Does ApoE4 affect aspects of Alzheimer’s disease other than amyloidosis? Animal studies have hinted as much, and now several brain imaging studies seem to agree. In a preprint posted to medRχiv on October 8, researchers led by Mark Bondi at the University of California, San Diego, report that at a given level of tau pathology, Alzheimer's Disease Neuroimaging Initiative participants with an ApoE4 allele perform worse on memory tests than noncarriers. The findings imply that ApoE4 amplifies the toxicity of tangles. Meanwhile, researchers led by Vijay Ramanan at the Mayo Clinic in Rochester, Minnesota, examined whether ApoE genotype exerts direct effects on tangles, independent of amyloidosis. In the October 23 JAMA Network Open, they reported finding few, although among cognitively healthy people with amyloid plaques, ApoE4 carriers did accumulate more tangles in the entorhinal cortex than did noncarriers. In that study, women of any genotype appeared to be more susceptible to the consequences of tangles than men, with worse brain metabolism at a given tangle burden. Other recent work suggests that ApoE4 may pack the biggest punch in women, with female carriers accumulating more tangles and having worse memories than male carriers.

  • ApoE4 carriers have worse memories than noncarriers with the same tau burden.
  • ApoE4 carriers with amyloid plaques have more tangles than do noncarriers.
  • The effects of ApoE4 are worse in women than men.

David Holtzman at Washington University in St. Louis said more such imaging studies are needed to sort out how ApoE affects events downstream of amyloidosis. “The major prediction from the animal studies is that relative to other ApoE isoforms, ApoE4 will be linked with greater inflammation and greater progressive neurodegeneration once tau pathology develops,” he wrote to Alzforum (full comment below).

Holtzman’s work first implicated ApoE4 in exacerbating tau. In mouse tauopathy models, his group found that mice expressing a human E4 allele developed more phosphorylated tau, inflammation, and neurodegeneration than did those with other isoforms. Some data hinted the findings could apply to people. In a large observational cohort followed for 10 years, ApoE4 carriers declined faster cognitively than did noncarriers (Sep 2017 news). 

Bondi and colleagues followed up on this by examining how the ApoE4 allele affects cognition. First author Alexandra Weigand analyzed data from 297 ADNI participants with an average age of 76. Just over a quarter had mild cognitive impairment, the rest were cognitively healthy, and one-third of the cohort carried ApoE4. As expected, carriers accumulated more amyloid and had more tau tangles in the medial temporal lobe (MTL) than noncarriers. Also as expected, after controlling for tau levels, the amount of amyloid a person had did not independently affect his or her cognitive performance. Tau pathology did. After controlling for amyloid, MTL tau associated with worse performance on tests of attention, executive function, language, and memory.

Unexpectedly, ApoE4 modified this association. At a given tau burden, ApoE4 carriers scored worse than noncarriers on memory tests, though not in the other cognitive domains. ApoE4 had this effect even in people at Braak stage I/II, when tangles are limited. Even among people without any amyloid plaques, ApoE4 carriers had worse memories than noncarriers with the same tau burden. “Our findings suggest that ApoE may exert deleterious effects on cognition through specific interactions with tau pathology, and these effects may occur independently of and prior to amyloidosis,” the authors noted.

Ramanan and colleagues took a different approach. They examined tau pathology and brain metabolism, not cognition, in ApoE4 carriers, analyzing data from 325 participants in the Mayo Clinic Study of Aging whose average age was also 76. Most were cognitively healthy, and 29 percent carried an ApoE4 allele. Across the whole cohort, tau pathology was no different between carriers and noncarriers, after adjusting for amyloid burden. However, when the researchers analyzed only people with a positive amyloid scan, ApoE4 carriers did have more tangles in the entorhinal cortex than noncarriers. “This suggests that ApoE4 may accelerate tau pathology in key AD regions, but only in the presence of amyloid, a conclusion consistent with prior postmortem neuropathology data in the ROS/MAP cohorts,” Ramanan wrote to Alzforum.

