Sam Gandy led this live discussion on 5 December 2001. Readers are invited to submit additional comments by using our Comments form at the bottom of the page.

Live discussion held 5 December 2001, noon-1 p.m. (EST).

Participants: Dick Lloyd, Sam Gandy, June Kinoshita, Richard Bowen, Nico Stanculescu, Seth Shaw, Mark Smith, Steve Fiander, Gunnar Gouras, Larry Tusick, Alan Lerner, Alexei R. Koudinov, Mike Shelanski, Rena Li, Gabrielle Strobel, Craig Atwood, and the mysterious Guest 2.

Note: The transcript has been edited for clarity and accuracy.

June: First, let me welcome you all to today's live discussion.

Sam Gandy: Thanks for organizing it, June.

June: I know that Richard Bowen has been eager to discuss some of the points in your discussion. Richard, would you like to ask the first question?

Richard L. Bowen: Yes, June: I wanted to know if Sam has looked at the ovariectomized mice and beta amyloid study, and have you had a chance to look at gonadotrophin levels?

Sam Gandy: We haven't studied gonadotrophin levels under any circumstances. A paper on ovariectomy effects on Ab load in transgenic mice is in press at J. Neurochem. Co-authors are Karen Duff, Suzana Petanceska, Huaxi Xu, and others. Maybe Gunnar was involved, too. Anyway, there are no gonadotrophin data there, either.

Mark Smith: Are there plans to study gonadotrophins? Seems like an issue worth exploring given Dr. Bowen's hypothesis?

Sam Gandy: Not in my group. I don't know about Karen, Suzana, Huaxi et al. I think it's up to Richard to study it.

Mark Smith: Richard?

Sam Gandy: I'm more interested in finding out how gonadal steroids control Ab than I am in dissecting hypothalamic-pituitary-end organ mechanisms.

Richard L. Bowen: Given that the gonadotrophins are the primary hormones regulating reproductive function, [we think] that anyone studying reproductive hormones and AD should consider evaluating gonadotrophins along with steroid hormones. We plan to correlate the steroid levels to gonadotrophins in our study.

Alan Lerner: Sam -- any thoughts about brain aromatase levels and the levels of steroids that cells are experiencing?

Sam Gandy: I think about studying aromatization in order to find out whether testosterone works directly or must be converted to E2 first, but I haven't thought much more about it. I would like to try DHT.

Alan Lerner: Here's a reference on aromatase in brain (Wozniak, 1998).

Craig Atwood: Also, are there testosterone receptors in hippocampal neurons?

Sam Gandy: Craig, I have not encountered much literature on androgen receptors in brain. That might also help answer the aromatization question: If there are no brain androgen receptors, then testosterone probably acts via E2.

Craig Atwood: Yep.

Richard L. Bowen: Gonadotrophin levels correlate much better with the incidence of AD than do sex steroid levels.

Alan Lerner: My thought is that androgen deficiency may also mean at least relative estrogen deficiency. Perhaps this may explain the slightly lower incidence of AD in men. I have not looked in a while at clinical studies of low androgen levels in AD patients (I know this is slightly off topic).

Sam Gandy: Alan, Gunnar has shown that testosterone controls Ab generation in cultured neurons, so I'd guess that testosterone is as important in men as E2 is in women.

Alan Lerner: Thanks.

Richard L. Bowen: Since estrogen and testosterone are low prior to puberty, why don't we see alterations in APP processing and Ab generation [in children]?

Sam Gandy: Pre-pubertal molecular endocrinology is in no way equivalent to postmenopausal/postandropausal molecular endocrinology. Withdrawal of decades of hormones causes many changes never seen before puberty.

Richard Bowen: [Post-chat addendum] I certainly agree that sex hormone levels after puberty have different effects than prior to puberty. One of the primary differences is that prior to puberty the hypothalamic-pituitary-gonadal axis has not been activated. After puberty, which encompasses the postmenopausal/postandropausal period, the loss of circulating sex steroid hormones results in elevated pituitary gonadotropin secretion. Since gonadotropin receptors are present in the brain with the highest density found in the hippocampus and they affect ApoE metabolism in other tissues, gonadotropins would seem to be a rational target for study.

