. Proteogenomic analysis of human cerebrospinal fluid identifies neurologically relevant regulation and implicates causal proteins for Alzheimer's disease. Nat Genet. 2024 Dec;56(12):2672-2684. Epub 2024 Nov 11 PubMed.

Recommends

Please login to recommend the paper.

Comments

  1. In this study, Western et al. leveraged CSF-based proteomics via the SOMAscan 7k platform and genetic data from approximately 3,500 individuals to conduct the largest investigation of CSF protein quantitative trait loci (pQTLs) to date. This study provides new insights into the genetic regulators of protein levels, which is a particularly compelling extension of existing human genetic data sets, given that most therapeutic approaches target proteins. By integrating pQTLs with AD GWAS data via standard approaches (co-localization, Mendelian Randomization or MR, and protein-wide association studies or PWAS), the authors provide several insights that are valuable for drug discovery, namely nominations for:

    • Genes responsible for AD GWAS signals, to inform target identification (via co-localization)
    • Proteins that are causally associated with AD, thereby nominating disease intervention points and informing target identification (via MR)
    • Directionality of protein levels and genetic associations (e.g., lower protein levels are associated with AD risk), which can inform therapeutic strategies, such as agonism versus inhibition (via MR and PWAS)

    This study nominated 38 proteins using at least two of these methods, while eight proteins were detected by all three (co-localization + MR + PWAS). In most cases, these proteins displayed intuitive relationships to disease. For example, the well-established microglial gene TREM2 was negatively associated with disease risk, i.e., lower protein levels, increased risk, which is expected based on known TREM2 loss-of-function coding variants that increase AD risk (Guerreiro et al., 2013; Jonsson et al., 2013). Moreover, GRN was found to be similarly negatively associated with AD risk. This is also an expected finding given that GRN mutations resulting in progranulin deficiency are associated with frontotemporal dementia (Baker et al., 2006), and a common variant signal driven by a known progranulin-lowering SNP (rs5848) was discovered in a recent AD GWAS (Bellenguez et al., 2022).

    Like TREM2 and GRN, PILRA was nominated across all three pQTL-GWAS integration methods. This is an exciting finding and adds to the emerging evidence that PILRA is the causal gene in the ZCWPW1/NYAP1 AD GWAS locus (Weerakkody et al., Research Square preprint 2024), including 1) the PILRA G78R variant (rs1859788) accounts for AD GWAS signal in this locus, as demonstrated via conditional analysis (Novikova et al., 2021) and 2) a burgeoning genetic connection between this PILRA variant and APOE genotype (Lopatko Lindman et al., 2022; Monroe et al, 2024; Belloy et al., 2023). However, the negative association of PILRA protein levels with AD risk appears to be at odds with the fact that the G78R variant, associated with reduced ligand binding (Rathore, et al., 2018), is associated with reduced AD risk and delayed AD age of onset (He et al., 2021). 

    Several characteristics of this study may reconcile these somewhat counterintuitive observations. First, this study nominated PILRA via pQTL data that only captures SNP effects on protein levels and not metrics of protein activity, for example, ligand binding, receptor signaling, co-receptor interactions, etc. Second, CSF levels of PILRA protein may not correspond to levels in brain. Therefore, while this study is useful in connecting AD GWAS regions to specific proteins and articulating a direction of association, further studies are necessary to understand the mechanistic basis of these connections and confirm their relationship to function.

    Finally, two additional characteristics of the proteomics platform used in this study are important to consider. First, the SOMAscan aptamer-based approach may not be sensitive enough to discern between proteins with a high degree of sequence homology, or between isoforms. Second, as the authors point out, many proteins remain unaccounted for in the current iteration of the SOMAscan assay. We might still be missing protein data for many AD GWAS loci, and there may be even more AD-relevant proteins to be discovered with future versions of the platform. Despite these technical limitations, we find this work to be an exciting step forward for elucidating mechanisms of genetic risk factors and their impact in disease.

    References:

    . TREM2 variants in Alzheimer's disease. N Engl J Med. 2013 Jan 10;368(2):117-27. Epub 2012 Nov 14 PubMed.

    . Variant of TREM2 associated with the risk of Alzheimer's disease. N Engl J Med. 2013 Jan 10;368(2):107-16. Epub 2012 Nov 14 PubMed.

    . Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17. Nature. 2006 Aug 24;442(7105):916-9. PubMed.

    . New insights into the genetic etiology of Alzheimer's disease and related dementias. Nat Genet. 2022 Apr;54(4):412-436. Epub 2022 Apr 4 PubMed.

    . Integration of Alzheimer's disease genetics and myeloid genomics identifies disease risk regulatory elements and genes. Nat Commun. 2021 Mar 12;12(1):1610. PubMed.

    . PILRA polymorphism modifies the effect of APOE4 and GM17 on Alzheimer's disease risk. Sci Rep. 2022 Aug 2;12(1):13264. PubMed.

    . PILRA regulates microglial neuroinflammation and lipid metabolism as a candidate therapeutic target for Alzheimer’s disease. Research Square, Feb 15, 2024 Research Square

    . Charting the genetic architecture of Alzheimer’s disease across APOE*4 and sex. https://doi.org/10.1002/alz.078638 Alzheimer's & Dementia

    . Paired Immunoglobulin-like Type 2 Receptor Alpha G78R variant alters ligand binding and confers protection to Alzheimer's disease. PLoS Genet. 2018 Nov;14(11):e1007427. Epub 2018 Nov 2 PubMed.

    . Exome-wide age-of-onset analysis reveals exonic variants in ERN1 and SPPL2C associated with Alzheimer's disease. Transl Psychiatry. 2021 Feb 26;11(1):146. PubMed.

    View all comments by Kathryn Monroe

Make a Comment

To make a comment you must login or register.