Notably, regardless of genotype, women with more entorhinal cortical tau had more sluggish brain metabolism, as seen by FDG PET, than men. To the authors, this suggests that women may succumb faster to the toxic effects of tangles. “This adds to a growing literature on the role of sex differences in AD, an exciting area that is getting a lot of attention,” Ramanan wrote.

Regarding women, a previous analysis of ADNI data found that p-tau levels in cerebrospinal fluid rise faster in female ApoE4 carriers than in noncarriers and men, suggesting a sex-specific effect of the E4 allele. Other research indicates that women with ApoE4 lose more white matter at menopause than do noncarriers, perhaps rendering their brains more prone to decline (Aug 2018 conference news; Buckley et al., 2019; Feb 2019 news). Other studies have reported a similar sex effect for ApoE4 (Altmann et al., 2014; May 2018 news). 

Since then, researchers led by Yun Zhou at WashU and Rongfu Wang at Peking University First Hospital, Beijing, analyzed 108 ADNI participants with mild cognitive impairment. In the July Theranostics, they reported that the 15 female ApoE4 carriers had higher tau signals throughout the brain than did noncarriers and men, contradicting the Mayo Clinic study. The difference was most pronounced in entorhinal cortex, amygdala, parahippocampal gyrus, and fusiform gyrus. Female carriers also had more p-tau and total tau in CSF.

In a new paper in the October Neurology, Bondi and colleagues add to the evidence of a sex difference among ApoE4 carriers. Erin Sundermann at the University of California, San Diego, led this study, which focused on whether setting sex-specific cutoffs for verbal memory would improve the clinical detection of mild cognitive impairment. This is necessary because women routinely score better on tests of verbal memory than do men, and indeed, raising the cutoff for women and lowering it for men in a cohort of 985 ADNI participants sharpened diagnostic accuracy for MCI by 20 percent. Importantly, the authors found that women with memory loss were five times as likely to have an ApoE4 allele as were women whose memory was stable. Among men, there was no difference in ApoE4 prevalence among those with poor or preserved memory.

“This raises the possibility that the vulnerability to developing amnestic MCI in female ApoE4 carriers may have been underestimated in studies reporting no sex difference in the effect of the ApoE4 allele on AD risk,” the authors note. This study had no tau PET data, but CSF p-tau tracked with memory, being elevated in people with MCI, and somewhat higher in MCI women than men.—Madolyn Bowman Rogers


  1. Two interesting papers being published look at the effects of APOE genotype on different aspects of AD, in particular tau pathology and cognitive progression of disease.

    Bondi and colleagues studied 297 older adults without dementia being followed in ADNI. They found that after adjusting for amyloid PET, medial temporal lobe tau PET was associated with cognitive abnormalities and the strongest effects were in APOE4 carriers and with high levels of tau. Interestingly, this association persisted even in amyloid-negative individuals. This suggests that APOE4 is affecting tau pathology and cognitive progression in the presence or absence of amyloid deposition.

    Ramanan and colleagues studied 325 individuals in the Mayo Clinic Study of Aging 65 years of age and older who were cognitively unimpaired. They found that while APOE4 was associated with high tau deposition as seen on Tau PET in the entorhinal cortex and a lower FDG-to-tau ratio, these associations appeared to be attenuated when controlling for amyloid deposition.

    It is clear that APOE genotype has a huge effect on amyloid deposition in the brain, with APOE4-positive individuals having significantly more amyloid deposition on average than is present with APOE3, and even less amyloid deposition is seen with APOE2. Since amyloid deposition has been shown in both human and animal studies to in some way drive tau pathology, it has been difficult to untangle whether APOE may also be influencing tau pathology and the progression of cognitive decline and dementia in AD independent of the effect of APOE on amyloid.

    We recently found in the PS19 mouse model of tau pathology (in the absence of amyloid), which develops both significant tau accumulation as well as marked neurodegeneration, that APOE is involved in driving tau-mediated neurodegeneration, with APOE4 having the strongest effect and the lack of APOE abrogating the effect of neurodegeneration almost entirely (Shi et al., 2017). 

