sign in sign up
<<
Home , Corticosteroid, Glucocorticoid

6 Annals of the Rheumatic 1995; 54: 6-16

REVIEW Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from Sex , and autoimmunity: facts and hypotheses

Jose Antonio P Da Silva

Increasing knowledge on the regulation of rejection and depressed tumour-associated immune responses over the past decade has immunity in animals and humans.' '-" A made it clear that the is thorough discussion of the vast evidence for involved in complex interactions with most, if immunomodulatory effects of sex in not all, of the biological systems responsible both animals and humans can be found in the for normal . The evidence for reviews cited above. Overall, oestrogens depress important interplays between the immune and mediated and stimulate mediated endocrine systems has launched the whole new immune responses, while and field of neuroimmunoendocrinology which has depress both components of the attracted the interest of scientists and clinicians immune system.3 alike. The mechanisms involved in the inter- There are a few experimental examples of actions between adrenal and gonadal steroids gender differences in autoimmune diseases and the immune system are poorly understood, which are dependent, not on sex hormones, however, and further clarification could have but on sex chromosome linked immune mech- important implications in the understanding anisms. The male BXSB mouse susceptibility of the pathogenesis and treatment of auto- to like is related to the Y immune diseases. chromosome linked 'autoimmune accelerating' yaa which is, intriguingly, capable of suppressing collagen induced , another Gender dimorphism in immune model of . 16 responses The pathogenic relevance of the immune Human and experimental autoimmune diseases effects of sex steroids is illustrated by their present marked gender differences. In humans, influence on experimental and human auto- females are more commonly affected, with a immune diseases. Androgens reduce and

female to male ratio that varies widely: 2-4:1 oestrogens enhance the incidence and severity http://ard.bmj.com/ in , 5-13:1 in systemic of experimental murine lupus.'7 18 Similarly, in lupus erythematosus, 3:1 in , 9:1 human SLE, oral contraceptives have been in Sjogren's syndrome, 4-8:1 in Graves' associated with disease exacerbation,'9 whereas disease.' 2 Similar gender related differences in the analogue offers some susceptibility to autoimmune diseases are degree of benefit.20 Orchidectomy increases observed with a large number of experimental and androgens decrease the susceptibility of and are to well rats whereas models,3 attributed the male to autoimmune thyroiditis,2' on September 26, 2021 by guest. Protected copyright. established gender dimorphism in immune administration of oestrogen increases the responses-a subject that has been object of production of against injected increasing interest over recent years, reflected thyroglobulin.22 Androgens were also shown in a number of excellent reviews.3-5 Overall, to reduce the incidence of autoimmune females have stronger humoral and cellular haemolytic anaemia in mice23 and rats,24 and to immune responses than males. Female animals decrease the severity of experimental Sjogren's from different species have higher serum syndrome-like lesions in mice.25 The relative concentrations of immunoglobulins6 and show resistance of male Lewis LEW/N rats to a greater and more sustained production of streptococcal cell wall induced arthritis is antibodies after immunisation and .' 7 abolished by castration and treatment with Cell mediated immune responses are also oestradiol.26 Interestingly, oestrogens have ben- Department of stronger in females, as demonstrated by a more eficial effects in a number of T cell dependent Experimental efficient rejection of allografts8 and relative experimental diseases such as collagen induced Pathology, to immunotolerance.9 27 induced 28 and St Bartholomew's resistance arthritis, adjuvant arthritis, Hospital Medical Sex hormones play a central role in this experimental autoimmune .29 College, gender dimorphism. Gender differences in anti- We have recently performed a series of London ECIM 6BQ, body production become apparent only after studies on the effects of sex hormones using United Kingdom Jose A P Da Silva sexual maturity'0 and are greatly reduced by an in vivo model of induced Correspondence to: gonadectomy." Oestrogens have been shown cartilage degradation in BALB/c mice. This Dr J A P Da Silva, to stimulate the B cell response, while both model allows the quantitation of cartilage Servico de Medicina III/ 13 Reumatologia, androgens and progesterone depress it.12 destruction by juxtaposed cotton-induced Hospitais da Universidade, graft rejection time is reduced by granulomatous tissue, the development of 3000 Coimbra, Portugal. gonadectomy in both male and female mice,8 which is T cell dependent.30 3' We have demon- Accepted for publication while treatment with oestrogens, androgens, strated that female mice degrade cartilage 5 September 1994 and progesterone results in delayed transplant faster than males and that gonadectomy was Sex hormones and immune response 7

associated with accelerated cartilage destruction hormones, growth , and prolactin. in both males and females, demonstrating a Treatment of rats with high doses of sex protective role for sex hormones in both hormones results in marked of the Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from sexes.32 Oestrogen replacement in castrated thymus50 and several lines of evidence suggest females completely reversed the effects of that effects of sex hormones in this gland ovariectomy and treatment with physiological may contribute to their actions on the immune concentrations of androgens resulted in signifi- response: a) oestrogens reduce and testos- cant cartilage protection in both castrated terone enhances the production of thymic males and females. Beneficial effects of sex hormones (such as thymulin) involved in intra- hormones were also recently reported in human thymic T cell differentiation;5' 52 b) CD4 and rheumatoid arthritis (RA), both with oestrogen CD8 thymic cells are depleted by oestrogen replacement in postmenopausal women33 and treatment;53 and c) the thymus was shown to with androgen therapy in male patients.34 be indispensable for some of the immune In summary, females have a stronger effects of sex steroids in vivo. II54 Growth immune reactivity than males which is trans- hormone and prolactin have important lated in a higher susceptibility to a variety of immunostimulatory effects. autoimmune diseases. Decrease in circulating has been shown to stimulate delayed cutaneous androgens as a result of orchidectomy increases hypersensitivity reactions, helper cell function male susceptibility, an effect reversed by of thymocytes, and haemaglutinin titres in androgen administration. Oestrogens, in con- hypophysectomized animals.55 56 It is con- trast, can enhance autoantibody production, sidered essential for development thereby increasing the severity of diseases and function.57 Female animals have greater depending on B cell hyperactivity, but they basal concentrations and lower pulse heights of have also been shown to suppress experimental growth hormone than males,58 probably as a diseases mediated by T cells. result ofhigher levels of oestrogens and proges- The mechanisms involved in terone.59 60 Prolactin has been demonstrated interaction with the immune system are to reconstitute humoral and cell mediated complex and poorly understood.5 35 In vitro immune responses in hypophysectomized studies have demonstrated the potential for animals55 61 and prolactin concentrations are direct effects of sex hormones on immuno- again greater in the female as a result of the competent cells. Oestrogens have been shown stimulating effects of oestrogens on its to suppress mitogen and antigen induced secretion,62 whereas has stimulation of T cells and depress T cell the opposite effect.63 The influence of sex suppressor activity.'2 15 3' and steroids on the production of thymic hor- progesterone inhibit mitogen induced lympho- mones, growth hormone, and prolactin have all cyte proliferation and enhance T cell been integrated in complex interaction suppressor activity.37-3 Sex hormones have schemes proposed to explain the gender

also been shown in vitro to modulate the dimorphism and influence of sex hormones in http://ard.bmj.com/ production of a variety of involved the immune system.4 64 Figure 1 presents a in immune responses, including interleukin simplified diagram of these interactions. (IL)-1,40 41 IL-6,42 IL-2,43 IL-4, IL-5, inter- Given the potent immunosuppressive and feron gamma,44 and transforming growth anti-inflammatory effects of glucocorticoids, factor f8 (TGFI.)5 The possibility of direct the potential interference of sex hormones with effects of sex hormones on the immune system the -pituitary-adrenal (HPA) axis is also supported by the evidence of oestrogen could have particularly important implications. on September 26, 2021 by guest. Protected copyright. receptors in human synovial macrophages46 However, the possibility that immune effects of and peripheral CD8 , although sex steroids in vivo may be related to inter- not in the helper/inducer subset.47 48 Androgen actions with the adrenal response is scarcely receptors have also been described in human addressed in the literature and mostly ignored thymocytes, but not in peripheral lympho- in reviews.35 64 65 cytes.43 4 However, the binding of sex steroids Glucocorticoids are the main endogenous was very low in these reports and studies using anti-inflammatory agents in vivo, interfering monoclonal antibodies to sex receptors with virtually every step of immune and have failed to demonstrate receptors in lympho- inflammatory responses,66 67 and are commonly cytes.3 Despite these interesting findings used in the treatment of autoimmune diseases. in vitro, caution should be used in their inter- The physiological circadian variation of pretation, as supraphysiological concentrations has been implicated in the diurnal fluctuation were used in most of these studies'2 15 and ofthe symptoms ofRA.68 69 Symptomatic auto- oestradiol has been shown to have variable or immune disease has also been reported even opposing results upon lymphocytes, following adrenalectomy for Cushing syndrome depending on concentration.4' The potential in women. o The importance of endogenous importance of carryover effects from in vivo glucocorticoids in the control of immune concentrations of sex steroids upon cultured conditions is also exemplified by the increased cells cannot be excluded49 and the presence of mortality associated with adrenalectomy in rats sex hormones in fetal calf serum used in cell with experimental allergic enchephalomyelitis.7' culture is seldom accounted for. Investigations using the cotton pellet model The actions of sex hormones on the immune of cartilage degradation have shown that system in vivo may be essentially indirect, glucocorticoids reduce the development of mediated by interactions with other immuno- granulomatous tissue and induce a significant modulatory factors such as the thymus protection of cartilage.72 73 8 Da Silva

