The HLA-DRA variation rs3129882 is not associated with Parkinsons disease in Sweden

(1)

The HLA-DRA variation rs3129882 is not

associated with Parkinsons disease in Sweden

Caroline Ran, Thomas Willows, Olof Sydow, Anders Johansson, Peter Söderkvist, Nil

Dizdar, Ahmad Ahmadi, Lars Olson and Carmine Belin

Linköping University Post Print

N.B.: When citing this work, cite the original article.

Original Publication:

Caroline Ran, Thomas Willows, Olof Sydow, Anders Johansson, Peter Söderkvist, Nil

Dizdar, Ahmad Ahmadi, Lars Olson and Carmine Belin, The HLA-DRA variation rs3129882

is not associated with Parkinsons disease in Sweden, 2013, Parkinsonism & Related

Disorders, (19), 7, 701-702.

http://dx.doi.org/10.1016/j.parkreldis.2013.03.001

Copyright: Elsevier. Under a Creative Commons license

http://www.elsevier.com/

Postprint available at: Linköping University Electronic Press

(2)

Letter to the Editor

The HLA-DRA variation rs3129882 is not associated with Parkinson

’s

disease in Sweden

Keywords: Parkinson’s disease HLA-DRA PARK18 Association study SNP

There is an increasing evidence for genetic contribution to Par-kinson’s disease (PD). Genes responsible for monogenetic heredi-tary PD have been discovered to carry less devastating mutations associated with“idiopathic” PD, suggesting that common molecu-lar pathways are involved in the pathogenesis of monogenetic and “idiopathic” PD. Recently, genome wide association studies (GWAS) have revealed new PARK loci, potentially harboring genes confer-ring an increased or decreased risk of developing PD. We have stud-ied PARK18, for which HLA-DRA, a class II human leukocyte antigen, was proposed as a candidate gene. Hamza et al.[1]reported that the G allele of the single nucleotide polymorphism (SNP) rs3129882 (A/G) in HLA-DRA, associates with increased risk for PD in an American cohort. Attempts to replicate the association have shown conflicting results[2–5]; not only were there lack of as-sociations in several cohorts[2–5], but the GG genotype has also been found to be protective [3]. Variable findings in different studies point to the presence of geographical and ethnical differ-ences and hence the importance of validating GWASfindings in

in-dependent homogenous case–control materials. We have

performed a replication study on the HLA-DRA SNP rs3129882 in a homogenous Swedish case–control material in order to further investigate the possible contribution of rs3129882 to PD.

Blood samples were collected at the Karolinska and Linköping University Hospitals after informed consent and approval of the Regional Ethical Review Board. DNA was obtained from samples using standard protocols. We genotyped rs3129882 in 511 PD patients and 636 neurologically healthy controls (Table 1). A subset of the patients (172) and controls (279) from Linköping was previously genotyped for rs3129882 in another study[5]. Control individuals were neuro-logically healthy spouses of patients (4%), individuals visiting hospi-tals for non-neurological disorders (54%) and from the SNAC-K project (The Swedish National Study on Aging and Care in Kungshol-menhttp://www.snac-k.se/) (42%). All samples are unrelated Cauca-sians from urban areas from the rather small population of Sweden, likely to represent a homogenous case–control material. Genotyping was performed by real-time polymerase chain reaction (PCR) (7500 fast system using standard cycling conditions, TaqManÒgenotyping master mix and Genotyping Assay C_2455644_10, Applied

