Steroids‐Dopamine Interactions in the Pathophysiology and Treatment of CNS Disorders

SUMMARY

Introduction: Dopamine cell loss is well documented in Parkinson's disease and dopamine hypofunction is proposed in certain depressive states. At the opposite, dopamine hyperactivity is an enduring theory in schizophrenia with extensive supporting evidence. Aims: This article reviews the sex differences in these diseases that are the object of many studies and meta‐analyses and could be explained by genetic differences but also an effect of steroids in the brain. This article then focuses on the extensive literature reporting on the effect of estrogens in these diseases and effects of the other ovarian hormone progesterone as well as androgens that are less documented. Moreover, dehydroepiandrosterone, the precursor of estrogens and androgens, shows effects on brain dopamine neurotransmission that are reviewed. To investigate the mechanisms implicated in the human findings, animal studies are reviewed showing effects of estrogens, progesterone, and androgens on various markers of dopamine neurotransmission under intact as well as lesioned conditions. Discussion: For possible future avenues for hormonal treatments in these central nervous system diseases, we discuss the effects of selective estrogen receptor modulators (SERMs), the various estrogen receptors and their specific drugs as well as progesterone drugs. Conclusion: Clinical and experimental evidence supports a role of steroid–dopamine interactions in the pathophysiology of schizophrenia, depression and Parkinson's disease. Specific steroidal receptor agonists and SERMs are available for endocrine and cancer treatments and could find other applications as adjunct treatments in central nervous system diseases.

Introduction

The contribution of gonadal hormones in the prevalence and symptomatology of neurodegenerative diseases such as Parkinson's disease (PD) and mental diseases such as schizophrenia, depression, premenstrual syndrome, and postnatal depression is the object of many studies but the literature on this subject is not always consistent. Furthermore, estrogens play an important role in other neurodegenerative disorders, as well as plasticity and cognition; this will not be discussed in this review and the reader is referred to previous publications [1, 2, 3]. This review of the sex steroids modulation of dopamine (DA) neurotransmission surveys relevant observations in normal and central nervous system (CNS) diseases in humans and animals. Focus is mainly on schizophrenia, PD, and depression because of their prevalence, the documented effect of estradiol and the implication of DA in these conditions. The sex differences and the implication of gonadal steroids are reviewed. This is then compared to DA neurotransmission and its modulation by gonadal steroids (Figure 1) in animal models. Finally, future research avenues with sex steroids for brain neuromodulation and neuroprotection of DA activity will be discussed.

Figure 1.

Schematic steroidal synthetic pathways of steroids mostly investigated for their activities on brain DA. Dotted arrows indicate that several synthetic steps and intermediate steroids were involved in the biosynthesis of the product. P450scc, cytochrome P450 sidechain cleavage; P450c17, cytochrome P450 17α‐hydroxylase/C17,20‐lyase; 3α‐HSD, 3α‐hydroxysteroid dehydrogenase; 3β‐HSD, 3β‐hydroxysteroid dehydrogenase; 5α‐R, 5α‐reductase; 17β‐HSD, 17β‐hydroxysteroid dehydrogenase.

Sex Differences, Steroids and Dopamine (Development, Pregnancy, Post‐Partum, Menopause, Aging)

Throughout life, estrogens are involved in various biological actions, including actions in the CNS [4, 5]. Estrogens are essential for normal brain function, in pregnancy, and continuing into the aging process [4, 5]. During the critical period of brain sexual differentiation, estrogens have permanent effects related to organizational differences between males and females in the CNS [4, 5]. During development, estrogens play a role as a growth factor in modulating plasticity and neuronal differentiation [6]. Estrogens continue to modulate brain function in adulthood [4, 5]. Through the entire lifespan, estrogens intervene to help to maintain brain function, modulate locomotion, cognition, and mental status [7]. In women, fluctuating plasma concentrations of estrogen as occurs during the menstrual cycle, impact on cognitive performance, and locomotor activity [8, 9, 10, 11]. Normal men respond to an amphetamine challenge by releasing more striatal DA than do normal women [12]; this greater sensitivity of the DA system could render males more prone to develop the striatal hyperdopaminergia that underlies psychosis [13].

Dehydroepiandrosterone (DHEA) and its sulfate ester, DHEAS, together represent the most abundant steroid hormones in the human body and can be synthesized de novo in the brain [14]. DHEA is a precursor of estradiol and testosterone (Figure 1). Levels of DHEA decrease gradually with age [15]. In men, plasma DHEAS concentrations decrease by an average of 1–4% per year between the ages of 40 and 80 and 2% per year in women [16]. Whether or not endogenous concentrations of DHEA and DHEAS are abnormal in various neuropsychiatric illnesses, it is possible that exogenous DHEA supplementation could have therapeutic benefits [14].

Schizophrenia, Steroids, and Dopamine

Schizophrenia is a complex disease implicating an interaction between genetic and environmental factors [17]. It is characterized by episodic positive symptoms such as delusions, hallucinations, paranoia, and psychosis and/or persistent negative symptoms such as flattened affect, impaired attention, social withdrawal, and cognitive impairments [18]. This illness mainly affects the mesocortical limbic system, at several brain structures, the most important seem to be the prefrontal cortex, involving multiple neurotransmitter systems that appear to be unbalanced [19]. Although the hypothesis of hyperactivity of the DA system in schizophrenia still prevails, other neurotransmitters such as glutamate and serotonin are involved [20]. Estrogens have a psychoprotective action that appears to be mediated by central dopaminergic and serotoninergic mechanisms [21]. Although the neurochemical origins of schizophrenia do not necessarily lie in DA dysregulation, present evidence supports that this operates as the final common pathway [13]. Table 1 summarizes the relevant literature supporting the implication of gonadal steroids in schizophrenia. This disease appears at the beginning of adulthood and a sex difference is observed [22]. Women develop the disease later, have a second peak of incidence around the age of menopause, need smaller doses of neuroleptics, except at menopause, and are more sensitive to them [22, 23]. A meta‐analysis study concludes that the male sex is a major risk factor for a more severe and more easily recognizable form of schizophrenia [24].

Table 1.

Human data on steroidal modulation in CNS disorders implicating DA

Disease	Observations of possible steroidal influence	References
Schizophrenia
Sex differences
	Higher prevalence in men than in women	[24, 290, 291]
	No difference in incidence between men and women	[292]
	Age at onset later in women	[22, 28, 290, 293]
	Different manifestation in men and women, women develop emotional symptoms and men symptoms of isolation or aggressive	[22, 28, 40, 293]
Hormonal therapy
Estrogens (specific molecule not stated)	Therapeutic effects, via neuromodulatory and neuroprotective activity	[25, 26]
Estradiol	Higher circulating estradiol levels provide an antipsychotic effect for women	[36]
Endogenous estrogens
Menarch	Increased negative symptoms	[31]
Menstrual cycle	Worsen symptoms in periods of low estrogen	[27, 29, 30]
Pregnancy	Decreased symptoms	[27]
Postpartum	Increased relapse	[27]
Menopause	Increased incidence	[22, 290, 293]
Depression
Sex differences	Higher prevalence in women than in men	[68, 69, 294, 295, 296, 297, 298]
	Age at onset later in men	[299]
Hormonal therapy
Estrogens (specific molecule not stated)	Effective for the treatment of postnatal depression	[73]
17β‐estradiol	Effective for the treatment of postnatal depression	[81]
Estradiol	Effective for the treatment of postnatal depression	[70]
Estrogen (specific molecule not stated)	Effective for the treatment of premenstrual depression	[73, 86]
Estradiol	Effective for the treatment of premenstrual depression	[70, 83, 87, 88]
Estrogen (specific molecule not stated)	Effective for the treatment of perimenopausal depression	[73, 82, 84]
Estradiol	Effective for the treatment of perimenopausal depression	[70, 85]
Endogenous estrogens
Bilateral oophorectomy	Increased risk	[72]
Premenstrual	Increased risk	[71, 73]
Postpartum	Increased risk	[71, 73, 300, 301]
Menopause	Increased risk	[71, 73, 302, 303]
Parkinson's disease
Sex differences
	Higher susceptibility in men than in women	[94, 95, 97, 98, 99, 100, 102, 103, 304, 305]
	No difference in incidence between men and women	[104]
	Age at onset later in women	[94, 105, 306, 307, 308, 309, 310]
	Age at onset later in men	[311, 312]
Hormonal therapy
Estrogen (specific molecule not stated) or conjugated estrogens	Increased risk in women with a hysterectomy	[108, 111]
Progestin (specific molecule not stated)	Increased risk	[112]
Estrogen (specific molecule not stated)	Reduced risk	[109]
Conjugated estrogens or estrogen (specific molecule not stated) in combination with progestin (MPA)	Reduced risk in women with natural menopause	[111]
Oral contraceptives	Reduced risk	[110, 112]
Estrogen (specific molecule not stated)	No effect on risk among women with natural menopause	[108]
Estrogen (specific molecule not stated) or conjugated estrogens	No effect on risk	[106, 110, 112, 313]
Conjugated estrogens or estrogen (specific molecule not stated) alone or in combination with progestin	Improved symptoms	[119, 122]
Progestin (specific molecule not stated)	Worsen symptoms	[120]
17β‐estradiol	No effect on symptoms	[120]
Endogenous estrogens
Hysterectomy	Increased risk	[106]
Early time of menopause	Reduced risk	[107]
Longer fertile lifespan	Reduced risk	[108]
Shorter fertile lifespan	Increased risk	[107]
Higher parity	Increased risk	[107, 108]
Premenstrual and menstrual periods	Worsen symptoms	[123, 124, 125, 126]
Pregnancy	Worsen symptoms	[127, 129, 130, 314]

