Characterisation of the Function of a SINE-VNTR-Alu Retrotransposon to Modulate Isoform Expression at the MAPT Locus

Abstract

SINE-VNTR-Alu retrotransposons represent one class of transposable elements which contribute to the regulation and evolution of the primate genome and have the potential to be involved in genetic instability and disease progression. However, these polymorphic elements have not been extensively analysed when addressing the missing heritability of neurodegenerative diseases, including Parkinson’s disease (PD) and amyotrophic lateral sclerosis (ALS). SVA_67, a retrotransposon insertion polymorphism, is located in a 1.8 Mb region of high linkage disequilibrium, called the MAPT locus, which is known to contribute to increased risk of developing PD, frontotemporal dementia and other tauopathies. To investigate the role of SVA_67 in directing differential gene expression at this locus, we characterised the impact of SVA_67 allele dosage on isoform expression of several genes in the MAPT locus using the datasets from both the Parkinson’s Progression Markers Initiative and New York Genome Center Consortium Target ALS cohort. The Parkinson’s data was from gene expression in the blood and the ALS data from a variety of CNS regions and allowed us to demonstrate that SVA_67 presence or absence correlated with both isoform- and tissue-specific expression of multiple genes at this locus. This study highlights the importance of addressing SVA polymorphism in disease genetics to gain insight into a better understanding of the role of these regulatory domains to a variety of neurodegenerative diseases.

Deciphering the role of a SINE-VNTR-Alu retrotransposon polymorphism as a biomarker of Parkinson’s disease progression

“…In our previous studies, we characterised the impact of SVA RIPs using transcriptomic, whole genome sequencing and clinical data from the Parkinson's Progression Markers Initiative (PPMI) cohort highlighting the association of seven RIPs with the progression of PD and with differential gene expression 32 . Moreover, our recent study also showed the association of the presence or absence of a particular SVA RIP, termed SVA_67, with differential isoform expression of several genes within the MAPT locus 33 . SVA_67, located about 12 kb upstream of the KAT8 regulatory NSL complex subunit 1 (KANSL1) gene, represents a 5´ truncated SVA element with a length of 701 bp (hg38) and tags H1 while its absence is associated with H2 32,33 (Fig.…”

“…Moreover, our recent study also showed the association of the presence or absence of a particular SVA RIP, termed SVA_67, with differential isoform expression of several genes within the MAPT locus 33 . SVA_67, located about 12 kb upstream of the KAT8 regulatory NSL complex subunit 1 (KANSL1) gene, represents a 5´ truncated SVA element with a length of 701 bp (hg38) and tags H1 while its absence is associated with H2 32,33 (Fig. 1).…”

“…The H1 inversion is associated with an increased risk of developing PD, however, the genetic variant(s) responsible for this risk remains unclear. We previously showed the association of SVA_67 presence/absence with differential gene and isoform expression of several genes in the MAPT locus, however, as SVA_67 is present in H1 and absent in H2, this effect cannot be fully differentiated from the haplotype speci c in uence on gene expression 32,33 . A study by O'Brien et al combined genome-wide genotyping with transcriptomic data and showed that differential expression of 13/21 genes at the MAPT locus can be "largely explained" by the H1/H2 inversion status 34 .…”

“…3, Supplementary Fig. 2), containing a minimum of four variants in addition to the previously described RIP for presence/absence in the genome 32,33 . The MAPT locus is contained within a large genomic inversion termed H1/H2.…”

“…We rst analysed SVA_67 associated expression levels between baseline (BL/V0) and four visits indicated by months after BL (6 months/V02, 12 months/V04, 24 months/V06, 36 months/V08). For analysis, we combined available data from CO and PD individuals similar to our previous studies on the SVA_67 RIP 32,33 . When addressing LRRC37A, expression levels varied strati ed by VNTR genotype (P < 2.2E-16), but did not change over the time course (P = 8.77E-01) (Fig.…”