Sex steroid hormones and brain function: PET imaging as a tool for research

Abstract Sex steroid hormones are major regulators of sexual characteristic among species. These hormones, however, are also produced in the brain. Steroidal hormone‐mediated signalling via the corresponding hormone receptors can influence brain function at the cellular level and thus affect behaviour and higher brain functions. Altered steroid hormone signalling has been associated with psychiatric disorders, such as anxiety and depression. Neurosteroids are also considered to have a neuroprotective effect in neurodegenerative diseases. So far, the role of steroid hormone receptors in physiological and pathological conditions has mainly been investigated post mortem on animal or human brain tissues. To study the dynamic interplay between sex steroids, their receptors, brain function and behaviour in psychiatric and neurological disorders in a longitudinal manner, however, non‐invasive techniques are needed. Positron emission tomography (PET) is a non‐invasive imaging tool that is used to quantitatively investigate a variety of physiological and biochemical parameters in vivo. PET uses radiotracers aimed at a specific target (eg, receptor, enzyme, transporter) to visualise the processes of interest. In this review, we discuss the current status of the use of PET imaging for studying sex steroid hormones in the brain. So far, PET has mainly been investigated as a tool to measure (changes in) sex hormone receptor expression in the brain, to measure a key enzyme in the steroid synthesis pathway (aromatase) and to evaluate the effects of hormonal treatment by imaging specific downstream processes in the brain. Although validated radiotracers for a number of targets are still warranted, PET can already be a useful technique for steroid hormone research and facilitate the translation of interesting findings in animal studies to clinical trials in patients.

gonadothropin-release hormones (GnRH) at the hypothalamus. GnRH induces the release of luteinising hormone (LH) and follicle-stimulating hormone (FSH) in the pituitary, which activate the secretion of steroidal sex hormones from the gonads ( Figure 1A). Peripheral sex hormones are present in the plasma, where they are mainly bound to plasma proteins such as sex hormone binding globulin (SHBG) or corticosteroid binding globulin (CBG). 6 SHBG has high affinity for both oestrogens and androgens, whereas progesterone is bound by CBG.
These globulins protect steroid hormones against metabolic degradation and, consequently, the fraction of free steroid hormones in plasma is small. Yet, this small fraction of unbound steroid hormones can readily cross the blood-brain barrier by passive diffusion as a result of the lipophilic nature of steroids. However, there is also a significant contribution of de novo synthesised steroid hormones in the brain because the brain itself contains the enzymes needed for the synthesis of these steroids. 7 Sex hormones produced in the brain include 17β-oestradiol, testosterone and progesterone, along with other neuroactive steroids such as pregnenolone, dehydroepiandrosterone and allopregnanolone. 8 In recent decades, the specific receptors for sex steroid hormones were found to be expressed in the brain. 9 Currently, most information has been obtained from animal experiments, which cannot easily be translated to humans, as well as from post-mortem analysis of human brain tissue. 10,11 In most studies, western blotting and in situ hybridisation have been used to quantify hormone receptors in the brain. 9,12 Such techniques would not allow research on the biology of steroid hormones and their receptors in the living human brain. One approach with respect to non-invasively investigating sex hormone receptors in the brain is the use of positron emission tomography (PET) with radiolabelled receptor ligands. PET allows the quantification of functional parameters, such as receptor density and occupancy. 13 PET imaging of steroid receptors is already widely used in oncology to visualise receptor expression and receptor occupancy in hormone-sensitive tumours such as breast and prostate cancer. 14 By contrast, sex hormone receptor imaging in the brain is still in its infancy. 15 Sex steroid receptor imaging in neuroscience suffers from some additional hurdles, such as the low receptor expression in some brain regions 16 and a poor penetration of radioligands through the blood-brain barrier.
F I G U R E 1 Effects of sex steroids at both physiological and cellular levels. (A) The regulatory processes for the synthesis of sex steroids by the hypothalamic-pituitary-gonadal (HPG) axis. The hypothalamus regulates the production of luteinising hormone (LH) and follicle-stimulating hormone (FSH) via the release of gonadotrophin-releasing hormone (GnRH). Both LH and FSH stimulate the synthesis and release of oestrogens and progesterone from the ovaries in females, as well as testosterone from the testis in males. At the same time, these sex steroids can regulate the release of GnRH from the hypothalamus, as well as LH and FSH from the pituitary. (B) General scheme of sex steroid effects at cellular level. Sex hormones can bind to either cytoplasmatic receptors or membrane-associated receptors. When the molecules bind to membrane receptors, the receptor (coupled to G protein subunits complex: Gα, Gβ and Gγ) activates phospholipase C (PLC) to exert a rapid nongenomic responses via the second messengers inositol phosphate 3 (IP 3 +) and diacylglycerol (DAG). On the other hand, when they bind to cytoplasmatic receptors, the complex is translocated to the nucleus (with the help of different co-activators) to exert genomic effects In this review, we survey the available literature about the use of PET imaging in the field of neuroendocrinology, in which imaging data are directly or indirectly correlated with sex steroid hormone (receptor) levels. We discuss the role of sex steroids in brain function and behaviour, give an overview of the tracers that are currently available for PET imaging of hormone receptors and their applicability in brain research, and summarise the results of PET imaging of the downstream effects of sex steroids in the brain. Based on these data, we propose that PET is a promising technique for future translational research in this field.