MPA: medroxyprogesterone acetate.

Accumulating evidence suggest that estrogens may have therapeutic effects in schizophrenia, via neuromodulatory and neuroprotective activity [25]. Estrogens are hypothesized to be protective for women against the early onset of severe symptoms of schizophrenia [26]. Epidemiological studies report that relative to men, women show an initial delay (4–7 years) in age of onset of schizophrenia with a second peak after 44 year [27, 28]. Clinical research shows that symptoms in women frequently vary during the menstrual cycle, worsening during low estrogen phases [27, 29, 30]. Later age at menarche is associated with higher negative symptoms scores and greater functional impairment [31]. Higher estrogen serum levels are associated with lower positive symptom scores [32]. Lower serum estradiol, estrone, and testosterone are observed in men acutely suffering from schizophrenia suggesting a role for estrogens in both sexes [33]. Pregnancy is often a less symptomatic time for women but postpartum relapses are frequent [27]. Younger women are reported to require lower antipsychotic dosages than men and following menopause they require higher dosages [27] but a study found that women require lower doses that men for all ages suggesting a negative role for testosterone [23]. In a randomized study, 36 women were assigned to a 100 or 50 mg estrogen patch, or placebo, with a standardized antipsychotic medication; women in the 100 mg group were significantly better [34]. A study of 32 women found that psychotic symptoms were worse during the low‐estrogen phase of their menstrual cycles [35]. This suggests that higher circulating estradiol levels provide an antipsychotic effect for women with schizophrenia [36]. This gender‐bound age difference in onset of illness is suggested to be due to the protective antidopaminergic effect of estrogens in women [37, 38, 39].

Men and women do not develop the same symptoms in schizophrenia [22, 40]. The prognosis in women is better than men because they tend to develop more frequently positive symptoms, better treatable by neuroleptics, and development of less severe forms (disorganized and paranoid types) [39]. Also women tend to develop more hallucinations [41, 42, 43] and men more negative symptoms less treatable by neuroleptics [44, 45, 46]. Women require lower doses of neuroleptics than men in sporadic episodes and in relapse prevention [23], and respond better to psychosocial and pharmacological treatment [47]. Around the age of menopause, the status of women tends to deteriorate more quickly than men of a similar age [48]. In addition, they require higher doses of neuroleptics and are at higher risk of developing tardive dyskinesia following treatment with typical neuroleptics, the prevalence increases by 24% in men over 70 years and 49% in women of this age [49, 50]. In addition, women develop forms of tardive dyskinesia tougher than men [50]. Some clinical studies in dyskinetic women show an improvement in their condition after estradiol treatment [51, 52, 53, 54] whereas others showed no beneficial effect of estradiol on tardive dyskinesia [51, 55]. A study also revealed that coadministration of estradiol with neuroleptics in women with schizophrenia improves response to antipsychotics [56]. Although estrogen appears to be a useful treatment for schizophrenia, further research is required to determine the correct dose and duration of use of estradiol [26].

Schizophrenia is linked to alteration of DHEA and DHEAS levels in many studies, but with findings in opposite directions, elevations, and abnormally low concentrations [14]. There are studies reporting low DHEA serum concentrations in schizophrenic patients [57, 58, 59]. In a placebo‐controlled study, DHEA decreased negative symptoms, anxiety, and depression in patients with schizophrenia [60]. In schizophrenic patients, DHEA decreased parkinsonian symptoms but not akathisia [61]. Changes in blood concentrations of DHEA were negatively associated with changes in parkinsonian symptoms, such that higher increases in DHEA concentrations were associated with greater decreases in parkinsonian symptoms and extrapyramidal symptoms (EPS) ratings [61]. However, another study found that DHEA was not superior to placebo in treating negative or positive symptoms of schizophrenia or EPS [62].

Depression, Steroids, and Dopamine

The DA hypothesis of depression proposes in certain depressive states a dopaminergic hypofunction with psychomotor retardation, loss of initiative, and anhedonia whereas manic episodes are the result of dopaminergic hyperactivity [63]. Depression is the most common neuropsychiatric disorder found in patients with PD [64]. Costa et al. [65] proposed that alteration of prefrontal and limbic cortical areas could be associated with major depression in PD. Recent studies using neurophysiological and functional imaging techniques indicate clearly that dopaminergic mechanisms in structures of the limbic system are related to depression in patients with PD [66, 67].

The literature on the implication of gonadal steroids in depression is summarized in Table 1. The prevalence of depression is higher in women than in men with a 2:1 ratio [68, 69]. It is estimated that 33% of women experience depression in their lives [70]. The risk of developing depression is increased in women when their plasma estrogen levels are low, such as during some phases of the menstrual cycle, postpartum, or menopause [71]. In women, a bilateral oophorectomy before menopause was reported to increase long‐term risk of depressive or anxiety symptoms [72]. The sex difference in depression begins at puberty but there are also three life stages that display increased risk of depression related to hormonal changes: premenstrual, postpartum, and climacteric [73]. The sex difference could be due to plasma concentrations of testosterone in men that are 1000 times greater than the highest plasma concentration of the principal estrogen, 17β‐estradiol, in women [74]. Even assuming a small level of aromatase activity, an enzyme that catalyses the transformation of testosterone into estradiol, 17β‐estradiol levels in the male brain would be just as high as those reached in females [74]. Moreover, female aromatase knockout mice deficient in estradiol were shown to exhibit “depressive‐like” symptoms that were not reversed with estradiol treatment in adulthood [75], also supporting a beneficial organizational role in development of estradiol related to depression. Furthermore, aging men do not have abrupt decline in plasma hormone levels such as occurring at menopause [76]. Various clinical and preclinical studies have suggested an antidepressant action of estrogens and related substances [77, 78, 79, 80]. Estrogen therapy was shown to be effective to treat postnatal depression, premenstrual depression, and perimenopausal depression (the triad of hormone‐responsive mood disorders) [70, 73, 81, 82, 83, 84, 85, 86, 87, 88].

Women with a first onset of major or minor depression during their perimenopause were reported to have low morning plasma DHEA and DHEAS concentrations [89]. Lower plasma DHEA concentrations during pregnancy and during the postpartum period were associated with higher postpartum ratings of depression [71]. Although few clinical trials of DHEA treatment for depression were conducted, they consistently suggest beneficial effects [14]. DHEA treatment resulted in antidepressant effects in un‐medicated middle‐aged to elderly patients with major depression [90]. This study was followed by a double blind, placebo‐controlled trial in which 22 depressed patients received DHEA or a placebo, 5 of 11 DHEA‐treated patients showed greater than 50% improvement in depression ratings [91]. Subsequently, another research group conducted a double blind placebo‐controlled randomized study and reported that DHEA treatment produced a robust antidepressant response [92]. In another double blind placebo‐controlled trial, DHEA monotherapy was associated with antidepressant effects in patients with both major and minor depression [93].