Mark Smith: Has anyone looked at effects of chemical ex-rapists become demented?

Sam Gandy: The prostate cancer patients that we studied should answer your rapist query: chemically castrate & plasma Aβlevels double.

Richard Bowen: [Post-chat addendum] Although Mayeux's data did show that individuals with higher A-β levels were more likely to develop AD, most data do not show a correlation between serum A-β levels and AD (Mehta PD et al., 2001); Vanderstichele H., 2000).

Craig Atwood: Sam, is that due to decreased estrogen/testosterone or increased gonadotrophins per Dr. Bowen's hypothesis?

June: Is there an increased risk of AD in these prostate cancer patients?

Mark Smith: So is there an epidemiological relation of prostate cancer and AD? We did this with Medicare database and found not.

Sam Gandy: I don't know about the long-term effects of chemical castration. There is actually very little in the way of literature on neurobiological effects of tamoxifen, etc., even in animals, though some studies are underway now.

Sam Gandy: Re: prostate cancer and cognitive decline, a prospective study is underway involving men undergoing chemical castration, following plasma hormones, plasma Ab 40 and 42, and cognitive function.

Alan Lerner: There is a small literature on steroids as treatment for behavior -- including DES, E2, leuprolide and medroxyprogesterone.

Sam Gandy: I doubt that the acute effects of steroids on behavior is mediated through Ab. Gonadal steroids have many effects via GABA receptor modulation etc. that to me seem more likely to explain acute neurological effects of steroids.

No Gender Difference in AD Risk?

Richard L. Bowen: Regarding the gender risk factor in AD, it is interesting to note that although sex steroid levels in Down Syndrome individuals are the same as in the general population, males with Down Syndrome get AD earlier [than females].

June: Richard, what about brain androgen receptors in Down patients?

Richard L. Bowen: June, I do not know. We have tried to get that data, but have had no luck thus far.

alexei r. koudinov enters

mshelanski enters

Sam Gandy: Hi, Mike.

Sam Gandy: I'm not convinced there's gender difference in incidence of AD. Prevalence is different due to dropout of men with coronary artery disease.

Alan Lerner: I agree with Sam -- there is not a convincing gender difference in incidence; many studies would systematically exclude those with strokes, more likely to be men. You would need an autopsy-based series with pathological confirmation.

Alan Lerner: Maybe Richard can give us some more information on his studies.....

Richard L. Bowen: Sam, are you saying that you do not think there is a gender difference in the incidence of AD in the general population?

Sam Gandy: Richard, yes. And I have this impression from a prominent neuroepidemiologist, but I don't have permission to quote him. [See Ruitenberg et al. 2001, for example, in which the authors found a similar incidence of AD onset in men and women while following 7,046 people of the Rotterdam prospective cohort for a mean of 5.7 years.]

Richard L. Bowen: Yes, Sam, please send me those. That would be great. [Post-chat addendum] Sam, most studies evaluating sex steroids and AD cite gender differences in prevalence as the reason for the study. If there is no gender difference in AD, what would be the rationale for researching sex steroids?

Alan Lerner: The problem with the gender ratio and hormones is to some extent a low signal to noise ratio - I don't know if we can conclude much from it.

Alan Lerner: Richard -- I would add that any of the many proposed environmental risk factors such as education, head injury, smoking, etc. as well as ApoE all contribute to gender ratio "bias".

Richard L. Bowen: Alan, I agree and all of those factors may affect gonadotrophin levels. Do you know if they do or not?

Sam Gandy: For the record, ERT appears to protect ApoE4+ women as well as E4- women.

Alan Lerner: Yes -- the best study (Tang, 1996) showed no effect of ApoE on estrogen-related AD incidence protection.

Richard L. Bowen: [Post-chat addendum] Again, since ERT lowers circulating levels of gonadotropins (Gn), one cannot infer that the benefits of ERT are solely due to the direct action of estrogen. The benefits could also be due to the lowering of Gn.

Craig Atwood: But does ERT act only prophylactically or as a treatment?

Gabrielle Strobel: An anecdotal question: I know two women whose mothers had 9/10 children, went into menopause right afterwards before age 40 and developed AD in their early 50s. Is there any data on number of births, age of menopause, and AD risk?

