07 Feb 2020

ApoE4’s rap sheet just grew longer. Not only does ApoE’s notorious allele accelerate Aβ and tau pathology, it also drives the toxic aggregation of α-synuclein, according to two studies published February 5 in Science Translational Medicine. One was led by Guojun Bu at the Mayo Clinic in Jacksonville, Florida; the other, by David Holtzman at Washington University in St. Louis. Both reported that ApoE4 aggravated α-synuclein phosphorylation, worsened motor and memory problems, and ramped up neurodegeneration in different mouse models. Once again, the goody-two-shoes allele, ApoE2, was protective, Holtzman and colleagues reported. People with Lewy body dementia who inherited an ApoE4 allele had more phosphorylated α-synuclein and faster cognitive decline than noncarriers.

Patrik Brundin of the Van Andel Research Institute in Grand Rapids, Michigan, said the studies further blur the line between Alzheimer’s and Lewy body dementias, in terms of both genetic risk and neuropathology. “The findings suggest that there might be opportunities to develop [single] therapies that target more than one of these neurodegenerative disorders in the future,” he wrote. 

Dementia with Lewy bodies (DLB) and Parkinson’s disease dementia (PDD) both fall under the umbrella of Lewy body dementias (LBDs), which hit patients with a double whammy of movement problems and cognitive impairment. Lewy bodies chock-full of aggregated α-synuclein underlie these disorders, but many people with LBD also have Aβ and tau pathology.

ApoE4, the strongest genetic risk factor for late-onset Alzheimer’s, also increases the risk for cognitive decline in people with PD, and is a risk factor for DLB as well (Mata et al., 2014; Dec 2015 news; Guerreiro et al., 2018). Studies hinted that in LBDs, ApoE4 works in much the same way as it does in AD, by ramping up Aβ and/or tau pathology (Irwin et al., 2017; Sep 2016 news). However, others have reported that ApoE4 promotes α-synuclein pathology independently (Tsuang et al., 2013; Dickson et al., 2018). 

To directly assess this, researchers led by Bu used a mouse model of synucleinopathy. First author Na Zhao injected adeno-associated viruses expressing human α-synuclein into both lateral ventricles of newborn ApoE-targeted replacement mice. These express human ApoE2, ApoE3, or ApoE4 in place of the endogenous mouse gene.

Nine months later, the scientists used two antibodies—one that recognizes pathological conformations of α-synuclein in people with LBD, and the other specific for α-synuclein phosphorylated at serine-129—to measure α-synuclein pathology in multiple brain regions. They found roughly double the amount of both pathological conformers and phosphorylated α-synuclein in several brain regions of αSyn-ApoE4 compared with αSyn-ApoE2 or αSyn-ApoE3 mice. Immunoreactivity of the two antibodies correlated tightly.

This uptick in α-synuclein pathology had consequences. All the αSyn mice were unbalanced and uncoordinated compared with controls, but ApoE4 mice fared worst. They also spent more time in an exposed area, suggesting they had lost some of their normal fear and inhibition. While αSyn-ApoE2 and αSyn-ApoE3 mice readily remembered sounds that signaled an impending foot shock, αSyn-ApoE4 mice had trouble on this memory task.

Normal neuron and synaptic protein levels in E2 and E3 mice stood in contrast to 10 percent fewer neurons in ApoE4s and a dip in levels of the postsynaptic protein PSD95. E4 mice had astrogliosis and microglial activation that were absent in αSyn mice expressing the other alleles. However, the researchers found no correlation between the burden of pathological conformers of α-synuclein in a given mouse and its degree of astrogliosis or microglial activation, suggesting the stepped-up neuroinflammation in ApoE4 mice was unrelated to the increased α-synuclein pathology.

A transcriptomic analysis revealed lower expression of genes involved in lipid metabolism and less glycogen synthase activity in ApoE4 mice, as well as suppression of genes needed for synaptic signaling, protein metabolism, and social behavior.

Does ApoE4 influence α-synuclein pathology in LBD? To find out, Zhao and colleagues measured pathologic conformations and p-Ser129 α-synuclein in postmortem brain samples from 44 people who had had pure LBD with little to no Aβ pathology. Half were ApoE4 carriers. On average, they had more p-Ser129 α-synuclein than age-matched noncarriers, but not more of the pathological conformation. The researchers detected no differences in astrocytic or microglial markers between carriers and noncarriers, nor did glial activation markers correlate with the burden of p-Ser129 α-synuclein across all participants. Together, these findings suggested that ApoE4 exacerbates α-synuclein pathology in people with LBD but does not necessarily stoke neuroinflammation.