Parkinson's Disease, Steroids, and Dopamine

PD is a neurodegenerative disorder, affecting approximately 1% of the population over the age of 65. PD is mainly characterized by the selective and progressive loss of DA neurons in the substantia nigra. As a result, striatal DA decreases, leading to an imbalance in basal ganglia processes and the manifestation of clinical motor symptoms such as tremor, rigidity and bradykinesia. As summarized in Table 1, most studies have observed a higher susceptibility of the disease in men [94, 95, 96, 97, 98, 99, 100, 101, 102, 103], whereas no difference was also reported [104]. The age at onset in women is reported to be delayed as compared to men in most of the studies whereas no such sex difference was also observed [105]. This sex difference suggests a protective effect of estrogens in PD.

Clinical reports have investigated the relationship between endogenous and exogenous estrogens and the risk of PD (Table 1). The risk to develop PD seems to increase with conditions causing an early diminution in endogenous estrogens. In women with PD, an early time of menopause, a higher frequency of hysterectomy and less occurrence of estrogens therapy is reported [106]. An association between factors reducing estrogen stimulation during life and the development of PD was found [107]. The Women's Health Initiative Observational Study (WHI‐OS) confirmed these results in a large group of 83,482 women, showing that longer fertile lifespan among women who experienced natural menopause was associated with decreased risk of PD whereas higher parity conferred increased risk of the disease [108]. The use of postmenopausal estrogen therapy was associated with a reduced risk of PD [109] but a similar risk among women using or who never used hormones was also observed [110]. An increase risk of PD in women with hysterectomy who used estrogen therapy but not in women who experienced natural menopause is reported [108, 111]. A recent prospective study reports no association between endogenous and exogenous exposure to estrogen and the risk of PD [112]. Moreover, in an exploratory analysis, women using progestin‐only hormone had an increased risk of PD, but this was based on only four cases [112].

Sex differences on the evolution of symptoms and response to levodopa treatment are described [113, 114, 115], with men exhibiting more severe parkinsonian motor features than women and a greater improvement of motor function in women following levodopa therapy. Estrogen modulates symptoms of PD and levodopa‐induced dyskinesia; an amelioration of parkinsonian and dyskinesia symptoms is observed with estrogen therapy or high levels of endogenous estrogen [54, 116, 117, 118]. In postmenopausal women with early PD not yet taking levodopa, a lower symptom severity score was reported with the use of estrogen therapy [119]. However, estrogen therapy had no effect at later stage of the disease on symptom severity [120]. In a 2‐week double‐blind cross‐over study of high‐dose transdermal 17β‐estradiol conducted in postmenopausal women with mild to moderate PD, a reduction in the antiparkinsonian threshold dose of levodopa required to improve motor symptoms was observed but dyskinesia scores were unchanged [121]. In a placebo‐controlled, randomized, double blind trial, no dopaminergic effect of estradiol on motor function in postmenopausal women was found [120]. In contrast, results from a double blind, parallel‐group, prospective study showed that treatment with Premarin (conjugated estrogens) reduces motor disability in postmenopausal women with PD associated with motor fluctuations [122].

Approximately 2–4% of women diagnosed with PD are under the age of 50 and are still experiencing regular menstrual cycles [123]. A worsening of parkinsonian symptoms was associated with premenstrual and menstrual periods, when estrogens and progesterone levels are low [123, 124, 125, 126]. However, no significant correlation was found between PD severity and estrogen and progesterone levels [124]. A reduction in the effectiveness of levodopa was also observed in young PD women before and during the menstruation period [117, 125]. A more rapid progression of parkinsonian symptoms associated with pregnancy and worsening of symptoms during or shortly after delivery was reported in PD women [127, 128, 129, 130], supporting a complex role of ovarian hormones in the modulation of PD symptom.

Steroids and Neuromodulation of Dopamine Neurotransmission in Animal Models

Sex steroid hormones influence DA systems of the hypothalamus as well as extrahypothalamic regions of the brain in controlling movement and behavior in humans and animals [116]. Several lines of evidence from animal and clinical literature demonstrate that DA activity declines with age [131]. Estrogens modulate the DA system, producing pro‐ or antidopaminergic effects at different levels, behavioral and cellular and at various steps of DA neurotransmission: enzymes involved in the synthesis and degradation of DA, DA release, DA transporter (DAT) and vesicular monoamine transporter 2 (VMAT₂), DA receptors, as well as coupling of DA receptors to their second messengers (Table 2).

Table 2.

Effects of steroids on nigrostriatal and mesolimbic dopamine (DA) neurotransmission in animal models