Biosystems/Life Technologies, Carlsbad, California, USA). Due to a non-negligible level of heterogeneity in the read-out in a subset of the Linköping samples with lower DNA concentration, these samples were re-sequenced with the pyrosequencing technique to confirm genotyping results. A PCR was run with Taq polymerase enzyme and two primers, one of which was biotinylated at the 50end (Thermo Fisher Scienti_{fic, Ulm, Germany), (sequences are available upon} request). The PCR product was purified according to manufacturer’s instructions, incubated with a sequencing primer and analyzed by PYRO-sequencing (PSQ 96MA system, BIOTAGE AB Uppsala, Sweden) using PyroMark Gold Q96 Reagents (QIAGEN AB, Solna, Sweden). Three PD patients were excluded from the analysis since they were carriers of the pathogenic G2019S mutation LRRK2. Association was evaluated with Chi-square (

c

2) and Fisher’s exact test, signiﬁcance level 5%, two-sided P-values (GraphPad Prism 5.03, GraphPad Soft-ware, San Diego California USA). Power and sample size software PS 3.0 was used for power analysis (http://biostat.mc.vanderbilt.edu/ wiki/Main/PowerSampleSize). With our sample size we can detect an odds ratio of<0.712 and >1.390 with a power of 80% (

a

¼ 0.05).

We genotyped a total of 511 PD patients and 636 neurologically healthy controls from Sweden and searched for possible genotypic and allelic association between rs3129882 and PD. Rs3129882 was in Hardy Weinberg equilibrium for all subjects (data not shown). Weﬁrst analyzed the entire material for which we found no geno-typic (data not shown) or allelic association for rs3129882 and PD (Table 2). In order to investigate variability in subgroups of pa-tients, we stratiﬁed the material with respect to age of onset,

Table 1

Demographic information.

Parkinson’s disease Control

Total no. 511 636

Age (years) 68.5 72.5

Age of onset (years) 61.0 Na Female no. (%) 194 (38.0) 354 (55.7) Early onset no. (%) 92 (18.0) Na Heredity no. (%) 39 (7.6) Na Na¼ not applicable, early onset ¼ 50 years or less at diagnosis.

Contents lists available atSciVerse ScienceDirect

Parkinsonism and Related Disorders

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / p a r k r e l d i s

Parkinsonism and Related Disorders 19 (2013) 701–702

(3)

sex and self-reported family history. Female and male patients were compared to the whole control group since genotype and allele frequencies were similar in the control groups (P> 0.5). Pa-tients with maximum oneﬁrst-, second- or third-degree relative were included in the group designated as having no known hered-ity, while familial PD was deﬁned as patients with two or more relatives with PD. Applying these stringent selection criteria corresponding to those used in the report by Hamza et al. [1], we discovered no further association for rs3129882 with PD (Table 2).

Previous replication studies on HLA-DRA show very

heteroge-neous results [1–5]. These variations might be explained by

ancestry; allele frequencies for rs3129882 vary greatly in different geographical regions. Hamza et al.[1]reported a signif-icant substructure in American subjects of European decent correlating with the country of origin of the subjects_{’ ancestors.} Association was stronger in subjects with ancestors from North-ern Europe. However, no replication has yet been reported in Eu-ropean cohorts. We found no association in our exclusively

Swedish material which is in agreement with recentﬁndings in

a Scandinavian (Sweden, Norway) population [5]. According to

Hamza et al.[1], association was stronger in sporadic PD patients with late onset. To mimic their study we performed an age of onset stratified analysis, but failed to detect any further signifi-cant differences. Therefore, our results do not suggest a role for HLA-DRA variants for PD in Sweden. Variable results in genetic replication studies can also be explained by different Linkage Disequilibrium (LD). Association can be found with an SNP which is in LD with the true disease causing SNP, while in another pop-ulation these two SNPs are not in LD. HLA genes, being closely located on chromosome 6, could exert a joint effect. Rs3129882 is for example believed to influence expression levels of

HLA-DRA, HLA-DQA2 and HLA-DRB5 [1]. Published data on HLA-DR

associations with PD are so far more in favor of association with HLA-DRB loci, which are also more polymorphic, than with the rather invariable HLA-DRA.

