Sex Differences in the Brain: The Estrogen Quandary
The Estrogen Quandary

by Brenda Patoine

January, 2005


The Rockefeller University

Q: You’ve devoted your career to studying hormonal effects on the brain, including sex differences. What surprises you most about how male and female brains differ?

McEwen: There are fundamental sex differences between males and females that go well beyond reproduction. The more people look, the more differ­ences are found in both the structure and function of cells throughout the brain. It’s just mind-boggling when you see the complexities. There is recent data suggesting that glial cells from male and female neonates respond differently to estrogen, as if there’s already been some programming that keeps the male glial cells from responding to estrogen the same way as the female cells. We have evidence, as do others, that the hippocampus, a memory-related organ unrelated to reproduction, responds differently to estrogens. The female responds to estrogen by forming new synaptic connections in the hippocampus, while the male does not. But if you block the actions of testosterone in the male at birth, then the male will respond to estrogens to induce these synapses. There are many other examples. The differences include cerebellum, the autonomic nervous system, cerebral cortex, and hypothalamus. The more we look, the more sex differences we discover.

“There are fundamental sex differences between males and females that go well beyond reproduction.”

Q: Has science done enough to take into account sex differences in basic and clinical research?

A: Science has not done enough to take into account sex differences in either basic or clinical research. It’s very clear, for example, that psychotropic medications and many other drugs work differently in males and females. Some of it is related to sex differ­ences in how the liver clears drugs from the body, but almost any drug that affects the brain is going to work somewhat differently in the male and female, depending on the gender first and secondly on the hormonal status. The NIH now has an office of Women’s Health Research and there are a number of private and university-sponsored programs in women’s health research. This has revitalized the study of sex differences, meaning, in large part, the study of whole animals and how they differ behaviorally and physiologically. You can’t just assume that what works in the male is going to work the same way in the female.

Q: The basic science on estrogen's effects on the brain seems so tangled. Why has it been so difficult to nail down these effects?

A: The basic science on estrogen’s effects on the brain is tangled, because there are so many things estrogen does. We and others have uncovered the fact that estrogens, and probably all steroid hormones, work not only through receptors in the cell nucleus (nuclear receptors), which directly regulate gene expression, but they also work through non-nuclear receptors, which could be in the cytoplasm of a glial cell, in the spine of a hippocam­pal neuron, or in a microglial cell, etc. These non-nuclear receptors can stimulate important signaling pathways, such as cyclic AMP, very much like neurotransmitters and other hormones that work on the surface of cells. The myriad of possible mechanisms and effects has grown enormously, for estrogens as well as for other steroids.

As a result, the pharmacology of estrogen antagonists is confusing. The antagonists that have traditionally blocked the nuclear receptor effect don’t necessarily block the non-nuclear effect. So when you give an estrogen antagonist or partial agonist, such as tamox­ifen or raloxifene, you don’t know what it’s going to do to the brain. It may actually stimulate some actions in the brain produced by estrogen and block others. It becomes a matter of trade-offs between getting a beneficial effect, for example in bone or protecting against uterine or ovarian cancer, versus possibly having a negative effect elsewhere. The SERM’s (selective estrogen response modulators) appear to protect bone and protect against cancer but they precipitate hot flashes because they are partial estrogen antagonists as well as agonists.

Q: What have we learned from and since the Women’s Health Initiative about hormone replacement therapy?

A: When the Women’s Health Initiative (WHI) tested Prempro, a combination of Premarin [estrogen made from horse urine] and Provera [a synthetic progestin], all it really did was to show that that’s not a very good combination. But the results have been misinterpret­ed as saying that all forms of hormone therapy are bad. We simply don’t know. We have to go back to the drawing board and perhaps investigate other types of therapies that involve more natural hormones, and are given in more physiological ways, which means sequentially and intermittently. We know that low-dose is always the best idea, and that sequential administration of estradiol, the natural hormone, and progesterone, the natural hormone, may be effective. A recent paper by Karen Prestwood in the Journal of the American Medical Association, suggests that low doses of natural estradiol protect bone density in post-menopausal women. The jury’s still out on whether these kinds of treatments will have beneficial effects on the brain and other body systems, but they may offer alternatives to Prempro and combined therapies.