Assay	Steroid	Species and sex	Brain region	Reported activity	References
DA metabolism
	17β‐estradiol	OVX rat	Striatum	↑ Basal, AMPH‐ or KCl‐ or cocaine evoked DA release	[134, 315, 316, 317, 318]
	17β‐estradiol	OVX rat	Striatum	↑ DA turnover	[135, 319, 320, 321]
	17β‐estradiol	OVX rat	Nucleus accumbens	− DA turnover	[321]
	17β‐estradiol	Female rat	Striatum	↑ Nicotine‐evoked DA release	[322]
	17β‐estradiol	Female rat	Striatum	− On DA levels	[166]
	17β‐estradiol	Male rat	Striatum	↓ Nicotine‐evoked DA release	[322]
	17β‐estradiol	GDX male rat	Striatum	− On DA release	[134]
	17β‐estradiol	Male rat	Striatum	− On AMPH‐stimulated DA release	[134]
	17β‐estradiol	Female mice	Striatum	↑ Basal‐ and KCl‐stimulated DA release	[323]
	17β‐estradiol	Male mice	Striatum	− On basal‐ and KCl‐stimulated DA release	[323]
	17β‐estradiol	OVX monkey	Putamen	↑ 3‐MT levels	[137]
	17β‐estradiol benzoate	Young and old OVX rat	Striatum	↑ Basal DA release ↓ DA levels	[324]
	17β‐estradiol benzoate	Old OVX rat	Striatum	↓ AMPH‐evoked DA release − KCl‐evoked DA release	[324]
	17β‐estradiol benzoate	Young OVX rat	Striatum	↓ KCl‐evoked DA release − AMPH‐evoked DA release	[324]
	17β‐estradiol hemisuccinate	E2‐primed OVX rat	Nucleus accumbens	↓ KCl‐evoked DA release	[133]
	Estrogen (specific molecule not stated)	OVX mice GDX and intact male mice	Striatum	↓ MA‐evoked DA release	[325]
	Estrone	OVX rat	Striatum	− AMPH‐evoked DA release − DA levels	[187, 326]
	Progesterone	Male and OVX rat	Striatum	↑ Spontaneous DA release	[132, 156, 158, 327, 328]
	Progesterone	OVX rat	Striatum	− Spontaneous or KCl‐evoked DA release ↑ DA turnover	[158, 159, 160, 320, 327]
	Progesterone	Intact female rat	Striatum	↑ NMDA‐evoked DA release in proestrus	[329]
	Progesterone	Female Rat	Striatum	↓ (Low dose) and ↑ (higher dose) KCl‐evoked DA release	[159]
	Progesterone	Male rat	Striatum	↑ DA levels and turnover	[160, 330]
	17β‐estradiol + Progesterone	Male and OVX rat	Striatum	↑ DA turnover	[320]
	Testosterone	Male rat	Striatum	↑ DA turnover	[164]
	Testosterone	Female rat	Striatum	↓ DA levels	[166]
	Testosterone	Male and OVX mice	Striatum	− MA‐evoked DA release	[325]
	Testosterone	GDX male mice	Striatum	↓ KCl‐evoked DA release	[167]
	Testosterone	Male rat	Striatum Nucleus accumbens	↑ DA levels	[163]
	DHEA	Male rat	Striatum	↓ DA turnover	[172]
	DHEA	Male and female rat	Striatum	− DA levels	[166, 331]
	DHEA	Male rat	Nucleus accumbens	− DA turnover	[172]
DAT
	17β‐estradiol	OVX rat	Striatum	↑ DAT density (acute and chronic treatment)	[151, 161]
	17β‐estradiol	OVX rat	Striatum Nucleus accumbens SN VTA	− On density Restore DAT mRNA levels 3 months but not 2 weeks after OVX in SN only	[149]
	17β‐estradiol (2 weeks and 4 months after OVX)	OVX rat	Striatum	↑ DAT density	[150]
	17β‐estradiol	OVX rat	Striatum	↓ DAT density	[152]
	17β‐estradiol	OVX monkey	Caudate nucleus Putamen	↑ DAT density	[137]
	17β‐estradiol + Progesterone	OVX rat	Striatum	↓ DAT density	[152]
	17β‐estradiol + Progesterone	OVX rat	Striatum	↑ DAT density	[151]
	Estrone and estriol	Male mice	Striatum SN	− On DATdensity and mRNA levels	[187]
	Progesterone	OVX rat	Striatum	− On DATdensity (acute) ↑ DAT density with no effect on affinity (chronic)	[151, 161]
	Progesterone	OVX rat	Striatum	↓ DAT density	[152]
	α‐methyltestosterone	GDX male rat	Striatum	− On DAT density	[152]
	DHEA	Male and female rat pups	Striatum Nucleus accumbens	↑ DAT immunodensity	[174]
VMAT2
	17β‐estradiol	OVX rat	Striatum	− On VMAT2 density	[150]
	17β‐estradiol	OVX rat	SN Middle striatum Nucleus accumbens	− On VMAT2 mRNA ↓ VMAT2 density in striatum and nucleus accumbens	[154]
	Progesterone	OVX rat	SN Middle striatum Nucleus accumbens	↓ VMAT2 density in striatum and nucleus accumbens ↓ VMAT2 mRNA levels	[154]
	17β‐estradiol + Progesterone	OVX rat	SN Nucleus accumbens	− On VMAT2 mRNA − On VMAT2 density	[154]
	Estrone and estriol	Male mice	Striatum SN	− On VMAT2 density and mRNA levels	[187]
D1 Receptor
	17β‐estradiol	OVX rat	Striatum	↑ D1 receptor density	[332, 333, 334]
	17β‐estradiol	Male rat	Striatum Shell and core regions Nucleus accumbens	− On D1 receptor density	[146]
D2 Receptor
	17β‐estradiol	OVX rat	Striatum	↑ D2 receptor density ↓ Ratio of high to low agonist affinity binding states − On D2 mRNA levels	[148, 319, 334, 335, 336, 337]
	17β‐estradiol	OVX rat	Striatum (anterior part) Nucleus accumbens Olfactory tubercle	↑ Density of D2 in striatum only	[145, 332]
	17β‐estradiol	Male rat	Striatum Nucleus accumbens	↓ D2 receptor mRNA levels − on D2 receptor density	[146]
	17β‐estradiol	OVX rat	Midbrain	↓ D2 receptor density (long isoform)	[147]
D3 Receptor
	17β‐estradiol	OVX rat	Island of Calleja Nucleus accumbens Anterior striatum VTA	↑ D3 receptor density	[145, 147]
	17β‐estradiol	OVX rat	Midbrain	↓ D3 receptor mRNA levels	[147]
TH
	17β‐estradiol benzoate	OVX rat	SN	↑ TH immunoreactive neurons	[338]
	17β‐estradiol	Male rat	Olfactory bulb	↑ TH protein content	[339]
	Progesterone	Male rat	Olfactory bulb	↑ TH protein content	[339]
	Estrone and estriol	Male mice	SN	− on TH mRNA levels	[187]
	17β‐estradiol	OVX monkey	Caudate nucleus Putamen	↑ TH immunoreactive neurons in dorsal regions	[340]
	Progesterone	OVX monkey	Caudate nucleus Putamen	↑ TH immunoreactive neurons in dorsal regions	[340]
	17β‐estradiol + progesterone	OVX monkey	Caudate nucleus Putamen	↑ TH immunoreactive neurons in dorsal regions	[340]
	17β‐estradiol	Male mice	SN pars compacta	↑ TH immunoreactive neurons	[341]

↑ indicated an increase as compared to non‐treated animal; ↓ indicated a decrease as compared to non‐treated animal; ― indicated lack of steroidal effect; 3‐MT, 3‐methoxytyramine; AMPH, amphetamine; DA, dopamine; DAT, dopamine transporter; DHEA, dehydroepiandrosterone; E2, 17β‐estradiol; GDX, gonadectomized; MA, methamphetamine; NMDA, N‐methyl‐d‐aspartate; OVX, ovariectomized; SN, substantia nigra; TH, tyrosine hydroxylase; VMAT₂, vesicular monoamine transporter 2; VTA, ventral tegmental area.

Estrogens

Estrogens administered in an acute, chronic, or pulsatile mode, facilitate the release in DA induced by administration of amphetamine or apomorphine in ovariectomized (OVX) rat striatum [132] and nucleus accumbens [116, 133]. Numerous earlier studies in rats have shown that various steps of dopaminergic neurotransmission vary during the estrous cycle [116]. High plasma concentration of estrogen is correlated with an increased release and degradation of striatal DA induced by amphetamine [133, 134]. Ovariectomy induces a decrease in the release of DA that is corrected by estradiol [134, 135]. There is also a sex difference, males responding with less intensity to amphetamines [136]. Chronic 17β‐estradiol treatment increased 3‐methoxytyramine (3‐MT) concentrations in the intact and lesioned caudate nucleus and putamen of OVX monkeys and increased DA concentrations in the lesioned putamen [137]. The increased 3‐MT levels in OVX monkeys are consistent with previous studies in OVX rats, showing that 17β‐estradiol treatment increases striatal DA turnover [135]. 3‐MT has been shown to be a good index of DA release [138].

Of the five DA receptors characterized, estradiol modulation was reported for D₁, D₂, and D₃ receptors. Striatal D₁ receptor density was shown to fluctuate during the estrous cycle [139]. Ovariectomy induces a decrease in the density of D₁ receptors [140], corrected by chronic treatment with 17β‐estradiol. This is observed when estradiol treatment is initiated shortly after ovariectomy, whereas after a long‐term ovariectomy the receptor is no longer responsive [140, 141].

Clinical and experimental studies have shown that D₂ receptors fluctuate during the menstrual (estrous) cycle [142, 143, 144]. Ovariectomy in rats reduces the density of striatal D₂ receptors corrected by estradiol treatment [116, 145]. This increased density of D₂ receptors by estradiol was also observed in animals lesioned with 6‐hydroxydopamine (6‐OHDA), indicating that estradiol modulates postsynaptic receptors [116]. In these studies, no difference in D₂ receptor mRNA levels was observed [116, 145]. Lammers et al. [146], found no increase in striatal D₂ density in male rats treated chronically with estradiol associated with decreased striatal D₂ receptor mRNA levels. In the midbrain, where DA neurons are located, a study shows that estradiol increased mRNA levels of the long isoform of D₂ receptors [147]. In acute treatment, estradiol does not alter the density of D₂ receptors but induces a change in proportion of high to low affinity agonist sites [148].

In rat, ovariectomy had no effect on D₃ receptor density, measured by autoradiography [145]. Estradiol left unchanged D₃ receptor, mRNA, and protein levels in male rats [146]. Estradiol decreases D₃ receptor density in the Island of Calleja, the nucleus accumbens and the lateral striatum [145]. In the midbrain, a study showed that D₃ receptor levels are reduced by estradiol [147].