    In addition, we also just found that the effect of APOE is completely dependent on microglia (Shi et al., 2019), as did another group (Mancuso et al., 2019). Of note is that while APOE and APOE4 in these animal studies had some effect on tau accumulation, the biggest effect was on the ability of APOE and APOE4 to influence the amount of microglial activation and neurodegeneration.

    If these new findings were to translate to the human brain, what one might expect is that the presence of APOE4 might be associated with somewhat greater tau pathology in either the absence or presence of amyloid deposition, but that a much bigger effect of APOE4 would be on the amount of microglial activation that occurred over time with tau pathology (expected to be larger with E4) and that the progression of cognitive decline would be faster.

    The study of Bondi et al. did find greater tau pathology in the presence of E4 even in the absence of amyloid, as well as greater cognitive progression. The study by Ramanan et al. did not find a greater effect of E4 on tau pathology once the amount of amyloid was corrected for. To fully understand in humans the effect of APOE on microglial activation and progression of neurodegeneration, longitudinal studies need to be done with the types of imaging biomarkers used in these studies but also markers of neuroinflammation (when available), amyloid, tau, and also structural MRI and cognitive assessments.

    The major prediction from the animal studies is that relative to other APOE isoforms, APOE4 will be linked with greater inflammation and greater progressive neurodegeneration once tau pathology develops. These two studies are beginning to get at the amyloid and tau part of the story and to some extent cognitive change in the Bondi et al. paper. Further studies will be needed to sort out these other issues in humans during normal aging, preclinical AD, AD, and in primary tauopathies.


    . ApoE4 markedly exacerbates tau-mediated neurodegeneration in a mouse model of tauopathy. Nature. 2017 Sep 28;549(7673):523-527. Epub 2017 Sep 20 PubMed.

    . Microglia drive APOE-dependent neurodegeneration in a tauopathy mouse model. J Exp Med. 2019 Nov 4;216(11):2546-2561. Epub 2019 Oct 10 PubMed.

    . CSF1R inhibitor JNJ-40346527 attenuates microglial proliferation and neurodegeneration in P301S mice. Brain. 2019 Oct 1;142(10):3243-3264. PubMed.

  2. We and Weigand et al. both take advantage of the availability of tau PET to ask important—but quite different—questions about AD pathophysiology. APOE ɛ4 is well-known to be associated with amyloidosis, but whether this allele may directly impact tau has been less clear.

    The ADNI study is primarily focused on cognition. Weigand et al. confirms prior work that medial temporal lobe tau burden is likely more proximally related to cognition than cortical amyloid load. The study also reports that APOE ɛ4 carriers display worse memory than noncarriers for the same level of medial temporal tau and after adjusting for amyloid. As a result, the authors conclude that APOE ɛ4 may interact with tau independently of amyloid. One challenge here is that cognition is a late endpoint in AD and is likely influenced by multiple factors including but not limited to amyloid and tau pathology. The finding that APOE ɛ4 strengthens the negative association between tau and cognition may be related to a mechanistic impact of APOE on tau as the authors hypothesize, but may alternatively reflect effects and factors downstream of tau. Nevertheless, it’s an interesting study which effectively calls more attention to the topic.

    Our Mayo Clinic Study of Aging work was focused on investigating factors that may be related to resistance, i.e., avoiding and conversely susceptibility to tau, and resilience, i.e., coping with tau. As part of this work, we directly tested whether APOE ɛ4 was associated with regional tau burden using a population-based sample. In fact we found that global amyloid load was the factor most strongly associated with regional tau deposition, and that after accounting for this effect, there were no main associations of APOE ɛ4 with regional tau burden.

    Interestingly, we did find that in the presence of substantial amyloidosis, for the same level of amyloid, APOE ɛ4 carriers displayed higher levels of tau in the entorhinal cortex compared to noncarriers. Taken together, this suggests that APOE ɛ4 may accelerate tau pathology in key AD regions, but that this would appear to be only in the presence of amyloid (a conclusion consistent with prior postmortem neuropathology data in the ROS/MAP cohorts).

