06 Sep 2024

Decades before memory loss becomes a concern, the biological processes that lead to Alzheimer’s disease are underway in the brain. Could proteins in the blood sound the alarm for this pending crisis? Answering this question is a goal of proteomics, a rapidly advancing field that is being used to plumb biofluids for disease-related proteins.

Three hundred plasma proteins predicted dementia among Icelandic elders.

Fewer than half did so regardless of ApoE genotype; most were neuronal.
ApoE4-dependent and -independent associations point to different paths toward AD.

Case in point: Researchers led by Valborg Guðmundsdóttir and Vilmundur Gudnason of the University of Iceland, Reykjavik, measured about 4,000 proteins within the plasma of more than 5,000 older Icelandic people who were cognitively normal at baseline. Concentrations of some 300 proteins tracked with an AD diagnosis over the following decade. Fewer than half of these did so regardless of a person’s ApoE genotype, and were involved in neuronal processes such as axon development.

A group of 17 plasma proteins were tied to AD only in ApoE4 carriers, and their expression appeared regulated by ApoE4. Curiously, four of these were downregulated by ApoE4, yet upregulated as AD onset crept closer among noncarriers. The findings point to biological pathways that converge on AD, via still-mysterious contributions of ApoE4 to disease.

“The study is elegant and underscores the wisdom of combining a well-defined prospective sample set from a geographically delimited population with a robust proteomic analysis platform,” wrote Gerold Schmitt-Ulms of the University of Toronto. “Improvements in signal-to-noise afforded by these choices can be expected to have been instrumental for bringing to the fore a whole host of new LOAD-associated proteins.”

In recent years, proteomics techniques have improved, allowing researchers to probe thousands of proteins churning within plasma, cerebrospinal fluid, and brain tissue. They have uncovered individual proteins and groups of functionally related proteins that correlate with ongoing AD, or predict its rise in the future (Aug 2023 news; Feb 2024 news; Feb 2024 news). Many proteins track with different AD endophenotypes. They hint at cellular mechanisms afoot at different stages of disease, and at biologically distinct subtypes of AD (Dec 2023 news).

Converging evidence from proteomics studies casts ApoE4 as powerfully influencing proteins found in biofluids, yet how these relate to AD risk had not been investigated (Goodjonsson et al., 2022; Cruchaga et al., 2023).

For this study, first author Elisabet Frick and colleagues took advantage of the Age, Gene/Environment Susceptibility—Reykjavik Study. AGES enrolled 5,127 older Icelandic adults who were free of cognitive impairment at baseline, and tracked them for multiple health outcomes including late-onset AD. Over an average follow-up of 12.8 years, 655 participants developed AD, including 115 who were diagnosed at their five-year follow-up visit.

The researchers applied SomaScan—a proteomic technique that uses single-stranded DNA molecules called aptamers to tag proteins—to measure the abundance of more than 4,100 proteins in baseline plasma samples. After adjusting for age and sex, they identified 303 proteins that significantly correlated with future AD. Many of them were involved in development of the shape of neurons and of the axon. Just under half—130—were tied to future AD no matter a participant’s ApoE alleles. These, too, contained predominantly neuron-related proteins.

Of the 173 proteins whose association with AD vanished once ApoE genotype was accounted for, the scientists zeroed in on 17 that were strongly tied to impending AD only among ApoE4 carriers. Serum levels of these 17 were regulated by the ApoE4 allele, which ramped up five and turned down the other 12. The researchers saw no functional commonalities among this E4-dependent lot, but the proteins physically interact with proteins involved in microtubule and centromere functions, neuronal response and development, neuroinflammation, and AD.

Neither did the proteins all hail from a single tissue. Four—LRRN1, FAM159B, NEFL and HBQ1—were more highly expressed in the brain than in other tissues, and one—TMCC3—is to do with oligodendrocytes. The remaining eight were ubiquitously expressed. Oddly, though these 17 proteins associated with AD in an E4-dependent way, they did not appear to mediate the higher AD risk conferred by the ApoE4 allele.

Four of the E4-dependent proteins—TBCA, ARL2, S100A13, and IRF6—threw the researchers for a loop. Among carriers, levels of these proteins were decreased, and this dip was tied to a future AD diagnosis. However, among noncarriers, the association reversed, such that higher levels of these four proteins were tied to AD. This flip-flop also cropped up in the Alzheimer Center Barcelona (ACE) cohort, which includes people with MCI due to AD at baseline. Relative to the Icelandic AGES cohort, disease is more advanced in ACE participants. Here, the direction switch was more extreme, such that these four proteins were decreased among ApoE4 carriers yet substantially boosted among noncarriers who later advanced from MCI to AD.

The findings suggest that while ApoE4 somehow turns down expression of these proteins in the blood, biological processes relating to AD have a countereffect, raising their levels. In support of this interpretation, the researchers found that, among people in AGES with a high burden of AD risk variants other than ApoE4, these four proteins were elevated in the serum, whereas the opposite was true among E4 carriers (image below).

Among Icelanders, Blood Proteome Predicts Dementia, Parses Role of ApoE4

Quick Links

Tools

Comments

Make a Comment

References

News Citations

Paper Citations

Other Citations

Further Reading

Papers

Annotate