Bu said the lack of a direct link between the burden of α-synuclein pathology and glial activation was unsurprising. As an intracellular protein, α-synuclein is less likely to incite glial cells than extracellular aggregates, such as Aβ.

In the second paper, first author Albert Davis and colleagues used a different mouse. They crossed A53T-αSyn transgenic mice with the same ApoE-targeted replacement mice Bu and colleagues used. Compared with the viral approach, the A53T-αSyn mice develop more severe disease, becoming paralyzed by 12 months of age. While Bu’s injected mice had α-synuclein pathology in many brain regions, the vast majority of α-synuclein inclusions in the A53T-αSyn mice developed in the brainstem, with sparse pathology in the neocortex.

Analyzing brainstem lysates from 12-month-old mice, Davis detected the highest levels of insoluble α-synuclein, as well as p-Ser129 α-synuclein, in ApoE4 mice. ApoE3 and ApoE knockouts had progressively less, and in lysates from ApoE2 mice, insoluble and phosphorylated forms of α-synuclein were virtually undetectable. Immunohistochemistry of brainstem sections told a similar story, with ApoE2 mice having no detectable phosphorylated α-synuclein, while ApoE4 had the most. Interestingly, p-Ser129 α-synuclein only appeared in mice with end-stage paralysis, which afflicted many ApoE4 and ApoE knockout mice by 12 months of age. ApoE2 mice did not develop severe paralysis until much later, around 18 months, suggesting that the allele staved off both α-synuclein pathology and neurodegeneration.

The scientists also compared transcriptomes. Both the amount of α-synuclein pathology and end-stage paralysis tracked with higher expression of proinflammatory genes, but not with ApoE genotype alone. This suggested that neuroinflammation arose in response to α-synuclein pathology and neurodegeneration, as opposed to being caused by ApoE4. Davis also picked up modules of myelination and anti-apoptotic genes—these were more highly expressed in ApoE2 mice.

Compared with Bu’s study, Davis’ found stronger ties between the burden of α-synuclein pathology and neuroinflammation. The reason for this discrepancy is unclear, but both researchers believe it could come down to differences between mouse models. Because the A53T-αSyn mice express much higher levels of α-synuclein than the virally infected model, the former could evoke a more pro-inflammatory component, Bu suggested.

Davis next investigated the relationship between ApoE genotype and cognition in several cohorts of PD patients. Among 251 people with PD in the Parkinson’s Progression Markers Initiative, the researchers found faster decline on the Montreal Cognitive Assessment in the ApoE4 carriers. While people with abnormal CSF Aβ42 or p-tau also declined faster on this screen, ApoE4 correlated independently. In two other PD cohorts, the researchers found ApoE4 carriers declined more sharply on the mini-mental state examination.

While the two papers came to slightly different conclusions in some respects, their main conclusion was the same, Bu said: “ApoE4 can drive α-synuclein pathology in the complete absence of amyloid.”

Exactly how ApoE does that remains unclear. ApoE is an extracellular protein, whereas the vast majority of α-synuclein resides inside, attached to synaptic vesicles. Still, small pools of α-synuclein have been detected outside of cells and, when aggregated, it can spread between neurons via a templated misfolding mechanism, at least in mice. It is possible the two lipophilic proteins meet up in this extracellular milieu, Bu said. In support of this, Davis also reported that preformed fibrils of α-synuclein injected into the mouse striatum spread more readily into neighboring substantia nigra when the animals expressed ApoE4.

Thomas Montine of Stanford University favors the idea that ApoE isoforms sway α-synuclein aggregation indirectly, via their role as immune modulators. He noted that, in general, ApoE4 promotes pro-inflammatory responses, while ApoE2 does the opposite.

If ApoE4 drives α-synuclein aggregation, then why is it a risk factor for LBD, but not PD? Davis said the answer could come down to where in the brain ApoE4 influences α-synuclein aggregation. Perhaps the allele is most impactful in cortical and limbic regions involved in cognition, he speculated. These regions are more affected by α-synuclein pathology in LBD than in PD, in which α-synuclein aggregates primarily pop up in the midbrain. Davis plans to use mouse models with more widespread α-synuclein aggregation to investigate this. Montine concurs that neuropathological studies suggest that ApoE4 strongly affects α-synuclein aggregation in cortical and limbic regions (Dickson et al., 2018). Another possibility is that ApoE4 promotes the propagation of α-synuclein pathology from the midbrain into these areas, Davis said.