Oestrogens Androgens, Prog (?) centrations of circulating glucocorticoids, to values similar to those of intact males. + I - Conversely, orchidectomy results in greater Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from +I- I I+"-I - responses to a variety of stresses.77 80 82 83 Replacement of oestradiol in ovariectomized females induces an increase in Pituitary Thymus| baseline and stimulated concentrations of , while androgens have the + opposite effect, reducing the Basal GH % Thymic hormones Prolactin T cell development response to activating stimuli.77 79 82 Oestrogens and androgens have been shown to influence the output ofadrenocorticotrophic hormone (ACTH) by the pituitary.83 84 The male and female pituitary contain similar numbers of ACTH immunoreactive cells.85 This suggests that there is a difference either in the secretion of corticotrophin releasing hormone (CRH) by the hypothalamus83 84 86 or in the sensitivity of the pituitary to CRH.87 88 In fact, both oestrogens and androgens have been shown to modulate the expression and sensitivity of glucocorticoid receptors in the and pituitary.89 90 Recent investi- gations support the concept that oestrogens increase the responsiveness of the HPA axis by downregulating the ofgluco- corticoids upon the hypothalamus.9' Overall, the effects of sex hormones upon the HPA axis Figure 1 Proposed mechanisms for sex hormone interactions with the immune system, seem to be mediated the including direct effects on immunocompetent cells and actions upon growth hormone, through hypothalamic prolactin, and thymus functions. Prog = progesterone; GH= growth hormone. production of CRH. The possibility that sex hormones affect not only cortisol secretion but also its catabolism has not been addressed in Changes in the physiological regulation of the literature. the glucocorticoid response to inflammation could, therefore, have very important conse- quences for both experimental and human SEX HORMONES AND GLUCOCORTICOID disease. Moreover, potential interactions of sex RESPONSE TO INFLAMMATION hormones with the glucocorticoid pathways Immune and inflammatory reactions elicit a http://ard.bmj.com/ could have far reaching implications in under- systemic reaction which includes an increase in standing the marked predominance of female glucocorticoid secretion which, in turn, tends patients in almost all autoimmune diseases. to downregulate the immune and inflam- The aim ofthe following sections is to review matory response. The mechanism by which the evidence that interaction with the HPA axis inflammation activates the HPA axis is may play an essential role in the immuno- complex, but cytokines, such as IL-1, IL-6 modulatory effects of sex hormones in vivo, and tumor necrosis factor alpha (TNFax) on September 26, 2021 by guest. Protected copyright. and to explore the potential consequences in (all detectable in serum during the acute phase the pathogenesis and treatment of human response) are thought to play a central role in disease. stimulating CRH production by the hypo- thalamus, thereby inducing glucocorticoid secretion by the adrenals.92 93 IL-I may also Sex hormones influence the HPA axis in directly stimulate corticosteroid production by experimental animals the adrenal gland94 and ACTH is also released There is extensive experimental evidence by activated lymphocytes during immune demonstrating important effects of sex hor- reactions.95 mones upon glucocorticoid levels in animals, Given that sex hormones have been shown especially rodents. Female rats and mice have to affect the HPA axis directly but also to greater basal concentrations of circulating influence immunocompetent cells and corticosterone, greater corticosteroidogenesis production, the overall effect of sex steroids on in adrenal slices ex vivo and greater circulating the glucocorticoid response to inflammation concentrations of , compared with could derive from actions in either or both males.7477 Females have a greater glucocorticoid sides of this equation. response to a variety of stresses, including ether We have recently performed a series of anaesthesia and physical restraint.78-80 investigations addressing the issue of gender These gender differences can be directly dimorphism in the HPA axis and its response attributed to sex hormones. Plasma concen- to inflammation. Using cotton induced trations of corticosterone have been shown to granulomatous reaction in mice as a model of fluctuate during the oestrus cycle of females, chronic inflammation, we have demonstrated being higher during the pro-oestrus (maximal that the glucocorticoid response to inflam- oestradiol concentration).78 79 81 Ovariectomy matory stimuli is affected by the same gender results in depressed basal and stimulated con- differences and effects of sex hormones as Sex hormones and immune response

100- All animals injected with IL-I had a marked increase in glucocorticoid levels and, again, the

response was greater in females, decreased by Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from C 80- ovariectomy in females and increased by male castration, reproducing the pattern seen in a) cn chronic inflammation. 0 Such observations show that sex hormones o440 may change the glucocorticoid response to 0 inflammation by directly affecting the HPA axis responsiveness to activating stimuli. E Changes in the production of cytokines may also contribute to these effects. Given that cytokine production by the granulomatous tissue is also affected by endogenous gluco- Figure 2 Effects ofgender and gonadectomy on corticoids, discussion of these mechanisms corticosterone concentrations in the presence of cotton risks a circular argument. However, inter- wrapped implants in BALBIc mice. Serum was collected under free conditions from sham operated and pretation of the simultaneous effects of sex gonadectomised male andfemale mice three weeks after they hormones upon glucocorticoid concentrations, received two subcutaneous implants ofcotton wrapped male cartilage degradation and production ofIL-I in femoral head cartilagesfrom male Wistar rats. Serum corticosterone was measured. Results are expressed as mean this model may help clarify the interactions. values with SEM bars; n = 26 to 28for all groups. The table summarises results from such a = Ovariectomy; 0 = orchidectomy; * =female sham; series of experiments. They actually lead to the = male sham. ++p < 0 001 vfemale sham; concept that and ***p < 0 001 v male sham (ANOVA). oestrogens androgens affect the glucocorticoid response to inflammation, not only in opposite directions, but also previously described in response to ether and through different mechanisms. other non-inflammatory conditions.80 In the Ovariectomy resulted in increased release presence of granulomatous reaction, females of IL-1a by granulomatous tissue,96 which had greater corticosterone concentrations than coincides with the observation that human males, ovariectomy resulted in a diminished ovariectomy is followed by an increased glucocorticoid response, while castrated males secretion of IL-1 and TNFao by peripheral had greater glucocorticoid concentrations than mononuclear cells, an effect reversed by intact males (fig 2). Androgen treatment hormone replacement.97 Ovariectomy also decreased glucocorticoid concentrations in resulted in decreased concentrations of gluco- both males and females, while hormone replace- corticoid. If the actions on inflammatory cells ment in castrated female mice enhanced the were the primary phenomenon, an increase in glucocorticoid response. glucocorticoids should be expected in response

