موجز عن البحث:
Parkinson’s disease has been reported in a small number of
patients with chromosome 22q11.2 deletion syndrome. In this study, we
screened a series of large, independent Parkinson’s disease case-control
studies for deletions at 22q11.2.
We used data on deletions spanning the 22q11.2 locus from four
independent case-control Parkinson’s disease studies (UK Wellcome Trust Case
Control Consortium 2, Dutch Parkinson’s Disease Genetics Consortium, US
National Institute on Aging, and International Parkinson’s Disease Genomics
Consortium studies), which were independent of the original reports of
chromosome 22q11.2 deletion syndrome. We did case-control association
analysis to compare the proportion of 22q11.2 deletions found, using the
Fisher’s exact test for the independent case-control studies and the
Mantel-Haenszel test for the meta-analyses. We retrieved clinical details of
patients with Parkinson’s disease who had 22q11.2 deletions from the medical
records of these patients.
We included array-based copy number variation data from 9387
patients with Parkinson’s disease and 13 863 controls. Eight patients with
Parkinson’s disease and none of the controls had 22q11.2 deletions
(p=0·00082). In the 8451 patients for whom age at onset data were available,
deletions at 22q11.2 were associated with Parkinson’s disease age at onset
(Mann-Whitney U test p=0·001). Age at onset of Parkinson’s disease was lower
in patients carrying a 22q11.2 deletion (median 37 years, 95% CI 32·0-55·5;
mean 42·1 years [SD 11·9]) than in those who did not carry a deletion (median
61 years, 95% CI 60·5-61·0; mean 60·3 years [SD 12·8]). A 22q11.2 deletion
was present in more patients with early-onset (p<0·0001) and late-onset
Parkinson’s disease (p=0·016) than in controls, and in more patients with early-onset
than late-onset Parkinson’s disease (p=0·005).
Clinicians should be alert to the possibility of 22q11.2
deletions in patients with Parkinson’s disease who have early presentation or
features associated with the chromosome 22q11.2 deletion syndrome, or both.
UK Medical Research Council, UK Wellcome Trust, Parkinson’s UK,
Patrick Berthoud Trust, National Institutes of Health, “Investissements
d’Avenir” ANR-10-IAIHU-06, Dutch Parkinson Foundation (Parkinson
Vereniging), Neuroscience Campus Amsterdam, National Institute for Health
Research, National Institute on Aging, National Institutes of Health.
Deletions at 22q11 have been associated
with a heterogeneous range of clinical syndromes which include DiGeorge
syndrome, Velo-cardio-facial syndrome and conotruncal anomaly face syndrome.
These different diagnostic categories are variable outcomes resulted from a
single genetic mechanism, thus they are collectively called Chromosome 22q11
Deletion Syndrome (22q11DS). The phenotype of 22q11DS is complex and people
with 22q11DS have a greatly elevated risk of developing psychosis
particularly schizophrenia. The vast majority of deletions span a large 3Mb
region at 22q11 (97% of cases), however; smaller deletions span ~1.5Mb region
which contains at least 28 genes are detected in 3% of the deletion carries.
It is still not clearly illustrated how heterozygous microdeletions on
chromosome 22q11 lead to this diverse spectrum of behavioural phenotypes and
more attention has been shedding on the complex pathogenic mechanisms
conferred by chromosomal abnormalities. ‘‘Second-hit’’ mutations are one
hypothesis encompassing a range of complexity from polygenic mechanisms to
explain these heterogeneous phenotypes and high risk of developing
psychiatric illness in 22q11 deletion carriers.
For our study, we tested the two-hit
hypothesis where diverse phenotypes and additional risk is conferred through
a second CNV. We identified large (>100 kb) CNVs in 68 VCFS cases and 32
control siblings. The results show there is a high rate of additional large
CNVs (> 300kb) in probands than in siblings (0.1912 and 0.1562 respectively).
However; there is a remarkable increasing in the rate of larger CNVs (>
400kb) in probands (0.1471) while it drops within the siblings group
(0.0625). Filtering common CNVs which occur in > 1% in the population
increases the additional CNVs (> 300kb) rates ratio in probands and
siblings by two folds. Interestingly, in larger CNVs no additional rare large
CNVs in siblings while the rate of additional large rare CNVs in probands is
0.1029. Four large rare CNVs (20%) are shown to be potentially implicated in
autism and other psychiatric diseases and that might give an explanation for
psychiatric illness manifestation in a group of 22q11.2 deletion carriers and
its absence in other group.
Our results suggest the two-hit hypothesis
may be relevant to the phenotype heterogeneity in VCFS patients but sample
sizes are small and further studies required. Additionally, more study is
needed to understand if these additional large rare CNVs were inherited or de
with chromosome 22q11.2 deletion syndrome (22q11.2DS) have an increased risk
of developing a diverse spectrum of behavioural phenotypes. However,
there is a lack of understanding the mechanisms explaining how
variable psychiatric illnesses could arise by a relatively
homogenous microdeletion on chromosome 22q11.2. As 22q11.2DS patients
carry the same deletion it is more likely
that haploinsufficiency of dosage sensitive genes spanning the
deleted region influence the pathogenesis of the resulting psychiatric
phenotypes. It is expected that DNA copy number on chromosome
22q11.2 have an impact on the level of gene expression
of happoinsufficient genes which could be reduced by
50% (Williams NM., 2011). Consequently, reduction in the
expression of these genes would change their biological function. These genes
are candidate loci for psychiatric phenotypes in 22q11.2 DS
patients. The approach to identify dosage sensitive genes is
differential gene expression analysis of 22q11.2 deletion
region and genome wide genes in 22q11.2 deletion group and a normal
of this study is to investigate the effect that 22q11 deletions have on
gene expression by assessing haploinsufficient genes located within
22q11.2. As not all genes are dosage sensitive the identification of
genes that are differentially expressed between 22q11.2 deletion carriers and
non-deleted controls will highlight those whose expression are influenced by
the deletion. It has been reported that the expression of genes away from the
CNV boundaries can be directly influenced the CNV (Stranger B., et al. 2007).
This phenomenon is called a ‘position effect’ in which genes within the same
chromosome could be altered (in CIS) or potentially on other
chromosomes (in TRANS). This will be investigated by performing
genome-wide gene expression analysis to assess the expression of
genes located outside 22q11.2.
gene expression analysis RNA samples of 22q11.2DS cases (n=33) and
controls (n=36) were analysed by Illumina HumanHT-12 v4 Expression BeadChip which
targets 47,231 probes. Samples and probes were subjected to quality control
assessments. Subsequent 34,461 good quality probes were
corrected for batch effects in 22q11.2DS (n=33) and controls (n=35)
after quality control procedure. Differential gene expression analysis was
applied by fitting a linear model and defining a contrast comparing
cases to controls using Limma package on R.
result of differential gene expression analysis shows ~35% of genes
located within the 3MB 22q11.2 region are significantly
differentially expressed between 22q11.2 deletion cases and
controls (FDR < 0.05). There is a minor in
CIS position effect of 22q11.2 deletion on nearby genes.
However; the deletion has a negligible effect genome
wide, 2.2% genes are significantly differentially expressed with
FDR < 0.05. As these studies have used RNA samples which were
extracted from peripheral blood, future studies will be required to
establish how they reflect gene expression in the human brain.