June: Gabrielle, maybe the number of children, not the early menopause, raises the risk of dementia!

Richard L. Bowen: If that is the case, my wife is DOOMED!

Gabrielle Strobel: Yes, my kids make me demented. But seriously, could many births "exhaust" the ovaries early on and early removal of estrogen's neuroprotective effect play out as accelerating incipient AD?

June: Gabrielle, I would think studying women with premature ovarectomies would give cleaner data than looking at women who have huge families -- too many confounders there!

Sam Gandy: The hormone neurobiology field would agree that the surgical ovarectomy population is more straightforward to study than those undergoing naturally occurring menopause

Richard L. Bowen: Since reproduction is the most important function of any organism with regard to the survival of the species and gonadotrophins regulate reproductive function, it would not be surprising that once reproductive function diminishes, gonadotrophins could play a role in planned senescence.

Craig Atwood: Weren't the gonads at one stage thought to be part of the CNS!?

Alan Lerner: Whose?

Richard L. Bowen: There are a number of similarities in the gonads and the brain with respect to AD. Presenilin and APP are highly expressed in the gonads. LRP receptors are also highly expressed.

June: I always thought the brain was just a big sex organ!

Richard L. Bowen: Especially in men!

Mark Smith: Could say something sexist here but resisted!

June: Good.

Mark Smith: You would only have edited it out!

ERKsome Questions

June: Sam, you mention wanting to know how the ERK signaling pathway affects Ab metabolism. Any thoughts out there?

Craig Atwood: Does ERK signal to modulate APP processing?

Mark Smith: ERK...ears prick up. YES!!!!!!!!!!!!!!!!!!!

Sam Gandy: Our model has been that ERK, like PKC, modulates biogenesis of constitutive secretory vesicles that carry APP to the plasma membrane. It seems that if this pathway is hyperstimulated--by phorbols, by transmitters linked to PKC--APP-alpha release goes up and Ab generation goes down.

Richard L. Bowen: Gonadotrophins also modulate ERK, for what it is worth.

Sam Gandy: The PKC work was from our group. Julia Mills and Peter Reiner showed that ERK does the same thing in intact cells: i.e., APPs-alpha generation goes up and Ab generation goes down.

Mark Smith: Sam...were these in primary cells?

Sam Gandy: PKC and ERK data are both from primary cortical neuronal cultures prepared by Gunnar Gouras.... In cell-free systems, Huaxi Xu (a member of our group at Cornell, now at Rockefeller with Paul Greengard,) showed that PKC stimulated budding of AβPP transport vesicles (Abstract). It's a bit of a mystery to explain how the competition occurs (between alpha & beta pathways) since even with activated alpha cleavage, at least half of the new APP molecules are trashed without making any discrete fragments.

Guest 2: There is a J. Neurosci paper describing ERK (MAPK) regulation of Ab trafficking/production.

Sam Gandy: that's Mills, 1997 re: ERK and Ab

Mark Smith: We have great data showing that Ab activates ERK...And inhibition of ERK activation protects cells.

June: Are you looking at whether ERK inhibitors increase Ab deposition in Tg mice?

Sam Gandy: We hadn't thought of putting ERK inhibitors into mice since there would be no therapeutic there, but still, it would be interesting to know whether the ERK inhibition had any effect in brain in vivo.

Castrati, etc.

Seth Shaw: Has Mark Smith's question re: rapists and Richard Bowen's question re: Downs been answered?

Mark Smith: Dr. Shaw...thanks for bringing us back on topic!

Gunnar Gouras enters

June: Welcome back, Gunnar!

Sam Gandy: For the answer to the rapist question, see our JAMA Letter on chemical castration and plasma Ab (Gandy, 2001 ). For the record, my lab's collaboration with Ralph Martins & Osvaldo Almeida at UWA in Perth, where the study was performed, didn't castrate rapists; we castrated men with prostate cancer. What was the Down question again?

Richard L. Bowen: The Down question is that even though sex steroid hormone levels are the same in Down as in the general population, the males with Down Syndrome get AD more than females, which is the reverse as seen in the general population. Gonadotrophin levels could explain this, since they are higher in males with Downs, but in the general population gonadotrophins are higher in females. That was more of a statement than a question, really. How would you explain it, Sam?

