ApoE ranks as the strongest genetic risk factor for late-onset Alzheimer’s, but scientists are still unclear about how the lipoprotein affects the brain. A study in the September 28 Journal of Neuroscience describes a unique mouse model that could help clarify. Scientists led by Courtney Lane-Donovan and Joachim Herz at the University of Texas Southwestern Medical Center, Dallas, created a mouse that expresses no ApoE in the brain but a normal amount in the periphery. Synapses faltered and died away in their brains. Surprisingly, these animals dodged memory deficits seen in complete ApoE knockouts. The authors interpret this to mean that while brain ApoE protects synaptic health in the central nervous system, plasma ApoE helps preserve learning and memory. The authors suggest peripheral ApoE may prevent breakdown of the blood-brain barrier or maintain lipid homeostasis that is essential to brain function.

“This is an elegant approach,” said Daniel Michaelson of Tel Aviv University in Israel. “It’s exciting because it suggests that future treatments targeting ApoE in the blood can help the brain.”

Some studies hint that reducing the ApoE may protect against AD. In mice overexpressing human APP, replacing endogenous ApoE with just one copy of human ApoE3 or ApoE4 reduces accumulation of the Aβ peptide (Dec 2011 news on Kim et al., 2011). Similarly, injecting an anti-ApoE antibody into the peritoneum of an AD mouse model reins in amyloid deposition and improves learning and memory (see May 2014 news on Liao et al., 2014). On the other hand, knocking out the ApoE gene altogether in mice degrades synapses and leads to learning impairments (Masliah et al., 1995). Scientists are not clear what causes these learning problems. Plasma lipid levels shoot through the roof in the ApoE knockouts, causing dyslipidemia and atherosclerosis that may affect the brain. If researchers could restore normal ApoE expression in the periphery of these mice, they might be able to separate out the effects of brain and peripheral ApoE on the central nervous system.

Herz and colleagues got a rare opportunity to do just that with a transgenic mouse they inadvertently created. While replacing mouse ApoE with the human version, the transgene randomly integrated into a spot in the genome where the regulatory elements for expression in the liver were active, but those for production in astrocytes were silent. Those are the cells in the periphery and blood, respectively, that produce ApoE. The authors realized what had happened when they analyzed plasma and brain extracts. The mice, which still made murine ApoE, produced human ApoE only in the plasma. By crossing this mouse with a complete ApoE knockout, they aimed to create a model that expressed human ApoE only in the plasma, serving essentially as a brain ApoE knockout (bEKO).

Synapse Loss.

Synaptophysin puncta (red) sparsely populate the neocortex if ApoE is absent from the brain. [Courtesy of Lane-Donovan, et al. The Journal of Neuroscience 2016.]

Lane-Donovan first used ApoE immunohistochemistry to confirm the genetic cross worked. While the antibody bound to astrocytes and in the interstitial space between brain cells in ApoE3 targeted replacement mice, none bound in the brain of the ApoE KO mice or to astrocytes in the bEKO mice. The antibody did detect a trace of ApoE in the interstitial space in the bEKO animals. The authors were uncertain why, but think ApoE in the blood may have crossed into the brain through leaks in the blood-brain barrier. To see if lack of astrocytic ApoE affected synapses, the researchers probed sections from the neocortices and hippocampi of seven- to eight-month-old mice with an anti-synaptophysin antibody. Compared with the ApoE3 controls, both bEKO and ApoE KO mice bound much less antibody in the neocortex (see image at left). In the hippocampus, synaptophysin levels were similar among all the mice. 

Despite normal synaptophysin in the hippocampus, electrophysiological experiments indicated synaptic plasticity and synapse strength took a hit there in bEKO mice, just as they did in ApoE knockouts. However, in hippocampal slices from ApoE KO mice, the AMPA/NMDA ratio ran lower than in controls, whereas it was normal in the bEKO animals. Maintaining that ratio might explain why the bEKO mice remembered normally, the authors suggested. In the Morris water maze, the bEKO mice spent as much time as controls in the target quadrant in a probe trial, while ApoE KO wasted more time in other quadrants. Interestingly, when the authors parsed the data by gender, they found that the female ApoE KO mice drove the behavioral differences. All males performed equally well, regardless of ApoE status. Michaelson found this gender effect interesting and said that it paved the way for a wider study on how peripheral ApoE affects distinct cognitive parameters such as short- and long-term memory, which are linked to different brain areas and mechanisms. 

