Elsevier

Experimental Neurology

Volume 259, September 2014, Pages 44-56
Experimental Neurology

Review
Sex differences in Parkinson's disease and other movement disorders

https://doi.org/10.1016/j.expneurol.2014.03.010Get rights and content

Highlights

  • Estrogen helps maintain the dopaminergic neuron population in the basal ganglia.

  • Estrogen and related compounds are neuroprotective in toxin-mediated animal models of Parkinson’s disease.

  • Estrogen’s neuroprotective effects are at least partially mediated by estrogen receptors, specifically ERa.

  • Caffeine lowers risk of PD in males more than females, and estrogen may limit caffeine’s neuroprotective effect in females.

  • Both the X and Y chromosomes may independently contribute to the development of dopaminergic pathways.

  • In tics and Tourette’s syndrome, modulation of the androgen system in males may improve symptoms.

Abstract

Movement disorders including Parkinson's disease (PD), Huntington's disease (HD), chorea, tics, and Tourette's syndrome (TS) display sex differences in disease susceptibility, disease pathogenesis, and clinical presentation. PD is more common in males than in females. Epidemiologic studies suggest that exposure to endogenous and exogenous estrogen contributes to these sex differences. There is extensive evidence that estrogen prevents dopaminergic neuron depletion induced by neurotoxins in PD animal models and therefore is neuroprotective. Estrogen may also decrease the efficacy of other neuroprotective substances such as caffeine in females but not males. Sex chromosomes can exert effects independent of sex steroid hormones on the development and maintenance of the dopamine system. As a result of hormone, chromosome and other unknown effects, there are sexual dimorphisms in the basal ganglia, and at the molecular levels in dopaminergic neurons that may lead to distinct mechanisms of pathogenesis in males and females. In this review, we summarize the evidence that estrogen and selective estrogen receptor modulators are neuroprotective in PD and discuss potential mechanisms of action. We also briefly review how sex differences in basal ganglia function and dopaminergic pathways may impact HD, chorea, and tics/Tourette's syndrome. Further understanding of these sex differences may lead to novel therapeutic strategies for prevention and treatment of these diseases.

Introduction

Movement disorders are a diverse group of neurologic conditions that can be grouped broadly into hyperkinetic and hypokinetic disorders. There are significant sex differences in the pathophysiology, epidemiology and clinical manifestations of many of these diseases. The unifying pathophysiology of these various diseases relates to dysfunction of the basal ganglia and interconnected pathways. The basal ganglia are compromised of the caudate and putamen, also known together as the striatum, the globus pallidus, the subthalamic nucleus, and the substantia nigra pars compacta and pars reticulata (SNpc and SNpr), which contain dopaminergic neurons. Dopamine (DA) is one of the major regulatory neurostransmitters of the basal ganglia, and dopaminergic system dysfunction can manifest in highly divergent clinical presentations.

The most common disease of dopamine dysfunction is Parkinson's disease (PD). There are sex disparities in PD, with a higher incidence and prevalence of PD in men. Sexual dimorphisms in non-diseased basal ganglia and substantia nigra may partly explain this sex-specific risk (Beyer et al., 1991). Estrogen may also account for some of these differences, and we will review the evidence that estrogen is neuroprotective to the dopaminergic system. In addition, chromosome differences may contribute to the sex differences noted in PD, with interplay between chromosomal factors and gonadal hormone factors. We will review the evidence available that implicates the male Y chromosome in increased risk of PD in men. Tourette's syndrome (TS) and tics, both hyperkinetic movement disorders, are also much more common in men than women. Conversely, the genetic movement disorder of dopa responsive dystonia occurs more often in women due to higher penetrance, and chorea associated with pregnancy occurs exclusively in women. We will also briefly review the limited literature that describes potential mechanisms by which sex may impact disease susceptibility, disease pathogenesis, and clinical presentation of these other movement disorders.

PD occurs more often in men than in women, with a meta-analysis reporting an increased relative risk of 1.5 (Wooten et al., 2004). PD incidence rates are twice as high in men compared to women at all ages in an Italian population, and 91% higher in men in the Kaiser Permanente Medical Care Program in Northern California (Baldereschi et al., 2000). Because estrogen is known to exert effects on dopamine synthesis and function, clinical studies have focused on the correlation between estrogen exposure and PD risk. Rocca and colleagues evaluated the risk of PD (or parkinsonism) in the Mayo Clinic Cohort Study of Oophorectomy and Aging, which included over 2000 patients and controls. In women who had undergone unilateral oophorectomy, the risk of PD was increased, but was only statistically significant if the surgery was performed before age 42. In women who underwent bilateral oophorectomy, there was a significantly increased risk of PD with a hazard ratio of 1.8 (Rocca et al., 2008). In a case-control study, PD was associated with a lower cumulative estrogen exposure during life, or shorter fertile lifespan (R S-P et al., 2009, Ragonese et al., 2004). Similarly, increased length of endogenous estrogen exposure was associated with older age of onset and less severe motor impairment in a cross-sectional study of women with PD (Cereda et al., 2013). Women with PD were less likely to have used HRT, and postmenopausal use of HRT correlated with lowered PD risk in another case-control study (Currie et al., 2004). However, this was not the case in a group of women studied by Marder and colleagues where the use of HRT did not decrease the risk of PD, but did decrease the risk of developing PD with dementia (Marder et al., 1998). Similarly, Simon and colleagues did not find an increased risk of PD associated with any endogenous or exogenous estrogen exposure marker in the Nurses' Health Study (Simon et al., 2009). Therefore, although both endogenous and exogenous estrogen can be potentially protective, it remains unclear to what degree estrogen exposure contributes to the risk of developing PD in the majority of women with this disease.

