Oestrogen, brain function, and neuropsychiatric disorders
- Division of Psychological Medicine, Box P050, Institute of Psychiatry, De Crespigny Park, London SE5 8AF, UK
- Correspondence to: R Norbury;
Oestrogen has multiple effects on brain function
There is an increasing amount of research on the neurobiological effects of oestrogen. Also, health professionals are being asked for guidance on whether women should be prescribed oestrogen and progestogen hormone replacement therapy (HRT) not only to treat vasomotor instability and reduce bone loss, but also in various neuropsychiatric disorders. However, it is controversial whether oestrogen is indicated in the treatment of disorders such as depression, Alzheimer’s disease, and schizophrenia. A recent large scale study examining the effects of HRT, funded by the National Institutes of Health (NIH) in the USA, was prematurely terminated owing to increased rates of breast cancer, heart disease, and stroke.1 Shortly afterwards, the WISDOM trial funded by the MRC in the United Kingdom was also terminated. This has reinforced the need to have solid indications for the use of oestrogen-only replacement therapy.
OESTROGEN AS A NEUROPROTECTANT
Oestrogens affect the development and aging of brain regions that are crucial to higher cognitive functions (like memory) and are implicated in neuropsychiatric disorders such as Alzheimer’s disease. For example, oestrogens increase synaptic and dendritic spine density in the hippocampus. In rats, oophrectomy results in a decrease in dendritic spine density in CA1 pyramidal cells, but this is prevented by the administration of oestrogens. Moreover, synaptic spine density is related to circulating oestradiol levels.2 Until recently it was unclear how these oestrogen induced dendritic changes affected neuronal function. However, it has now been shown that oestrogen induces an increase in N-methyl-D-aspartate (NMDA) receptors in rat hippocampal neurones in the same region where an increase in dendritic spines is found, suggesting that the “new” oestrogen-induced spines are excitatory.3,4
Among the most biologically plausible explanations why HRT might ameliorate age associated deficits in memory are modulatory effects on the cholinergic system in brain regions affected by healthy aging and Alzheimer’s disease (for example, the hippocampus and the association neocortex).5 The basal forebrain nuclei and hippocampus contain receptors for sex steroids.6 Further, in rats oestrogen reverses memory deficits, and ovariectomy decreases cholinergic uptake and choline acetyltransferase activity in the hippocampus and frontal cortex.7,8
There are also indications that oestrogen affects central cholinergic function in healthy female humans. For example oestradiol modulates the growth hormone response to pyridostigmine throughout the menstrual cycle,9 and there is a significant correlation between oestradiol levels and cholinergic responsivity.10 We recently reported that cholinergic responsivity is greater in women who had received oestrogen replacement therapy than in those who had not.11 Moreover, among long term users of oestrogen replacement therapy there was a positive correlation between enhanced cholinergic neurotransmission and longer duration of oestrogen exposure. In contrast, Smith et al found no significant between-group difference in density of the vesicular acetylcholine transporter located in presynaptic terminals (as measured using SPET and the radiotracer [123I]IBVM).12 However, they did report a significant relation between length of HRT treatment and synaptic concentration in the frontal, temporal, and parietal cortices. That study was an important first step and suggests that although an overall effect of HRT was not found, it may influence the survival or plasticity of cholinergic cells. Taken together, these findings suggest that in older postmenopausal women oestrogen may be involved in the normal maintenance and physiological regulation of the cholinergic projections, and that oestrogen replacement can enhance the functional status of these cholinergic projections.
Various studies have addressed the relation between oestrogen and the serotonergic (5-hydroxytryptamine, 5-HT) system. The amygdala receives 5-HT projections from the raphé nuclei (at least in the rat)13 and oestrogen receptor α mRNA is abundantly expressed in the amygdala14 (oestrogen receptors occur in two isoforms, α and β; both subtypes are expressed in the brain, but with varying patterns of distribution—for a comprehensive review, see Hall et al 15). In rats, the density of 5-HT2A receptors increases during pro-oestrus in the frontal and cingulate cortex.16 Similarly, in ovariectomised rats there is an increase in central 5-HT2 receptors but a decrease in 5-HT1 receptors following oestrogen treatment.17 Thus oestrogen can modulate 5-HT receptor density and this may be of relevance to the action of antidepressants and atypical antipsychotics. Oestrogen also increases tryptophan hydroxylase mRNA levels in rhesus macaques18 and decreases monoamine oxidase activity in rat brain.6 Therefore oestrogen may both encourage the synthesis of 5-HT and decrease its catabolism. In humans, short term oestrogen replacement therapy increases 5-HT2A receptor density, as measured by positron emission tomography (PET).19 Also, we showed that oestrogen replacement therapy modulates the age related reduction in serotonergic responsivity in healthy women.20 Thus there is increasing evidence that oestrogen interacts with the 5-HT system at multiple levels and this may provide a theoretical role for oestrogen in the regulation of mood.