To clarify if these observations were the to greater concentrations of cytokines released. http://ard.bmj.com/ result of changes in cytokine release by the These observations support the concept that inflammatory tissue, we attempted to measure oestrogens influence the inflammatory process the serum concentrations of IL-1 in these through the release of and that animals. Unfortunately, these were persistently the HPA is the primary site of oestrogen below the detection limit of our assay. How- influence in vivo. A similar mechanism may ever, granulomatous tissue from females had a underly the immunosuppressive effects of

greater content of IL-l o than those from female sex hormones in vivo. In fact, overall on September 26, 2021 by guest. Protected copyright. males, and orchidectomy was associated with effects of glucocorticoids upon the immune increased concentrations of this cytokine in the system in vivo are similar to those attributed to inflammatory tissue from males.32 Similarly, oestrogens: they also depress T cell mediated resident peritoneal from female responses but may actually increase immuno- mice were shown to produce significantly globulin production at physiological concen- greater amounts of IL- 1 when cultured ex vivo, trations.66 98 Actions of oestrogens upon the than those from males.80 This would suggest thymus in vivo are also markedly similar to that modulation of cytokine production may those of glucocorticoids.4 play a role in modulation of the glucocorticoid This hypothesis has been sparsely addressed response by sex steroids. To investigate direct in the literature. Toivanen et al28 reported that effects of sex hormones on the HPA axis the protective effects of oestrogens on adjuvant response to inflammatory mediators, sham induced arthritis were also observed in operated and castrated male and female adrenalectomised rats. However, the animals mice were injected with either saline or IL-i[ were maintained on supraphysiological con- (12-5 ,ug/kg) and serum concentrations of centrations of both corticosterone and oestrone, corticosterone were assessed two hours later. making interpretation of the results difficult. Recently the effect of oestrogen upon in vivo Summary ofthe effects ofovariectomy and orchidectomy (v sham operatedfemale and male leucocyte production and trafficking in mice mice, respectively) upon inflammatory tissue and glucocorticoid response32 80 96 103 was found to be independent of endogenous Ovariectomy Orchidectomy glucocorticoids,99 but the authors monitored only cortisol, a steroid of minor importance in Inflammation-induced cartilage degradation t t rodents. 100 Granulomatous tissue IL-1 t t In contrast, the lesser production of IL-1 by Corticosterone response to inflammation t Corticosterone to IL-1 granulomatous tissue and peritoneal macro- response t phages in males suggests that androgens may 10 Da Silva

act primarily at the inflammatory site, reducing Immune/inflammatory [- the production of inflammatory mediators that process I reach and activate the hypothalamus. This is supported by the observations of an increased Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from production of IL-I by granulomatous tissue in 11en castrated males, coinciding with increased ,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ circulating concentrations of glucocorticoid. If Cytokines the actions of androgens on the HPA axis were IL-1; IL-6, TNF, ...II the primary phenomenon, a reduction in IL-I I production and cartilage damage would be expected in response to higher levels of corti- + costerone in castrated males. Hypothalamus It is not clear how androgens bring about this effect in inflammatory tissue. In our experi- ence, orchidectomy does not significantly Pituitary influence the ex vivo release ofIL-1 by resident ACTH mouse peritoneal macrophages.80 However, macrophages in an inflammatory environment +1 may show a different response, and the com- plex interactions that surround these cells in Adrenal inflammatory tissue provide a large number of Glucocorticoids pathways for indirect modulation, for example Li through T cells and lymphokine production." The possibility of direct action by androgens Figure 3 Proposed mechanismsfor differential interaction on these cells is supported by the recent ofandrogens and oestrogens with the immune and description of androgen receptors in macro- inflammatory response. phage like cells in normal and rheumatoid synovia'°l and in cotton induced granulo- To investigate the hypothesis further, we matous tissue.32 This would suggest a direct assessed the effects of adrenalectomy on anti-inflammatory role for androgens, supported granuloma induced cartilage degradation in by the previous findings of reduced inflam- male and female mice (fig 4). Adrenal- mation and cartilage erosion after intralesional ectomised females showed accelerated administration of testosterone in models of air cartilage destruction compared with intact pouch and antigen induced arthritis in mice.'02 females. The effects of adrenalectomy were Androgens have also been shown to ameliorate similar to those of ovariectomy. In practice, the RA in male patients, an effect associated with ovariectomy works as a partial adrenalectomy, a significant increase in circulating CD8 because of the loss of the priming actions of

lymphocytes.34 oestradiol on the HPA axis. However, http://ard.bmj.com/ Overall, both female and male sex hormones adrenalectomy failed to induce any significant have the ability to reduce immune responses change in cartilage destruction in males, and to protect cartilage from inflammation whereas orchidectomy consistently resulted in induced degradation. However, while oestrogens more severe cartilage damage. These obser- increase the glucocorticoid response to a vations suggest that endogenous gluco- variety of inflammatory and non-inflammatory corticoids are more important in controlling have the opposite effect- inflammation in female than in male animals. stimuli, androgens on September 26, 2021 by guest. Protected copyright. they depress glucocorticoid levels. Taken Androgens, however, seem capable of reducing together, these observations support the inflammation to a greater extent than endogen- concept that the immunosuppressive effects of ous glucocorticoids, given that orchidectomy oestrogen in vivo may be mediated, at least in accelerates cartilage destruction and increases part, by their ability to enhance the gluco- corticoid response. In contrast, similar 80 - ** immunosuppressive actions of androgens are Ci I probably independent from glucocorticoids. 0

Figure 3 presents a simplified diagram of the 0 pathways for interaction between androgens L- and oestrogens with the immune and gluco- 0 corticoid response proposed here. CO cn 0 POSSIBLE GENDER DIFFERENCES IN THE 0 SENSITIVITY TO GLUCOCORTICOIDS This becomes pertinent in face of the evidence that females, despite having greater levels of Female Male glucocorticoid, show a stronger immuno- Figure 4 Effects ofadrenalectomy on inflammation induced cartilage degradation in male andfemale mice. reactivity, produce greater amounts of IL-1 BALBIc mice received a subcutaneous implant ofcotton and show a greater ability to degrade articular wrappedfemoral head cartilages immediately after cartilage by juxtaposed inflammatory tissue.32 adrenalectomy ( J) or sham operation (O). Implants were recoveredfor analysis three weeks later and This raises the possibility that glucocorticoids glycosaminoglycan (GAG) lossfrom cartilage calculated by are more effective in male than in female comparison withfrozen controls. +p < 0 01 vfemale sham animals. (ANO VA). Sex hormones and immune response I1I

IL-I production by granulomatous tissue, would then act on a hypothalamus with despite resulting in circulating concentrations diminished negative feedback. However, the

of glucocorticoids that are similar to those fact that ovariectomy resulted in increased Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from observed in intact females.80 We also found that production of IL-1 by the granulomatous androgen treatment of castrated females tissue, but decreased corticoid secretion, and males resulted in significant cartilage suggests that any potential peripheral protection and reduction of tissue IL-1, antagonism is of less importance than the despite reduced levels of endogenous gluco- hypothalamic changes. corticoids.96 103 Moreover, the facts that male Evidence for direct interactions between sex animals show a weaker immune response hormones and glucocorticoids outside the despite having smaller concentrations of corti- hypothalamus is scarce. Additive or potentiating costeroids, and that orchidectomy results in effects of androgens and corticosteroids have enhanced immune response despite increasing been demonstrated using human breast glucocorticoid levels, suggest that endogenous cell lines,104 rat ovary granulosa cellsl05 and androgens are more effective than endogenous salivary glands of the rat,'06 but no reports corticosteroids in the control of the immune regarding immunocompetent cells could be response in the male animal. found. Testosterone does not interact with Overall females seem to be more dependent glucocorticoid receptors.'07 108 However, the on glucocorticoids than males to control glucocorticoid, androgen, and progesterone inflammatory and immune responses. receptors share a high degree of homology in How could this be explained? It is humans,'09 translated into the competitive possible that direct immunosuppressive and binding of molecules such as the , anti-inflammatory effects of androgens in ."0 It is conceivable, therefore, that males result in a lesser degree of inflam- androgenic steroids, other than testosterone, mation and, therefore, a reduced need for could activate glucocorticoid receptors. The endogenous glucocorticoid actions. Alterna- potency ofglucocorticoid effects in the absence tively, it could be suggested that glucocorticoid of interacting substances depends on the effects are either antagonised in females number of receptors in the target cell. 1"' or potentiated in males, probably by sex Sex related differences in the expression hormones. of glucocorticoid receptors have been reported One hypothesis encompassing these different in rats. Females have fewer receptors in observations is that androgens may potentiate the pituitary,89 , and thymus"'2an glucocorticoid effects both at the hypo- effect dependent on oestrogens. Androgens, thalamus and at the inflammatory site. Such including testosterone ciprionate, have been actions upon the hypothalamus would justify shown to increase the reduction of corticoid secretion in all of the expression in the hipoccampus of adrenal- three systems used (granuloma, IL-1, and ectomised rats."3 Dihydrotestosterone has