An epidemiologic phenomenon possibly related to both PD and estrogen exposure is pesticides. Exposure to various pesticides may increase the risk of PD (Brown et al., 2006). Moreover, many pesticides either mimic or block estrogen. While a causal relationship between pesticide exposure, estrogen and PD risk has not been established, the underlying mechanisms may be elucidated by studying the effect of pesticides on estrogen pathways and sex differences in exposure-related risk.

Estrogen may also affect the clinical presentation of women with PD. Estrogen has been shown to both improve and worsen symptoms. Much of the older literature in this field points to estrogen as an anti-dopaminergic agent (Koller et al., 1982, Quinn and Marsden, 1986). A case was reported in which PD symptoms improved with pharmacologically induced menopause (Session et al., 1994). On the other hand, there are more recent and numerous reports of estrogen improving the motor symptoms of PD. In a randomized placebo-controlled study of oral conjugated estrogen, patients receiving estrogen had significantly more “on” time and a clinically significant decrease in motor scores over 8 weeks of treatment (Tsang et al., 2000). In the POETRY study, a pilot study of HRT in postmenopausal women with PD, there was a trend toward improvement in motor scores, however this did not reach statistical significance and the study was limited by under-enrollment (Anon, 2011).

Estrogen is likely a contributor to the sex differences observed in PD prevalence. As discussed above, longer estrogen exposure during a female's lifetime may decrease the risk of PD. Most women develop PD after menopause, which suggests that estrogen withdrawal may be related to the pathogenesis of the disease. Furthermore, estrogen appears to protect against dopaminergic neuron loss in both disease and non-disease states. It has been shown that dopaminergic neuron loss occurs after ovariectomy in rats and primates (Le Saux and Di Paolo, 2006) and this can be reversed with administration of estrogen compounds. A study in monkeys demonstrated that estradiol altered DA metabolism and transporter uptake in the brain after surgically induced menopause (Morissette and Di Paolo, 2009). Interestingly, this post-ovariectomy DA loss also manifests clinically as decreased spontaneous locomotor activity in rats, which can be reversed with administration of exogenous estrogen (Ohtani et al., 2001). Although the clinical manifestations of post-menopausal DA loss have not been studied in non-parkinsonian human females, a small pilot study showed that estrogen replacement therapy in non-parkinsonian women increased putamenal dopamine active transporter (DAT) as measured by TRODAT SPECT scan (Gardiner et al., 2004). With evidence pointing more consistently to estrogen as a pro-dopaminergic agent, as well as exciting implications for estrogen as a neuroprotective agent in ischemia and other neuropathologic processes, there has been a great deal of research into estrogen's potential neuroprotective effects on dopaminergic neurons. The majority of this research has utilized animal neurotoxin-mediated models of PD. 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) is a by-product of meperidine synthesis that is well known to cause parkinsonism in humans, and is probably the most widely utilized animal model for PD. Male mice are more sensitive to MPTP, in terms of striatal dopaminergic neuron loss, than female mice (Dluzen et al., 1996). Methamphetamine (MA) also causes degeneration of striatal dopaminergic neurons in animals and humans, and has a greater negative effect on male mice compared to female mice (Miller et al., 1998). If estrogen accounts for this differential susceptibility, then exogenous estrogen administration should be able to further protect from DA neuron loss in these PD animal models. This hypothesis has been studied extensively and the results are summarized in Table 1.1, Table 1.2, Table 1.3, Table 1.4. Overall, certain formulations of estrogen have been shown to have a neuroprotective effect against a variety of toxic substances (MPTP, 1-methyl-4-phenylpyridinium (MPP), MA, 6-hydroxydopamine (6-OHDA)). Neuroprotection after neurotoxin exposure in these animal studies is defined in two ways: 1) prevention of reduction in striatal dopamine and its related metabolites dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA); and 2) maintenance of dopamine neuronal integrity as measured by dopamine active transporter (DAT), vesicular monoamine transporter 2 (VMAT2) specific binding and mRNA, or tyrosine hydroxylase (TH) immunoreactivity. Animals were treated with estrogen compounds at different time points including before, simultaneously with, and after toxin treatment. Results demonstrate that estrogen treatment needs to precede toxin administration to observe a protective effect. Treatment after the toxic insult cannot reverse the injury. The majority of the studies utilized male mice or female mice that had been ovariectomized at 6 weeks old. Neuroprotective effects have been shown most consistently with 17β-estradiol. This stereoisomer has the highest affinity for the estrogen receptor, suggesting that the neuroprotective effects are receptor-mediated (Kuiper et al., 1997). Other estrogen compounds with lower affinity for the estrogen receptor include 17α-estradiol, estriol and estone. Our review of the literature yielded 31 studies, of which 27 demonstrated that 17β-estradiol has some neuroprotective benefit (Table 1.2). The majority of successful experiments administered low doses of 17β-estradiol mimicking physiologic levels (1 μg, usually as a slow release pellet) 1–10 days before toxin administration. One study in contrast reported a neuroprotective effect using high but not low dose 17β-estradiol (Ookubo et al., 2008). While striatal dopamine and dopamine metabolite concentrations were more consistently protected, the few studies using markers of SNpc DA neuronal loss did not consistently show a protective effect. For instance, Ferraz and colleagues did not find a decrease in SNpc dopamine neuron loss after treatment with 6-OHDA (Ferraz et al., 2003, Ferraz et al., 2008). Estrogen may therefore alter the release or metabolism of dopamine in the striatum without directly affecting SN dopamine neuronal loss. Further research is needed to distinguish this neuromodulatory effect from a true neuroprotective effect. The results in males were also less consistent than in females. All studies using females ovariectomized at 6 weeks showed neuroprotection when 17β-estradiol was administered before the toxin exposure. One study using females ovariectomized before puberty was negative (Anderson et al., 2005) and another study using two year old intact females was negative (Miller et al., 1998). Interestingly, studies using a similar protocol but using the MA toxicity model instead of MPTP showed different results for male and female mice. Low dose pellet formulation of 17β-estradiol protected DA concentrations in females but not males (Anderson et al., 2005, Dluzen et al., 2001, Dluzen et al., 2002, Gao and Dluzen, 2001b). Therefore, it seems that MA has different mechanisms of action in males and females that can be differentially affected by estrogen administration resulting in neuroprotection in females only.

