Article In Brief
In two separate studies and papers, investigators found that APOE4 had a direct effect on alpha-synuclein levels and disease progression in a genetically-modified mouse model of Lewy body dementia, but also in a virally-delivered model, an alpha-synuclein fibril injection model, and in human populations as well.
Raising its pathogenic profile, apolipoprotein E (APOE) has now been shown in two studies to directly regulate alpha-synuclein levels and disease progression in Lewy body dementias independent of amyloid-beta or tau pathologies.
Previous studies had established that the APOE4 genotype is strongly associated with Alzheimer's disease (AD) and that it affects amyloid-beta and tau pathologies. APOE has also previously been linked to the risk of Lewy body dementias, which often have mixed amyloid-beta and tau pathologies in addition to the hallmark Lewy bodies composed of alpha-synuclein. The question was whether APOE4 exerts its effect in Lewy body dementias via amyloid-beta and tau pathology, or whether it directly worsens alpha-synuclein aggregation.
The direct effect was seen not only in a genetically-modified mouse model, but also in a virally-delivered model, an alpha-synuclein fibril injection model, and in human populations as well. The two papers were published simultaneously in the February 5 issue of Science Translational Medicine.
“It's nice when you do science separately with different models but reach similar conclusions,” said Guojun Bu, PhD, senior author of one of the papers and the Mary Lowell Leary Professor and chair of neuroscience at Mayo Clinic in Jacksonville, FL.
But not all the findings of the two papers were consistent, Dr. Bu noted.
“The other group found that APOE2 is protective against alpha-synuclein pathology,” he said. “In our paper we did not see that. Our APOE2 and APOE3 mice were similar.”
The first author of the other paper pointed out another difference between the findings of the two papers.
“We measured the expression of about 700 genes in our mouse model, many of which go up in the context of inflammation,” said Albert (Gus) Davis, MD, PhD, assistant professor of neurology at Washington University School of Medicine in St. Louis. “At 12 months, about half of the APOE4 and APOE knockout-mice had developed alpha-synuclein pathology. The expression of inflammatory-associated genes seemed to be associated with whether or not they had developed the pathology or not. That was interesting and a bit unexpected.”
But, Dr. Davis emphasized, “The findings of our two groups are more similar than they are different. The way to think of our results is to view APOE as a direct regulator of the principal pathology that occurs in Lewy body dementia.”
Patrik Brundin, MD, PhD, director of the Center for Neurodegenerative Science and the Jay Van Andel Endowed Chair at the Van Andel Research Institute in Grand Rapids, MI, said the papers show that the lines between AD, Lewy body dementia, and Parkinson's disease are becoming “increasingly blurred.”
“This is truly fascinating, and the findings suggest there might be opportunities to develop therapies that target more than one of these neurodegenerative disorders in the future, which is really hopeful,” Dr. Brundin said.
In one of the studies, Dr. Davis's group crossed a line of mice specially bred to develop alpha-synuclein pathology (A53T mice) with either an APOE knockout (EKO) or with one of three human APOE knock-in backgrounds: APOE2, E3, or E4.
At 12 months, the E4 mice had accumulated higher amounts of insoluble and phosphorylated alpha-synuclein compared with the EKO and E3 mice. In the E2 mice, soluble and phosphorylated alpha-synuclein was entirely undetectable. Immunohistochemistry testing found that the E4 mice had a higher burden of reactive gliosis compared with the E2 mice. What's more, the E2 mice survived longer and did not show the pattern of motor deficit seen in the other APOE genotypes.
In what was deemed a model of alpha-synuclein “spreading,” Dr. Davis's group also injected alpha-synuclein preformed fibrils into the striatum of E2, E3, E4, and EKO mice. They found greater accumulation of alpha-synuclein pathology in the substantia nigra of the E4 mice compared with the E2 or EKO mice.
Finally, in two different cohorts of human patients with Parkinson's disease, those with two copies of the E4 allele displayed the fastest cognitive decline of their peers.
“Our results demonstrate that APOE genotype directly regulates alpha-synuclein pathology independent of its established effects on Abeta and tau, corroborate the finding that APOE e4 exacerbates pathology, and suggest that APOE e2 may protect against alpha-synuclein aggregation and neurodegeneration in synucleinopathies,” concluded the paper, whose senior author was David M. Holtzman, MD, FAAN, the Andrew B. and Gretchen P. Jones professor and chair of the department of neurology at Washington University.
The study led by Dr. Bu took a different approach to interrogating the role of APOE in alpha-synuclein pathology. Rather than using mice bred to develop high levels of alpha-synuclein, his group used an adeno-associated virus gene delivery of the protein. They injected that into mice expressing either human APOE2, APOE3, or APOE4.
They found that the mice expressing E4, but not E2 or E3, had increased alpha-synuclein pathology, impaired behavior, greater neuronal and synaptic loss, and increased astrogliosis at nine months. They also found altered lipid and energy metabolism and synapse-related pathways when they profiled the E4's transcriptome.
Dr. Bu and colleagues examined brains from people who had Lewy body dementia from a brain bank maintained by Mayo Clinic. “We looked at the sub-cohort of brains where there is very little amyloid-beta,” Dr. Bu said. “We then asked whether having E4 increases alpha-synuclein pathology in those brains. The answer was yes.”