ether) by potentiation of the negative feedback been reported to modulate the expression of http://ard.bmj.com/ of glucocorticoids. Potential synergy at the glucocorticoid receptors in the thymus and inflammatory site could account for the pro- bursa of immature chickens."'4 Post-receptor tection of cartilage and decreased production mechanisms could also be involved. Gluco- of IL-1. The findings that adrenalectomy did corticoids and androgens share the same not result in accelerated cartilage damage'in hormone response element in sensitive males would argue against this possibility, as ,"15 suggesting that these hormones

the adrenals are the only recognised source of may have synergising effects upon gene on September 26, 2021 by guest. Protected copyright. glucocorticoids. However, prolonged experi- expression. An example of such inter- ments with adrenalectomy are always associated actions is given by investigations of the with a high mortality rate. The levels of corti- hormone regulation of costerone are not routinely assessed to verify acetyltransferase gene expression."'6 The the efficiency of the operation. This raises the possible interactions between sex steroids possibility that the animals that survive (those and glucocorticoids on inflammatory and for which results are included in the final immune mechanisms seem worthy of further analysis) are actually the ones with incomplete investigation. adrenalectomy resulting in diminished, but not absent, circulating glucocorticoids. Ifthis is the case, it is conceivable that the potentiation of Potential implications of gender HPA the remaining corticoids could be enough to dimorphism in human inflammatory and afford cartilage protection in adrenalectomised autoimmune diseases males with intact gonads. This possibility is not To discuss the potential importance of the sex taken into account in most published reports hormone effects upon the HPA axis in human on the effects of adrenalectomy upon different immune and inflammatory diseases, two main experimental systems. questions will be addressed: is there any A potential antagonism between oestradiol evidence for the involvement of HPA axis and glucocorticoids is also compatible with some dysfunction in the pathogenesis of these of the results revised above. Such antagonism conditions, and is this gender dimorphism in is demonstrated at the hypothalamic level by glucocorticoid response also observed in the reduction of the glucocorticoid negative humans? This discussion will be centred feedback induced by oestrogens.9' A similar around models of experimental arthritis and effect at the inflammatory site could result human rheumatoid disease, which have been in increased production of mediators, which more extensively investigated. 12 Da Silva

DISREGULATION OF ENDOGENOUS that rheumatoid patients have a peripheral GLUCOCORTICOIDS AND AUTOIMMUNE DISEASE defect that could lead to glucocorticoid A role of HPA axis dysfunction has but the that these pathogenic resistance, possibility Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from been clearly demonstrated in a large number of abnormalities are a consequence of the disease experimental models of autoimmune diseases. cannot be excluded. The Lewis LEW/N rat has a remarkable Information regarding HPA axis function in susceptibility to a large variety of experimental RA is limited. Myles et al'32 found normal basal autoimmune diseases, in contrast with other and peak cortisol concentrations in response to histocompatible strains. After repeated injections insulin induced hypoglycaemia in patients of streptococcal cell walls, the female Lewis rat with RA. However, the activation of the HPA develops an arthritis characterised by a rapid axis by hypoglycaemia is probably mediated by onset acute phase followed by a chronic mechanisms that differ from those involved in proliferative and destructive course which inflammation. The observations with the Lewis mimics human RA."17 In contrast, the histo- rat led to a surge of interest in the compatible Fisher rat (F344/N) develops only of the HPA axis in humans with RA. Cash a mild and transient acute phase without et all33 reported signs of mildly deficient chronic inflammation. A series of studies adrenocortical response after injection of CRH have demonstrated that these differences are in RA patients, but the hypothalamic function attributable to an innate inability of the was not tested. A clearly defective gluco- LEW/N rat to generate an appropriate gluco- corticoid response to inflammatory stimuli in corticoid response to the immune and RA has been reported recently.'34 Patients inflammatory challenge,"18 resulting from a with the disease showed a diurnal cortisol deficient production of corticotrophin releasing secretion at the lower limit of normal com- hormone by the hypothalamus."'9 The deficient pared with both normal controls and osteo- glucocorticoid response was demonstrated in myelitis patients, with a comparable degree of response to different activating stimuli, inflammation. Furthermore, patients with RA including IL-1."8 In contrast, the Fisher rat failed to increase cortisol concentrations after shows a strong glucocorticoid surge after replacement surgery, although they immune challenge but develops severe experi- presented greater concentrations of IL-1 and mental arthritis if treated with the steroid IL-6 than osteomyelitis patients. However, the antagonist RU 486.' 18 Conversely, treatment response to CRH was normal in RA, of the LEW/N rat with physiological doses of suggesting a defect in the hypothalamic pro- glucocorticoids resulted in attenuated arthritis, duction of this hormone, with normal pituitary with a pattern similar to that of the untreated and adrenal function. Such observations Fisher rat. Similar findings have been reported strongly suggest that RA is associated with an regarding other autoimmune diseases and abnormality in the HPA axis response to different animal species.'20 121 inflammatory stimuli which does not result

The first suggestion that similar mechanisms from the inflammatory process per se, as osteo- http://ard.bmj.com/ could be operating in human disease was given myelitis patients with a similar disease duration by the finding of an increased incidence of and acute phase response showed appropriate arthritis in females with panhypopituitarism.'22 cortisol increases. These abnormalities may be Rheumatoid arthritis has been associated an important and hitherto unsuspected factor with abnormalities in cortisol secretion and in the pathogenesis of rheumatoid disease. It is function. Most published studies of cortisol in tempting to suggest that they are genetically serum concen- RA patients have reported determined, as in the Lewis rat-possibly on September 26, 2021 by guest. Protected copyright. trations within the normal range,'23-125 with few in parallel with rheumatoid predisposing showing either low'26 or high'27 mean serum haplotypes. values. However, work by Neeck et al'28 demonstrated that the normal cortisol concen- tration in RA was inappropriate for the degree SEX HORMONE EFFECTS ON THE HPA AXIS IN of ongoing inflammation and that the normal HUMANS AND THEIR POTENTIAL SIGNIFICANCE circadian rhythm was markedly reduced in The demonstration of gender dimorphism in patients with an increased erythrocyte sedi- the HPA axis response to inflammatory stimuli mentation rate. Morand et al'29 reported that in humans would represent a major, but as yet mononuclear cells from RA patients produce unrecognised, factor for the greater female smaller amounts of lipocortin-1 (an important susceptibility to autoimmune and inflam- mediator of glucocorticoid effects) in response matory diseases. Although serum cortisol to cortisol administration than do mono- concentrations are usually described as similar nuclear cells from healthy controls. RA in normal males and females, Schoneshofer patients receiving oral glucocorticoids were and Wagner135 observed that women in the also shown to have antilipocortin antibodies, follicular phase of the had resulting in resistance to the therapeutic lower serum corticosteroids than those effects of intravenous .130 observed in the luteal phase, or in age matched Peripheral lymphocytes from RA patients were males. Other indications that oestrogens found to have significantly smaller numbers stimulate glucocorticoid secretion in humans of glucocorticoid receptors than those from comes from the increased concentrations of controls.'25 Abnormalities of cortisol catabolism cortisol seen in the late stages of pregnancy136 have also been identified in lymphocytes and in patients taking oral contraceptives.'37 138 from patients with RA and systemic lupus More clear evidence is given by the findings erythematosus (SLE).'3' These findings suggest that high dose oestrogen treatment given to Sex hormones and immune response 13

young women resulted in a marked increase humans in a pattern similar to that observed in in cortisol concentrations,'30 while post- experimental animals. The hypothesis that