Estrogen compounds with lower affinity for the estrogen receptor include 17α-estradiol, estriol and estrone. Although 17α-estradiol has previously demonstrated neuroprotective effects in other experimental models of brain injury, it failed to show neuroprotection in male and female MPTP treated mice (Table 1.3). A single study showed protection of all markers assayed with estradiol in ovariectomized females (Liu et al., 2008). Two studies report that estrone does not protect striatal dopamine concentration, DAT binding or substantia nigra VMAT2 mRNA levels (Jourdain et al., 2005, Ookubo et al., 2008). Estriol failed to protect DAT and VMAT2 specific binding in another study (Jourdain et al., 2005). There were 7 total studies of estradiol benzoate, and all that used the compound after toxin administration reported a neuroprotective effect (Table 1.3). Furthermore, Yu and colleagues showed that estradiol benzoate protected DA and DOPAC concentrations in males or females gonadectomized at 6 weeks but not at 4 weeks (Yu et al., 2002). Thus, the timing of exposure to endogenous steroids during development appears to influence the response to exogenous steroids later in life. Progesterone acts on its own receptors, but depending on tissue location and estrogen levels, it has interconnected effects on estrogen pathways. In MPTP models, progesterone protected against decreases in striatal DA, DOPAC and HVA concentrations (Grandbois et al., 2000) and DAT specific binding (Callier et al., 2000). Morissette and colleagues combined physiologic doses of 17β-estradiol and progesterone and demonstrated that while there was no additive neuroprotective effect, progesterone did not block the neuroprotective effect of 17β-estradiol (Morissette et al., 2008). In MA models, progesterone prevented striatal DA loss at low doses in males gonadectomized at 6 weeks, but only at high doses in ovariectomized females (Yu et al., 2002). Progesterone may therefore be a promising neuroprotective agent, either in isolation at physiologic doses or in combination with estrogen. This needs to be studied further to elucidate the different effects between dosage regimens, sexes and neurotoxin models.

Tamoxifen and raloxifene are selective estrogen receptors modulators (SERMs) used clinically to treat breast cancer and osteoporosis, respectively. These compounds act as estrogen receptor antagonists in the target tissue of these diseases, but can also act as estrogen receptor agonists in other tissues. There are also experimental compounds that target estrogen receptor α or β specifically, and new compounds are currently being developed that could selectively stimulate human estrogen receptors without causing unwanted side effects that would limit clinical utility. The current literature supports both tamoxifen and raloxifene as neuroprotective agents in PD animal models; 10 of 13 articles reported protective effects (Table 1.1) A single study reported that tamoxifen blocked the neuroprotective effect of 17β-estradiol, however this study used higher doses of tamoxifen than other studies reviewed (Gao and Dluzen, 2001). Interestingly, this finding has clinical relevance. In a cohort study of 15,440 women with breast cancer, women on tamoxifen had an increased hazard ratio for PD (HR = 5.1), but only within a delayed period of 4 to 6 years after initiation of treatment (Latourelle et al., 2010). Thus, tamoxifen may increase the risk of PD when used in higher dosage.

In summary, the neuroprotective or neurotoxic effects of estrogens and SERMs depend on several variables including specific estrogen or SERM compound, dose and duration of treatment, and sex. Another key factor is timing of hormone administration relative to gonadal development or gonadectomy. It is therefore important whether these steroid compounds are given early or late in life and pre- or post-menopause. Furthermore, long terms effects have yet to be evaluated, as most of the animal studies involve sacrificing the subjects within days of toxin exposure. Finally, animal neurotoxin models of PD have an entirely distinct mechanism of disease pathogenesis from idiopathic PD in humans. Estrogen may be protective in these neurotoxin models specifically by decreasing the uptake of the toxin into dopaminergic neurons (Le Saux and Di Paolo, 2006). Studies using genetic animal models would be helpful to characterize the ability of estrogen to protect dopamine neurons in the presence of disease-causing or disease-susceptibility mutations. In order to design human clinical trials, the effects on the healthy and diseased dopaminergic system also needs to be more clearly defined, and risks of peripheral systemic side effects of hormonal therapies considered. It therefore remains unclear how estrogen's neuroprotective effects may be translated from animal models to humans.

Androgens including testosterone and dehydroepiandrosterone (DHEA) failed to show neuroprotection (Table 1.4). In both MA and MPTP-treated mice, pre-treatment with androgen compounds actually increased the detrimental effects on markers of striatal DA integrity in males but not females (Ekue et al., 2002). The detrimental effect of testosterone on striatal dopamine is eliminated by gonadectomy, but restored by subsequent estrogen treatment, suggesting that testosterone aromatized to estrogen in the brain is actually responsible (Gillies and McArthur, 2010). From a clinical perspective, testosterone supplementation has been investigated as an adjunct therapy for motor and non-motor symptoms of PD. Men with PD were found to have similar to higher rates of testosterone deficiency than the general population, but supplemental testosterone failed to show significant improvement in motor or non-motor outcomes (Okun et al., 2006). However, this study was designed to look at symptomatic benefit using a patient questionnaire, and was not intended to look at neuroprotection.