| SEX STEROID HORMONES AND BRAIN FUNCTION
Oestrogens can exert their effects through either intracellular or membrane-associated oestrogen receptors (ERs); in particular, the intracellular receptors ERα and ERβ, and membrane-associated Gprotein regulator motifs. Upon binding of oestrogen to the ER, the ligand-receptor complex dimerises and migrates to the nucleus, where the dimer can bind to hormone response elements (HRE) in the promotor region of oestrogen-responsive genes. Activation of the HRE leads to the induction or the repression of gene transcription. In addition to this genomic signalling pathway, sex steroids can act via nongenomic signalling ( Figure 1B) (for a review, see Kawata et al 17 ).
Oestrogen signalling can affect various aspects of brain function and behaviour. Most information about the relationship between oestrogens and brain disorders was obtained from studies in female animals or women demonstrating behavioural differences between the different stages of the menstrual cycle. There is ample evidence for a role of oestrogens in anxiety and depression, both from animals and humans. 18 Women are vulnerable to depression when the concentration of sex hormones changes markedly. This can lead to pre-menstrual dysphoric disorder, post-partum depression and perimenopausal or postmenopausal depression. 19 Oestrogens have antidepressant effects when they are administered either alone or in combination with antidepressants 20,21 and, consequently, oestrogen replacement therapy can be used to prevent the development of depression in individuals who are at risk. 22 Oestrogens can also have neuroprotective effects. High levels of circulating oestrogens are associated with less ischaemia-induced brain injury. 23 A similar effect is also observed when high levels of endogenous oestrogens are synthesised in the brain. 24 Oestrogens were found to play a role in neuronal plasticity and spine synapse formation. 25,26 Furthermore, many studies have shown positive effects of oestrogens on cognition. [27][28][29] In Alzheimer's disease, oestrogens have been shown to protect neurones against the toxicity of amyloid plaques. 30 Nevertheless, more studies are necessary 31 because investigators from the Women's Health Initiative Memory Study found that therapy with a combination of oestrogen and progestin increased the risk for dementia in postmenopausal women and did not improve their performance in mild cognitive tasks. 32 For this reason, the contribution of oestrogens and the molecular dynamics of their interaction with other hormones and neurotransmitters should be determined to obtain a better understanding of the role of these steroids in brain function and neuroprotection.
Progestins can exert their effects through both intracellular progestin receptors (PR-A and PR-B) and membrane-associated PRs. In addition, these neuroactive steroids can also interact with several other receptors and ion channels. 33 For example, several steroid hormones, including progesterone, were found to bind to sigma-1 receptors. 34 Progesterone can act as a sigma-1 receptor antagonist. 35 Under ischaemic conditions, progesterone antagonism of sigma-1 receptors can be neuroprotective because it attenuates the NMDA-induced influx of Ca 2+ via the NMDA receptor ion channel. 36 Progestins can also interact with oestrogens in the brain, such as in the regulation of synapse formation. 37 Progestins are also involved in processes such as maintenance of the structural integrity of myelin, 38 regulation of spinogenesis, synaptogenesis, neuronal survival and dendritic growth. [39][40][41] There is evidence indicating that the administration of exogenous progesterone in animal models of traumatic brain injury and ischaemia can decrease the lesion volume in the brain 42 and decrease cognitive deficits. 43 Likewise, progestins can exhibit a neuroprotective effect in spinal cord injury. 44 Evidence has also been presented suggesting a neuroprotective effect of progestins in other brain disorders, such as peripheral nerve injury, demyelinating disease, motoneurone diseases, seizures, depression and Alzheimer's disease. 18,[45][46][47] Androgens exert their effects through the androgen receptor (AR) subtypes AR-A and AR-B. Androgens are known to affect various brain functions and behaviour. The most common behavioural role of androgens is related to aggression. An excess of circulating androgens induces aggressive behaviour in both males and females. 48 Androgens are also involved in depression and anxiety-like disorders, especially after menopause in women and during hypogonadism in men. 49 Alterations of testosterone levels were associated with an increased risk of mood disorders and psychosis. 50 Anabolic abuse and hyper-or hypoandrogenism are related to mood changes 51 and the incidence of depression. 52 On the other hand, androgens can also have a neuroprotective role. Long-term exposure to androgens increases hippocampal neurogenesis and modulates the survival of new neurones. 53 Androgens also play a role in synapse formation and they are capable of inducing the formation of spine synapses, 54 which appears to be mediated by NMDA activity. 55