In rodents, DAT density was shown to fluctuate during the estrous cycle [116]. A short‐ and long‐term ovariectomy decreased striatal DAT density [149] while in the nucleus accumbens, ovariectomy decreased DAT density only in long term OVX rats. In OVX rats chronically treated with 17β‐estradiol, striatal DAT density is increased [150, 151]. In the substantia nigra pars compacta and ventral tegmental area of these animals, ovariectomy and treatment with 17β‐estradiol had no effect on DAT mRNA levels [149]. By contrast, Attali et al. [152] showed that ovariectomy induced an increase of DAT in striatum, restricted to synaptosomes; this increase was corrected by chronic treatment with estradiol. Zhou et al. [147] found no modulation by estradiol of the DAT in the amygdala, hypothalamus and midbrain, nucleus accumbens, and ventral tegmental area. Supporting the results in rats showing an increase of striatal DAT with estradiol treatment, an increase of DAT specific binding was also observed after a chronic 1 month treatment with 17β‐estradiol in subregions of the caudate nucleus and putamen of OVX monkeys [137]. In addition, a study in postmenopausal women receiving estrogen replacement therapy reported an increase of DAT density in the left anterior putamen [153].

The effect of estrogens on VMAT₂ is less documented. A study reported a decrease of striatal VMAT₂ density by a chronic estradiol treatment [154] and an other no effect [150]. Generally the VMAT₂ is not modulated by gonadal steroids and serves as a biomarker of progression of PD [155].

Progesterone

Early studies have shown that progesterone might modulate DA systems. A microdialysis study in intact male and OVX female rats showed that progesterone at a physiological dose increases striatal spontaneous DA and its metabolites 3,4‐dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) release [156]. In vitro and in vivo stimulated DA release by progesterone was observed in OVX estrogen‐primed female but not in non‐estrogen treated rats [132, 157, 158]. In vitro modification of striatal [³H]DA release in an estrous cycle‐dependant way by progesterone was reported [159]. A physiological dose of progesterone modulates striatal DA activity in female and male rats as shown by a rapid increase in DA, DOPAC, and HVA content [160]. Chronic treatment with a pharmacological dose of progesterone increases DAT density in striatum and substantia nigra pars compacta in OVX female rats [151] whereas acute treatment with a physiological dose of progesterone lacks this effect [161]. Investigation of D₂ DA receptor agonist binding shows that chronic progesterone treatment of OVX rats left striatal [³H]spiperone density and affinity unchanged [148] whereas a decrease in binding density was observed with an acute treatment [162].

Testosterone

Intranasal and subcutaneous administration of testosterone in male rats is reported to increase DA levels in the neostriatum whereas only intranasal administration enhances DA in the nucleus accumbens [163]. Another study showed that chronic treatment with testosterone in male rats left unchanged striatal DA content but increased DOPAC and HVA concentrations as well as DA turnover [164]. In OVX female, but not in gonadectomized (GDX) male rats, striatal DA content was decreased with testosterone treatment whereas both DOPAC and HVA levels showed an increase [165]. Tomas‐Camardiel et al. [166] reported a decrease of both striatal DA and DOPAC contents in female rats treated with testosterone. In GDX male mice treated with testosterone, potassium‐ and reserpine‐induced DA outputs are altered, suggesting that testosterone might modulate the storage and/or the uptake of DA [167]. In GDX male rats, no effect of testosterone administration is observed on DAT density and affinity in prefrontal cortex, sensorimotor cortex and striatum [168]. Treatment with testosterone between P20 and P35 in intact female rats increases striatal D₁ but not D₂ receptors evaluated at P40 whereas no such effect was observed in male rats [169]. Testosterone treatment in GDX male rats reversed the effect of castration on D_2L/D_2S receptor ratio in the pituitary and olfactory tubercle but not in the substantia nigra [170].

Dehydroepiandrosterone (DHEA)

DHEA and DHEAS also influence DA neurotransmission. DHEAS produces an increase in DA release in a dose‐dependent manner in hypothalamic cells [171]. Intrastriatal injection of DHEA did not change DA and HVA contents while DOPAC concentrations were decreased [166]. Similarly, a reduction in DOPAC content and DOPAC/DA ratio is reported in the rat striatum, but not in nucleus accumbens, while DA concentrations remained unchanged [172]. The acute effect of DHEA on monoamine oxidase (MAO) activity has been investigated in striatum and nucleus accumbens both in vivo and in vitro. In the nucleus accumbens, DHEA reduced total MAO activity, an effect not observed in the striatum [173]. When assessed in vitro, DHEA reduced MAO‐A and MAO‐B in both striatum and nucleus accumbens [173]. In postnatal rats, an increase in DAT immunodensity was found in the striatum and in the core of nucleus accumbens with DHEA treatment [174].

Steroids and Protection of the Nigrostriatal Dopamine System in Lesioned Animal Models

Estrogens

Experimental studies in animals highlight that steroids have neuroprotective effects against brain injuries (summarized in Table 3). Sex differences are reported in the 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine (MPTP) model of PD, in methamphetamine (MA) toxicity, and in 6‐OHDA‐induced unilateral lesions of the nigrostriatal DA system; male rodents showing greater sensibility to the toxins as observed by more extensive striatal DA depletion, greater decrease in DAT specific binding in striatum and higher loss of tyrosine hydroxylase positive cells in substantia nigra as compared to females [175, 176, 177, 178, 179, 180]. Estrous cycle variations also support a beneficial role of estrogen in lesioned mice. Female BALB/c mice exhibit a more severe DA loss when MA was administered at diestrus and a lower DA depletion at proestrus whereas this difference is not observed in the C57BL/6J strain of mice [180]. In female rats, striatal DA content was preserved in female exposed to 6‐OHDA at proestrus, whereas female lesioned at diestrus showed a reduction in DA concentrations [181]. Moreover, the loss of tyrosine hydroxylase positive cells in the substantia nigra induced by 6‐OHDA was higher in female rats lesioned at diestrus compared to proestrus, when estrogen levels are elevated [181].

Table 3.

Effect of steroids on dopamine (DA) neurotransmission in animal models of neurodegeneration