Gunnar Gouras: I'm interested in the reference for Down syndrome males with earlier onset? [Editor: it is Schupf 1998.]

Sam Gandy: Nicole Schupf is paying a lot of attention these days to the role of hormone deficiency in the onset of Alzheimer's in the Down population. She tells me there is a good correlation between falling hormone levels and onset of dementia (Patel, 2001)

Mark Smith: Sam...does that explain male/female ratio?

Sam Gandy: I'd want to see more replicated data on gender effects in Down syndrome before I'd comment. The "n" in Down syndrome studies is typically very small, and many of the published studies on such small numbers are flawed I doubt that there is a real gender difference in Alzheimer's incidence in normal humans, and I'd be skeptical about claims to that effect in Down's. I'd have to see the data.

Mark Smith: In Down syndrome it is pretty clear that bad stuff is happening early at least 20 years before the onset of pathology...Do hormone or gonadotrophin differences explain?

Sam Gandy: Individuals with Down syndrome have a genetic overdose of AbPP. They don't require hormonal changes to get pathology started early.

Richard L. Bowen: Sam, individuals with DS have higher gonadotrophin levels from conception throughout their lives. Maternal HCG levels are one of the ways that the decision to perform amniocentesis is made.

Craig Atwood: I believe gonadotrophins are altered in Down syndrome (Hsiang, 1987).

Mike Shelanski: A critical issue is where one should measure the "effective" Ab levels. The CSF, the tissue, the thin part, the thick part, serum or plasma? Our data show that low concentrations of Ab inhibit PKA and LTP by increasing the concentration of the PKA regulatory subunit. This is reversible by agents that raise cAMP. It also results in a loss of spine motility in pyramidal cells in slices. We hypothesize that the "only" reason Down syndrome patients are retarded is high Ab levels in development. Does estrogen work only by decreasing Ab, or does it directly regulate synaptic strength via trophic factors?

Sam Gandy: Mike, I'm sure estrogen does many things. I would never point to Ab as the only parameter affected by E2. Re: fluid of interest, I'd say it's the Ab concentration in the interstitial space of the brain, but that is unknowable.

Alexei R. Koudinov: To extend Dr. Shelanski's point - in my view it is well possible that gonadotrophins affect synaptic plasticity, and Ab is somehow involved... The Society for Neuroscience meeting had several important abstracts on Ab and synaptic plasticity. [Search for SFN abstracts.]

Sam Gandy: The problem that I have with acute Ab toxicity studies is that soluble Ab concentration in body fluids changes very little over the lifespan, even during Alzheimer's. Accelerated deposition or impaired clearance seems more likely to me.

Rena Li: I have a question for Dr. Gandy. Do we know anything about estrogen receptors in Ab processing?

Sam Gandy: Estradiol & testosterone control Ab generation from cultured neurons (Xu 1998; Gouras, 2000), and estradiol controls brain Ab levels in vivo (Petanceska 2000; Duff et al., J Neurochem, in press). That J Neurochem paper in press has Karen Duff as last, not first. I forget who is first.

Rena Li: There is a paper, Savaskan 2001, showing estrogen receptor beta immunoreactivity is higher in AD hippocampus than in age-matched controls. Many in vitro studies provide multiple evidence of estrogen's neuroprotective action via estrogen receptor beta. What do you think about the finding from AD brain and the results from the in vitro studies?

Sam Gandy: Estrogens have many genomic and nongenomic actions, the latter being particularly difficult to explain. I think that the best way to sort out whether estrogen receptor alpha or estrogen receptor beta is involved would be to study neuronal cultures from knockout mice. So I would go from the human result to knockout mice and test whether either known receptor was involved.

Sam Gandy: Just to finish the thought: Progesterone receptor was just found to bind and activate SRC, another protein kinase that regulates the ERK pathway. This was a totally unexpected result, but may be relevant to the effects that we are studying. (Boonyaratanakornkit et al. 2001).

Closing Remarks

Mark Smith: I gotta go teach...look forward to the continued dialogue...finally, remember that amyloid is protective, not toxic...but that's another debate!

Craig Atwood: Great.

June: Bye Mark. Yes, let's have another debate!

June: We're approaching the end of our hour. Seems like we're just getting off the ground! To wrap, Sam, what are the critical questions that you'd like to see answered?