menopausal women were reported to have male and female humans have a different Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from reduced concentrations of cortisol, which glucocorticoid profile in response to immune are increased by hormone replacement.'40 141 and inflammatory stimuli remains untested, Androgens have also been shown to decrease but could have important implications in glucocorticoid secretion in humans. In males understanding the preponderance offemales in with prostate carcinoma, both castration and most autoimmune disease populations. oestradiol treatment resulted in increased There is now good evidence to suggest that circulating cortisol.'42 143 Testosterone treat- female animals are more dependent on the ment also resulted in decreased cortisol con- glucocorticoid response than males to control centrations in ovariectomised humans. 144 the immune and inflammatory processes, These observations are in agreement with the probably because of protective effects of experimental findings reviewed above. androgens in males. If this is shown to be true The cortisol response to ACTH and CRH in humans, a defect in glucocorticoid response are described as similar in both sexes.'45 How- (as reported in RA), perhaps genetically deter- ever, gender differences and effects of sex mined, would be expected to contribute to a steroids in the hypothalamus, as suggested by greater susceptibility to immune and inflam- experimental evidence, would not be detected matory processes in females. In contrast, by these tests. The responses to ACTH and gonadal insufficiency would be expected to CRH in RA patients reported by Chikanza play a more important role in males, in keeping et al '14 were not analysed separately for each with the observations ofhypogonadism in male sex. No reports were found regarding cortisol RA and SLE patients and greater incidence of response to CRH in male and female animals the disease with ageing and declining con- and, although the male LEW/N rat is relatively centrations of androgens. Such interactions resistant to streptococcal cell wall arthritis, would also have important implications for its glucocorticoid response to the immune the therapeutic approach to autoimmune and challenge has not been tested. The only inflammatory diseases. standard investigation of HPA axis function in humans that includes the hypothalamus is the 1 Ansar Ahmed S, Penhale W J, Talal N. Sex hormones, immune responses, and autoimmune diseases. Mech- hypoglycaemia test. Responses to this test are anisms of sex hormones action. Amn J Pathol 1985; 121: also reportedly similar between the genders. '45 531-51. 2 Felson D T. Epidemiology of the rheumatic diseases. However, hypothalamic secretion of CRH can In: McCarty D J, Koopman W J, eds. Arthritis and be activated by a variety of different mech- allied conditions. A textbook of , 12th edn. Philadelphia: Lea & Febiger, 1993; 17-48. anisms which will depend on the nature of the 3 Ansar Ahmed S, Talal N. Sex hormones and the immune stimulus involved.'46 Gender differences may system-part 2. Animal data. Bailliere's Clin Rheumatol 1990;4:13-31. exist in response to inflammatory stimuli 4 Grossman C. Possible underlying mechanisms of sexual

activating the HPA axis by a different pathway, dimorphism in the immune response, fact and hypothesis. http://ard.bmj.com/ J Steroid Biochem 1989; 34: 241-5 1. as suggested by findings that patients with an 5 Schuurs A H W M, Verheul H A M. Effects of gender and impaired corticosteroid response to hypo- sex steroids on the immune response. J Steroid Biochem 1990; 35: 157-72. glycaemia may have a normal response to 6 Butterworth M B, McClellan B, Alansmith M. Influence surgery or infection.141 of sex on immunoglobulin levels. Nature 1967; 214: 1224-5. Bliss et al'48 proposed the assessment of the 7 Brick J E, Wilson D A, Walker S E. Hormonal modulation glucocorticoid response to 'bacterial pyrogens' of responses to thymic independent and thymic- dependent antigens in autoimmune NZB/W mice. as a method to investigate the HPA axis in J'Immunol 1985; 136: 3693-8. on September 26, 2021 by guest. Protected copyright. humans and it is almost certain that the 8 Graff R J, Lappe M A, Snell G D. The influence of the gonads and adrenal glands on the immune response to response to this pyrogen is mediated by IL-1 skin grafts. Transplantation 1969; 7: 105-11. and other cytokines, offering an opportunity to 9 Dresser D W. Specific inhibition of production. I. over-loading paralysis. 1962; 5: 161-8. study the HPA axis activation by inflammatory 10 Blazkovec A A, Orsini M W. Ontogenetic aspects of sexual mediators. Serum glucocorticoids achieved dimorphism and the primary immune response to sheep erythrocytes in hamster from prepuberty through maximal values three hours after the . Int Arch Appl Imlmunol 1976; 50: injection-a timing compatible with cytokine 55-67. 11 Castro J E. Orchidectomy and the immune response. II. mediation. None ofthe published reports using Response of orchidectomized mice to antigens. Proc R Soc this method analysed the response of normal Biol (Lond) 1974; 185: 437-51. 12 Stimson W H, Hunter I C. An investigation into the males and females separately, and only a few immunosuppressive properties of . J Endocrinol offer results in normal controls. However, in 1976; 69: 42-3P. 13 Paavonen T. Hormonal regulation of function. the report by Staub et al'49 the corticosteroid Med Biol 1987; 65: 229-40. response to pyrogen was significantly greater in 14 Ablin R J, Bhatti R A, Guinan P D, Khin W. Modulating effects of estrogen on immunological responsiveness. II. two normal females than in the five males Suppression of tumour-associated immunity in patients tested (means 387 v 261 ,ug/l two hours after with prostatic carcinoma. Clin Exp Immunol 1979; 38: 83-91. injection; p < 0.05 by Mann-Whitney U test), 15 Kincl F A, Ciaccio L A. Suppression of immune response while there were no differences in the response by progesterone. Endocrinologia Experimentalis 1980; 14: 27-32. to vasopressin or insulin induced hypo- 16 Jansson L, Holmdahl R. The Y chromosome-linked glycaemia. However, Jenkins and Elkington,'50 'autoimmune accelerating' yaa gene suppresses collagen- induced arthritis. EurJnImmunol 1994; 24: 1213-7. whose study included seven male and seven 17 Roubinian J R, Papoian R, Talal N. Androgenic hormones female normal controls showed no gender modulate autoantibody responses and improve survival in murine lupus.7Clin Invest 1977; 59: 1066-70. differences in the response to pyrogen or 18 Ansar Ahmed S, Talal N. Sex steroids, sex steroid . receptors and autoimmune diseases. In: Sheridan P S, et al, eds. Steroid receptors and disease: cancer, autoimmnune, Overall, there is strong evidence that sex and circulatory disorders. New York: Marcel Dekker, hormones affect glucocorticoid secretion in 1988; 289-316. 14 Da Silva