Assay	Steroid	Species and sex	Toxin	Brain region	Effects	References
DA concentration
	17β‐estradiol	C57BL/6 intact and GDX male mice	MPTP	Striatum	↑	[176, 182, 184, 185, 186, 187, 188, 189, 210, 342, 343]
	17β‐estradiol	C57BL/6 OVX female mice	MPTP	Striatum	↑	[176, 177, 203, 344]
	17β‐estradiol	C57BL/6 female and male mice	MPTP	Striatum	―	[190]
	17β‐estradiol	Wistar intact female rats	MPP+	Striatum	↑	[166]
	17β‐estradiol	Sprague‐Dawley OVX female rats	6‐OHDA	Striatum	↑	[178, 193]
	17β‐estradiol	Sprague‐Dawley GDX male rats	6‐OHDA	Striatum	↓	[178]
	17β‐estradiol	Wistar OVX female rats	6‐OHDA	Striatum	↑	[198]
	17β‐estradiol (post‐treatment)	Wistar OVX female rats	6‐OHDA	Striatum	―	[198]
	17β‐estradiol	CD‐1 intact and OVX female mice	MA	Striatum	↑	[194, 196, 202, 345]
	17β‐estradiol	CD‐1 intact and GDX male mice	MA	Striatum	↓ or ―	[196, 200]
	17β‐estradiol (post‐treatment)	CD‐1 OVX female mice	MA	Striatum	―	[202]
	17α‐estradiol	C57BL/6 intact male and OVX female mice	MPTP	Striatum	―	[182, 184, 189, 203, 210]
	Estradiol benzoate	C57BL/6 OVX female mice	MPTP	Striatum	↑	[346]
	Estradiol benzoate	C57BL/6 OVX female mice	MA	Striatum	↑	[211, 212]
	Estradiol benzoate	C57BL/6 GDX male mice	MA	Striatum	―	[211]
	Estradiol benzoate	CD‐1 OVX female mice	MA	Striatum	↑	[191, 192, 195, 197]
	Estradiol benzoate (post‐treatment)	CD‐1 GDX female and male mice	MA	Striatum	↓	[195, 215]
	Estrone	C57BL/6 male mice	MPTP	Striatum	―	[187, 190]
	Estriol	C57BL/6 male mice	MPTP	Striatum	―	[187]
	DHEA	C57BL/6 male mice	MPTP	Striatum	↑	[185]
	DHEA	Wistar intact female rats	MPP+	Striatum	↑	[166]
	Progesterone	C57BL/6 male mice	MPTP	Striatum	↑	[183, 188, 210]
	Progesterone	C57BL/6 GDX female and male mice	MA	Striatum	↑	[211, 212]
	17β‐estradiol + progesterone	C57BL/6 male mice	MPTP	Striatum	↑	[188]
	Estradiol benzoate + progesterone	C57BL/6 OVX female mice	MA	Striatum	↑	[212]
	Testosterone	C57BL/6 intact and GDX male mice	MPTP	Striatum	―	[176, 186]
	Testosterone	Wistar intact female rats	MPP+	Striatum	―	[166]
	Testosterone	CD‐1 intact and GDX female and male mice	MA	Striatum	↓ or ―	[194, 196, 214]
	Dihydrotestosterone	C57BL/6 male mice	MPTP	Striatum	―	[186]
	Dihydrotestosterone	Sprague–Dawley GDX male and female rats	6‐OHDA	Striatum	―	[178, 199]
DAT protein
	17β‐estradiol	C57BL/6 female and male mice	MPTP	Striatum	↑	[183, 186, 187, 190, 343]
	17β‐estradiol	C57BL/6 male mice	MPTP (extensive lesion)	Striatum	―	[182]
	17β‐estradiol	C57BL/6 male mice	MPTP	SN	↑	[187]
	17α‐estradiol	C57BL/6 male mice	MPTP	Striatum	―	[182]
	Estradiol benzoate	CD‐1 OVX female mice	MA	Striatum	↑	[191, 192]
	Estradiol benzoate	CD‐1 OVX female mice	MA	SN	*	[192]
	Estrone	C57BL/6 male mice	MPTP	Striatum	↑ or ―	[187, 190]
	Estrone	C57BL/6 male mice	MPTP	SN	―	[187]
	Estriol	C57BL/6 male mice	MPTP	Striatum	―	[187]
	Estriol	C57BL/6 male mice	MPTP	SN	―	[187]
	Progesterone	C57BL/6 male mice	MPTP	Striatum	↑	[183]
	Testosterone	C57BL/6 male mice	MPTP	Striatum	―	[186]
	Dihydrotestosterone	C57BL/6 male mice	MPTP	Striatum	―	[186]
DAT mRNA
	17β‐estradiol	C57BL/6 male mice	MPTP	SN	↑	[185, 186]
	Estradiol benzoate	C57BL/6 OVX female mice	MPTP	SN	↑	[346]
	Estradiol benzoate	CD‐1 OVX female mice	MA	SN	↑	[191]
	Estrone	C57BL/6 male mice	MPTP	SN	*	[187]
	Estriol	C57BL/6 male mice	MPTP	SN	*	[187]
	DHEA	C57BL/6 male mice	MPTP	SN	↑	[185]
	Progesterone	C57BL/6 male mice	MPTP	SN	―	[183]
	Testosterone	C57BL/6 male mice	MPTP	SN	↓	[186]
	Dihydrotestosterone	C57BL/6 male mice	MPTP	SN	―	[186]
TH protein
	17β‐estradiol	C57/blk6 GDX male mice	MPTP	Striatum	↑	[347]
	17β‐estradiol	C57BL/6 female and male mice	MPTP	Striatum	↑	[190, 348]
	17β‐estradiol	C57BL/6 male mice	MPTP	SN	↑	[348]
	17β‐estradiol	Wistar intact female rats	MPP+	Striatum	↑	[166]
	Estradiol benzoate	C57BL/6 OVX female mice	MPTP	SN	↑	[346]
	Estrone	C57BL/6 male mice	MPTP	Striatum	―	[190]
	Estradiol benzoate	Long‐Evans OVX female rats	6‐OHDA	Striatum	↑	[349, 350]
	Estradiol benzoate	Long‐Evans OVX female rats	6‐OHDA	SN	↑	[349, 350]
	17β‐estradiol	Wistar OVX female rats	6‐OHDA	SN	―	[351]
	DHEA	Wistar intact female rats	MPP+	Striatum	↑	[166]
	Testosterone	Wistar intact female rats	MPP+	Striatum	―	[166]
TH mRNA
	17β‐estradiol	C57BL/6 male mice	MPTP	SN	↑	[185]
	Estradiol benzoate	C57BL/6 OVX female mice	MPTP	SN	↑	[346]
	Estrone	C57BL/6 male mice	MPTP	SN	*	[187]
	Estriol	C57BL/6 male mice	MPTP	SN	*	[187]
VMAT2 specific binding
	17β‐estradiol	C57BL/6 male mice	MPTP	Striatum	↑	[186, 187]
	17β‐estradiol	C57BL/6 male mice	MPTP	SN	↑	[187]
	Estradiol benzoate	CD‐1 OVX female mice	MA	Striatum	↑	[192]
	Estrone	C57BL/6 male mice	MPTP	Striatum	―	[187]
	Estrone	C57BL/6 male mice	MPTP	SN	―	[187]
	Estriol	C57BL/6 male mice	MPTP	Striatum	―	[187]
	Estriol	C57BL/6 male mice	MPTP	SN	―	[187]
	Testosterone	C57BL/6 male mice	MPTP	Striatum	―	[186]
	Dihydrotestosterone	C57BL/6 male mice	MPTP	Striatum	―	[186]
VMAT2 mRNA
	17β‐estradiol	C57BL/6 male mice	MPTP	SN	↑	[187]
	Estrone	C57BL/6 male mice	MPTP	SN	↑	[187]
	Estriol	C57BL/6 male mice	MPTP	SN	―	[187]

↑ indicated an increase as compared to lesioned animal; ↓ indicated a decrease as compared to lesioned‐animal; ― indicated lack of steroidal effect; *Indicated no lesion effect. 6‐OHDA, 6‐hydroxydopamine; DA, dopamine; DAT, dopamine transporter; DHEA, dehydroepiandrosterone; GDX, gonadectomized; MA, methamphetamine; MPP+, 1‐methyl‐4‐phenylpyridium; MPTP, 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine; OVX, ovariectomized; SN, substantia nigra; TH, tyrosine hydroxylase; VMAT₂, vesicular monoamine transporter 2.

Pretreatment with a low dose (physiological levels) of 17β‐estradiol prevents MPTP‐induced striatal DA depletion in female and male mice [176, 177, 182, 183, 184, 185, 186, 187, 188, 189], whereas one study reported no protection [190]. By contrast, a high‐dose administered prior and after MPTP lesion did not protect striatal DA [189, 190] (for a more extensive review on dose and treatment, see [175]). A protective effect of 17β‐estradiol and estradiol benzoate against MA‐induced DA toxicity and 6‐OHDA‐induced lesion was also observed with a pretreatment in female rodents [175, 178, 191, 192, 193, 194, 195, 196, 197, 198]. However, 17β‐estradiol or estradiol benzoate does not produce a neuroprotective effect against MA and 6‐OHDA toxicity in males. In GDX 6‐OHDA lesioned male rats, 17β‐estradiol treatment increases 6‐OHDA toxicity [199]. 17β‐estradiol treatment failed to protect MA‐treated GDX male mice [200] and in intact male produces a severe acute toxicity to MA, with a high incidence of mortality [194]. Estradiol administered after the lesion failed to provide a protective effect against MA and 6‐OHDA toxicities in female rodents [195, 201, 202]. Thus, estradiol can prevent but not reverse striatal neurotoxicity induced by MA and 6‐OHDA in female but not male rodents.

The stereoisomer 17α‐estradiol has less affinity for estrogen receptors (ERs) and shows no protective effect against MPTP in vivo[182, 189, 203]. Estriol lacks DA neuroprotective activity whereas estrone shows weak protection (Table 3) [187].

