporin A. Brain Behav Immun 2:1-10, 1988 69. Ferrari C, Boghen M, Paracchi A. et al: Thyroid autoimmunity in hyperprolactinaemic disorders. Acta Endocrinol (Copenh) 104:35-41, 1983 70. Ishibashi M, Yamaji T: Immunological abnormalitied with hyperprolactinemia. The Endocrine Society. 71 Annual meeting, pp 407, 1989 (abstr) 71. Vidaller A, Llorente L, Larrea F, et al: T-cell dysregulation in patients with hyperprolactinemia: Effect of bromocriptine treatment. Clin Immunol lmmunopathol 38337.343, 1986 72. Alcocer-Varela J, Alarcon-Segovia D: Decreased production of and response to interleukin 2 by cultured lymphocytes from patients with systemic lupus erythematosus. J Clin Invest 69:1388-1392, 1982 73. Kammer GM, Stein RL: T lymphocyte immune dysfunctions in systemic lupus erythematosus. J Lab Clin Med 115:273-282, 1990 74. Stevens FM, McCarthy CF, Craig A: Is prolactin trophic to the intestine in coeliac disease? Gut 19:A992, 1978 75. Conconi MV, Walker AM: Chloroquine affects prolactin secretion and Golgi morphology in the mammotroph. Endocrinology. 114:725-734, 1984 76. Rabinovich CE, Schanberg LE, Kredich DW: Intravenous immunoglobulin and bromocriptine in the treatment of refractory neuropsychiatric systemic lupus erythematosus. Arthritis Rheum 33:R22, 1990 (abstr) 77. McMurray R, Keisler D, Walker S: Bromocriptine (Br) decreases early mortality in female NZBl’w mice. Arthritis Rheum 33:R30, 1990 (abstr) 78. McCain G, Tilbe KS: Diurnal hormone variation in fibromyalgia syndrome: A comparison with rheumatoid arthritis. J Rheumatol 16:154-157, 1989 (suppl 19) 79. Ferraccioli G, Cavalieri F, Salaffi F, et al: Neuroendocrinologic findings in primary fibromyalgia (soft tissue chronic pain syndrome) and in other chronic rheumatic conditions (rheumatoid arthritis, low back pain). J Rheumato1 17:869-873, 1990 (editorial) 80. Hedner LP, Bynke G: Endogenous iridocyclitis relieved during treatment with bromocriptine. Am J Ophthalmol 100:618-619, 1985 81. Jimenez-Balderas FJ, Tapia-Serrano R, MaderoCervera J. et al: Ovarian function studies in active ankylosing spondylitis in women. Clinical response to estrogen therapy. J Rheumatol 17:497-502, 1990 82. Palestine AG, Muellenberg-Coulombre CG, Kim MK, et al: Bromocriptine and low dose cyclosporine in the treatment of experimental autoimmune uveitis in the rat. J Clin Invest 79:1078-1081. 1987 83. Blank M, Palestine A, Nussenblatt R, et al: Downregulation of autoantibody levels of cyclosporine and bro-

JARA ET AL

mocriptine treatment in patients with uveitis. Clin Immunol ImmunophatoI54:87-97, 1990 84. Palestine AG, Nussenblatt RB, Gelato M: Therapy for human autoimmune uveitis with low-dose cyclosporine plus bromocriptine. Transplant Proc 20:131-135, 1988 85. Dougados M, Duchesne L, Amor B: Bromocriptine and cyclosporine A combination therapy in rheumatoid arthritis. Arthritis Rheum 31:1333-1334, 1988 (letter) 86. Weber G, Frey H: Zur behandlung der Psoriasis arthropathica mit Bromocriptin. Z Hautkr 61:1456-1466, 1986 87. Larson DF, Copeland JG, Russell DH: Prolactin predicts cardiac allograft rejection in cyclosporine immunosuppressed patients. Lancet 2:53, 1985 (letter) 88. Carrier M, Emery RW, Wild-Mobley J, et al: Prolactin as a marker of rejection in human heart transplantation. Transplant Proc 19:3442-3443, 1987 89. Carrier M, Russell DH, Cork RC, et al: Analysis of

risk factors for acute allograft rejection after heart transplantation. J Heart Transplant 9:372-375,199O 90. Wilner M, Ettenger R, Koyle M, et al: The effect of hypoprolactinemia alone and in combination with cyclosporine on allograft rejection. Transplantation 49:264-267,199O 91. Carrier M, Wild J, Pelletier C, et al: Bromocriptine as an adjuvant to cyclosporine immunosuppression after heart transplantation. Ann Thorac Surg 49:129-32, 1990 92. Sternberg EM, Hill JM, Chrousos GP, et al: Inflammatory mediator-induced hypothalamic-pituitary-adrenal axis activation is defective in streptococcal cell wall arthritissusceptible Lewis rats. Proc Nat1 Acad Sci USA 86:23742378, 1989 93. Sternberg EM, Scott Young III W, Bernardini R: A central nervous system defect in biosynthesis of corticotropinreleasing hormone is associated with susceptibility to streptococcal cell wall-induced arthritis in Lewis rats. Proc Nat1 Acad Sci USA 86:4771-4775, 1989