Genetic and functional characterization of candidate genes for complex psychiatric diseases using next-generation sequencing and cellular uptake assays

Complex phenotypes are the result of a complex interplay between
genes and environmental factors. Extensive linkage, candidate and
genome-wide association studies (GWASs) have been carried out to
unravel genetic risk variants for human diseases. The
identification of genes, involved in the pathomechanism of a
disease, might be beneficial for its diagnosis, treatment and
prognosis. While GWASs allowed the identification of a large number
of common variants robustly associated with common complex
diseases, the heritability, which can be explained by these
variants, is small. The discrepancy between the estimated
heritability from twin, family and adoption studies and the
heritability obtained from GWAS was termed “missing heritability”
and led to the investigation of additional factors that might also
contribute to disease susceptibility, including gene-environment
interactions, gene-gene interactions, structural variants and rare
variants. In this thesis, the role of less common and rare variants
in susceptibility to common complex diseases was investigated. In
order to accomplish this, a candidate gene for panic disorder (PD)
and a possible risk gene for major depressive disorder (MDD) were
screened for the presence of common and rare variants using
next-generation sequencing in a pooled approach. In a previously
published GWAS, a haplotype containing two common intronic variants
in the transmembrane protein 132D (TMEM132D) gene was associated
with PD. Another GWAS identified solute carrier family 6 member 15
(SLC6A15), which encodes an amino acid transporter, as a risk gene
for MDD. A common intergenic variant about 600 kilobase downstream
of this gene was shown to decrease SLC6A15 gene expression in
lymphoblastoid cell lines and hippocampus. Susceptibility genes for
complex diseases, identified in GWAS, are promising candidates for
the search of rare variants as genes harbouring common variants are
likely to contain also rare variants. Pooled targeted re-sequencing
of the exonic regions of TMEM132D in 300 anxiety disorder patients,
mostly suffering from PD (84.7%), and 300 healthy controls allowed
the detection of 371 genetic variants. Of these variants, 24.0%
were common (minor allele frequency (MAF) > 5.0%), whereas the
vast majority was less common (MAF 1.0 – 5.0%) to rare. 247
variants had not been reported before, including 12 novel
non-synonymous variants leading to an amino acid exchange in the
protein. While common variants associated with PD were not
identified, an overrepresentation of non-synonymous variants and
variants with predicted changes on splicing in healthy controls
compared to PD patients was observed. These putatively functional
relevant variants were distributed along a broad MAF spectrum,
ranging from 0.17 to 30.0%. In addition, a higher rate of private
non-synonymous variants, which were only present in either cases or
controls in this study, but not in over 7,500 individuals with
different ethnic backgrounds from other publicly available
re-sequencing datasets, in patients compared to controls was seen.
Combined with the data from the previous GWAS study in which the
association with PD was carried by common variants, this pooled
re-sequencing study suggests that not only common or rare variants
alone, but a combination of both contributes to the development of
anxiety-related phenotypes. Re-sequencing the whole SLC6A15 locus
in 400 MDD patients and 400 healthy controls, 405 genetic variants
were identified, including twelve non-synonymous variants. Only
15.0% of the detected variants were common. While none of the
non-synonymous variants was significantly associated with MDD, two
rare non-synonymous variants were identified to influence protein
function. In contrast to the TMEM132D protein whose molecular
function has still to be discovered, SLC6A15 is known to transport
neutral amino acids into predominantly neuronal cells. The cellular
uptake of neutral amino acids such as proline is thus a measurable
property that associates with function. The uptake experiments
identified two rare variants to be associated with a significant
increase in proline uptake in HEK cells. This result suggests that
rare variants in SLC6A15 might influence the biochemical function
of its amino acid transporter and thus downstream neuronal function
and possibly the risk for MDD and other stress-related psychiatric
disorders. In addition, this study highlights that functional
exploration of genetic variants might be promising to identify
putatively disease-relevant variants as statistically significant
associations for rare variants might only be achieved in extremely
large samples.