    Our work also suggested that women may be vulnerable to greater neuronal damage (measured by FDG-PET) than men at the same level of entorhinal cortex tau burden. This adds to a growing literature on the role of sex differences in AD, which is an exciting area that is getting a lot of attention.

  3. We thank Alzforum for covering this important area of research on the relationship between APOE genotype and tau PET. We’d like to provide a note on the seemingly contradictory findings between Ramanan et al. and our group (Weigand et al.). The primary outcome of the Ramanan et al. study, to examine differential effects of ε4 genotype on tau PET level, was not a direct goal of our study. Notably, we did find higher tau PET levels among ε4 carriers when examining group differences, but we did not assess whether this effect occurred independently of amyloid. Rather, the focus of our study was on associations between tau and cognition, as well as the moderating effect of ε4 carrier status on those associations, with all effects examined independently of amyloid.

    We found that tau PET level was associated with all cognitive domains, whereas amyloid PET exhibited no such associations. Further, we found that ε4 carriers had a stronger association between tau PET and memory independently of amyloid PET level, and this moderating effect persisted when we included only those individuals with negative amyloid PET scans. Thus, methodological differences in model specification may explain discrepancies between our findings and those of Ramanan et al. It’s possible that ε4 genotype may not directly increase tau accumulation, but rather may interact with tau pathology to exert compounding consequences on neuronal and cognitive integrity.

    This interactive effect may be largely influenced by inflammatory mechanisms, as has been demonstrated in Shi et al., 20172019, in which ε4 genotype exacerbated tau-mediated neurodegeneration primarily via microglial activation (see Oct 2019 news). This effect occurred independently of amyloid, as did our findings of ε4 carrier status strengthening the association between tau and memory.

    Thus, while tau pathology and ε4 genotype may not directly influence one another (although this remains to be determined), they may have additive negative effects on neuronal and cognitive outcomes. However, further converging evidence using multiple methods (e.g., animal models, neuropathological studies, PET) are needed to better disentangle the complex interactions between APOE, amyloid, and tau.


    . ApoE4 markedly exacerbates tau-mediated neurodegeneration in a mouse model of tauopathy. Nature. 2017 Sep 28;549(7673):523-527. Epub 2017 Sep 20 PubMed.

    . Microglia drive APOE-dependent neurodegeneration in a tauopathy mouse model. J Exp Med. 2019 Nov 4;216(11):2546-2561. Epub 2019 Oct 10 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. ApoE4 Makes All Things Tau Worse, From Beginning to End
  2. Do Brain Changes at Menopause Make Women More Prone to Alzheimer’s?
  3. Is a Woman’s Brain More Susceptible to Tau Pathology?
  4. Study Finds Sex Influences CSF Tau Levels in ApoE4 Carriers

Paper Citations

  1. . Associations between baseline amyloid, sex, and APOE on subsequent tau accumulation in cerebrospinal fluid. Neurobiol Aging. 2019 Jun;78:178-185. Epub 2019 Mar 7 PubMed.
  2. . Sex modifies the APOE-related risk of developing Alzheimer disease. Ann Neurol. 2014 Apr;75(4):563-73. Epub 2014 Apr 14 PubMed.

External Citations

  1. medRχiv 

Further Reading

Primary Papers

  1. . Association of Apolipoprotein E ɛ4, Educational Level, and Sex With Tau Deposition and Tau-Mediated Metabolic Dysfunction in Older Adults. JAMA Netw Open. 2019 Oct 2;2(10):e1913909. PubMed.
  2. . Sex-specific norms for verbal memory tests may improve diagnostic accuracy of amnestic MCI. Neurology. 2019 Nov 12;93(20):e1881-e1889. Epub 2019 Oct 9 PubMed.
  3. . Sex modulates the ApoE ε4 effect on brain tau deposition measured by 18F-AV-1451 PET in individuals with mild cognitive impairment. Theranostics. 2019;9(17):4959-4970. Epub 2019 Jul 9 PubMed.