In addition to direct effects on neurones, oestrogens also act with neurotrophins to stimulate nerve cell growth indirectly. Oestrogen and neurotrophin receptors are co-expressed on rodent neurones in forebrain, hippocampus, and cerebral cortex, and this co-localisation may be important for neuronal survival.21 In addition, oestrogen can also protect against neurotoxins that boost free radical production,22 it can reduce the neuronal generation of β amyloid,23 and it can act as an antioxidant.24
In summary, there is increasing evidence that oestrogen modulates the aging of brain systems that are both crucial to higher cognitive function and implicated in mood and neuropsychiatric disorders such as Alzheimer’s disease.
OESTROGEN AND COGNITIVE FUNCTION
The most robust effect of oestrogen on cognitive function is most probably on verbal memory. Prospective randomised studies of HRT versus placebo following total abdominal hysterectomy and bilateral salpingo-ophorectomy report a significant positive effect of estrogens on verbal memory.25,26 Performance in some cognitive tasks also varies as a function of the menstrual cycle in healthy premenopausal women. During the luteal phase (characterised by high levels of oestrogen and progesterone), verbal articulation is improved whereas spatial ability is decreased,27 a pattern that is reversed during the follicular phase (when there is relatively low oestrogen and progesterone). A similar pattern of cognitive performance is also observed if subjects are tested during the preovulatory oestradiol surge (to control for the potential effects of progesterone on cognitive performance), suggesting that oestrogen rather than progesterone is responsible for the observed cognitive effects.
Functional imaging techniques have also been employed to assess the effects of oestrogen on networks subserving various aspects of cognitive function (see Maki and Resnick28 for a review). A recent randomised, placebo controlled crossover study using functional magnetic resonance imaging29 found oestrogen induced alterations in brain activation patterns during encoding and retrieval of both verbal and non-verbal stimuli. More recently, Maki and Resnick30 used PET and 15O to examine longitudinal changes in regional cerebral blood flow (rCBF) over a two year interval in women on and off HRT (both with and without adjuvant progesterone therapy). Significant differences in rCBF were found in the right hippocampus, the parahippocampal gyrus, and the left middle temporal gyrus—regions crucial to memory. We recently reported that oestrogen reduces age related differences in neuronal membrane breakdown (as measured by 1H magnetic resonance spectroscopy) in the hippocampus and parietal lobe, and this was related to memory function.31
Thus there is increasing evidence from in vivo brain imaging studies that estrogen modulates cognitive function, cerebral blood flow, and membrane breakdown. However, further prospective randomised studies are required.
OESTROGEN AND ALZHEIMER’S DISEASE
Epidemiological studies have reported that the prevalence of Alzheimer’s disease is significantly decreased in women on HRT, and that those women with Alzheimer’s disease who were taking HRT had milder disease than those who were not.32 A recent longitudinal study reported that prolonged use of HRT decreased the risk and delayed the onset of Alzheimer’s disease (relative risk = 0.40; 95% confidence interval, 0.22 to 0.85); moreover, the use of oestrogen for longer than one year reduced the risk of developing Alzheimer’s disease by 5%.33 However, results from such studies are often difficult to interpret. For example, women who use oestrogen are often better educated, generally healthier, and less depressed than non-users—a factor known as “the healthy user bias”. Thus we cannot exclude the possibility that oestrogen use reflects an as yet unidentified bias that accounts for the effects observed.33
Results from early clinical trials of HRT in people with Alzheimer’s disease suggested that it may also benefit women with established disease. For example, women with Alzheimer’s disease who were using oestrogen had better scores on the Alzheimer’s disease assessment scale (ADAS-Cog, a standard instrument used in clinical trials on Alzheimer’s disease) than their counterparts who did not take oestrogens.34 However, results from recent large randomised double blind, placebo controlled trials are less optimistic.35–37 These studies reported no beneficial effects of HRT on cognition, mood, or functional outcomes in Alzheimer’s disease. In contrast, a smaller study by Asthana et al reported a significant treatment benefit after a four week treatment period with oestradiol patches in subjects with Alzheimer’s disease.38 However, the effects were temporary and diminished once treatment ceased. The differences in results may be explained by differences in the type of HRT used (Asthana used an oestradiol patch, whereas Mulnard et al employed oral conjugated equine oestrogen (CEE) in 120 hysterectomised women with mild to moderate Alzheimer’s disease for a period of one year39). A significant improvement on the mini-mental state examination was also reported by Mulnard after two months’ treatment with CEE. However, the benefit did not persist with prolonged treatment. Overall, oestrogen replacement therapy for one year did not slow disease progression or improve cognitive function.