Gabrielle Strobel: For the physicians and lay women in our audience who have to decide about HRT before results of the primary prevention trials come in: does research to date on estrogen and AD suggest any new downsides to HRT, in addition to the previously known ones?

Sam Gandy: The single most interesting piece of data will be the outcome of the primary prevention trials with estrogen. The 5-year data are available in 2003, the 10-year in 2008. If they fail, this whole line of investigation will dry up.

Craig Atwood: I guess we'll have to wait!

Richard L. Bowen: On leaving, I just would like to make the point that gonadotrophin levels and sex steroid hormone levels are interdependent and that any study evaluating sex steroid hormone levels and AD should also evaluate gonadotrophin levels.

Alan Lerner: I don't know about gonadotrophin levels per se -- I can send you the references re: androgens, aging and cognition. There is some interesting work on estrogen and steroids with a phenoloic a ring as anti-oxidants, and DHEA and DS levels in aging and human experiments on their replacement, although not in the AD population to my knowledge (J Clin Endocrinology and Metabolism 84(10):3416-3419). I think this is a fascinating are for clinical research.

Richard L. Bowen: Thanks Alan.

June: Sam, you get the final word... Re: Gabrielle's question, what words of advice do you have for women today who are contemplating HRT as a preventive measure?

Alan Lerner: It's an important private health as well as public health issue. I suppose if you have the follow-up screening (pap, mammograms) there is no special added risk. From a public health perspective we've got a ways to go.

Sam Gandy: HRT is now, and will remain, an individual decision between a women and her primary physician or gynecologist. If the prevention study succeeds, then modifying AD risk will enter that discussion.

June: Thanks, Sam, and thanks to all of you for participating.

Nico: Bye everyone and thanks!

Sam Gandy: So long, everyone! Signing off…

June: Time for lunch!

Alexei R. Koudinov: … or for dinner, across Atlantic at the Holy Land.

June: Our next live chat is December 11 at noon, EST. We'll be discussing hypotheses of ApoE's role in AD risk.


See also recent ARF news stories
Neuroprotection From Testosterone? (And A Link Between Hysterectomy And Parkinson's
A New Candidate Neuroprotective Mechanism For Estrogen


Background Text
By Sam Gandy

Many independent lines of evidence implicate β amyloidosis in brain as the key event in the development of Alzheimer's disease (AD). Strongest among this evidence is the linkage of cerebral amyloidosis and the clinical phenotype of autosomal dominant, completely penetrant familial AD to pro-amyloidogenic missense mutations in the amyloid precursor protein (APP) or in one of the presenilins, key regulators of the β-amyloid-generating gamma secretases (Gandy, 1999). Less well defined is how cerebral amyloidosis is initiated or propagated when identifiable mutations are absent, as is the case for the disease that we now know as typical, late-onset, sporadic AD.

One metabolic risk factor that controls the age-at-onset of AD may be gonadal senescence: i.e., menopause in women and andropause in men. Gonadal hormones appear to control neuronal amyloid beta peptide (A-β) metabolism in cultured cells (Xu, 1998; Gouras, 2000), in the brains of experimental animals (Petanceska, 2000), and in the circulation (Gandy, 2001) and cerebrospinal fluid of human subjects (Schonknecht, 2001). Elevated levels of circulating A-β 42 have been associated with an increased risk for AD (Mayeux, 1999; Ertekin-Taner, 2000). Regulated A-β metabolism may underlie these phenomena, perhaps via the protein kinase C-regulated pathway (PKC; Buxbaum, 1993) or extracellular-signal-regulated protein kinase-regulated pathway (ERK; Mills, 1997; Singh, 2000). Direct interaction between one member of the steroid receptor family and the protein kinase src has recently been described (Boonyaratanakornkit, 2001), providing novel evidence for a direct link between hormone receptor signaling and signal transduction via protein phosphorylation.

Relationships might also exist involving hormone withdrawal, A-β metabolism and programmed cell death, or apoptosis. A classical experimental model for apoptosis involves withdrawal of the neurotrophic factor NGF from cultured neurons (Hamburger and Yip, 1984). It is well established that estrogens play key roles in regulating the levels of NGF receptors (Sohrabji, 1994), raising the possibility that estrogen withdrawal might mimic some of the features of trophic factor withdrawal (Zhang, 2001).