19 Jungers P, Dougados M, Pelissier C et al. Influence of oral 48 Stimson W H. Estrogen and human T lymphocytes: contraceptive therapy on activity of systemic lupus presence of specific receptors in the T-suppressor/ erythematosus. Arthritis Rheum 1982; 25: 618-23. cytotoxic subset. Scand7Immunol 1988; 28: 345-50. 20 Angello V, Parisier J, Gell J, Gelfand J, Turksoy R N. 49 Glover J F, Irwin J T, Darbre P D. Interaction of phenol Preliminary observations on danazol therapy of systemic red with estrogenic and antiestrogenic action on growth Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from lupus erythematosus. J Rheumatol 1983; 10: 682-7. of human breast cancer cells ZR-75-1 and T-47-D. 21 Ansar Ahmed S, Penhale W J. The influence of testos- Cancer Res 1988; 48: 3693-7. terone on the development of autoimmune thyroiditis in 50 Sobhon P, Jirasattham C. Effects of sex hormones on the thymectomized and irradiated rats. Clin Exp Immunol thymus and lymphoid tissue of ovariectomized rats. Acta 1982; 48: 367-74. Anatomica 1974; 89: 211-25. 22 Okayasu I, Kong Y M, Rose N R. Effect of castration and 51 Allen L S, McClure J E, Goldstein A L, Barkley M S, sex hormones on experimental autoimmune thyroiditis. Michael S D. Estrogen and thymic hormone inter- Clin Immunol Immunopathol 1981; 20: 240-5. actions in the female mouse. J Reprod Immunol 1984; 61: 23 Steinberg A D, Smathers P A, Boegel W B. Effects of sex 25-37. hormones on autoantibody production by NZB mice and 52 Hall R N, Goldstein A L. Endocrine regulation of host modification by environmental factors. Clin Immunol immunity. In: Fenichel R L, Chirigos M A, eds. Immuno- Immunopathol 1980; 17: 562-72. modulation agents and their mechanisms. New York: Marcel 24 Milch D R, Gershwin M E. Murine autoimmune hemolytic Dekker, 1984: 533-63. anemia via xenogeneic erythrocyte immunization. III. 53 Screpanti I, Morrone S, Meco D, et al. Steroid sensitivity Differences of sex. Clin Immunol Immunopathol 1981; 18: of thymocyte subpopulations during intrathymic 1-11. differentiation. Effects of 1713 and dexa- 25 Ariga H, Edwards J, Sullivan D A. Androgen control of on subsets expressing T cell antigen receptor autoimmune expression in lacrimal gland of MRL-MP/ or IL-2 receptor. Immunol 1989; 142: 3378-83. lpr-lpr mice. Clin Immunol Immunopathol 1989; 53: 54 Stimson W H, Hunter I C. Estrogen-induced immuno- 499-508. regulation mediated through the thymus. J Clin Lab 26 Allen J B, Blatter B, Calandra G B, Wilder R L. Sex Immunol 1980; 4: 27-33. hormonal effects on the severity ofstreptococcal cell wall- 55 Nagy E, Berczi I, Friesen H G. Regulation of immunity induced polyarthritis in the rat. Arthritis Rheum 1983; 26: in rats by lactogenic and growth hormones. Acta 560-3. Endocrinol (Copenh) 1983; 102: 351-7. 27 Larson P, Holmdahl R. Estrogen-induced suppression of 56 Pandian M R, Talwar G P. Effect of growth hormone on collagen arthritis. II. Treatment of rats suppresses the of thymus and on the immune response development of arthritis but does not affect the anti- against sheep erythrocytes. JT Exp Med 1971; 134: type II collagen humoral response. ScandJrImmunol 1987; 1095-113. 27: 579-83. 57 Wilder R L, Stemnberg E M. Neuroendocrine hormonal 28 Toivanen P, Sukala H, Laiho P, Paavilaines T. Suppression factors in rheumatoid arthritis and related conditions. of adjuvant arthritis by in adrenalectomized and Curr Opin Rheumatol 1990; 2: 436-40. ovariectomized rats. Experientia 1967; 23: 560-1. 58 Jansson J-O, Ekberg S, Isaksson 0 G P, Eden S. Influence 29 Arnason B G, Richman D P. Effect of oral contraceptives of gonadal steroids on age- and sex-related secretory on experimental demyelinating disease. Arch Neurol 1969; patterns of growth hormone in the rat. 21: 103-6. 1984; 114: 1287-94. 30 Willoughby D A, Ryan G B. Evidence for a possible 59 Plotnick L P, Thompson R G, Beitins I, Blizzard R M. endogenous antigen in chronic inflammation. J Pathol Integrated concentrations of growth hormone correlated 1970; 101: 233-9. with stage of puberty and estrogen levels in girls. 7 Clin 31 De Brito F B, Moore A R, Holmes M J G, EndocrinolMetab 1974; 38: 436-9. Willoughby D A. Cartilage damage by a granulomatous 60 Fukata J, Martin J B. Influence of sex steroid hormones reaction in a murine species. Br J Exp Path 1987; 68: on rat growth hormone-releasing factor and somatostatin 675-86. in dispersed pituitary cells. Endocrinology 1986; 119: 32 Da Silva J A P, Larbre J-P, Seed M P, et al. Gender 2256-61. differences in inflammation-induced cartilage degradation 61 Berczi I, Nagy E, Kovacs K, Horvath E. Regulation of in rodents. The influence of sex steroids. J Rheumatol in rats by pituitary hormones. Acta 1994; 21: 330-7. Endocrinol (Copenh) 1981; 98: 506-13. 33 Hall G M, Daniels M, Huskisson E C, Spector T D. A 62 Shull J, Gorski J. Estrogen regulation of prolactin gene randomized controlled trial of the effect of hormone in vivo; paradoxical effect of 17,3-estradiol replacement therapy on disease activity in post- dose. Endocrinology 1989; 124: 279-85. menopausal rheumatoid arthritis. Ann Rheum Dis 1994; 63 Giguere V, Meunier H, Veilleux R, Labrie F. Direct effect 53: 112-6. of sex steroids on prolactin release at the

34 Cutolo M, Balleari E, Giusti M, Intra E, Accardo S. level; interactions with dopamine, thyrotropin-releasing http://ard.bmj.com/ Androgen replacement therapy in male patients with hormone and siobutyl-methyl-. Endocrinology rheumatoid arthritis. Arthritis Rheum 1991; 34: 1-5. 1982; 111: 857-62. 35 Grossman C J, Roselle G A, Mendenhall C L. Sex steroid 64 Golsteyn E J, Fritzler M J. Review: the role of the thymus- regulation of autoimmunity. J Steroid Biochem Mol Biol hypothalamus-pituitary-gonadal axis in normal immune 1991;40: 649-59. processes and autoimmunity. .7 Rheumatol 1987; 14: 36 Paavonen T, Andersson L C, Adlercreutz H. Sex hormone 982-90. regulation of in vitro immune response. Estradiol 65 Bhalla A K. Hormones and the immune response. Ann enhances human B-cell maturation via inhibition of Rheum Dis 1989; 48: 1-6. suppressor T cells in pokeweed mitogen-stimulated 66 Cupps T R, Fauci A S. Corticosteroid-mediated immuno- cultures.J Exp Med 1981; 154: 1935-45. regulation in man. Immun Rev 1982; 65: 133-55. 37 Wyle F A, Kent J R. by sex steroid 67 Steinberg E M, Wilder R L. Corticosteroids. In: on September 26, 2021 by guest. Protected copyright. hormones. I. The effect upon PHA- and PPD-stimulated McCarty D J, Koopman W J, eds. Arthritis and allied lymphocytes. Clin Exp Immunol 1977; 27: 407-15. conditions. A textbook of rheumatology, 12th edn. 38 Hou J, Zheng W F. Effect of sex hormones on NK and Philadelphia: Lea & Febiger, 1993: 665-82. ADCC activity of mice. IntrjImmunopharmacol 1988; 10: 68 Saldanha C, Touzas G, Grace E. Evidence for the anti- 15-22. inflammatory effects of normal circulating plasma 39 Holdstock G, Chastenay B F, Krawitt E L. Effect of cortisol. Clin Exp Rheumatol 1986; 4: 365-6. testosterone, oestradiol and progesterone on immune 69 Kirkham B W, Panayi G S. Diurnal periodicity of cortisol regulation. Clin Exp Immunol 1982; 47: 449-56. secretion, immune reactivity and disease activity in 40 Polan M L, Daniele A, Kuo A. Gonadal steroids modulate rheumatoid arthritis: implications for steroid treatment. human interleukin- 1 (IL-1) activity. Fertil Steril BrJRheumatol 1989; 28: 154-7. 1988; 49: 964-8. 70 Takasu N, Komiya I, Nagasawa Y, Asawa T, Yamada T. 41 Hu S-K, Mitcho Y L, Rath N C. Effect of estradiol on Exacerbation of autoimmune thyroid dysfunction after interleukin I synthesis by macrophages. Int J Immuno- unilateral adrenalectomy in patients with Cushing's pharmacol 1988; 10: 247-52. syndrome due to an adrenocortical adenoma. N Engl. 42 Gueme P-A, Carson D A, Lotz M. IL-6 production by Med 1990; 322: 1708-12. human articular chondrocytes. Modulation of its 71 Macphee I A M, Antoni F A, Mason D W. Spontaneous synthesis by cytokines, growth factors, and hormones in recovery of rats from experimental allergic encephalo- vitro.JImmunol 1990; 144: 499-505. myelitis is dependent on regulation of the immune system 43 Kovacs W J, Olsen N J. Androgen receptors in human by endogenous adrenal corticosteroids. .7 Exp Med 1989; thymocytes._ Immunol 1987; 139: 490-3. 169: 431-45. 44 Araneo B A, Dowell T, Diegel M, Daynes R A. 72 De Brito F B, Holmes M J G, Carney S L, Dihydrotestosterone exerts a depressive influence on the Willoughby D A. effects on a novel model of production of interleukin-4 (IL-4), IL-5 and gamma- connective tissue breakdown. Agents Actions 1987; 21: interferon but not IL-2 by activated murine cells. Blood 287-9. 1991; 78: 688-99. 73 Bottomley K M K, Griffiths R J, Rising T J, Steward A. 45 Komm B S, Terpening C M, Benz D J, et al. Estrogen A modified air pouch model for evaluating the effects of binding, receptor mRNA, and biologic response in compounds on granuloma induced cartilage degradation. oestoblast-like osteosarcoma cells. Science 1988; 241: 81-4. BrJ'Pharmacol 1988; 93: 627-35. 46 Cutolo M, Accardo S, Villaggio B, et al. Presence of 74 Kitay J I. Pituitary-adrenal function in the rat after estrogen-binding sites on macrophage-like synoviocytes gonadectomy and gonadal hormone replacement. and CD8+, CD29+, CD45RO+ lymphocytes in normal Endocrinology 1963; 73: 253-60. and rheumatoid synovium. Arthritis Rheum 1993; 36: 75 Gala R, Westphal U. Further studies on the corticosteroid- 1087-97. binding globulin in the rat: proposed endocrine control. 47 Cohen J H M, Danel L, Cordier G, et al. Sex steroid Endocrinology 1966; 79: 67-76. receptors in peripheral T cells: absence of androgen 76 Allen-Rowlands C F, Allen J P, Greer M A, Wilson M. receptors and restriction of estrogen receptors to OKT8- Circadian rhythmicity ofACTH and corticosterone in the positive cells. J Immunol 1983; 131: 2767-7 1. rat. I Endocrinol Invest 1980; 4: 371-7. Sex hormones and immune response 15

