Cholesterol clears the brain once metabolized to 24-hydroxycholesterol, but that is not the only way 24HC can be beneficial, according to Silvia Maioli at the Karolinska Institutet in Stockholm. In the January 24 Science Advances, Maioli and colleagues reported that 24HC drives estrogen receptor expression in neurons. This likely explains why female, but not male, mice overexpressing the cholesterol 24-hydroxylase gene, CYP46A1, sported bulkier synapses and remembered better than controls.

The authors believe these sex differences might translate to people, and could affect ongoing trials of efavirenz, an allosteric activator of the 24-hydroxylase enzyme. In women, but not men, with Alzheimer’s disease, high cerebrospinal fluid levels of 24HC correlated with low CSF total tau, implying that increased brain cholesterol metabolism is neuroprotective only in women.

“This paper … defines some of the molecular pathways by which sex, through differences in cholesterol handling toward neurosteroid hormones, might contribute to AD risk and pathogenesis,” wrote Rik van der Kant at Vrije University, Amsterdam (comment below).  

To better understand the role of brain cholesterol metabolism in cognition, first author María Latorre-Leal overexpressed human CYP46A1 in wild-type mice. A modified β-actin promoter drove the enzyme’s expression in neurons and several peripheral organs, increasing 24HC in the brain and plasma. The authors did not measure by how much human CYP46A1 was overexpressed in the brain.

Latorre-Leal found that CYP46A1-overexpressing females outperformed controls in Y-maze and water maze tests, indicating better working and spatial memory. In their hippocampi, they had similar dendritic spine density as wild-type females but up to 75 percent more postsynaptic PSD95 and NMDA receptors 1 and 2A (image below). Conversely, CYP46A1-overexpressing males had worse memories and spent less time in the open field than did controls, suggesting they were more anxious. They had fewer dendrites and synaptic proteins than wild-type males.

Sex and the Synapse. In female mice, neuronal and dendritic morphology were similar (seen by Golgi silver stain) whether they overexpressed CYP46A1 (top right) or not (top left). Male CYP46A1-overexpressing mice (bottom right) had fewer dendrites than wild-type (bottom left). [Courtesy of Latorre-Leal et al., Sci Adv, 2024.]

Do hormonal changes explain the sex difference? CYP46A1-overexpressing females had as much estrogen in their brains as did wild-type females but expressed 25 percent more estrogen receptor β in their hippocampi. ERβ protein levels also increased by one-quarter in cultured wild-type mouse hippocampal neurons given 24HC but not in cells also given testosterone. This receptor promotes neurogenesis and synaptic plasticity, and activation of ERβ has been reported to increase synaptic protein production and dendrite density (reviewed by Vargas et al., 2016; Engler-Chiurazzi et al., 2017; Liu et al., 2008). Notably, CYP46A1-overexpressing males had about 1.5 times more testosterone in their brains than wild-type.

In women, estrogen production slows after menopause. To test how this might factor in, the scientists removed the gonads from 2-month-old mice. Six months later, CYP46A1 females still did better in the Y and water mazes than controls, which now had faulty memories, a known effect of ovariectomy. The findings suggest that 24HC can compensate for too little estrogen, perhaps by making better use of the hormone via more abundant receptors. They also hint that CYP46A1 upregulation might shield against cognitive decline seen after menopause. On the other hand, removing testes restored anxiety, learning, and memory in CYP46A1 males to normal, fitting with the idea that 24HC conspires with testosterone to worsen neural deficits. The scientists did not measure brain levels of sex hormones in these mice.

All told, overexpressing CYP46A1 benefitted synapses and memory in females, possibly due to better estrogen signaling in the brain, while harming memory in males.

The scientists did not study CYP46A1 overexpression in mice with amyloid or tau pathology. Previous work reported reduced Aβ burden and improved spatial memory in female amyloidosis models overexpressing CYP46A1 in their hippocampi (Hudry et al., 2010). 