The increase DA levels observed in lesioned animals treated with estradiol could be caused by enhanced synthesis or reduced metabolism by a hormonal effect on enzymes in the metabolic pathway of DA. However, this explanation cannot be extended to all the changes observed of the DA markers measured. The DAT and the VMAT₂ play an important role in MPTP‐ and MA‐induced toxicity [204, 205, 206, 207]. These transporters are commonly used to evaluate DA terminals and cell bodies integrity. Estradiol treatments prevent the decrease in [¹²⁵I]RTI‐121 binding to DAT and [³H]TBZ‐OH binding to VMAT₂ in striatum and substantia nigra induced by MPTP and MA in mice [183, 187, 191, 192]. The mRNA levels of these transporters in substantia nigra are also preserved by estradiol treatment [187]. Moreover, the decrease in DAT specific binding or mRNA levels in substantia nigra is lower than that observed in DAT specific binding in striatum of MPTP and MA treated mice, suggesting that the presynaptic terminals of DA neurons in the striatum are more affected than their cell bodies in substantia nigra [186, 208]. With the administration of lower concentrations of toxins such as MPTP and 6‐OHDA, the loss of tyrosine hydroxylase‐immunolabeled dendrites in substantia nigra is smaller than at higher toxins’ concentrations where the dendritic loss is more extensive and associated with a disintegration of DA cell bodies [209].

The early stage of degeneration, where DA neurons in substantia nigra are injured but not dead, provides an appropriate period for neuroprotection by steroids. With an extensive MPTP lesion in mice, 17β‐estradiol failed to protect against the loss of DAT specific binding induced by MPTP [182]. In human, the beneficial effect of estrogens is observed in the early stages of PD, before initiating levodopa therapy [119].

Progesterone

Few studies have investigated the neuroprotective action of progesterone on animal models of PD (Table 3). Our laboratory has shown that pretreatment with a low dose of progesterone prevents the decrease of striatal DA content in MPTP‐treated male mice [183, 188, 210]. Interestingly, when a low dose of 17β‐estradiol and progesterone was coadministered, progesterone did not oppose the beneficial effect of estradiol on MPTP toxicity [188]. In MA treated‐mice, pretreatment with progesterone used at a high dose shows neuroprotective properties in OVX female mice on striatal DA concentrations whereas the low dose failed to induce any beneficial effect [211, 212]. In contrast, the low dose of progesterone administered in pretreatment displayed a beneficial effect on MA‐induced DA depletion in GDX male mice [211]. Moreover, a two‐day pretreatment with a low dose of estradiol benzoate followed by a 1‐day treatment with a high dose of progesterone resulted in a beneficial effect on DA content in MA‐treated OVX female mice [212]. To our knowledge no study investigated the potential protective properties of progesterone against 6‐OHDA toxicity.

Androgens

Testosterone is one of the main androgenic steroid synthesized by the testis biotransformed in the brain into estradiol by an aromatization process (Figure 1). Testosterone can also be biotransformed in dihydrotestosterone but this steroid cannot be converted into 17β‐estradiol. A pretreatment with testosterone failed to protect MPTP male mice, suggesting that testosterone is not converted in the brain into estradiol in sufficient concentration to achieve neuroprotective levels [186, 213] (Table 3). Androgenic activity, evaluated with dihydrotestosterone, is not able to protect striatal DA against MPTP toxicity in male mice [186] and in 6‐OHDA lesioned GDX female and male rats [178, 199]. In the MA model of degeneration, pretreatment with testosterone did not protect against MA‐induced striatal DA loss in intact and OVX female [196, 214] and acute treatment with testosterone in male mice enhances striatal DA depletion induced by MA [214].

Dehydroepiandrosterone (DHEA)

In male mice, a pretreatment with DHEA shows neuroprotective properties against MPTP toxicity [185] (Table 3). A study with rats lesioned with 1‐methyl‐4‐phenylpyridium (MPP+) also observed a protective effect of DHEA on striatal DA content, on tyrosine hydroxylase positive fibers and an attenuation of microglia activation induced by the toxin [166]. When tested in a condition of impaired nigrostriatal dopaminergic system, that is, after MA administration, DHEA failed to enhance neurodegenerative or neuroprotective effects against a second MA invasion, suggesting that DHEA cannot exert a neuroprotective effect when the brain is damaged [215].

New Pharmacological Approaches

A variety of steroidal and nonsteroidal compounds that interact with ERs as contraceptives and for the treatment of breast cancer, uterine dysfunction, and other reproductive disorders are available. A role of these drugs in nontarget tissues such as skeleton, the cardiovascular system, and the CNS was recognized [216] but the WHI studies casts doubts on their benefits over the risk [217, 218]. Several shortcomings of the WHI studies have let to additional studies to investigate, among other important factors, the possible “window of opportunity”[219] to have beneficial hormonal effects after menopause and if all estrogen and progestative compounds are equal (likely not) [220]. Hence, ERs have been targeted for therapeutic approaches to prevent cognition decline with aging or the development of affective and cognitive disorders after menopause [221]. Finding new drugs effective in reducing symptoms associated with menopause without increasing the cancer risk is of great interest.

SERMs

Selective estrogen receptor modulators (SERMs) either synthetic or natural, such as phytoestrogens, may represent an alternative to estradiol for treatment or prevention of neurodegenerative disorders. SERMs are estrogenic molecules developed to reproduce the beneficial effects of estrogen in the prevention and treatment of osteoporosis, for the cardiovascular system and to affect mood without adverse effects on hormone‐dependent cancers. SERMs have the same beneficial effect as estrogen in skeleton and cardiovascular systems but act as antagonists in breast and uterus [74]. SERMs could be an alternative to estradiol for treatment in both men and women. Synthetic SERMs, such as tamoxifen, raloxifene, or bazedoxifen [183, 197, 222, 223, 224, 225, 226, 227, 228, 229] and natural SERMs such as genistein [230], are neuroprotective in vitro and in vivo. New SERMs might be developed that would lack feminizing effects and would preferentially target the nervous system (NeuroSERMs) to promote cognitive function and to reduce the risk of neurodegenerative diseases [229]. Tamoxifen and raloxifene modulate and protect DA systems [74, 145, 210, 231] and these drugs may have a different activity on DA systems by specific ER subtype activation.

Tamoxifen is a first generation SERM. It has little affinity for ERs but is metabolized into 4‐hydroxytamoxifen that has high affinity for ERs [232]. Tamoxifen is used in prevention and treatment of breast cancer, it has estrogen antagonist activity in the mammary gland [233]. For 25 years, more than 10 million women have been treated with this SERM [233]. Most studies on the impact of tamoxifen therapy on quality of life in breast cancer survivors show that the frequency of depression in women taking tamoxifen was similar to the general population [234, 235, 236, 237, 238], except one study reporting that clinical depression is a side effect of tamoxifen therapy [239]. Tamoxifen, but not estradiol, was found to be a weak competitive inhibitor of the DA antagonist site of D₂ receptors in striatal membranes [240, 241]. In animals treated with estradiol, tamoxifen antagonizes both the increases in [³H]spiperone binding sites to D₂ receptors and stereotyped behavior induced by apomorphine [242]. Tamoxifen also mimics some of the estradiol activities on DA systems. In pituitary, tamoxifen increases the number of adenohypophysal DA receptor binding labeled with [³H]spiperone [242]. McDermott et al. [243, 244, 245, 246] investigated the interactive effects of tamoxifen and estrogen upon the nigrostriatal DA system. Their findings suggest that tamoxifen, such as estradiol, can alter DA output through direct, nongenomic effects upon striatal neurons [243, 244, 245, 246]. In addition, an indirect effect of tamoxifen on DA activity has been shown, because tamoxifen reduces nigral glutamic acid decarboxylase activity [247]. Tamoxifen is also shown to increase striatal extracellular levels of DA and DOPAC in freely moving male rats [248].