| PET TRACERS FOR STEROID HORMONE RECEPTORS IN BRAIN RESEARCH
Despite the increasing knowledge on the roles of sex steroid hormones, many aspects of the functions and mechanisms of action of sex steroid hormones in the brain are still incompletely understood and require further research. Non-invasive imaging tools such as PET could facilitate such research. Several PET tracers for steroid hormone receptors are available and have been successfully used in oncology, although, to date, there are only few studies in which they have been used for brain research. Most studies with PET tracers for steroid hormone receptors use either autoradiography or ex vivo tissue counting.
So far, only a few studies have measured the in vivo distribution of steroid receptor ligands in rodents, whereas imaging studies of the human brain are lacking.

The most frequently used PET tracer for imaging oestrogen receptors is 16α-[ 18 F] fluoro-17ß-oestradiol ([ 18 F]FES). [ 18 F]FES has been
successfully used in both preclinical and clinical studies, mostly in breast cancer. 56 [ 18 F]FES was the first PET tracer to be applied for quantitative ex vivo assessment of oestrogen receptors in the brain. The brain of female rats was dissected and radioactivity in different brain areas was measured ex vivo with a γ counter. By applying different distribution times, information about the kinetics of the tracer in the rat brain was obtained. 16 Specific binding of the tracer was observed only in brain regions with high ER density, such as the pituitary and hypothalamus. Specific binding could be quantified both by equilibrium and dynamic kinetic analysis. 16 Two years later, [ 18 F]FES PET was successfully used to identify ER expression in the tumour of six patients with brain meningiomas. 57 Later studies, including our own, have shown that [ 18 F]FES PET is able to detect ER-expression in brain metastases of ER-sensitive tumours such as breast cancer.
More than a decade after the experiments of Moresco et al, 57 our group investigated whether ER in the rat brain could be quantified in vivo using [ 18 F]FES with a dedicated small-animal PET scanner. 58 The results obtained were in agreement with the ex vivo data of Moresco et al. 16 Specific binding was observed in the pituitary and hypothalamus, which are both brain regions with a high ER density, but not in other parts of the brain. 58   Our study showed that [ 18 F]FDHT is metabolised very rapidly in rats, and its uptake in the brain is very low. 69 This results in a poor signalto-noise ratio, which precludes the detection of AR in the rat brain. By contrast to rats, humans express SHBG, which can protect steroids such as [ 18 F]FDHT from metabolic degradation. 70 Despite the disappointing results obtained in rats, the stabilising effect of SHBG in men would still warrant investigation of the ability of [ 18 F]FDHT PET to visualise AR receptors in the brain of humans.
In conclusion, it can be proposed that PET has potential as a noninvasive tool for assessing the expression of steroid receptors in the brain, provided tracers become available that can penetrate the bloodbrain barrier and have higher affinity and metabolic stability.

| PET IMAGING OF AROMATASE AS A BIOMARKER FOR OESTROGEN SYNTHESIS
Aromatase is a key enzyme in the biosynthesis of oestrogens; it catalyses the conversion of testosterone into oestradiol. 71 Aromatase is expressed in a wide variety of tissues, including ovaries, adipose tissue, skin, testicles, muscle, liver and the central nervous system.
Aromatase has been suggested as a biomarker for neuroprotection because it increases the local levels of oestrogens in injured neurones in the brain. 72 Aromatase is not expressed constitutively in the brain but can be induced by testosterone or dihydrotestosterone. 73 Brain aromatase is involved in, amongst others, the regulation of sexual behaviour, emotional behaviour, aggression, cognition, memory and neuroprotection, 73 making this enzyme an interesting target for the study of sex steroid hormones in the brain.

| USE OF PET TRACERS TO STUDY SEX STEROID HORMONE-INDUCED CHANGES IN BRAIN FUNCTION
PET imaging using radioligands of receptors related to sex steroid hormone signalling may provide valuable information about the interaction of these hormones with other signalling systems in the brain, as well as the possible behavioural outcome of that interaction, thus offering a wide range of possible studies. PET imaging may also be used to study the impact of steroid hormones on physiological or metabolic biomarkers. Below, we discuss some studies in which the downstream effects of hormonal changes were evaluated by PET imaging.

| Impact of sex steroid hormones in cerebral blood flow
A general approach for studying the impact of steroid hormones is to detect activation of specific brain regions by measurement of the regional cerebral blood flow (rCBF  to investigate the effect of sex steroid hormones on brain glucose metabolism. The first study in this specific field aimed to determine the neural correlates of sexual competition in male rhesus macaques.