Are the findings relevant to human aging? Latorre-Leal analyzed CSF from 30 adults with subjective cognitive decline, 30 with mild cognitive impairment, and 30 with AD from the Karolinska University Hospital memory clinic. Their average age was 65; half were women. CSF 24HC did not vary by age or sex. However, high levels of the metabolite correlated with decreased CSF total tau, p-tau181, phospho-tau181/Aβ42 ratio, and neurofilament light—only in women. These results hint that 24HC might protect against tau pathology in women.

Others have suggested that 24HC, or efavirenz, hastens tau propagation in the mouse brain by enabling tau entry into neurons (May 2022 news).

Conversely, van der Kant and colleagues found that efavirenz reduces phospho-tau accumulation in human neurons. They believe boosting 24HC production limits formation of cholesteryl esters, which hobble the proteasome and allow toxic forms of tau to accumulate (Feb 2019 news). A Phase 1 trial for MCI was attempted but only recruited five people. Van der Kant and colleagues are planning a Phase 2a study in both men and women. “It seems important for possible later trials to take sex into account when evaluating the effects of pharmacological CYP46A1 activation on cognition,” van der Kant noted.

Getting rid of the related cholesterol 25 (not 24) hydroxylase gene ablated 25-HC and also prevented tau from aggregating in mice (Oct 2023 conference news).—Chelsea Weidman Burke


  1. This is an interesting study by the Maioli lab, with possibly important implications for stratifying patients in AD trials with cholesterol-targeting interventions. Over the last year, our understanding of the links between cholesterol, AD risk genes, and AD pathogenesis has increased, but there are still many unknowns. One major unknown is how altered cholesterol metabolism in the brain affects steroid hormones. In the brain, cholesterol at the ER can be stored (as cholesteryl esters) or secreted (as 24-hydroxycholesterol). Alternatively, cholesterol enters mitochondria, where it is an essential building block for many (less studied) bioactive compounds including sex hormones.

    Thus, the logical prediction is that, when cholesterol metabolism is changed—by AD risk genes or through cholesterol-targeting interventions as in this manuscript—these steroid hormones will also change. Indeed, the authors find such a correlation and give an example on how this linkage between cholesterol metabolism and hormones can affect the outcome of an intervention targeting cholesterol metabolism. They show that CYP46A1 activation can enhance estrogen signaling and rescue cognitive impairments induced by ovariectomy in females, but has little effect in gonadectomized males. Also important for clinical trial designs is that CYP46A1 overexpression improved cognition in female mice, but worsened cognition in males and caused anxiety-like behavior.

    How well this translates to humans is of course the big question. CYP46A1 overactivation is a very strong way of inducing conversion of cholesterol to 24-hydroxycholesterol, and it is unclear how this relates to pharmacological activation of CYP46A1 by, e.g., efavirenz. We are currently pursuing CYP46A1 activation by low-dose efavirenz in an early dose-finding Phase 2a clinical trial. Based on the data in this paper, it seems important for possible later trials to take sex into account when evaluating the effects of pharmacological CYP46A1 activation on cognition.

    I also find this paper interesting as it defines some of the molecular pathways by which sex (through differences in cholesterol handling towards neurosteroid hormones) might contribute to AD risk and pathogenesis.

  2. This is an interesting study to obtain additional insight into the higher prevalence of Alzheimer’s disease in women than men. Mice with transgenic CYP46A1 overexpression were used, and sex-specific behavioral changes were correlated with dendritic spine morphology, the levels of synaptic proteins, and activation of estrogen signaling in female brain during aging. Subsequent analyses of the human CSF samples suggested female-specific effects of the CYP46A1 product 24-hydroxycholesterol on t-tau.

    The data obtained align well with our previous finding that efavirenz-mediated activation of CYP46A1 in 5XFAD mice, hence brain 24-hydroxycholesterol levels, affect steroidogenesis in the brain, at least in male mice (Mast et al, 2020). 