Thus current evidence does not support a role for oestrogen in the treatment of established Alzheimer’s disease. However, oestrogen may be a neuroprotectant in healthy brain and delay the onset of Alzheimer’s disease. Perhaps in the established disease the potential therapeutic window has been missed and the remaining neurones are refractory to the beneficial effects of oestrogen, although oestrogen may still be useful as an augmentation strategy in those taking acetylcholinesterase inhibitors.40
OESTROGEN AND SCHIZOPHRENIA
There are sex differences in premorbid functioning, age of onset, symptomatology, and outcome of schizophrenia. For example, men have a single peak of disease onset in their early twenties, whereas women have a later age of onset and a second peak in incidence between the ages of 45 and 55 years. In addition, women are more likely to have a family history of schizophrenia, display atypical or affective features, and have a seasonal pattern of hospital admission. Because oestrogen has putative antidopaminergic/antipsychotic actions, it has been suggested that in women oestrogens may be responsible for the delayed onset of the first schizophrenia peak, and that the second peak may reflect the decline in oestrogen levels at the menopause. There is evidence that oestrogen protects the nigrostriatal dopaminergic system against the neurotoxic effects of MPTP in rats.41–43 An antipsychotic action of oestrogens is supported by clinical studies reporting that women have increased admission rates for psychosis around the menses42 and that psychotic symptomatology varies with phase of the menstrual cycle.43,44 Also, women with schizophrenia may have reduced oestradiol concentrations compared with controls.44 However, there is currently little evidence to support the therapeutic use of oestrogen in schizophrenia. For example, when women with schizophrenia are treated with adjunctive oestrogen there is a slight increase in speed of recovery, but no improvement overall as compared with antipsychotic drug treatment alone.45 Despite the lack of clear evidence for the efficacy of oestrogen as an antipsychotic agent, it remains plausible that oestrogen replacement therapy might protect against late onset schizophrenia in postmenopausal women by reducing age related changes in brain structure and neurochemistry (for example, in the hippocampus).
OESTROGEN AND DEPRESSION
Epidemiological studies suggest that, in addition to psychosocial factors, times of change in oestrogen levels make depression more likely in a vulnerable subgroup of women. For example, after puberty, women are around twice as likely to suffer depression than men46 and in the postpartum period there is a peak of incidence.47 However, although the perimenopausal period brings with it an increase in mild depressive symptoms, there is no increase in depressive disorders46 and there is evidence that the prevalence of depression decreases postmenopausally.48
In the light of the above findings, it has been suggested that oestrogen may have a role as an antidepressant. However, methodological difficulties affect the few studies in this area. These include small numbers of subjects, the lack of control groups, and (in the studies of menopausal depression) a variable or inadequate definition of the menopause and the use of multiple HRT preparations. In postpartum depression, oestrogen therapy may be useful both as prophylaxis in vulnerable individuals49 and as a treatment.50 In high doses, oestrogen has been reported as being useful in non-postpartum treatment of resistant depression in women,51 but this small study has yet to be replicated. There is also some evidence that women taking oestrogen replacement therapy may respond better to fluoxetine.52 In the perimenopause,oestrogen replacement therapy is effective in reducing mild depressive symptoms46 and it has been reported to be an effective treatment for depression.53,54 However, it is difficult to determine from these studies whether the oestrogen is treating menopausal symptoms such as sleep deprivation and anergia or the depression per se.
Currently, there is little evidence to suggest that oestrogen is a useful treatment for depression during the menopause or at other times. Indeed it is our opinion that oestrogen replacement or HRT should not be used as a first line treatment for depression in any group of depressed women, although in non-depressed individuals it can foster a sense of “psychological wellbeing” in the perimenopausal period.46 Oestrogen may, however, have a role as a time limited (brief) adjunct to antidepressants in treatment resistant depression, or in the alleviation of mild mood symptoms (which are nonetheless distressing) in peri-postmenopausal women.
Basic research indicates that oestrogen has multiple effects on brain function, modulating aspects of neurotransmitter function, glucose metabolism, synaptogenesis, and brain aging. In line with this, epidemiological studies implicate oestrogen in the aetiology of neuropsychiatric disorders. However, these encouraging findings do not at present translate into the use of oestrogen as a treatment. Current evidence suggests that oestrogen alone has no role in the treatment of established Alzheimer’s disease, but may delay its onset. The evidence base for the use of oestrogen as an adjunct to neuroleptics in schizophrenia or in the treatment of postnatal and perimenopausal depression is currently too weak to merit a change in clinical practice. Larger prospective studies are required to establish whether oestrogen has a role in the prophylaxis and treatment of these disorders.
Oestrogen has multiple effects on brain function