Despite myriad unknowns, the new findings cast ApoE as a potential therapeutic target not only in AD, but also in LBD, Bu and Davis agreed. Montine noted that the protective effect of ApoE2 reported in Davis’s study is important, especially in light of recent reports that the allele strongly protects against AD (Feb 2020 news). A therapy that mimics the behavior of ApoE2 could benefit patients with LBD and AD, he said.—Jessica Shugart

Comments

1. • Sonja Scholz
     National Institutes of Health
   • Posted: 11 Feb 2020

   Genetic studies have shown that the APOE ε4 allele not only markedly increases risk for Alzheimer’s disease but also strongly associates with risk for developing Lewy body dementia (LBD). The exact role of APOE in LBD, however, is poorly understood. One school of thought is that APOE drives amyloid co-pathology in LBD; alternatively, it is also possible that APOE directly triggers α-synuclein aggregation.

   To study the effect of APOE on α-synuclein, these two groups generated transgenic synucleinopathy mouse models. In the Zhao et al. study, intracerebral injection of AAV-α-synuclein into animals with different APOE backgrounds was performed, while the Davis et al. study used A53T SNCA transgenic mice on either knockout or APOE knock-in backgrounds. Remarkably, both studies found that animals with an APOEε4 background exhibited increased α-synuclein pathology with associated reactive gliosis compared with APOEε2 mice. Additionally, APOEε4 mice exhibited behavioral deficits.

   These findings are important observations as they support the hypothesis that APOEε4 not only drives Alzheimer’s disease pathology, but it also stimulates α-synuclein aggregation. Equally notable is the finding that, contrary to APOEε4, the ε2 allele appears to impart resilience to protein aggregation. What is less clear is what the exact molecular cascades are that lead to the deposition of these different proteins. Further, we do not know whether pathological β-amyloid, tau, and α-synuclein proteins are interacting and exacerbating their respective accumulations or whether these proteins aggregate independently from each other. Advancing our understanding of the genetic factors that are driving concomitant proteinopathies is crucial for the development of disease-modifying interventions.

   View all comments by Sonja Scholz
2. • Patrik Brundin
     Associate Director of the Van Andel Research Institute, and Director of the Center for Neurodegenerative Science
   • Posted: 11 Feb 2020

   These two papers support each other, and, coming from two different teams and using different approaches, that is clearly a great strength.

   When it comes to neuropathology and genetic risk, the lines between AD, DLB, and PD (especially PDD) are becoming increasingly blurred the more that we learn about these diseases. It is truly fascinating, and these two papers add important information to this developing story. One implication of the findings is that several of these neurodegenerative disorders might share common therapeutic targets. Thus, there is hope that in the future it might be possible to develop single therapies that provide disease-modifying benefit to more than one of the disorders. The emerging pattern that inflammation is a potential driver (as opposed to just a consequence of neurodegeneration) is particularly interesting. Also, the papers also touch upon the notion that proteinopathies depend on interactions with lipid homeostasis, which is another exciting emerging theme in the research field. These two insights might say something about where future disease-modifying treatments could be developed.

   The collective data in the two papers are strong and very convincing, but, as is often the case with large studies, all the datasets are not equally conclusive. For example, the data on α-synuclein preformed fibril (PFF) injections into the striatum (Fig. 6 in the Davis et al. paper) leave several unanswered questions. The authors eloquently discuss some shortcomings, and, for example, mention that they only examined one short (three months) survival time point. In this model, it would have been desirable to follow animals for six months for α-synuclein pathology and neuronal loss to fully develop in the substantia nigra. Furthermore, it appears that nigral neurons were only counted on four sections, as opposed to unbiased stereology throughout the substantia nigra. More insight would also have been gained if the nigral neurons had been stained with an additional method beyond tyrosine hydroxylase immunohistochemistry. This would clarify if cells were dying, or just displaying reduced levels of tyrosine hydroxylase in a subset of neurons.

   Finally, the use of the word “spreading” of pathology is misleading in the context of this experiment. When injecting PFFs into the striatum and monitoring aggregates in the substantia nigra, or loss of neurons in the nigra, one is only assessing the uptake and retrograde transport of α-synuclein assemblies that act as seeds for endogenous α-synuclein. To address “spreading” would require examining brain regions that are “one further synapse away.”

