Comparing exome sequencing data from over 16,000 patients with Alzheimer’s disease (AD) and controls, using gene-based burden analysis instead of the more commonly used approach of genome-wide association studies (GWAS), scientists at the Vrije University (VU) of Amsterdam, have identified significant associations of rare damaging genetic variants with AD risk, that are predictive or suggestive.

The findings were published in the journal Nature Genetics on November 21, 2022 “Exome sequencing identifies rare damaging variants in ATP8B4 and ABCA1 as risk factors for Alzheimer’s disease.”

“Our results provide additional evidence for a major role for amyloid-β precursor protein processing, amyloid-β aggregation, lipid metabolism and microglial function in AD,” the authors noted.

Heritability, the ratio of genetic variations to the variation in clinical manifestation, in AD is nearly 70%. However, GWAS studies, commonly used to assess the genetic risk factors for AD, do not capture the risk contributed by rare genetic variants. To date, GWAS have linked 75 most common genetic risk factors with AD in people of European ancestry, but individually these common variants display low effects. Other analytical approaches have identified rare damaging variants in the genes TREM2, SORL1 and ABCA7 that have higher effects on individual AD risk than GWAS approaches.

In addition to the known variants in TREM2, SORL1 and ABCA7, the current study observed a significant association of rare, predicted damaging variants in ATP8B4 and ABCA1 with AD risk, and a suggestive signal in ADAM10, as well as rare-variant burden in the genes RIN3, CLU, ZCWPW1 and ACE. The investigators also found that variants associated with the strongest effect on AD risk, particularly those variations that result in a loss of a functional gene product, are found at greater frequency in patients with early-onset AD.

New damaging variants  

The dataset used in this burden analysis consisted of independent individuals of European ancestry who were not related by family. The study reported that damaging mutations in ATP8B4—an ATPase enzyme that transports phospholipids in the cell membrane—occur in 3.6% early-onset patients, 3.1% late-onset patients and 2.1% individuals without dementia.

Holstege
Henne Holstege, PhD, an assistant professor of clinical genetics at Vrije University (VU) of Amsterdam, is the senior author of the study.

Senior author of the study Henne Holstege, PhD, an assistant professor of clinical genetics at VU said, “We find that missense mutations [in ATP8B4] associate with a higher increased risk (1.6-fold increased risk in early-onset AD cases compared to non-carriers) compared to truncating mutations (1.2-fold), which suggests that the deleterious effects may be due to gain-of-function missense mutations.”

A similar analysis showed damaging mutations in ABCA1—a major regulator of cellular cholesterol and phospholipid—occur in 1.5% early-onset patients, 1.1% late-onset patients, and 0.52% individuals without dementia.

Holstege said, “Here, truncating mutations associate with a higher risk of AD (4.7-fold increase) compared to missense mutations (2.7-fold), which suggests that damaging or losing protein function underlies the observed increased risk.”

On the other hand, the investigators observed damaging mutations in ADAM10—a cell surface enzyme that cleaves proteins—in only 0.23% early-onset patients, 0.05% late-onset patients, and 0.02% individuals without dementia.

“Carrying a damaging variant is associated with a 9-fold increased risk of AD,” said Holstege. “These variants include protein truncating and missense variants, suggesting that losing protein function or protein impairment underlies the increased risk.”

Previous studies on genetic variations that contribute to increased risk for AD, have identified variants in genes involved in lipid-turnover, the endo-lysosomal system, and the neuro-immune system.

“The newly identified genes fit this picture,” said Holstege. “ABCA1 maintains healthy cholesterol and phospholipid levels in the brain cells, and it is associated with lower levels of aggregated amyloid protein. Like ABCA1, ATB8B4 is involved in the transport of phospholipids, mainly in the brain’s immune cells. ADAM10 is also involved in the processing of the amyloid-β precursor protein, but in such a way that it prevents amyloid-β protein from being formed.”

Limitations

Holtege admits that the current study only gets at the low-hanging fruit in this large dataset. She believes the construction of even larger datasets through international collaborations will help identify additional genes that have a significantly higher burden of rare damaging variants in AD patients, compared to controls.

Another obstacle in the burden analysis was the presence of confounding batch effects in the sequence data.

“It took us an extremely long time to analyze this dataset, because the data was heavily affected by batch effects, as the data was generated at different sites, using different sequencing techniques, and different sequencing machines,” said Holstege. “This led to unique features in genetic sequences that have nothing to do with being an Alzheimer’s Disease patient.”

Holstege’s team designed and applied learning algorithms to eliminate these batch effects.

Amyloid and others

Despite numerous studies that support the increased deposition of β-amyloid in AD patients, the β-amyloid theory of AD has met with some harsh criticism due to the lack of efficacy of AD drugs that target β-amyloid deposition or degradation. The recent success of amyloid-clearing agents such as Aducanumab or Lecanemab might tip the balance in favor of the β-amyloid theory again.

Holstege believes the abundance of amyloid in the brain and the functional pleiotropy of the genes associated with AD risk, make it necessary that in addition to treatments that remove brain amyloid, scientists target core pathological mechanisms. Removing amyloid alone may not provide a solution, as the evidence indicates.

“Early treatment with Aducanumab or Lecanemab may be very important for effectivity,” said Holstege. “Additionally, the field needs to focus on generating treatments that ‘correct’ or support the endogenous mechanisms involved in protein processing and clearance. When applied to at-risk individuals before onset of disease such agents may prevent the load of amyloid or other aggregating proteins to accumulate to disease-associated levels.”

The findings of the current study uncover alternative genetic mechanisms that might lead to precision therapeutic approaches for AD patients.