Neuroprotective activities have now been discovered for testosterone, acting via the androgen receptor (Hammond, 2001) and for phytoestrogens (Wang, 2001). The hormone/neuroprotection/apoptosis data dovetail well with observations from others, indicating that activation of apoptosis increases A-β generation (LeBlanc, 1995; Gervais, 1999; Guo, 2001), as does oxidative stress (Olivieri, 2001). While it is worth noting that some of these relationships are controversial (Gervais, 1999; Soriano, 2001), the apparent association of these phenomena suggests a possible model for some of the pathways that cause the molecular neuropathology of AD. In such a scenario, propagation of A-β amyloidosis might occur in a situation of diminished protection against caspase activation and oxidative stress; caspase activation and oxidative stress might, in turn, stimulate A-β generation.

This model is summarized in the chart below. This model also provides a mechanism to explain how hormone replacement therapy (HRT) can apparently delay or prevent AD (Tang, 1996), since both A-β generation and caspase activation would be minimized by HRT.


Recent treatment trials involving the prescription of estrogen replacement therapy for existing AD have mostly failed (see reviews by Toran-Allerand, 2000, Marder and Sano, 2000), although a recent treatment trial was more promising (Asthana, 2001). As of this writing, then, the potentially useful therapeutic (or prophylactic) issue surrounding HRT and AD has to do with whether delay or prevention of AD might be an indication for HRT in asymptomatic subjects at high risk for AD. This question is under study in 5- and 10-year primary prevention trials: the results are eagerly awaited and will begin to become available in 2003 (M. Sano, personal communication).


Asthana S, et al. High-dose estradiol improves cognition for women with AD: Results of a randomized study Neurology 57, 605-612; 2001. Abstract.

Boonyaratanakornkit V, et al. Progesterone receptor contains a proline-rich motif that directly interacts with SH3 domains and activates c-src family tyrosine kinases. Molecular Cell 8, 269-280, 2001. Abstract

Buxbaum JD, et al. Protein phosphorylation inhibits production of Alzheimer amyloid ß/ A4 peptide. Proc Natl Acad Sci USA 90, 9195-9198 ; 1993. Abstract

Ertekin-Taner N, et al. Linkage of plasma A-β 42 to a quantitative locus on chromosome 10 in late-onset Alzheimer's disease. Science 290, 2303-2304; 2000. Abstract

Gandy S. Neurohormonal regulation of Alzheimer's beta amyloid precursor metabolism. Trends Endocrinol Metab 10, 273-279; 1999. Abstract

Gandy S, et al. Chemical andropause and amyloid-beta peptide. JAMA 285, 2195-2196; 2001. Abstract

Gervais FG, et al. Involvement of caspases in proteolytic cleavage of Alzheimer's amyloid-beta precursor protein and amyloidogenic A beta peptide formation. Cell 97, 395-406; 1999. Abstract

Gouras G, et al. Testosterone reduces neuronal secretion of Alzheimer's beta-amyloid peptides. Proc Natl Acad Sci USA 97, 1202-1205; 2000. Abstract

Guo Q, et al. Prostate apoptosis response-4 enhances secretion of amyloid beta peptide 1-42 in human neuroblastoma IMR-32 cells by a caspase-dependent pathway. J Biol Chem 276, 16040-16044; 2001. Abstract

Hamburger V, Yip JW. Reduction of experimentally induced neuronal death in spinal ganglia of the chick embryo by nerve growth factor. J Neurosci. 4, 767-74; 1984. Abstract

Hammond J, et al. Testosterone-mediated neuroprotection through the androgen receptor in human primary neurons. J Neurochem 77, 1319-1326; 2001. Abstract

LeBlanc A, et al. Increased production of 4 kDa amyloid beta peptide in serum deprived human primary neuron cultures: Possible involvement of apoptosis. J Neurosci 15, 7837-7846; 1995. Abstract

Mayeux R, et al. Plasma amyloid beta-peptide and incipient Alzheimer's disease. Annals of Neurology 46,412-416; 1999. Abstract