| Brain metabolism and sex hormones: [ 18 F]FDG
The study showed metabolic differences between male monkeys confronted with threats to their exclusive sexual access to a female mate and controls. The differences in brain glucose metabolism were correlated with differences in testosterone levels. 89   Alzheimer's disease. 99 [ 18 F]FDG PET has also been used to study the correlation between brain metabolism and oestradiol brain levels in postmenopausal women with Alzheimer's disease. In a small study, a direct linear correlation was found between hippocampal glucose metabolism and oestradiol levels in the cerebrospinal fluid. 100 [ 18 F] FDG PET was also able to reveal regional changes in brain glucose metabolism as a result of testosterone replacement therapy in two hypogonadal patients with Alzheimer's disease. 101 Furthermore, [ 18 F] FDG PET was able to demonstrate a compensatory effect of testosterone administration on brain hypometabolism in women with anorexia nervosa. 102 An example of [ 18 F]FDG imaging in the brain is provided in Figure 4.

| Sex steroid hormones and neurotransmitter activity regulation
Sex steroid hormones are known to participate in many developmental and regulatory processes in the brain. Most of these effects are  Sex hormone levels were found to be correlated with test scores for aggression and 5-HTR 1A tracer uptake in frontal areas. 108 Serotonin changes have also been studied in the brain of menopausal women treated with hormone therapy, but no significant differences in [ 11 C]-WAY100635 uptake were found between subjects treated with oestradiol alone or oestradiol + progesterone. 109 Another receptor of interest is the 5-HTR 2A . Longitudinal PET studies with the tracer [ 18 F]altanserin showed increased 5-HTR 2A binding in the whole brain and in specific brain regions (eg, the hypothalamus and cortex) of postmenopausal women that were first treated with oestradiol alone, and were later treated with the combination of oestradiol with progesterone. 110,111 Another study using the same radiotracer showed a positive correlation between cortical which is known to be related to brain processes affected in psychiatric disorders. 115  The aforementioned studies show that sex steroid hormones can have an effect on brain neurotransmitter systems and also that these effects can be monitored non-invasively with PET. So far, only few publication describe the use of this approach in neuroendocrinology studies, indicating an area of research that still remains unexplored. Most of the studies showed interactions of sex hormones with major neurotransmitter systems involved in psychiatric disorders such as serotonin and dopamine, positioning them as likely targets for future research in this field.

| CONCLUSIONS
There is ongoing research on the influence of sex steroid hormones on brain development and brain function. Although the expression of sex steroid hormone receptors in the brain has been demonstrated 9 and some roles of these hormones in the brain have been elucidated, [2][3][4] there is still a large gap in our knowledge of these hormone system.
This can be partly be ascribed to the lack of suitable techniques available for assessing the dynamics and interplay of these molecules in the living brain. Non-invasive imaging could offer a good opportunity to investigate the role of sex steroid hormones and their receptors in the brain in health and disease.
Specific radiotracers for PET imaging of oestrogen, progestin and androgen receptors have been developed, although only few of them have been tested in the brain. Some successful studies, especially using the ER tracer [ 18 F]FES, have been performed, although low uptake in brain areas with low receptor density, rapid tracer metabolism and unfavourable kinetics of many tracers limit the application of these tracers for the visualisation and quantification of changes of steroid receptor density in specific brain areas. Tracers with higher affinities and metabolic stability and better blood-brain barrier penetration are needed to expand this research field.
PET imaging can also be used to quantify the effects of sex steroids on brain perfusion and metabolism. Hormone treatment in conditions such as menopause, hypogonadism and steroid abuse appears to provide useful paradigms for studying the effect of steroid signalling on brain activity or examining the relationship between stress hormone levels and biological outcomes in humans. Studies of the correlations between hormone levels in plasma and regional tracer uptake may also provide useful information on the involvement of specific brain regions and regional connectivity. The use of animal models may also be useful because many experimental manipulations can be applied in animals but not in humans.
Furthermore, a plethora of PET tracers for specific neurotransmitter receptors and transporters are available. These tracers enable the investigation of the interaction between sex steroid hormones and various neurotransmitter systems. These studies could help to unravel the mechanisms that are responsible for the impact of sex steroid hormones on brain function and neuroprotection. An improved understanding of these effects could result in the improvement of existing hormone therapies. Studies could focus on, for example, discrimination of specific receptor functions in terms of fast and slow effects, sex differences and the mechanisms of action of steroids in diseases of the brain. We have reviewed PET studies related to the function of sex hormones in the brain. If the limitations identified can be overcome, PET may prove to be a promising non-invasive technique that can be applied in both experimental animals and human subjects, which would facilitate the translation of interesting findings from studies in experimental animals into clinical trials in humans.