Researchers have proposed several different pathogenic mechanisms in idiopathic PD, including a multiple hit hypothesis. These mechanisms include oxidative stress, mitochondrial dysfunction, glutamate excitotoxicity and dysfunction of organelle trafficking and degradation pathways. Estrogen's neuroprotective effects may be mediated by some of these same mechanisms, making it relevant not only to experimental neurotoxin-mediated PD models but also to idiopathic PD pathogenesis (Fig. 1). Estrogen produces downstream effects by both non-genomic actions, such as activation of signal pathways, and genomic effects involving gene transcription. Both ERα and ERβ receptors localize in the mouse striatum (Kuppers and Beyer, 1999). Al-Sweidi and colleagues studied ERα and ERβ knock out (KO) mice, and demonstrated that ERKOα mice were more sensitive to the striatal effects of MPTP than wild type or ERKOβ mice. 17β-estradiol failed to protect striatal DAT or VMAT2 specific binding in ERα or ERβ KO mice treated with MPTP (Al-Sweidi et al., 2011). They concluded that ERα plays the primary role in mediating neuroprotection and that ERβ plays a supportive role. This is further supported by studies utilizing propyl-pyrazole-triol (PPT) and diarylpropionitrile (DPT) as specific ERα and ERβ agonists, respectively, that found PPT but not DPT was neuroprotective (Baraka et al., 2011).

17β-estradiol acts on the MAPK/ERK and P13K/Akt pathways, and exerts neuroprotective effects from glutamate, H2O2, and β-amyloid toxicity through these pathways. These signal cascades function broadly in cell survival. MAPK/ERK activation leads to inhibition of pro-apoptotic proteins such as BAD and GSK3β. GSK3β in particular has been shown to mediate toxin-induced striatal neuron death. It is a constitutively active protein kinase that is inhibited by Akt via phosphorylation at serine residues. MPTP decreases GSKβ phosphorylation, thereby allowing GSK3β to remain active which leads to neuronal death. MPTP also decreases the Bcl-2 (pro-survival):BAD (pro-apoptotic) ratio. A specific ERα agonist (PPT), and to a lesser degree an ERβ agonist (DPT), were able to block both of these effects of MPTP (Bourque et al., 2009). The other neurotoxin models of PD, 6-OHDA and MA, also affect Akt and GSK3β. Importantly, there is evidence that these same signal cascade pathways play a role in genetic forms of PD such as parkin mutations. One study utilizing parkin null mice found that activation of these neuroprotective cellular pathways by estrogen was dependent on the presence of parkin. Treatment of wild-type but not parkin-null midbrain cell cultures with estradiol led to MAPK, Akt, and GSK3β activation, increased TH neuron counts, and decreased apoptosis (Rodriguez-Navarro et al., 2008). Of note, GSK3β is also implicated in the pathogenesis of neurodegenerative diseases with tau pathology, such as progressive supranuclear palsy (PSP). GSK3β phosphorylates tau, which leads to microtubule dysfunction, disrupted intracellular protein trafficking, formation of neurofibrillary tangles, and neuronal death (Goodenough et al., 2005). PSP also exhibits a male predominance clinically, so it is relevant that GSK3β is inhibited by estrogen via the P13K/Akt cascade. Another estrogen receptor-mediated mechanism of neuroprotection may be increased expression of neurotrophic factors, such as glial cell line-derived neurotrophic factor (GDNF). Campos and colleagues demonstrated that 17β-estrogen increases GDNF levels in SN and striatum, and that protection of dopaminergic neurons from 6-OHDA was dependent on GDNF expression (Campos et al., 2012).

In addition to these signal cascade effects, estrogen could impact PD pathogenesis via its influence on mitochondrial function and response to oxidative stress. 17β-estradiol reduces toxicity from glutamate, superoxide anions, and hydrogen peroxide in in-vitro ventral mesencephalic neuronal culture. Since this effect was independent of the estrogen receptor, investigators hypothesized that it occurred by sequestration of cytosolic calcium (Sawada et al., 1998). Another study demonstrated that estrogen pretreatment led to attenuation of the rise of intracellular free calcium concentrations induced by glutamate (Goodman et al., 1996). Estrogen also decreases production of reactive oxygen species by increasing expression of superoxide dismutase, glutathione peroxidase and glutaredoxin (Wang et al., 2001). There are several additional mechanisms by which estrogen may act to stabilize and preserve mitochondrial function in the face of various types of stress. Estrogen protects mitochondrial membrane potential, prevents ATP depletion, and reduces free radical production (Simpkins et al., 2010). Transcription effects via the estrogen receptor include increased glucose transporter expression and increased production of glycolytic pathway enzymes, overall leading to increased glucose utilization and decreased production of toxic free glutamate. Estrogen also increases transcription of mitochondrial DNA and may itself have direct antioxidant effects (Simpkins et al., 2010).

Another possible mechanism by which estrogen may be protective in PD involves α-synuclein, the pathologic protein aggregate that defines PD. It remains controversial whether these α-synuclein aggregates themselves are toxic to dopamine neurons leading to neuronal death, or whether sequestration of α-synuclein represents a protective defense mechanism by the neuron. If the former hypothesis is correct, then estrogen may also exert neuroprotective effects in idiopathic PD by preventing Lewy body formation. A single study showed that estrogen has both anti-aggregation and fibril destabilization properties in α-synuclein specifically (Hirohata et al., 2009). Inflammation may also play a role in the pathogenesis of PD, as is evidenced by microglial activation surrounding Lewy bodies, and this is yet another possible mechanism for estrogen neuroprotection. Estrogen appears to have anti-inflammatory effects, including decreasing levels of cytokines and other inflammatory modulators, leukocyte CNS entry, and microglia activation in-vitro and in-vivo (Pozzi et al., 2006). McFarland and colleagues demonstrated that a specific ERβ agonist prevented the increase in TNFα in peripheral blood monocytes and brain homogenate after treatment with 6-OHDA (Olsson, 2013).

