Page 109 - Read Online

Page 243 Ratnapriya. J Transl Genet Genom 2022;6:240-256 https://dx.doi.org/10.20517/jtgg.2021.54

STEPS TO CONNECT GWAS DISCOVERY TO BIOLOGY
GWAS unquestionably established a strong genetic component of AMD and provided a broad framework
for elucidating the contribution of genetic variants to AMD. It allowed a hypotheses-free discovery of
disease variants that revealed novel disease biology and provided a wealth of clues for biological
experimentation, risk prediction, disease progression, and biomarker and therapeutic target discovery
[Figure 1]. However, we must also acknowledge that progress in translating genetic findings into
understanding disease mechanisms and new cures and treatments for AMD has been slow to arrive. There
are no biomarkers for predicting the disease progression to either GA or CNV. Unfortunately, targeting
GWAS implicated pathways has not been effective in preventing/slowing disease progression or treatment
[44]
as many clinical trials that relied on these hypotheses have failed to ameliorate AMD .

Drug targets with human genetic evidence of disease association are twice as likely to succeed as approved
[45]
drugs . Thus, investments in the functional follow-up of genomic findings are likely to be beneficial for
AMD drug development. However, several factors have made it difficult to bridge the gap between the
statistical associations of genetic variations and a functional understanding of the biology underlying AMD
risk. First, GWAS variants are not causal. The causative SNP may lie anywhere within the linkage
disequilibrium (LD) block surrounding the associated SNP that can span over 100 kb and often contain over
a thousand individual SNPs. Second, a majority of associated variants reside in non-coding regions of the
genome, thus underlying causative genes or pathological mechanisms are not immediately obvious. Some of
the AMD-specific challenges are also worth noting. For example, modeling of AMD presents a substantial
challenge . AMD causes vision loss because of degeneration of photoreceptors in the central region of the
[46]
[47]
retina called the macula, which is a primate-specific structure shared only by humans, monkeys and apes .
Thus, mouse models fall short of recapitulating many aspects of AMD. Similarly, there is a dearth of cell
lines for ocular studies, and transformed cell lines such as ARPE-19 do not express key phenotypic features
of human RPE . Finally, we have a limited understanding of molecular and genetic underpinnings in the
[48]
early and intermediate stages of the disease.

There are four key steps in elucidating the disease mechanisms underlying AMD risk variants: (1)
identification of causal variant(s) and target gene(s) at the associated loci; (2) annotating variants and genes
at tissues and single-cell resolution; (3) identifying molecular and cellular pathways associated with the
disease; and (4) developing appropriate disease models to infer disease mechanisms. As a vast majority of
GWAS variants fall into non-coding regions and often overlap enhancers, promoters and open-chromatin
regions, gene expression regulation is emerging as a dominant mechanism in mediating disease risk. Thus,
here we will be focusing on various ways gene expression studies can be leveraged to understand the risk
variant-mediated disease causation.

TRANSCRIPTOME STUDIES IN DISEASE-RELEVANT TISSUES AND CELL TYPES
GWAS have established that > 90% of the risk variants associated with complex diseases including AMD
reside in the non-coding regions that are not likely to directly affect the coding region of the gene. This
brought attention to the annotation of the non-coding genome to understand its function. Enrichment of
GWAS variants within regulatory DNA marked by deoxyribonuclease I (DNase I) hypersensitive sites
[49]
(DHSs) led the focus on studying the gene expression regulation in the context of GWAS variants .
Additionally, these risk variants often disrupt binding sites for transcription factors (TFs), and these variable
TF-interaction are believed to be the primary driver of phenotypic variation . Additionally, cell context is a
[50]
key determinant of gene regulation. Thus, a comprehensive understanding of global transcriptome
regulation in disease-relevant cells and tissues represents the first logical step in functional understanding of
GWAS findings. Towards this goal, Genotype-Tissue Expression (GTEx) project was initiated to establish a