Neither group knew that they were both investigating similar questions until late in 2018. “Both of our teams presented some preliminary work at a Gordon Research Conference,” Dr. Davis said. “We discovered we were working on the same question and had some consistent results using different model systems, which I think adds to the power of the findings overall. We decided we would do better, and the scientific community would do better, if we worked in a coordinated fashion rather than competed with each other.”
The fact that Dr. Davis's group found increased alpha-synuclein pathology associated with inflammatory gene expression, while Dr. Bu's group found altered lipid and energy metabolism, may not be as contradictory as they appear, Dr. Davis said. “Those may speak to some of the nuances of the animal models, and not necessarily to a fundamental difference in the results,” he said.
Eric M. Reiman, MD, executive director of Banner Alzheimer's Institute, said the two papers, “conducted by leaders in the field, capitalized on experimental models and human cohorts to suggest that E4 may have an additional, direct impact on alpha-synuclein pathophysiology that is not mediated by amyloid-beta or tau pathology.”
He emphasized, however, that he would like to see the results confirmed in a larger cohort of Parkinson's disease and controls, “including those without comorbid Alzheimer's disease.”
In a paper published in January in Nature Communications, Dr. Reiman and colleagues examined the APOE-associated odds ratios in over 5,000 neuropathologically verified Alzheimer's disease cases and controls. In addition to their Alzheimer's disease findings, Dr. Reiman's group found that the E4 allele was associated with higher odds for Lewy body dementia (before and after adjustment for the presence or absence of Alzheimer's), but that the E2 allele was not associated with lower risk of the disorder than seen in those with E3.
Jeffrey Kordower, PhD, the Alla V. and Solomon Jesmer Professor of Neurological Sciences at Rush University, was particularly intrigued by Dr. Davis's findings regarding the role of inflammatory gene expression. In a paper published last year in Brain, Dr. Kordower and colleagues reported finding evidence of inflammation with activated microglia beginning at 18 months in the brains of patients with advanced Parkinson's disease who had received fetal brain implants 14 to 16 years earlier; he did not find alpha-synuclein aggregated, however, until 14 to 16 years after transplantation. That paper concluded it appears that “microglia play an integral role in the propagation and spread of alpha-synuclein pathology.”
“We looked at the sub-cohort of brains where there is very little amyloid-beta. We then asked whether having E4 increases alpha-synuclein pathology in those brains. The answer was yes.”
—DR. GUOJUN BU
“The inflammation occurs well before the Lewy body forms,” Dr. Kordower told Neurology Today.
Dr. Davis said his own paper's findings may suggest a reversed order: “We observed that inflammation tended to occur only in A53T mice that had alpha-synuclein pathology, so one possible conclusion is that alpha-synuclein pathology drives inflammation in this mouse model, but we could not determine exactly which process came first in our experiment. We and others in the field are going to look closely at that and follow up.”
Alice Chen-Plotkin, MD, the Parker Family associate professor of neurology at the University of Pennsylvania School of Medicine, said, “The data for E4 being bad is much stronger than for E2 being good.”
She pointed out that the APOE2 mouse model had a decreased risk of developing alpha-synuclein pathology compared with the knockouts.
Dr. Chen-Plotkin professed surprise, however, that the E4 genotype was found to raise the risk of cognitive decline independent of any intervening effect of beta amyloid.
“There is a lot of evidence that APOE has an effect on amyloid-beta, including from David Holtzman's group,” she said. “Many of us thought that the different isoforms of APOE cause their effects on alpha-synuclein by interacting differently with amyloid-beta. That could still be true in part. All this paper shows is that there is another mechanism independent of amyloid-beta.”
Dr. Chen-Plotkin added that she was struck by Dr. Davis's findings of an effect of APOE4 on reactive glia.
“The field is coming to think about more than just the neurons,” she said. “We also have to think about the support cells. This paper had a nice demonstration that there is something going on with the APOE genotype and the glia.”
Dr. Chen-Plotkin was the senior author of a paper published in Movement Disorders in 2018, which found that the E4 allele was associated with a 3.5 times higher risk of worsening cognitive diagnosis over time in 100 patients with Parkinson's disease.
Human trials targeting alpha-synuclein are already underway, including an open-label phase 1 gene therapy trial in which an adeno-associated virus carrying the gene for E2 is delivered intravenously to the CSF of 15 patients with very mild to severe Alzheimer's disease. The trial, which is expected to be completed by December 2021, is being led by Ronald G. Crystal, MD, of Weill Cornell Medicine.
While praising the rigor of the studies overall, Dr. Brundin objected to the use of the word “spreading” in the Davis paper.
“When injecting preformed fibrils into the striatum and monitoring aggregates in the nigra, or loss of neurons in the nigra, one is only assessing the uptake and retrograde transport of synuclein assemblies that act as seeds,” Dr. Brundin said. “To address ‘spreading’ would require examining brain regions that are ‘one further synapse away.’”
Drs. Bu, Davis, Kordower and Chen-Plotkin had not disclosures. Dr. Reiman has been a scientific advisor to several companies involved in the development of AD drugs and biomarkers, including Alzheon, Alkahest, Aural Analytics, Denali, Green Valley, MagQ, Takeda, United Neuroscience, and Zinfandel; he has been a scientific advisor to Roche and Roche Diagnostics (travel expenses only).