    Yet we found a decrease in efavirenz-treated 5XFAD male mice in the levels of pregnenolone, the precursor of all steroid hormones, and therefore, of dehydroepiandrosterone, the precursor of estradiol and testosterone, and estradiol. Testosterone was below the limits of detection by gas chromatography-mass spectrometry. This is in contrast to increased levels of dihydroxytestosterone in aged CYP46A1 transgenic male mice. Further studies are required to clarify if a different pattern of changes as a result of increased brain 24-hydroxycholesterol levels is due to differences between CYP46A1 transgenic expression versus pharmacologic CYP46A1 activation, steroid measurements in the brain hippocampus versus whole brain, age of 9 months versus 18 months, and use of ELISA versus gas chromatography-mass spectrometry for steroid quantifications.

    Also, the authors did not measure estradiol, which is well detected in the male brain, and did not consider steroid hormone sulfation, which affects the biological activities of many steroids, including estradiol.

    Of importance is that in mice, but not humans, Cyp46a1 mRNA is also detected in testes, yet the protein product was not detected (Lund et al, 1999). Accordingly, it is conceivable that transgenic CYP46A1 overexpression in mice leads to protein expression in testes and affects systemic levels of testosterone, consistent with the author’s data on decreased serum levels of 24-hydroxycholesterol in gonadectomized male mice. Thus, caution should be exercised while interpreting the data in male mice and relating them to humans.

    Despite certain issues that need to be clarified, the study’s main message–24-hydroxycholesterol effects should be assessed in both female and male mice; women and men–is well supported. We have a pharmacologic tool, low-dose efavirenz, to allosterically activate CYP46A1 and brain cholesterol turnover in the brain of humans (Lerner et al., 2022). 


    . CYP46A1-dependent and independent effects of efavirenz treatment. Brain Commun. 2020;2(2):fcaa180. Epub 2020 Oct 29 PubMed.

    . cDNA cloning of cholesterol 24-hydroxylase, a mediator of cholesterol homeostasis in the brain. Proc Natl Acad Sci U S A. 1999 Jun 22;96(13):7238-43. PubMed.

    . CYP46A1 activation by low-dose efavirenz enhances brain cholesterol metabolism in subjects with early Alzheimer's disease. Alzheimers Res Ther. 2022 Dec 29;14(1):198. PubMed.

Make a Comment

To make a comment you must login or register.


Therapeutics Citations

  1. Efavirenz

News Citations

  1. Membrane Border Patrol: Cholesterol Stymies Tau Uptake, Aggregation
  2. Cholesteryl Esters Hobble Proteasomes, Increase p-Tau
  3. Cracking the Cholesterol-AD Code: Metabolites and Cell Type

Paper Citations

  1. . The functions of estrogen receptor beta in the female brain: A systematic review. Maturitas. 2016 Nov;93:41-57. Epub 2016 Jun 4 PubMed.
  2. . Estrogens as neuroprotectants: Estrogenic actions in the context of cognitive aging and brain injury. Prog Neurobiol. 2017 Oct;157:188-211. Epub 2016 Feb 15 PubMed.
  3. . Activation of estrogen receptor-beta regulates hippocampal synaptic plasticity and improves memory. Nat Neurosci. 2008 Mar;11(3):334-43. PubMed.
  4. . Adeno-associated virus gene therapy with cholesterol 24-hydroxylase reduces the amyloid pathology before or after the onset of amyloid plaques in mouse models of Alzheimer's disease. Mol Ther. 2010 Jan;18(1):44-53. PubMed.

External Citations

  1. Phase 1 trial

Further Reading

Primary Papers

  1. . CYP46A1-mediated cholesterol turnover induces sex-specific changes in cognition and counteracts memory loss in ovariectomized mice. Science Advances, Jan 24, 2024 Science Advances