These numerous mechanisms for neuroprotection by estrogens could lead to targeted therapies for PD and a variety of neurologic conditions. If the specific mechanisms by which estrogen exerts its neuroprotective effects in PD are established, novel estrogen-like compounds with structural modifications may be developed to maximize neuroprotection while minimizing unwanted systemic effects.

While estrogen appears to have broad neuroprotective effects in the pathogenesis of PD, it becomes more complicated to determine how this effect interacts with all of the other endogenous and exogenous factors that contribute to PD susceptibility. One interesting and unexpected example of this interaction is between estrogen and caffeine. Higher caffeine and coffee intake are associated with a reduced risk of PD in several large prospective and case-control studies (Ascherio, 2001, Ascherio et al., 2003). However, this risk reduction appears to be modified by sex and by use of hormone replacement therapy in women. Ascherio and colleagues reported that there was a strong inverse correlation with coffee or caffeine intake in men. In women however, there was a U-shaped curve, with the only significant reduction in risk with 1–3 cups of coffee daily (RR = 0.6) (Ascherio, 2001). Palacios and colleagues utilized the ACS Cancer Prevention Study II Nutrition Cohort to show that men within the highest quartile of caffeine intake compared to the lowest quartile had a relative PD risk of 0.43 (CI 0.26–0.71), whereas women had a non-significant reduction in relative risk to 0.61 (CI 0.34–1.09) (Palacios et al., 2012). The interplay between estrogen and caffeine was further studied in women receiving HRT after menopause. Ascherio and colleagues found that PD risk was decreased in the group with HRT use and lowest coffee intake, but that PD risk was actually increased in the group with both HRT use and highest caffeine intake (Ascherio et al., 2003). Caffeine therefore appears to decrease the relative risk of PD in never users of HRT, but either loses this benefit or even increases the relative risk with concurrent estrogen exposure in users of HRT (Ascherio, 2004). There is corresponding evidence in PD animal models that estrogen interferes with the effects of caffeine on neuroprotection. Xu and colleagues studied the effect of caffeine in MPTP-treated male mice, ovariectomized females without estrogen replacement, and non-ovariectomized females. In males and ovariectomized females, caffeine had a dose-dependent protective effect on striatal dopamine levels. However, both non-ovariectomized and estrogen-treated female mice were much less sensitive to the protective effect of caffeine (Xu, 2006). Based on this epidemiologic and basic science data, the potential neuroprotective effect of caffeine can be negated or even reversed when endogenous or exogenous estrogen is present. Since there are currently clinical trials investigating the use of caffeine-like agents in the treatment of PD, it will be important to stratify the clinical and disease modifying effects by sex and hormone status. A great deal of research is underway to develop medications that could prevent or slow disease onset and progression, and sex differences in response to these medications would not be surprising based on the above literature. Although HRT has become less prevalent after the results of randomized trials showed increased stroke and cardiovascular risk, it will be useful to consider how HRT or SERM usage impacts the efficacy of potential neuroprotective medications.

There are sexual dimorphisms in the normal human basal ganglia, specifically within the DA system that may influence the development of diseases such as PD, chorea, and tics/Tourette's syndrome. Animal studies of both rats and monkeys demonstrated that females have a higher number of dopaminergic neurons in substantia nigra than males (Beyer et al., 1991). Sexual dimorphisms are also evident through functional neuroimaging of the striatum in normal human subjects. While two studies showed higher [123I]β-CIT uptake in the striatum in females compared to males (Lavalaye et al., 2000), another did not show a sex difference (van Dyck et al., 1995). In a study investigating 18F fluorodopa uptakes, women were found to have higher striatal uptake than men in the caudate more than the putamen (Laakso et al., 2002). Another functional neuroimaging study using DAT binding showed women younger than 60 years old had 8.4% higher striatal DAT binding compared to age-matched men. However, in subjects above 60 years old there was no difference in DAT binding (Wong et al., 2012). Overall, these studies suggest that women and especially younger women have a higher baseline number of dopamine neurons. This is relevant to PD, in which loss of dopamine neurons gradually leads to clinical symptoms only after 80% loss (Bernheimer et al., 1973). If women have a higher reserve of dopamine neurons, this may explain why women develop clinical symptoms later than men.

In addition to these differences in the number of dopaminergic neurons, there are sex specific differences in molecular function of dopamine neurons. Simunovic and colleagues showed that genes involved in cellular homeostasis and mitochondrial function are upregulated in males compared to females using DA neurons isolated from normal human postmortem brains (Simunovic et al., 2010). They hypothesized that these differences translate into a higher metabolic rate that accelerates aging and cellular demise in males. This study also investigated the postmortem brains of PD patients. The PD DA neurons exhibited dysregulated gene expression profiles in both males and females, but the gene expression profiles had little overlap between sexes. Of 36 genes associated with PD pathogenesis, only 2 exhibited similar dysregulation in males and females. PARK1, 6 and 7 genes were downregulated in males compared to females, suggesting that additional mechanisms of increased risk of disease are at play. Cantuti-Castelvetri and colleagues also investigated human dopaminergic neurons from PD patients and controls. They determined there were sex specific differences independent of PD status. In females, genes for signal transduction and neuronal maturation were upregulated, while in males, genes for oxidative phosphorylation and PD pathogenesis-associated genes were upregulated. In the PD subjects, females had alterations in gene expression profiles for genes involved in protein kinase activity, proteolysis, and WNT signaling, and males had alterations in vesicle-mediated transport, protein-binding proteins and copper-binding proteins (Cantuti-Castelvetri, 2007). In summary, there are sex-specific gene expression in normal DA neurons, and sex-specific dysregulation that occurs in dopaminergic neurons in Parkinson's disease, suggesting that the pathogenesis of the disease may differ between the sexes.