77 Lesniewska B, Miskowa B, Nowak M, Malendowicz L K. 104 Simard J, de Launoit Y, Haagensen D A, Labrie F. Additive Sex differences in adrenocortical structure and function. stimulatory action of glucocorticoids and androgens on XXVII. The effect of ether stress on ACTH and costi- basal and estrogen-repressed apolipoprotein-D messenger costerone in intact, gonadectomized, and testosterone- ribonucleic acid levels and secretion in human breast or estradiol-replaced rats. Res Exp Med 1990; 190: cancer cells. Endocrinology 1992; 130: 1115-21. Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from 95-103. 105 Jia X C, Ny T, Hsueh A J. Synergistic effects of 78 Baron S, Brush F R. Effects of acute and chronic restraint glucocorticoids and androgens on the hormonal induction and estrus cycle on pituitary adrenal function in the rat. of tissue plasminogen activator activity and messenger Horm Behav 1979; 12: 218-24. ribonucleic acid levels in granulosa cells. Mol Cell 79 Viau V, Meaney M J. Variations in the hypothalamic- Endocrinol 1990; 68: 143-51. pituitary-adrenal response to stress during the estrous 106 Antakly T, Zhang C X, Sarrieau A, Raquidan D. Cell- cycle in the rat. Endocrinology 1991; 129: 2503-11. specific expression of the glucocorticoid receptor within 80 Da Silva J A P, Peers S H, Perretti M, Willoughby D A. granular convoluted tubules of the rat submaxillary gland. Sex steroids affect glucocorticoid response to chronic Endocrinology 1991; 128: 617-22. inflammation and to interleukin-1. J7 Endocrinol 1993; 107 Cidlowski J A, Cidlowski N B. Regulation of gluco- 136: 389-97. corticoid receptors by glucocorticoids in cultured 81 Nichols D J, Chevins P F D. Plasma corticosterone HeLa S3 cells. Endocrinology 1981; 109: 1975-82. fluctuations during the oestrus cycle of the house mouse. 108 DiBattista J A, Martel-Pelletier J, Wosu L, Sandor T, Experientia 1981; 37: 319-20. Antakly T, Pelletier J-P. Glucocorticoid receptor 82 Kitay J I. Effects of estrogen and androgen on the adrenal mediated inhibition of interleukin-1 stimulated neutral cortex of the rat. In: Kemns K W, ed. Functions of the metalloprotease synthesis in normal human chondrocytes. , vol 2. New York: Appleton-Century-Crofts, J3 Rheumatol 1991; 72: 316-26. 1968: 775-811. 109 Brinkmann A 0, Faber P W, van Rooij H C J, et al. The 83 LeMevel J C, Abitbol S, Beraud G, Maniey J. Temporal human androgen receptor: domain structure, genomic changes in plasma adrenocorticotropin concentration organization and regulation of expression. J Steroid after repeated neurotropic stress in male and female mice. Biochem 1989; 34: 307-10. Endocrinology 1979; 105: 812-7. 110 Gaillard-Moguilewsky M. of 84 Buckingham J C. Effect of adrenocortical and gonadal and value of combining androgen suppression with anti- steroids on the secretion in vitro of corticotrophin and its androgen therapy. Urology 1991; 37 (suppl 2): 5-12. hypothalamic releasing factor. Endocrinol 1982; 93: 111 Catt K J, Harwood J P, Agullera J, Dufau M L. Hormonal 123-32. regulation of peptide receptors and target cell responses. 85 Malendowicz L K, Stachowiak A, Zabel M. Sex differences Nature 1979; 280: 109-11. in adrenocortical structure and function. XXV. 112 Endres D B, Milholland R J, Rosen F. Sex differences in Quantitative analysis of ACTH-immunoreactive cells in the concentrations of glucocorticoid receptors in rat liver the anterior pituitary of gonadectomized and gonadal and thymus.J Endocrinol 1979; 80: 21-6. hormone replaced male and female rats. Exp Clin 113 Ahima R S, Harlan R E. Regulation of glucocorticoid Endocrinol 1987; 90: 1-8. immunoreactivity in the rat hipoccampus by androgenic- 86 Haas D A, George S R. Estradiol and ovariectomy anabolic steroids. Brain Res 1992; 585: 311-4. decreases CRF synthesis in hypothalamus. Brain Res Bull 114 Coulson P, Skafar D, Seaver S, Thornthwaite J. 1989; 23: 215-8. Dihydrotestosterone modulation of glucocorticoid 87 Coyne M D, Kitay J I. Effect of orchiectomy on pituitary receptor levels in thymus and Bursa de Fabricius cells in secretion of ACTH. Endocrinology 1971; 89: 1024-8. immature chickens. In: Proceedings of the Endocrine Society 88 Miskowiak B, Lesniewska B, Nowak M, Malendowicz L K. 63rdAnnual Meeting, 1981: 255. Studies on the hypothalamo-pituitary corticoliberin 115 Beato M, Chalepakis G, Schauer M, Slater E P. DNA system. V. The effects of gonadectomy and sex hormones regulatory elements for steroid hormones. Steroid on plasma ACTH and on the reactivity of the anterior Biochem 1989; 32: 737-47. pituitary gland to CRF. Exp Clin Endocrinol 1988; 92: 116 Adler A J, Scheller A, Hoffman Y, Robins D M. Multiple 1-6. components of a complex androgen-dependent enhancer. 89 Turner B B. Sex difference in glucocorticoid binding in rat Mol Endocrinol 1991; 5: 1587-96. pituitary is estrogen dependent. Life Sci 1990; 46: 117 Wilder R L, Calandra G B, Garvin A J, Wright K D, 1399-406. Hansen C T. Strain and sex variation in the susceptibility 90 Ahima R S, Lawson A N L, Osei S Y S, Harlan R E. Sexual to streptococcal cell wall-induced polyarthritis in the rat. dimorphism in regulation of Type II corticosteroid Arthritis Rheum 1982; 25: 1064-72. receptor immunoreactivity in the rat hippocampus. 118 Stemnberg E M, Hill J M, Chrousos G, et al. Inflammatory Endocrinology 1992; 131: 1409-16. mediator-induced hypothalamic-pituitary-adrenal axis 91 Burgess L H, Handa R J. Chronic estrogen-induced activation is defective in stretococcal cell wall arthritis- 86: alterations in adrenocorticotropin and corticosterone susceptible Lewis rats. Proc Natl Acad Sci USA 1989; http://ard.bmj.com/ secretion, and glucocorticoid receptor-mediated func- 2374-8. tions in female rats. Endocrinology 1991; 131: 1261-9. 119 Stemnberg E M, Scott Young W III, Bernardini R, et al. 92 Besedowsky H 0, Del Rey A, Klusman I, Furukawa H, A central defect in biosynthesis of Monge Arditi G, Kabiersch A. Cytokines as modulators corticotropin-releasing hormone is associated with of the hypothalamo-pituitary-adrenal axis. Steroid susceptibility to streptococcal cell wall-induced arthritis in Biochem Mol Biol 1991; 40: 613-8. Lewis rats. Proc Natl Acad Sci USA 1989; 86: 4771-5. 93 Imura H, Fukata J, Mori T. Cytokines and endocrine 120 Schauenstein K, Fassler R, Dietrich H, Schwarz S, function: an interaction between the immune and Kromer G, Wick G. Disturbed immune-endocrine neuroendocrine systems. Clin Endocrinol 1991; 35: communication in autoimmune disease: lack of corti- 107-15. costerone response to immune signals in obese strain 94 Roh M S, Drazenovich K A, Barbose J F, Dinarello C A, chickens with spontaneous autoimmune thyroiditis. on September 26, 2021 by guest. Protected copyright. Cobb C F. Direct stimulation of the adrenal cortex by YImmunol 1987; 139: 1830-3. interleukin-1. Surgery 1987; 102: 140-6. 121 Mason D. Genetic variation in the stress response: 95 Smith E M, Highes T K, Leung M K, Stefano G B. The susceptibility to experimental allergic encephalomyelitis production and action of ACTH-related peptides in and implications for human inflammatory disease. invertebrate hemocytes. Adv Neuroimmunol 1991; 7: 7-16. Immunol Today 1991; 12: 57-60. 96 Da Silva J A P, Colville-Nash P, Spector T D, Scott D L, 122 Sugar M. Arthritis and panhypopituitarism. J Clin Endo- Willoughby D A. Inflammation-induced cartilage crinolMetab 1953; 13: 118-21. degradation in female rodents: protective role of sex 123 Spector T D, Perry L A, Tubb G, Silman A J, hormones. Arthritis Rheum 1993; 36: 1007-13. Huskisson E C. Low free testosterone levels in 97 Pacifici R, Brown C, Puscheck E, et al. Effect of surgical rheumatoid arthritis. Ann Rheum Dis 1988; 47: 65-8. menopause and estrogen replacement on cytokine release 124 Aroso Dias A, Lopes Vaz A, Hargreaves M, et al. from human blood mononuclear cells. Proc Natl Acad Sci Biomarks in secondary . Clin Rheumlatol USA 1991; 88: 5134-8. 1989; 8 (suppl 2): 89-94. 98 Emilie D, Crevon M C, Auffredou M T, Galanaud P. 125 Schlaghecke R, Kornely E, Wollenhaupt J, Specker C. Glucocorticosteroid-dependent synergi between inter- Glucocorticoid receptors in rheumatoid arthritis. Arthritis leukin-1 and interleukin-6 for human B lymphocyte Rheum 1992; 35: 740-4. differentiation. EurJrImmunol 1988; 18: 2043-7. 126 Hedman M, Nilsson E, de la Torre B. Low blood and 99 Josefsson E, Tarkowski A, Carlsten H. Anti-inflammatory synovial levels of sulphoconjugated steroids in rheumatoid properties of estrogen. I. In vivo suppression of leucocyte arthritis. Clin Exp Rheumatol 1992;19: 25-30. production in bone marrow and redistribution of 127 Spector T D, Perry L A, Tubb G, Huskisson E C. peripheral blood . Cell Immunol 1992; 142: Androgen status of females with RA. Br J Rheumnatol 67-78. 1987;25: 316-8. 100 Shimizu K, Amagaya S, Ogihara Y. Analysis of 128 Neeck G, Federlin K, Graef V, Rusch D, Schmidt K L. corticosterone in the serum of mice and rats, using high Adrenal secretion of cortisol in patients with rheumatoid performance liquid chromatography. Chromatogr 1983; arthritis. YRheumatol 1990; 17: 24-9. 272: 170-4. 129 Morand E F, Jefferiss C M, Goulding N J, Maddison P J. 101 Cutolo M, Accardo S, Villagio B, et al. Evidence for the Induction of lipocortin-1 by glucocorticoids is impaired presence of androgen receptors in the synovial tissue of in rheumatoid arthritis [abstract]. Arthritis Rheum 1991; rheumatoid arthritis patients and healthy controls. 34 (suppl 9): S155. Arthritis Rheum 1992; 35: 1007-15. 130 Podgorski M R, Goulding N J, Hall N D, Flower R J, 102 Steward A, Bayley D L. Effects of androgens in models of Maddison P J. Autoantibodies to lipocortin- 1 are rheumatoid arthritis. Agents Actions 1992; 35: 268-72. associated with impaired glucocorticoid responsiveness in 103 Da Silva J A P, Larbre J-P, Spector T D, Perry L A, rheumatoid arthritis. J Rheumatol 1992; 19: 1668-71. Scott D L, Willoughby D A. Protective effect of 131 Klein A, Buskilla D, Gladman D, Bruser B, Malkin A. androgens against inflammation-induced cartilage Cortisol catabolism by lymphocytes of patients with degradation in male rodents. Ann Rheum Dis 1993; 52: systemic lupus erythematosus and rheumatoid arthritis. 285-91. 3tRheumatol 1990; 17: 30-33. 16 Da Silva