“...The results of WHI have been misinter­preted as saying that all forms of hormone therapy are bad. We simply don’t know. We have to go back to the drawing board...”

The other part of the WHI story is that the women in the study were, by and large, 10-14 years post­menopausal when they were started on therapy. There is evidence from animal studies that the body in general, and the brain in particular, may lose sensitivity to estrogens if they are missing for too long. This may be why some women don’t experience hot flashes after a while. And, if the brain is continually reorganizing itself, it may adapt and not need these hormones, at least not in the same way. That suggests, and there is evidence, that if you treat around the time of menopause, perhaps even with Prempro, there may actually be some lasting protective benefit. It remains to be seen whether women should be treated around the time of the menopause and not later, or whether they have to be treated continuously for the rest of their lives, in the most physiologic manner possible. And we still don’t know what the most physiologic manner is. Some women are probably okay without the supplemental hormones, but if the hormones are given too late, according to WHI, they may make matters worse. 

Q: There seems to be this large “disconnect” between basic research suggesting a neuropro­tective role for estrogen and clinical data, like the WHIMS (Women’s Health Initiative Memory Study) study, showing that estrogen increases Alzheimer’s risk. What might account for this?

A: There is a disconnect. I think, unfortunately, clinicians ignored basic research in designing Prempro. In meetings where clinicians and basic researchers are present, it sometimes feels like nobody is listening to one another. There are still people that totally ignore the basic research and just continue on their own logic. The basic researchers have probably not been as sensitive as they should be to doing studies that mimic more closely what’s going on in the human, such as looking at whether there is a loss of sensitivity to estrogens with long-time absence of estrogens. So there’s fault on both sides, and one of the problems in dealing with this is to try to get people to really hear each other out. Of course, as a basic researcher, I feel that really the whole problem with Prempro was produced by a failure to recognize the basic facts on hormone action, and bowing to the pressure from pharmaceutical companies.

The Europeans, incidentally, did not buy any of this; most of them stayed away from Prempro like a plague. It was out of their experience that the estrogen/ estradiol skin patch was developed. It was also out of their experience that we now have progesterone gels, and even estradiol that can be applied topically. I think these offer hope for real benefits.



University of Michigan

Q: Your research focuses on sex differences in the brain, and how estrogen modulates neural systems. What surprises you most about how male and female brains differ? 

Becker: When I started investigating sex differences in the brain more than 25 years ago, most people thought they would be relatively minor and only in a few areas of the brain. It is now clear that there are sex differences all over the brain and the hormones important for reproduction are also important for nor­mal healthy sex-specific brain development and func­tion. What is surprising at this point is that we know that men and women have different brains, and yet we know very little about the consequences of these differences for neurological disorders, mental health, or drug abuse. 

Q: The basic science on estrogen's effects on the brain still seems so tangled. Why has it been so difficult to nail down what estrogen is doing in the brain? 

A: Estrogen is acting on the brain in many ways, and there isn’t a unified description of what estrogen is doing or how it is producing its effects.1 We now know that there are multiple receptors for estradiol, and that these receptors induce effects that are characteristic of the cellular environment in which they are found. In other words, in some cells estradiol does one thing, in other cells it does another, and in many cells the effect varies with the recent history of hormone exposure. 

For example, in the hypothalamic brain regions important for reproductive behavior in rats, it can take 48 hours of exposure to estradiol before a female rat will exhibit sexual receptivity. In contrast, in the striatum, estradiol acts extremely rapidly (within minutes) to induce a change in behavior triggered by drugs like cocaine or amphetamine. So, there are both long-term effects of estradiol as well as rapid immediate effects. This is one source of the confusion about what estradiol is doing in the brain.

