Effects of gonadotrophin-releasing hormone outside the hypothalamic-pituitary-reproductive axis.

GnRH binding has been detected in multiple extra-CNS sites ( 19 , 129 ) but only the presence of GnRH receptors on tumors has attracted considerable attention due to the therapeutic potential of co-opting their use to target the delivery of toxic substances to cancer cells ( 130 , 131 ). Apart from the noted reproductive tissues that express GnRH receptors ( 14 - 18 , 20 - 22 ), there are several other sites expressing GnRH receptors warranting significant further study. It is noteworthy that the presence of GnRH and GnRH receptors in the heart of lower vertebrates, especially fish, is well established ( 132 - 137 ). In an elegant study knocking down GnRH by blocking GnRH mRNA translation, cardiac development in the zebrafish was significantly impaired ( 138 ). Studies injecting biotinylated GnRH ( 89 ), radioactive GnRH ( 139 , 140 ) or GnRH agonists ( 141 - 143 ) have consistently reported GnRH binding in the rodent heart. GnRH receptor mRNA has also been detected in the human heart ( 19 ). Immunoreactive GnRH receptors have also been noted in the human heart, with highest GnRH receptor levels evident in the infarcted heart ( 144 ). Moreover, for cetrorelix, a GnRH agonist, the total amount of cetrorelix bound to the rat heart was nearly 50% of the amount bound to the pituitary gland ( 145 ). GnRH has been reported in the rat heart ( 89 , 146 , 147 ) and GnRH mRNA is measurable in the human ( 19 ) and mouse ( 148 ) heart. Men who are chemically castrated are at a significantly increased risk of a serious cardiovascular event ( 149 , 150 ). This may be due to the loss of testicular androgens: androgens are known to affect cardiac contractility ( 151 , 152 ) and low circulating androgen levels are linked to cardiovascular disease ( 153 ). However, a recent epidemiological study on 73000 men, which compared chemically vs surgically castrated men, strongly supports the hypothesis that the increased risk of cardiovascular disease is not due to the loss of androgens ( 154 ). Subsequent studies ( 155 - 158 ) have confirmed this seminal investigation. Our preliminary in vitro investigations demonstrate a direct effect of GnRH, as low as 1pg/ml, on the contractility of murine cardiomyocytes in serum-free, and thus androgen-free, media ( 159 ). Studies in the chemically castrated rat using the GnRH agonist, Zoladex, suggest this may translate into impaired cardiac function in vivo ( 160 , 161 ). It was not established whether this impaired cardiac function was due to the GnRH agonist per se rather than the loss of testosterone. In several species, including humans, binding of GnRH analogs or GnRH receptor mRNA have been reported in the mammalian adrenal ( 17 , 142 , 145 , 162 - 164 ). Administration of a GnRH analogue, Surfagon, induced morphological changes in the adrenal cortex of the male mouse and, importantly, these effects persisted in castrated animals ( 165 ). In castrate or ovariectomized ferrets, the GnRH agonist deslorelin significantly improved adrenocortical disease ( 166 ). Treating female rats for 3 months with the GnRH antagonist, Detirelix, caused a significant reduction in the adrenal/body weight ratio ( 167 ). However, whether these affects are directly on the adrenal gland or indirectly through modulation of gonadotropin release has not been established. The human bladder epithelium produces both GnRH and GnRH receptor ( 6 ). Following 3 H-GnRH injection significant accumulation of radioactivity was reported in the mouse bladder ( 139 ). The putative function of this paracrine GnRH system has been investigated in the dog ( 168 - 172 ). Ovariectomy causes incontinence in dogs. Treatment with the GnRH agonist, deslorelin, restored continence to all ovariectomized incontinent animals ( 169 ). With the loss of ovarian steroids and an absence of a relationship between gonadotropin levels and urodynamic function, the effect was considered directly due to GnRH ( 168 ) and confirmation of the GnRH receptor in the bladder of this species ( 171 , 172 ) supports this hypothesis. It is not known how GnRH affects bladder function but urethral closure pressure is unaffected by GnRH ( 170 ). There are other GnRH target sites that have received scant attention to date, such as the skin ( 142 , 173 ), lymphocytes ( 174 ), kidney ( 131 , 140 , 142 , 175 ) and liver ( 140 , 142 , 175 ). Several studies have shown that GnRH binding may occur in the liver and kidney ( 142 , 175 ) but discussion of these data has argued that these sites are involved in peptide degradation, despite evidence that GnRH is undetectable in jugular blood ( 176 ). Clearly, future work will be required to address the functional relevance, if any, of these novel putative GnRH targets.

Gonadotropin-releasing hormone (GnRH) was among an array of hypothalamic releasing factors discovered nearly four decades ago by the laboratories of Schally and Guillemin ( 1 ). Confirmation that GnRH was released into hypophyseal portal blood ( 2 , 3 ) cemented the contention that this decapeptide was unique to the reproductive hypothalamo-pituitary axis. However, there are occasional reports that GnRH has unexpected effects or is present in non-reproductive tissues forcing us to reconsider this restricted reproduction-only view. For example, GnRH has activity on the sympathetic ganglia of the frog ( 4 ), GnRH receptor expression is present in the cerebellum ( 5 ) and bladder ( 6 ), and GnRH is released in significant concentrations into cerebrospinal fluid ( 7 ), to potentially act outside the hypothalamus through volume transmission ( 8 ). Indeed, GnRH may have evolved with functions extraneous to reproduction. Studies on octopi provide evidence that GnRH has potent cardiovascular roles ( 9 ). More than 40 different GnRH precursors have been identified ( 10 , 11 ). Most evidence in mammals indicates that GnRH I and chicken GnRH II have been conserved in this Class although GnRH II has not been retained in all species ( 12 ). In mammals, two GnRH receptors have been identified, type I and type II, but the type II GnRH receptor has been silenced in several species ( 10 , 12 ). It is possible that the functions of GnRH II and the type II GnRH receptor have been assumed by GnRH I and/or the type I GnRH receptor. Except for an evolutionary context, this review will focus on GnRH I and the type I GnRH receptor. There is compelling evidence that GnRH and its receptor may perform fundamental roles in cancer cells (see ( 13 ) for recent review) but it is arguable that these tumor effects do not occur in “normal” physiology. There are also several studies showing local GnRH and GnRH receptor production in extra-pituitary reproductive tissues: endometrium ( 14 ), ovary ( 15 , 16 ), placenta ( 17 - 19 ), testis ( 20 , 21 ), prostate ( 22 ). Thus, the purpose of this short review is to summarize evidence that, in addition to its well-established reproductive roles, GnRH may affect multiple tissues not directly associated with the reproductive axis or cancer. Table 1 summarizes putative non-reproductive sites of action of GnRH in mammals. This review will address areas in which studies have attempted to address the physiological significance of GnRH effects.

GnRH is not just a reproductive hormone. Indeed, one of the first functions of GnRH in evolution may have been cardio-active, as shown powerfully in the octopus ( 9 ). The diverse location of GnRH receptors and/or ligand suggests that GnRH may be a major modulator of multiple physiological systems in addition to reproduction. Recent studies suggest that although GnRH may act through a common receptor at the pituitary and these novel sites, the intracellular signaling pathways employed may be different ( 13 , 177 ). Certainly, the presence of a receptor on a particular target does not establish that site as biologically important (for example, olfactory receptors are expressed in the heart ( 178 )). Nevertheless, reconsideration of the potential widespread action that this traditional reproductive hormone may exert, could lead to the generation of novel therapies and encourage due caution when investigating the potential targets of current GnRH therapies (e.g. prostate cancer, endometriosis).