We thank Professor Laura Fratiglioni for providing SNAC-K con-trol samples and Dr. Fengqing Xiang for technical assistance. This study was supported by the Swedish Research Council, The

Swedish Parkinson Foundation, The Swedish Brain Foundation, Swedish Brain Power, the Åke Wiberg Foundation, the Karolinska Institute Distinguished Professor Award, StratNeuro and Karolinska Institutet Funds.

References

[1] Hamza TH, Zabetian CP, Tenesa A, Laederach A, Montimurro J, Yearout D, et al. Common genetic variation in the HLA region is associated with late-onset spo-radic Parkinson’s disease. Nat Genet 2010;42:781–5.

[2] Chiang HL, Lee-Chen GJ, Chen CM, Chen YC, Lee CM, Liao MH, et al. Genetic analysis of HLADRA region variation in Taiwanese Parkinson’s disease. Parkin-sonism Relat Disord 2012;18:391–3.

[3] Puschmann A, Verbeeck C, Heckman MG, Soto-Ortolaza AI, Lynch T, Jasinska-Myga B, et al. Human leukocyte antigen variation and Parkinson’s disease. Parkinsonism Relat Disord 2011;17:376–8.

[4] Mata IF, Yearout D, Alvarez V, Coto E, de ML, Ribacoba R, et al. Replication of MAPT and SNCA, but not PARK16-18, as susceptibility genes for Parkinson’s dis-ease. Mov Disord 2011;26:819–23.

[5] Pihlstrøm L, Axelsson G, Bjørnarå KA, Dizdar N, Fardell C, Forsgren L, et al. Sup-portive evidence for 11 loci from genome-wide association studies in Parkin-son’s disease. Neurobiol Aging, published online 12 Nov 2012,http://dx.doi. org/10.1016/j.neurobiolaging.2012.10.019.

Caroline Ran*

Department of Neuroscience, Karolinska Institutet, Retzius väg 8, B2:4, 171 77 Stockholm, Sweden Thomas Willows Department of Neurology, Karolinska University Hospital Huddinge, 141 86 Stockholm, Sweden Olof Sydow, Anders Johansson Department of Neurology, Karolinska University Hospital Solna, 171 76 Stockholm, Sweden Peter Söderkvist Division of Cell Biology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, 581 85 Linköping, Sweden Nil Dizdar Division of Cell Biology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, 581 85 Linköping, Sweden Department of Neurology, University Hospital, Linköping University, 581 85 Linköping, Sweden Ahmad Ahmadi Division of Cell Biology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, 581 85 Linköping, Sweden Lars Olson, Andrea Carmine Belin Department of Neuroscience, Karolinska Institutet, Retzius väg 8, 171 77 Stockholm, Sweden * Corresponding author. Tel.: þ46 8 524 870 62; fax: þ46 8 323 742. E-mail address:caroline.ran@ki.se(C. Ran) 19 November 2012

Table 2

Allele frequencies for rs3129882 in Parkinson’s disease patients and control subjects analyzed with Fisher’s exact test.

PD (no.) Ctrl (no.) MAF PD MAF Ctrl P-value OR (95%CI) Strata Stockholm 314 206 0.425 0.422 0.91 1.02 (0.81–1.27) Linköping 194 330 0.381 0.398 0.6 0.93 (0.72–1.2) All subjects 508 636 0.408 0.410 0.99 1.00 (0.85–1.19) PD-associated strata No heredity 469 0.403 0.79 1.03 (0.87–1.22) Familial 39 0.474 0.28 0.77 (0.49–1.22) Late onset 414 0.402 0.77 1.03 (0.86–1.23) Early onset 91 0.434 0.57 0.90 (0.66–1.24) Female 193 0.402 0.81 1.03 (0.82–1.30) Male 315 0.413 0.92 0.99 (0.81–1.20) PD¼ Parkinson’s disease, Ctrl ¼ controls, MAF ¼ minor allele frequency, OR ¼ odds ratio, CI¼ conﬁdence interval.

Letter to the Editor / Parkinsonism and Related Disorders 19 (2013) 701–702 702