   View all comments by Patrik Brundin

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Genetics of DLB: Setting Up to Fill a Mostly Empty Canvas 29 Dec 2015
2. Tau Deepens Cognitive Trouble in Lewy Body Diseases 23 Sep 2016
3. Paper Alert: Two ApoE2s Provide “Exceptional” Protection Against AD 5 Feb 2020

Research Models Citations

1. APOE2 Targeted Replacement
2. APOE3 Targeted Replacement
3. APOE4 Targeted Replacement

Paper Citations

1. Mata IF, Leverenz JB, Weintraub D, Trojanowski JQ, Hurtig HI, Van Deerlin VM, Ritz B, Rausch R, Rhodes SL, Factor SA, Wood-Siverio C, Quinn JF, Chung KA, Peterson AL, Espay AJ, Revilla FJ, Devoto J, Hu SC, Cholerton BA, Wan JY, Montine TJ, Edwards KL, Zabetian CP. APOE, MAPT, and SNCA genes and cognitive performance in Parkinson disease. JAMA Neurol. 2014 Nov;71(11):1405-12. PubMed.
2. Guerreiro R, Ross OA, Kun-Rodrigues C, Hernandez DG, Orme T, Eicher JD, Shepherd CE, Parkkinen L, Darwent L, Heckman MG, Scholz SW, Troncoso JC, Pletnikova O, Ansorge O, Clarimon J, Lleo A, Morenas-Rodriguez E, Clark L, Honig LS, Marder K, Lemstra A, Rogaeva E, St George-Hyslop P, Londos E, Zetterberg H, Barber I, Braae A, Brown K, Morgan K, Troakes C, Al-Sarraj S, Lashley T, Holton J, Compta Y, Van Deerlin V, Serrano GE, Beach TG, Lesage S, Galasko D, Masliah E, Santana I, Pastor P, Diez-Fairen M, Aguilar M, Tienari PJ, Myllykangas L, Oinas M, Revesz T, Lees A, Boeve BF, Petersen RC, Ferman TJ, Escott-Price V, Graff-Radford N, Cairns NJ, Morris JC, Pickering-Brown S, Mann D, Halliday GM, Hardy J, Trojanowski JQ, Dickson DW, Singleton A, Stone DJ, Bras J. Investigating the genetic architecture of dementia with Lewy bodies: a two-stage genome-wide association study. Lancet Neurol. 2018 Jan;17(1):64-74. Epub 2017 Dec 16 PubMed.
3. Irwin DJ, Grossman M, Weintraub D, Hurtig HI, Duda JE, Xie SX, Lee EB, Van Deerlin VM, Lopez OL, Kofler JK, Nelson PT, Jicha GA, Woltjer R, Quinn JF, Kaye J, Leverenz JB, Tsuang D, Longfellow K, Yearout D, Kukull W, Keene CD, Montine TJ, Zabetian CP, Trojanowski JQ. Neuropathological and genetic correlates of survival and dementia onset in synucleinopathies: a retrospective analysis. Lancet Neurol. 2017 Jan;16(1):55-65. PubMed.
4. Tsuang D, Leverenz JB, Lopez OL, Hamilton RL, Bennett DA, Schneider JA, Buchman AS, Larson EB, Crane PK, Kaye JA, Kramer P, Woltjer R, Trojanowski JQ, Weintraub D, Chen-Plotkin AS, Irwin DJ, Rick J, Schellenberg GD, Watson GS, Kukull W, Nelson PT, Jicha GA, Neltner JH, Galasko D, Masliah E, Quinn JF, Chung KA, Yearout D, Mata IF, Wan JY, Edwards KL, Montine TJ, Zabetian CP. APOE ε4 increases risk for dementia in pure synucleinopathies. JAMA Neurol. 2013 Feb;70(2):223-8. PubMed.
5. Dickson DW, Heckman MG, Murray ME, Soto AI, Walton RL, Diehl NN, van Gerpen JA, Uitti RJ, Wszolek ZK, Ertekin-Taner N, Knopman DS, Petersen RC, Graff-Radford NR, Boeve BF, Bu G, Ferman TJ, Ross OA. APOE ε4 is associated with severity of Lewy body pathology independent of Alzheimer pathology. Neurology. 2018 Sep 18;91(12):e1182-e1195. Epub 2018 Aug 24 PubMed.

Further Reading

Papers

1. Morley JF, Xie SX, Hurtig HI, Stern MB, Colcher A, Horn S, Dahodwala N, Duda JE, Weintraub D, Chen-Plotkin AS, Van Deerlin V, Falcone D, Siderowf A. Genetic influences on cognitive decline in Parkinson's disease. Mov Disord. 2012 Apr;27(4):512-8. PubMed.

News

1. Lewy Pathology in DLB Spreads Fast, Maybe From the Nose 28 Dec 2015

Webinars

1. Betwixt and Intermixed—Dementia With Lewy Bodies 15 Jun 2009