Despite the substantial evidence for sex steroid hormone effects on sexual dimorphisms in brain neurodegenerative diseases, it has become clear that hormones alone do not entirely explain these sex differences. For instance, murine cells exhibit differential gene expression profiles very early in development, before any gonadal hormone influence (Dewing, 2003). Sex differences are also present in dopaminergic neurons grown in cultures with identical hormone environments (Raab et al., 1995, Reisert and Pilgrim, 1991). Another logical source of distinction between males and females is genetics, and in fact it does appear that both the X and Y chromosomes contribute to these differences independently.

Much of this research focused on genes unique to a single sex, and this has led to investigation of the SRY region of the Y chromosome. SRY encodes a transcription factor that is responsible for gonadal differentiation and therefore sex steroid hormone production. In male mice, SRY is expressed in the substantia nigra, and was found to localize exclusively in TH-positive neurons (Dewing, 2006). Czech and colleagues reported that SRY is expressed in substantia nigra pars compacta and reticulata, and ventral tegmental area of human males, and that expression co-localized with a subset of 43% of TH-positive neurons (Czech et al., 2012).

Researchers have proposed that SRY is an important positive regulator of dopamine synthesis and metabolism. SRY expression is associated with increased monoamine oxidase-A (MAO-A), DOPA decarboxylase, dopamine β-hydroxylase, D2 receptor and tyrosine hydroxylase (TH) levels (Tao et al., 2011, Wu et al., 2009). ERβ levels are either increased or decreased with SRY (Wu et al., 2009). From the above results, it remains unclear whether SRY has an overall pro- or anti-dopaminergic effect. When SRY is overexpressed, there is increased expression of MAO-A, which in isolation would lead to decreased dopamine levels due to increased metabolism. It is important to note that the MAO-A gene is located on the X chromosome, and the gene core promoter contains a SRY-binding site. This was the first X-linked gene to demonstrate regulation by SRY, and introduced a novel mechanism for sexual dimorphisms in normal and disease states. There also appears to be an SRY-binding site in the promoter region of the ERβ gene. SRY-overexpressed cells also demonstrated increased resistance to hydrogen peroxide-induced cytotoxicity through unclear mechanisms (Tao et al., 2011). On the other hand, SRY downregulation (via antisense oligodeoxynucleotides), led to decreased TH expression, and furthermore led to decreased spontaneous movement on the contralateral side when unilateral injections into substantia nigra were performed (Dewing, 2006). This phenotypic outcome suggests that SRY overall has a pro-dopaminergic effect. Furthermore, SRY may promote the initial development of the dopamine neuron population. Czech and colleagues noted that as neuronal precursor cells differentiated into dopamine neurons, a rise in SRY preceded a rise in TH. Further evidence that TH expression is under control of SRY was that SRY knockdown resulted in decreased TH expression, and that SRY was able to activate the TH promoter (Czech et al., 2012). The Y chromosome contains a number of genes in addition to the SRY region, which are responsible for sex determination. Studies in mouse embryo cells to determine the effect of the Y chromosome in the absence of testis development and sex steroid hormone production dissociated the SRY region from the Y chromosome (Carruth et al., 2002). In this study, cells that were XY but lacked SRY had higher levels of TH-positive neurons than cells that were XX. This suggests that there are mechanisms other than direct SRY-mediated upregulation of TH expression that are unknown.

The X chromosome copy number is another potential variable underlying sexual dimorphisms. Chen and colleagues reported that the complement of X chromosomes impacts gene expression in the striatum independently of gonadal sex (Chen et al., 2009). The experiments used mouse models that separated sex chromosome complement from gonadal sex. XX mice had increased levels of prodynorphin (Pdyn) in the striatum compared to XY of the same gonadal sex or XO mice, which suggests that the number of X chromosomes is the key variable independent of the presence or absence of a Y chromosome. This may relate to PD because it has been shown that genetic variation in the X chromosome associates with PD susceptibility, though no specific PD-related genes have been identified that are differentially expressed in this X chromosome-dependent manner (Pankratz et al., 2003). This X-dependent mechanism of gene expression has been called the “hemizygous exposure” effect, and may be due to genes that escape X inactivation. Therefore, in addition to sex steroid hormone effects, both the X and Y chromosomes individually may have distinct contributions to sex differences in the pathogenesis and clinical characteristics of PD.

Huntington's disease is a neurodegenerative disease caused by a polyglutamate triplet repeat expansion (CAG) on chromosome 4. It is a hyperkinetic disorder characterized by chorea, tremor, dystonia, and prominent neuropsychiatric and cognitive changes. As it is inherited in an autosomal dominant fashion, there are both equal penetrance in individuals and prevalence in the population in men and women. However, animal models of Huntington's disease as well as from epidemiologic cohorts suggest that sex may account for a portion of the variability in the disease between males and females. Similar to PD, there is some evidence that estrogen has neuroprotective effects on the mechanisms underlying Huntington's disease development, specifically oxidative stress and glutamate excitotoxicity (Weaver et al., 1997). Dorner and colleagues reported that striatal ascorbate release upon behavioral activation, a marker of functional reserve in the KI mouse model of Huntington's disease, was increased in female mice compared to males. While estrogen has been shown to modulate ascorbate release in other settings of oxidative stress such as ischemia, this specific effect has not been studied in Huntington's disease models specifically (Dorner et al., 2007). There are 2 additional reports of 17β-estradiol protecting against other markers of lipid peroxidation and oxidative stress in models of Huntington's disease (Heron and Daya, 2000, Tunez et al., 2006). One study by Bode and colleagues examined endogenous 17β-estradiol levels in the tgHD mouse model, in which females did not display significant motor dysfunction in the accelerod test during the first 14 months of life. There were significant correlations between 17β-estradiol levels and total number of DARPP32 + medium spiny neurons (MSN) as well as total striatal cell volume. ERα and ERβ were both expressed on DARPP32 + MSNs, thus they hypothesized that estrogen, through receptor-mediated effects, leads to neuroprotection that translates to less severe motor dysfunction in females with Huntington's disease (Bode et al., 2008).