Taken together, the results suggest that brain and peripheral ApoE have distinct effects on the CNS, the authors wrote. Brain ApoE is important for synapse development, while ApoE in the blood holds further sway over cognitive function, they conclude.

How could peripheral ApoE do that? By maintaining lipid homeostasis, the lipoprotein may help normalize blood flow to the brain, shore up the blood-brain barrier to keep out neurotoxic factors, or corral free fatty acids, which can impair cognitive function in mice (Heverin et al., 2015), the authors suggested (see image below). They could not exclude the possibility that some ApoE might cross into the brain through a leaky blood-brain barrier in the bEKO mice. This could explain the small amount of ApoE detected in the interstitial space, and could account for the memory advantage over complete knockouts. Gregory Cole, University of California, Los Angeles, suggested that restoring plasma ApoE could also benefit the brain by reducing inflammatory signaling in the circulating myeloid cells that then cross the blood-brain barrier.

ApoE’s Potential Roles. Without astrocytic ApoE, fewer synapses mature (top). Without plasma ApoE (bottom right), the blood-brain barrier may leak and toxins could enter the brain. [Courtesy of Lane-Donovan, et al. The Journal of Neuroscience 2016.]

Notably, this study says nothing about the role of the different isoforms of ApoE, Herz told Alzforum. While ApoE2 protects people from AD and ApoE4 puts them at risk for the disease, scientists still debate whether reducing or increasing ApoE4 in the brain would lower disease risk. “This study gives a more informed basis on which to consider such strategies,” said Herz. “Completely abolishing ApoE expression in the brain may not be the best approach, but reducing it, which has been shown to lead to reduced plaque accumulation in mice, may be useful to mitigate pathology.” Whether this will stave off cognitive impairment is still unclear, he said.

“This paper highlights the potential importance of high plasma lipids and how this can negatively impact brain function,” wrote David Holtzman, Washington University School of Medicine in St. Louis, to Alzforum (see full comment below). It would be interesting to assess other synaptic markers as well as the structure of synapses in these mice to see if they are altered, he said.—Gwyneth Dickey Zakaib 


  1. This work from the Herz lab presents an interesting set of findings. The mice they generated express ApoE in the periphery but not in the brain.

    The main finding is that there is a decrease in the synaptic marker synaptophysin in the neocortex in both full ApoE KO mice and in brain-specific ApoE KO mice.

    The ApoE KO mice also have decreased memory retention in the water maze (only in females), decreased LTP, and decreased EPSCs due to altered NMDA/AMPA ratios, none of which are seen in the brain-specific ApoE KO mice.

    These results suggest that in the setting of complete ApoE KO, which results in very high plasma cholesterol levels as well as no ApoE in plasma, that the very high lipids in some way influence brain function negatively.

    This could be via things such as excessive entry of certain lipids into the brain, damage to the BBB, or altered brain vasculature function. The only abnormality shown in the brain-specific ApoE KO mice was the decrease in synaptophysin staining in the neocortex (but not in the hippocampus). It is interesting this is found in the presence of normal behavioral performance. It is possible that this abnormality arose during development or in adult mice. It would be interesting to assess other synaptic markers in these mice as well as the structure of synapses to see if they are altered in any way. The absence of brain ApoE has little effect on the brain functions analyzed. It will be important to determine the effect of lowering ApoE in the setting of brain injury or disease to see whether positive or negative effects of ApoE are uncovered. This paper highlights the potential importance of high plasma lipids and how this can negatively impact brain function.