132 Myles A B, Schiller L F, Glass D, Daly J R. Single daily 141 Tazuke S, Khaw K T, Barrett-Connor E. Exogenous dose corticosteroid treatment. Ann Rheumn Dis 1976; 35: and endogenous sex hormones. Medicinie 73-6. (Baltimore) 1992; 71: 44-51. 133 Cash J M, Crofford L J, Gallucci W T, et al. Pituitary- 142 Pousette A, Carlstrom K, Stege R. Androgen during

adrenal axis responsiveness to ovine corticotropin different modes of endocrine treatment of prostatic Ann Rheum Dis: first published as 10.1136/ard.54.1.6 on 1 January 1995. Downloaded from releasing hormone in patients with rheumatoid arthritis cancer. UrolRes 1989; 17: 85-8. treated with low-dose . J Rheunmatol 1992; 143 Carlstrom K, Stege R. Adrenocortical function in prostatic 19: 1692-6. cancer patients: effects of orchidectomv or different 134 Chikanza I C, Petrou P, Kingsley G, Chrousos G, modes of estrogen treatment on basal steroid levels and Panayi G S. Defective hypothalamic response to immune on the response to exogenous adrenocorticotropic and inflammatory stimuli in patients with rheumatoid hormone. Urol Int 1990; 45: 160-3. arthritis. Arthritis Rheumii 1992; 35: 1281 -8. 144 de Weerdt 0, Gooren L J. Patterns of serum cortisol levels 135 Schoneshofer M, Wagner G C. Sex differences in in ovariectomized females with and without androgen corticosteroids in man. J Clin Endocrinol Metab 1977; 45: administration. Honn Metab Res 1992; 24: 82-4. 814-7. 145 Orth D N, Kovacs W J, DeBold C R. The adrenal cortex. 136 Nolton W E, Rueckert P A. Elevated free cortisol index In: Wilson V, Foster D W, eds. Williamiis textbook of in pregnancy. Possible regulatory mechanism. AmJn3 Obstet endocrinology, 8th edn. Philadelphia: W B Saunders Co, Gynecol 1981; 139: 492-8. 1992: 489-619. 137 Bulbrook R D, Herian M, Tong J L, Hayward J L, 146 Gillies G, Linton E, Lowry P. The physiology of Swain M C, Wang D Y. Effects of steroidal contraceptives corticotropin-releasing factor. In: De Groot L J, et al, eds. on levels of plasma androgen sulphates and cortisol. Endocrinology, 2nd edn. Philadelphia: W B Saunders Co, Lancet 1973; i: 628-31. 1989: 167-75. 138 Meulenberg P M M, Ross H A, Swinkels L M J W, 147 Wynn V. The assessment of hypothalamic-pituitary- Benraad T J. The effects of oral contraceptives on plasma- adrenal function in man. Menioirs of the Society, for free and salivary cortisol and . Clin Chinm Acta Endocrinology 1968; 17: 213-36. 1987; 165: 379-85. 148 Bliss E L, Migeon C J, Eik-Nes K, Sandberg A A, 139 Zachman M, Manella B, Eiholzer U, Bucher H, Prader A. Samuels L T. The effects of insulin, , bacterial Influence of estrogen in high and low doses on plasma pyrogen and the antabuse- reaction upon the levels steroid concentrations in girls with tall stature and of 17- in the peripheral blood of Turner syndrome. Acta Endocrinol (Copenh) 1984; 106: man. Metabolism 1954; 3: 493-501. 368-73. 149 Staub J J, Jenkins J S, Ratcliffe J G, Landon J. Comparison of 140 Helgason S, Carlstrom K, Damber M G, Damber J E, corticotrophin and corticosteroid response to lysine-vasopressin, Selstam G, von Scoultz B. Effects of various estrogens on insulin, and pyrogen in man. BMJ 1973; 1: 267-9. circulating androgens and cortisol during replacement 150 Jenkins J S, Elkington S G. Metyrapone and pyrogen in therapy in post-menopausal women. Maturitas 1981; 3: the assessment ofpituitary-adrenal function after removal 301-8. of . Lancet 1964; 2: 991 -4. http://ard.bmj.com/ on September 26, 2021 by guest. Protected copyright.