Mills J, et al. Regulation of amyloid precursor protein catabolism involves the mitogen-activated protein kinase signal transduction pathway. J Neurosci. 17, 9415-22 ; 1997. Abstract

Olivieri G, et al. Mercury induces cell cytotoxicity and oxidative stress and increases beta-amyloid secretion and tau phosphorylation in SHSY5Y neuroblastoma cells. J Neurochem. 74,231-236; 2000. Abstract

Petanceska S, et al. Ovariectomy and 17-beta estradiol modulate the levels of Alzheimer's amyloid beta peptides in brain. Neurology 54, 2212-2217; 2000. Abstract

Marder, K., Sano, M. Estrogen to treat Alzheimer's disease: Too little, too late? So what's a woman to do? Neurology 54, 2035-2037; 2000. Abstract

Schonknecht P, et al. Reduced cerebrospinal fluid estradiol levels are associated with increased beta-amyloid levels in female patients with Alzheimer's disease. Neurosci Lett 307, 122-124 ; 2001. Abstract

Singh M, et al. Estrogen-induced activation of the mitogen-activated protein kinase cascade in the cerebral cortex of estrogen receptor-alpha knock-out mice. J Neurosci. 20, 1694-700 ; 2000. Abstract

Sohrabji F, et al. Estrogen differentially regulates estrogen and nerve growth factor receptor mRNAs in adult sensory neurons. J Neurosci. 14, 459-71; 1994. Abstract

Soriano S, et al. The amyloidogenic pathway of amyloid precursor protein (APP) is independent of its cleavage by caspases. J Biol Chem 2001 Aug 3;276(31):29045-50. Abstract

Tang MX, et al. Effect of oestrogen during menopause on risk and age at onset of Alzheimer's disease. Lancet 348, 429-432; 1996. Abstract

Toran-Allerand CD. Estrogen as a treatment for Alzheimer disease. JAMA. 284, 307-308; 2000. Abstract

Wang CN, et al. The neuroprotective effects of phytoestrogens on amyloid beta protein-induced toxicity are mediated by abrogating the activation of caspase cascade in rat cortical neurons. J Biol Chem 276, 5287-5295; 2001. Abstract

Xu H, et al. Estrogen reduces neuronal generation of Alzheimer beta-amyloid peptides. Nature Medicine 4, 447-451; 1998. Abstract

Zhang Y, et al. 17-{beta}-estradiol induces an inhibitor of active caspases. J. Neurosci. 21, 176; 2001. Abstract


  1. Participants in the forum may be aware that there has been an ongoing
    dialogue between myself and Dr. Gandy on the role of gonadal hormones in the
    etiology of Alzheimer's disease and this discussion has been recently
    published as letters in JAMA. While epidemiological evidence implicates a
    role for estrogen/testosterone in AD, and estrogen and testosterone modulate
    APP processing in cell lines and mice, a number of observations indicate
    that the decrease or absence of circulating estrogen/testosterone cannot
    entirely explain AD.

    This is best exemplified by the lack of AD-like
    changes observed during pre-pubescence when circulating concentrations of
    sex steroids during this 12 to 14 year period are extremely low. The
    possible role of the intermediate hormones that regulate estrogen and
    testosterone production have been largely ignored with regards to AD. This
    is despite the facts that changes in sex steroid levels cause a reciprocal
    change in gonadotropin (Gn) levels, Gn's cross the blood brain barrier, and
    that Gn receptors are in the brain with the highest density found in the

    We recently reported a two-fold increase in circulating Gn
    (luteinizing hormone and follicle stimulating hormone) in individuals with
    AD compared with age-matched control individuals (Bowen et al., 2000). We
    are currently investigating potential mechanisms by which Gn's may be
    contributing to the pathogenesis of AD.

  2. Hormone reduction (gonadal, estrogen, testosterone) is one of the most
    salient changes in aging (amid energy and growth factor decline, metal
    imbalance, free radicals, etc.). So it is expected and has been shown
    to contribute to (though may not be solely responsible for) the
    generation of amyloid. I agree with this picture but also consider the
    next two questions crucial:

    1. Through what pathways can hormone reduction lead to amyloid plaques?
    Today everybody says amyloid is due to β- and γ secretases. But
    why and how can hormone reduction eventually activate these enzymes?