  2. This is a very interesting initial look at a new model with selective brain ApoE KO that still shows some synaptic deficits but not the supposedly related cognitive deficits. The authors view this as entirely due to correction of peripheral hyperlipidemia, but it may also be due to ApoE’s well-known impact on inflammatory signaling. ApoE promotes negative inhibitory feedback of miR146a over toll and IL-1 receptor NFkappB signaling in peripheral compartments, including vascular immune cells. We recently extended this to brain and showed ApoE4 targeted replacement mice have deficits in peripheral and central miR146a (Teter et al., 2016). So, restoration of peripheral ApoE should dampen vascular pro-inflammatory signaling that impacts the neurovascular unit where invading myeloid cells are most relevant to pericyte populations.

    It will be interesting to see how well exercise works in the aging brain exercise model where ApoE mediated exercise effects on neuroinflammation (Soto et al., 2016).  It will be interesting if Herz and colleagues study  neuroinflammation in the brains of these bEKO mice, particularly in the  context of aging or some proinflammatory drivers at earlier ages (high fat diet, LPS, or APP Tg/ amyloid). It seems likely that the immunoregulatory effects of peripheral ApoE will play a role in the CNS where the role of the immune system is increasingly relevant.


    . Apolipoprotein E isotype-dependent modulation of microRNA-146a in plasma and brain. Neuroreport. 2016 Aug 3;27(11):791-5. PubMed.

  3. The Herz laboratory generated an important novel mouse model (bEKO) that expresses ApoE normally in the periphery and none in the brain. They compared this model to ApoE knockouts (ApoE KO), which show a profound elevation in plasma cholesterol and triglyceride levels, and ApoE3 homozygous targeted replacement (TR) mice that express the protein everywhere. The bEKO model allows us to study the contribution of ApoE in and outside the brain to CNS function. bEKO mice look like ApoE KO mice in terms of synaptophysin in the neocortex, with lower levels in both as compared to those seen in ApoE3 TR mice. When synaptic physiology is analyzed in hippocampal slices, bEKO and ApoE KO mice both show impaired long-term potentiation (LTP) without alterations in paired-pulse facilitation, as compared to ApoE3 TR mice. However, hippocampal slices of ApoE KO, but not bEKO, mice show lower AMPA/NMDA ratios than those seen in hippocampal slices from ApoE3 TR mice.

    When ApoE KO mice are tested for spatial learning and memory in a water maze paradigm in which mice are first trained to locate a hidden platform, female, but not male, ApoE KO mice show an impaired ability to locate the hidden platform. This is consistent with our 1998 study, where female mice expressing ApoE4 in the brain showed impairments in the water maze task compared to males (Raber et al., 1998). These sex differences are interesting by themselves, considering the previously reported sex-dependent effects of ApoE4 on cognitive performance of ApoE KO mice that express ApoE4 in the brain and the ability of androgens (Raber et al., 2002) and selective androgen receptor modulators (Acevedo et al., 2008) to treat them, in addition to the sex differences in humans with ApoE4 in risk to develop AD (Farrer et al., 1997). 

    However, in another previous study, when ApoE KO mice were first trained to locate a visible platform, female ApoE KO mice showed no impairment when they were subsequently trained to locate a hidden platform (Raber et al., 2000). So it is possible that increased measures of anxiety and activation of the hypothalamic-pituitary-adrenal (HPA) axis in the ApoE KO mice (Raber et al., 2000; Robertson et al., 2005) and altered histamine receptor-mediated signaling (Bongers et al., 2003) might contribute to the poor performance of ApoE KO mice in the water maze version in which the mice are trained first to locate a hidden platform. For HPA axis activation, the deficiency of ApoE in the adrenal gland might be critical. Restraint stress-induced plasma corticosterone levels in ApoE KO mice expressing ApoE3 or ApoE4 only in the brain are not significantly different and comparable to those in ApoE KO mice without human ApoE expression (Raber et al., 2000). The bEKO model will allow us to study the role of ApoE in the periphery, including the adrenal gland, in modulating measures of anxiety and HPA axis activation when there is very little ApoE in brain. The ability to compare bEKO mice expressing different ApoE isoforms would be ideal for such studies.