This scientific evidence suggesting a protective effect of estrogen in Huntington's disease is preliminary, and the epidemiologic evidence is contradictory. There is conflicting information with some studies that suggest female sex is detrimental while others suggest that it is protective. In a large European cohort, women had a more severe disease phenotype and faster progression despite no difference in age of onset (Zielonka et al., 2013). Other studies have found that women had a later age of onset and longer course (Foroud et al., 1999). Since Huntington's disease has a clear inheritance pattern and a younger age of onset than PD, it is difficult to correlate age at onset and early clinical characteristics with lifetime estrogen exposure, and it would be rare for patients to have been on hormone replacement therapy. In summary, sex differences likely account for a small portion of the phenotypic variance in Huntington's disease and the multifactorial neuroprotective effects of estrogen may extend to this disease.

If estrogen possesses pro-dopaminergic properties as described above in PD, it would logically follow that estrogen may exacerbate conditions mediated by hyper-dopaminergic states. This, while oversimplified, describes the hyperkinetic movement disorders, including chorea. Chorea is a key symptom of Huntington's disease as discussed in the prior section, and some studies indeed have shown that women have more severe clinical symptoms including chorea (Zielonka et al., 2013). Another key example is Sydenham's chorea, which becomes persistent more frequently in females, and can recur during pregnancy or estrogen exposure (Fahn et al., 2011). Choreic movements during pregnancy, or chorea gravidarum, may represent a variant of Sydenham's disease, or it may be the first sign of systemic lupus erythematosus or the related antiphospholipid syndrome. Further research will be necessary to determine estrogen's role in chorea and its underlying mechanisms of action in the basal ganglia, so the data could be applied therapeutically.

Tics and Tourette's syndrome (TS) are thought to be caused by dopamine signaling dysfunction, in addition to other neurotransmitter abnormalities within the cortico-striatal-thalamocortical circuits. TS criteria include both motor and vocal tics. It is much more common in boys than girls with a prevalence ratio of approximately 4:1, despite a proposed autosomal dominant mode of transmission (Bortolato et al., 2013, Jankovic and Kurlan, 2011). There are several possible contributing factors to this sex-dependent susceptibility. First, there are baseline sexual dimorphisms in the basal ganglia pathways thought to be involved in tics and TS. One important structure in TS is the ventral striatum, which receives input from the sexually dimorphic amygdala and bed nucleus of the stria terminalis. Arginine vasopressin (AVP) mediates this pathway, and AVP levels are higher in males at baseline, are decreased after castration, and restored by exogenous testosterone therapy (Peterson et al., 1992). Therefore, structural differences that occur during brain development may be influenced later in life by gonadal hormones.

Sex steroid hormones also influence the severity of tics in both females and males, and may additionally play a role in disease pathogenesis. It is well known that tics tend to begin around the age of adrenarche, and often improve at puberty and even more in adulthood. Anabolic steroids worsen tic severity in males (Leckman and Scahill, 1990). Schwabe and colleagues found that 26% of 47 women experienced increased tic severity during the estrogenic phase of the menstrual cycle (Schwabe and Konkol, 1992). In another study, however, no correlation was found between tic severity and estrogen or progesterone levels in 8 women (Kompoliti et al., 2001).

A recent article by Bortolate and colleagues summarized the role of neuro-steroids in TS. Currently, there are clinical studies of 5α-reductase inhibitors in the treatment of TS. Briefly, 5α-reducatase inhibitors exert anti-dopaminergic effects that appear to be largely mediated through D1 receptors (these receptors are the main subtype implicated in TS). An open label trial of finasteride, a common 5α-reductase inhibitor, was successful in significantly decreasing tic severity and compulsive symptoms in adult men (Muroni et al., 2011).

Section snippets

Conclusions

In conclusion, sex differences play an important role in the susceptibility to, and pathogenesis of, various movement disorders. Sexual dimorphisms in structure and function appear during development, likely due to effects of both sex steroid hormones and sex chromosomes. Sex steroid hormones furthermore appear to influence the dopaminergic system in both healthy and disease states. Estrogen exerts a pro-dopaminergic effect, which may both decrease the risk of PD in women, as well as predispose

References (115)

  • D.E. Dluzen et al.

    Estrogen as a neuroprotectant against MPTP-induced neurotoxicity in C57/B1 mice

    Neurotoxicol. Teratol.

    (1996)
  • D. Dluzen

    Estrogen decreases corpus striatal neurotoxicity in response to 6-hydroxydopamine

    Brain Res

    (1997)
  • D.E. Dluzen et al.

    Methamphetamine-gonadal steroid hormonal interactions: effects upon acute toxicity and striatal dopamine concentrations

    Neurotoxicol. Teratol.

    (2002)
  • J.L. Dorner et al.

    Sex differences in behavior and striatal ascorbate release in the 140 CAG knock-in mouse model of Huntington's disease

    Behav. Brain Res.

    (2007)
  • A.C. Ferraz et al.

    Failure of estrogen to protect the substantia nigra pars compacta of female rats from lesion induced by 6-hydroxydopamine

    Brain Res.

    (2003)
  • X. Gao et al.

    Tamoxifen abolishes estrogen's neuroprotective effect upon methamphetamine neurotoxicity of the nigrostriatal dopaminergic system

    Neuroscience

    (2001)
  • X. Gao et al.

    The effect of testosterone upon methamphetamine neurotoxicity of the nigrostriatal dopaminergic system

    Brain Res.

    (2001)
  • G.E. Gillies et al.

    Independent influences of sex steroids of systemic and central origin in a rat model of Parkinson's disease: a contribution to sex-specific neuroprotection by estrogens

    Horm. Behav.