There can also be big effects depending on how the hormone is administered. These methodological differences matter a lot to the brain, but tend to confuse the public’s understanding of the results, because reports don’t always point out how the drug was given or why that might be important. In an area of the brain that I study, the striatum, low intermittent doses of estradiol (comparable to what the ovary would produce) enhance the effects of cocaine or amphetamine, while high doses or prolonged continu­ous exposure to estradiol can have the opposite effect. 

Finally, many of the effects of estradiol are permis­sive—they allow other things to happen, but they don’t make them happen. For example, estradiol can rapidly enhance the behavioral effects of cocaine or amphetamine, but it doesn’t rapidly induce changes in behavior all by itself. 

Q: You have advocated for increased attention to gender differences in basic neuroscience and clinical research. Is enough being done? If not, what needs to happen to improve the situation? 

A: I do not think that enough research is being done on gender differences in basic neuroscience, clinical research, or in pharmaceutical testing of drugs. A number of things need to be done. Scientists need to become more familiar with how to design experi­ments to determine whether there are male-female differences in the measures they are interested in. As a first step toward facilitating this, the Isis Network on Sex, Gender, Drugs and the Brain, a network sponsored by the Society for Women’s Health Research, has recently published a paper in Endocrinology that provides some guidelines for research on sex differences in basic and clinical research. Physicians also need to be more aware of sex differences in the brain and body. The only way that will happen is if the medical schools begin teaching how the brains of men and women differ, and the medical boards include questions to assess that this knowledge has been learned. Finally, drug companies need to conduct their clinical trials so that they can determine from the results whether there are sex differ­ences in drug responses. Currently they are only required to include women, but are not required to include enough subjects so as to determine if women and men react differently to a drug. 

“...drug companies need to conduct their clinical trials so that they can determine ... whether there are sex differences in drug responses.”

 Q: What have you learned about how estrogen relates to addiction vulnerability? What are the practical implications of this work? 

A: My research has shown that physiological doses of estradiol affect acquisition of cocaine-taking in female rats, so that they will start taking cocaine at lower doses than will males or females without estradiol. Estradiol doesn’t affect how much cocaine a female will take once the habit has been acquired, but it can affect how hard a female will work to get cocaine. On the other hand, progesterone may counteract the effects of estradiol on acquisition of cocaine-taking behavior. So, there may be times during the menstrual cycle—the time when a female is about to ovulate—when a woman’s hormones may enhance the acquisition of cocaine-taking or influence how much she is motivated to obtain cocaine. 

We have also been looking at an animal model of cocaine relapse. If cocaine-taking ceases, estradiol alone does not induce a female rat to start taking cocaine again. But, if a rat gets an injection of cocaine after receiving estradiol, they respond more intensely to cocaine (that is, they relapse more easily). So, there may be certain times during the menstrual cycle when a woman who has kicked a cocaine habit will be at greater risk for relapse if she is exposed to even a little cocaine. 

Q: There's so much confusion about hormone replacement therapy. What do you tell women?   

A: I tell women that if they are over 60 and have never taken hormone replacement therapy (HRT), they are probably not good candidates for HRT, according to the results of the Women’s Health Initiative (WHI) studies. On the other hand, a woman who is peri­menopausal should talk with her doctor and decide what is best for her. We know that estradiol promotes cognitive functions related to language and fine motor skills in women. The animal research is very clear that estradiol has beneficial effects on the brain. Estradiol induces nerve growth factor, which promotes survival of neurons important for learning and memory. Estradiol promotes activity in motor and motivational brain regions, and maintains the survival of neurons known to degenerate in Parkinson’s disease. In the absence of estradiol, these functions are lost. Long-term hormone withdrawal can make it impossible for estradiol to restore these functions. So, a woman should decide if she is going to take HRT before she is symptomatic and work with her doctor to determine the best formulation and dose for her. My personal opinion is that HRT is beneficial for cognitive function, and so women should seriously consider taking estradiol, with or without proges­terone (depending on whether they have had a hysterectomy). I believe that the adverse effects reported in the WHI study cannot be generalized to all women, but are specific for women who have been post-menopausal without HRT for many years.