Raloxifene is a second generation SERM affine for ERs [249]. It is used in postmenopausal women in prevention and treatment of osteoporosis [250] and unlike tamoxifen, it is not a partial estrogenic agonist at the uterine endometrium. Thus, taking long‐term raloxifene does not induce an increased risk of developing cancer of the uterus. In the mammary gland, raloxifene is an estrogen antagonist, as tamoxifen [233]. Postmenopausal women were shown to be responsive to raloxifene (3 years of treatment and ongoing) displaying a lower risk of mild cognitive impairment compared to those taking placebo [251]. OVX rats treated for 2 weeks with tamoxifen and raloxifene have increased striatal DAT specific binding [150].

Estrogen Receptors

The transcriptional activity of estrogens is mediated by activation of two nuclear receptors, ERα and ERβ[252] and numerous splice variants of these receptors are reported [229]. These receptors evoke principally a genomic action through target genes transcription activation [252]. Estrogens also induce fast responses within milliseconds, associated with the membrane and involving activation of second messenger mechanisms [252]. These rapid effects can be achieved through ERα or ERβ activation localized at the membrane [253]. ERα is abundantly expressed in many brain areas, including the hippocampus, hypothalamic/preoptic continuum, amygdala, midbrain, dorsal horn of the spinal cord and dorsal root ganglia [254]. ERβ is also expressed in these regions [254, 255, 256], and in other brain regions, such as the cerebellum, it is the predominant form of classical ER [254].

ER specific ligands provide a pharmacological approach to study implication of each ER. 4,4′,4″‐(4‐Propyl‐[1H]‐pyrazole‐1,3,5‐triyl)tris‐phenol (PPT) displays 400‐fold more binding affinity for ERα than for ERβ and is inactive on ERβ transcriptional activity [257]. In vivo, PPT, as estradiol, increases uterine weight of OVX rats [258]. Chronic administration of PPT reproduced the effect of estradiol on the prevention of weight gain and bone loss after ovariectomy, stimulates mRNA levels of complement C3 in the uterus [258]. PPT crosses the blood–brain barrier and reproduces the effect of estradiol in the hypothalamus, known to involve ERα, namely increased progesterone receptor mRNA levels, prevention of hot flashes and increased body temperature [258]. There is also methyl‐piperidinopyrazole (MPP), a receptor antagonist with 200 times greater affinity for ERα than ERβ that inhibits the effect of estradiol [257].

ERβ agonists include 2,3‐bis(4‐hydroxyphenyl)‐propionitrile (DPN) the specific ERβ agonist most effective displaying 100‐fold more affinity for ERβ than ERα and 170‐fold greater relative potency in transcriptional assays for ERβ than ERα[259]. This compound does not stimulate the uterus [260]; it crosses the blood–brain barrier and induces effects in the brain [261].

Activation of both ERα and ERβ with estradiol or specific agonists has been implicated in learning and memory [262]. In rats, both DPN and PPT treatment enhance memory [263]. In mice, estradiol and DPN improved performance on cognitive tests in wild type but not in ERβ knockout mice (ERKOβ) [264]. Also, ERα‐selective agonists are considered anxiogenic, whereas ERβ‐selective compounds are anxiolytic and antidepressive [261, 265, 266].

Our laboratory investigated ERα and ERβ specific agonists to delineate the implication of each receptor subtype in the effect of estradiol (for more details see review: [188]). Relevant to this review, D₂ receptors in the striatum and the core of the nucleus accumbens as well as DAT in the middle striatum are decreased 2 weeks after ovariectomy in rats [150, 267]. These decreases were prevented by treatment with 17β‐estradiol and the ERβ agonists DPN but not by ERα agonists PPT, suggesting the implication of ERβ. The uterus weight of these rats was not stimulated by DPN [150].

To support our results using specific ER agonists we investigated the effect of 17β‐estradiol in young and old OVX female and male ERKOβ mice [268]. Chronic treatment with 17β‐estradiol left unchanged DA, D₂ receptors, and DAT specific binding in the striatum and nucleus accumbens of old ERKOβ male and OVX female mice [268]. Our results suggest that modulation of D₂ receptors and DAT by 17β‐estradiol in the striatum and nucleus accumbens are through the activation of ERβ[268].

GPR30, now designated G protein‐coupled estrogen receptor 1 (GPER1), is proposed to function as a membrane ER [269, 270] but some controversy regarding its role as an ER still remains [271, 272, 273]. Immunohistochemical studies have revealed a high expression of GPER1 in brain regions such as striatum, substantia nigra, Islands of Calleja, hippocampus, and the hypothalamic–pituitary axis [274]. Cellular localization of GPER1 in the brain is found at the plasma membrane, endoplasmic reticulum, and Golgi apparatus [269, 275, 276]. GPER1 is able to mediate both rapid and transcriptional actions in response to estrogens in the brain and in periphery [277]. 17β‐estradiol binds to GPER1 with high affinity whereas 17α‐estradiol fails to bind to this receptor [270]. Competitive binding assays also demonstrated that estrone, estriol, progesterone and testosterone fail to bind GPER1 whereas tamoxifen, the ER antagonist ICI 182,780 and the phytoestrogen genistein display significant binding to GPER1 [270, 278]. No study has yet investigated the role of GPER1 on brain DA systems. Specific agonist [279] and antagonist [280] of GPER1 have been identified, with no detectable activity on either ERα or ERβ. These compounds, and the GPER1 knockout mice [281], could be useful to elucidate the physiological role of this receptor in brain function.

Discussion

The abundant literature reviewed shows a sex difference in schizophrenia, PD, and various depressive states. The effect of steroid treatments in these diseases supports a role of gonadal steroids in the sex differences in addition to a possible genetic effect. An extensive literature in animal models shows a modulatory role of estrogens on brain DA neurotransmission. This is also documented for progesterone and androgens. Hence, estrogens could play a beneficial role for brain DA activity. Estrogenic drugs are not all equal and the endogenous 17β‐estradiol should be a first choice. Progesterone may also be useful and should be considered first before the synthetic analogs such as medroxyprogesterone acetate that may not share similar activities in the brain [220]. However, estrogens and androgens may not be suitable for persons at risk of cancer. In addition, androgens may be abused. Then, DHEA could be an alterative for men and women and serve as a precursor of 17β‐estradiol and testosterone because the converting enzymes are present in the brain [282]. Moreover, DHEA is taken safely in high doses compared to estrogens and androgens and only partial conversion would be required to reach beneficial estrogen and androgen brain levels.

Alternatively, SERMs could be useful for their estrogenic activity in the brain in both men and women. Raloxifene was administered to men for several months without significant side‐effects [283]. Novel SERMs and NeuroSERMs are possible future alternatives for estrogenic activity in the brain. Agonists specific for an ER subtype may help dissociate central from peripheral estrogenic activity. ERβ agonists have raised interest in treatment of depression [284]. It is also important to think outside the classical mechanisms of action of steroids on their nuclear receptors and explore the membrane receptors either for the classical ERα and ERβ but also the more recently studied GPER1 [277]. The steroidal effects on growth factors as well as pro‐ and anti‐apoptotic factors in the mitogen‐activated protein kinase/extracellular signal‐regulated kinase (MAPK/ERK) and phosphatidylinositol‐3 kinase (PI3K) pathways are fertile areas of recent research [175, 285].

Progesterone is also a fertile field for future development. Progesterone in humans was shown to be beneficial after a trauma to accelerate recovery [286, 287]. Moreover, numerous progesterone analogs are available in the clinic for endocrine applications and their effect in the brain is not fully investigated [288]. In addition, several metabolites of progesterone show brain activity [289]. Progesterone was shown to be neuroprotective in male mice brain at lower concentrations than those used in endocrine studies [211], suggesting the implication of different mechanisms than activation of classical nuclear progesterone receptors. Membrane receptors for progesterone [288] have been characterized and may bring novel mechanisms of action in the brain for progesterone.

In conclusion, gonadal steroids modulate brain DA activity and this could be used in adjunct treatment of CNS diseases implicating this catecholamine. If the endogenous gonadal steroids are not chosen for various reasons, analogs are available already used in the clinic for endocrine or cancer treatments. Numerous drugs are under development and novel mechanisms of actions are explored thus providing good future therapies.

Conflict of Interest

There is no conflict of interest for any of the authors.

Disclosure

None of the authors had compensation from any source possibly influencing the objectivity of the report.