    (2010)
  • S. Goodenough et al.

    Glycogen synthase kinase 3beta links neuroprotection by 17beta-estradiol to key Alzheimer processes

    Neuroscience

    (2005)
  • M. Hirohata et al.

    Anti-aggregation and fibril-destabilizing effects of sex hormones on α-synuclein fibrils in vitro

    Exp. Neurol.

    (2009)
  • E. Kuppers et al.

    Expression of estrogen receptor-alpha and beta mRNA in the developing and adult mouse striatum

    Neurosci. Lett.

    (1999)
  • A. Laakso et al.

    Sex differences in striatal presynaptic dopamine synthesis capacity in healthy subjects

    Biol. Psychiatry

    (2002)
  • C. Lewis et al.

    Testosterone enhances dopamine depletion by methamphetamine in male, but not female, mice

    Neurosci Lett

    (2008)
  • L.X. Liu et al.

    Neuroprotective effects of genistein on dopaminergic neurons in the mice model of Parkinson's disease

    Neurosci. Res.

    (2008)
  • M. Morissette et al.

    Estrogen and SERM neuroprotection in animal models of Parkinson's disease

    Mol. Cell. Endocrinol.

    (2008)
  • H.E. Murray et al.

    Dose- and sex-dependent effects of the neurotoxin 6-hydroxydopamine on the nigrostriatal dopaminergic pathway of adult rats: differential actions of estrogen in males and females

    Neuroscience

    (2003)
  • R.E. Myers et al.

    Estrogen, but not testosterone, attenuates methamphetamine-evoked dopamine output from superfused striatal tissue of female and male mice

    Neuropharmacology

    (2003)
  • H. Ohtani et al.

    Chronic estrogen treatment replaces striatal dopaminergic function in ovariectomized rats

    Brain Res.

    (2001)
  • M. Ookubo et al.

    Effects of estrogens on striatal damage after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) neurotoxicity in male and female mice

    Mol. Cell. Endocrinol.

    (2008)
  • B.S. Peterson et al.

    Steroid hormones and CNS sexual dimorphisms modulate symptom expression in Tourette's syndrome

    Psychoneuroendocrinology

    (1992)
  • H. Raab et al.

    Effects of sex and estrogen on tyrosine hydroxylase mRNA in cultured embryonic rat mesencephalon

    Brain Res. Mol. Brain Res.

    (1995)
  • I. Reisert et al.

    Sexual differentiation of monoaminergic neurons—genetic or epigenetic?

    Trends Neurosci.

    (1991)
  • M.J. Schwabe et al.

    Menstrual cycle-related fluctuations of tics in Tourette syndrome

    Pediatr. Neurol.

    (1992)
  • D.R. Session et al.

    Estrogens and Parkinson's disease

    Med. Hypotheses

    (1994)
  • P.J. Shughrue

    Estrogen attenuates the MPTP-induced loss of dopamine neurons from the mouse SNc despite a lack of estrogen receptors (ERalpha and ERbeta)

    Exp Neurol

    (2004)
  • J.W. Simpkins et al.

    Mitochondrial mechanisms of estrogen neuroprotection

    Biochim. Biophys. Acta Gen. Subj.

    (2010)
  • Anon

    A randomized pilot trial of estrogen replacement therapy in post-menopausal women with Parkinson's disease

    Parkinsonism Relat. Disord.

    (2011)
  • A. Ascherio

    Prospective study of caffeine consumption and risk of Parkinson's disease in men and women

    Ann. Neurol.

    (2001)
  • A. Ascherio

    Coffee consumption, gender, and Parkinson's disease mortality in the Cancer Prevention Study II Cohort: the modifying effects of estrogen

    Am. J. Epidemiol.

    (2004)
  • A. Ascherio et al.

    Caffeine, postmenopausal estrogen, and risk of Parkinson's disease

    Neurology

    (2003)
  • M. Baldereschi et al.

    Parkinson's disease and parkinsonism in a longitudinal study: two-fold higher incidence in men. ILSA Working Group. Italian Longitudinal Study on Aging

    Neurology

    (2000)
  • C. Beyer et al.

    Dopamine content and metabolism in mesencephalic and diencephalic cell cultures: sex differences and effects of sex steroids

    J. Neurosci.

    (1991)
  • F.J. Bode et al.

    Sex differences in a transgenic rat model of Huntington's disease: decreased 17beta-estradiol levels correlate with reduced numbers of DARPP32 + neurons in males

    Hum. Mol. Genet.

    (2008)
  • M. Bortolato et al.

    The implication of neuroactive steroids in Tourette syndrome pathogenesis: a role for 5alpha-reductase?

    J. Neuroendocrinol

    (2013)
  • T.P. Brown et al.

    Pesticides and Parkinson's disease—is there a link?

    Environ. Health Perspect.

    (2006)
  • S. Callier et al.

    Stereospecific prevention by 17beta-estradiol of MPTP-induced dopamine depletion in mice

    Synapse

    (2000)
  • S. Callier et al.

    Neuroprotective properties of 17beta-estradiol, progesterone, and raloxifene in MPTP C57Bl/6 mice

    Synapse

    (2001)
  • F.L. Campos et al.

    GDNF contributes to oestrogen-mediated protection of midbrain dopaminergic neurones

    J. Neuroendocrinol.

    (2012)
  • Cantuti-Castelvetri

    Effects of gender on nigral gene expression and Parkinson disease

    Neurobiol. Dis.

    (2007)
  • L. Carruth et al.

    Sex chromosome genes directly affect brain sexual differentiation

    Nat. Neurosci.

    (2002)
  • Cited by (129)

    • Biomarkers of Oxidative Stress and Antioxidant Defense

      2022, Journal of Pharmaceutical and Biomedical Analysis
    View all citing articles on Scopus
    View full text