    Our previous study reported that ApoE KO mice showed no impairment either in spatial memory retention in the probe trial (platform removed), even when tested at 18 months of age. Age-matched ApoE KO mice expressing ApoE4 only in the brain did show impairment in spatial memory impairment in the water maze probe trial. Based on these results, increasing ApoE4 levels in brain is unlikely beneficial for cognitive performance.

    Although the ApoE KO mice show a loss of synaptic markers in the Lane-Donovan study, the cognitive impairments are subtle and sex-dependent, as acknowledged by the authors. However, in previous studies, ApoE KO mice showed profound impairments in cognitive tests other than the water maze that might have been more related to impaired cholinergic function than impaired glutamatergic function. The ApoE KO mice have a reduced number of cortical and hippocampal muscarinic acetylcholine receptors (mAChRs), are more sensitive to effects of scopolamine on delay eyeblink classical conditioning, and are unable to acquire performance on the five-choice serial reaction time tasks (5-SRTT) (Siegel et al., 2011). We should be open to the possibility that ApoE has other effects on cognition that aren’t examined here. 


    . Isoform-specific effects of human apolipoprotein E on brain function revealed in ApoE knockout mice: increased susceptibility of females. Proc Natl Acad Sci U S A. 1998 Sep 1;95(18):10914-9. PubMed.

    . Androgens protect against apolipoprotein E4-induced cognitive deficits. J Neurosci. 2002 Jun 15;22(12):5204-9. PubMed.

    . Selective Androgen Receptor Modulators Antagonize Apolipoprotein E4-Induced Cognitive Impairments. Lett Drug Des Discov. 2008;5(4):271-6.

    . Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA. 1997 Oct 22-29;278(16):1349-56. PubMed.

    . Apolipoprotein E and cognitive performance. Nature. 2000 Mar 23;404(6776):352-4. PubMed.

    . Hypothalamic-pituitary-adrenal dysfunction in Apoe(-/-) mice: possible role in behavioral and metabolic alterations. J Neurosci. 2000 Mar 1;20(5):2064-71. PubMed.

    . apoE isoforms and measures of anxiety in probable AD patients and Apoe-/- mice. Neurobiol Aging. 2005 May;26(5):637-43. PubMed.

    . Role of H3-receptor-mediated signaling in anxiety and cognition in wild-type and Apoe-/- mice. Neuropsychopharmacology. 2004 Mar;29(3):441-9. PubMed.

    . Hypothalamic-pituitary-adrenal dysfunction in Apoe(-/-) mice: possible role in behavioral and metabolic alterations. J Neurosci. 2000 Mar 1;20(5):2064-71. PubMed.

    . Selective Androgen Receptor Modulators Antagonize Apolipoprotein E4-Induced Cognitive Impairments. Lett Drug Des Discov. 2008;5(4):271-6.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Lowering ApoE Brings Down Amyloid in Mice
  2. Has ApoE’s Time Come as a Therapeutic Target?

Paper Citations

  1. . Haploinsufficiency of human APOE reduces amyloid deposition in a mouse model of amyloid-β amyloidosis. J Neurosci. 2011 Dec 7;31(49):18007-12. PubMed.
  2. . Anti-ApoE antibody given after plaque onset decreases Aβ accumulation and improves brain function in a mouse model of Aβ amyloidosis. J Neurosci. 2014 May 21;34(21):7281-92. PubMed.
  3. . Neurodegeneration in the central nervous system of apoE-deficient mice. Exp Neurol. 1995 Dec;136(2):107-22. PubMed.

Further Reading


  1. . A human apolipoprotein E mimetic peptide reduces atherosclerosis in aged apolipoprotein E null mice. Am J Transl Res. 2016;8(8):3482-92. Epub 2016 Aug 15 PubMed.
  2. . The Complex Role of Apolipoprotein E in Alzheimer's Disease: an Overview and Update. J Mol Neurosci. 2016 Nov;60(3):325-335. Epub 2016 Sep 19 PubMed.

Primary Papers

  1. . Genetic Restoration of Plasma ApoE Improves Cognition and Partially Restores Synaptic Defects in ApoE-Deficient Mice. J Neurosci. 2016 Sep 28;36(39):10141-50. PubMed.