    2. Hormone reduction occurs in all elderly, but why do only some of
    them, but not others, develop AD? This may be easily explained by an
    "excessive hormone reduction" in the patients. But what has caused the
    excessive reduction in the first place? Will AD be explained without
    answering these questions?

  3. Reply by Sam Gandy
    The mechanism(s) by which estradiol/testosterone control(s) Aβ levels are yet to be definitively elucidated. Our working model is that estradiol/testosterone activate ERK (MAPK), a signalling pathway well-known to be estrogen-sensitive (Toran-Allerand et al., 2000) and to modulate Aβ release (Mills et al., 1997).

    Assuming that "sporadic" AD is really "polygenic" AD, we would propose a model whereby menopause/andropause "tip the scales" toward amyloidogenesis in individuals who also have the phenotype of marginal abeta economy. This should be testable, since these individuals might have modest elevations in plasma abeta levels.

    I agree that it might be true that those individuals whose hormones fall farthest and/or fastest are probably at the most risk, but this is an opinion that I cannot support with data.


    . Novel sites and mechanisms of oestrogen action in the brain. Novartis Found Symp. 2000;230:56-69; discussion 69-73. PubMed.

    . Regulation of amyloid precursor protein catabolism involves the mitogen-activated protein kinase signal transduction pathway. J Neurosci. 1997 Dec 15;17(24):9415-22. PubMed.

  4. Reply by Sam Gandy
    We have no data on
    gonadotrophins and either Abeta metabolism or Alzheimer risk, though it
    should be relatively straightforward to obtain. In general, I would
    suspect that the neurobiology of hormone withdrawal following decades of
    their presence (menopause and andropause) to be very different from that
    found in the relatively hormone-naive pre-pubertal brain.

  5. Dr. Gandy proposes that estradiol/testosterone activate ERK, so that hormone
    withdrawal would decrease ERK, a reasonable scheme since many
    hormone-regulated pathways are reduced in aging.

    But Aβ is overly produced in the same period and whole world is developing inhibitors in order to reduce the activities of β- and γ-secretases. So the question here is why and how REDUCED ERK activity could OVERLY ACTIVATE β- and γ-secretases. Perhaps a more direct question is, should amyloid deposition be conceived to be due to something overly activated, or rather, due to something decreased? This may be a starting point for our reasoning.

Make a Comment

To make a comment you must login or register.


Webinar Citations

  1. Gonadal Hormone Withdrawal, Apoptosis, and Generation of Beta Amyloid Peptide: A Vicious Cycle of Amyloidogenesis and Neurotoxicity in Alzheimer's Disease

News Citations

  1. Neuroprotection from Testosterone? (And a Link Between Hysterectomy and Parkinson's)
  2. A New Candidate Neuroprotective Mechanism for Estrogen

External Citations

  1. Gandy, 1999
  2. Xu, 1998
  3. Gouras, 2000
  4. Petanceska, 2000
  5. Gandy, 2001
  6. Schonknecht, 2001
  7. Mayeux, 1999
  8. Ertekin-Taner, 2000
  9. Buxbaum, 1993
  10. Mills, 1997
  11. Singh, 2000
  12. Boonyaratanakornkit, 2001
  13. Hamburger and Yip, 1984
  14. Sohrabji, 1994
  15. Wang, 2001
  16. LeBlanc, 1995
  17. Gervais, 1999
  18. Guo, 2001
  19. Olivieri, 2001
  20. Soriano, 2001
  21. Tang, 1996
  22. Toran-Allerand, 2000
  23. Marder and Sano, 2000
  24. Asthana, 2001
  25. Abstract
  26. Abstract
  27. Abstract
  28. Wozniak, 1998
  29. Mehta PD et al., 2001
  30. Vanderstichele H., 2000
  31. Ruitenberg et al. 2001
  32. Tang, 1996
  33. Abstract
  34. Mills, 1997
  35. Gandy, 2001
  36. Schupf 1998
  37. Patel, 2001
  38. Hsiang, 1987
  39. Search for SFN abstracts
  40. Xu 1998
  41. Gouras, 2000
  42. Petanceska 2000
  43. Savaskan 2001
  44. Boonyaratanakornkit et al. 2001

Further Reading