Variants in the 10q26 Gene Cluster (LOC387715 and HTRA1) Exhibit Enhanced Risk of Age-Related Macular Degeneration along with CFH in Indian Patients

Abstract

purpose. Single nucleotide polymorphisms (SNPs) in the LOC387715 (rs10490924), HTRA1 (rs11200638), and CFH (rs1061170) genes have been implicated in age-related macular degeneration (AMD). The present study was undertaken to determine the involvement of the LOC387715 and HTRA1 in an AMD cohort from India.

methods. The coding region of LOC387715 (exon 1) and the promoter of HTRA1 were screened by resequencing in AMD cases and normal controls. Odds ratios were calculated to assess the risk of individual genotypes. Linkage disequilibrium (LD) and haplotype frequencies were estimated with Haploview software. Population attributable risk (PAR %) for the associated SNPs and their combined effects were calculated.

results. Resequencing revealed seven different SNPs in these genes, of which significant associations were noted with the risk alleles of rs10490924 (T allele; P = 5.34 × 10⁻¹²) in LOC387715, and rs11200638 (A allele; P = 4.32 × 10⁻¹²) and rs2672598 (C allele; P = 3.39 × 10⁻¹¹) in HTRA1 among the cases. Correspondingly, the homozygous risk genotypes TT, AA, and CC in these SNPs exhibited higher disease odds and PAR %. rs10490924 and rs11200638 were in tight LD (D′, 0.90; 95% CI, 0.84–0.93). G-C-T-A-C was the risk haplotype (P = 8.04 × 10⁻¹⁵), whereas the G-C-G-G-T haplotype was protective (P = 2.01 × 10⁻⁴). The combined effect of the CFH (CC) and LOC387715 (TT) risk genotypes exhibited a PAR of 93.7% (OR, 73.89; 95% CI, 8.69–628.13).

conclusions. The present data provided an independent validation of the association of LOC387715 and HTRA1 SNPs, along with their risk estimates among Indian patients with AMD. These associations underscore their significant involvement in AMD susceptibility, which may be useful for predictive testing.

Age-related macular degeneration (AMD; OMIM 603075; Online Mendelian Inheritance in Man; http://www.ncbi.nlm.nih.gov/Omim/ provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) leads to progressive impairment of central vision and is a major cause of vision loss among the elderly worldwide. ¹ It is a complex disorder with a multifactorial etiology, and genetic predisposition is a major risk factor in its pathogenesis. ^{2 3} With rapid demographic changes and increased life expectancy worldwide, managing AMD is a major global challenge. The global prevalence of AMD varies across populations but is higher in the age groups 65 to 74 and 75 to 84 years. ⁴ The estimated prevalence of AMD in Indians is similar to that observed in Western populations and increases significantly in the elderly. ^{5 6}  

Recent genetic association studies conducted on large case–control cohorts have indicated a single nucleotide polymorphism (SNP) in the complement factor-H (CFH; OMIM 134370) gene at 1q32 that regulates innate immunity, in AMD susceptibility. A Tyr402His variant (rs1061170) in CFH has been shown to increase the risk of AMD by severalfold, ^{7 8 9} and the high-risk allele exhibits similar involvement with the dry and wet stages of AMD. ¹⁰ Although the Tyr402His SNP of CFH has been significantly associated with AMD in most populations worldwide, ¹¹ it does not exhibit any major involvement among Japanese patients. ^{12 13 14 15}  

The second AMD locus mapped on 10q26 harbors three important candidate genes: PLEKHA1 (OMIM 607772), LOC387715 (OMIM 611313), and HTRA serine peptidase 1 (HTRA1; OMIM 602194). ¹⁶ An independent analysis of the 10q26 region indicated a strong association within a 60-kb region of high linkage disequilibrium (LD) that harbored the hypothetical LOC387715 and PLEKHA1 genes. Further analysis has revealed a significant association of the rs10490924 SNP of LOC387715 with AMD in two unrelated German cohorts. ¹⁷ This association was later replicated in various degrees among Caucasian, ^{18 19 20 21 22} Japanese, ^{23 24} and Russian ²⁵ patients with AMD. It has also been shown that the presence of the rs10490924 SNP, along with an associated history of smoking, strongly modifies the risk of AMD. ²¹ The combined effect of the rs10490924 SNP and smoking significantly enhanced the risk of AMD in some populations, ^{20 21 26} but this finding could not be replicated in a large dataset comprising the AREDS (Age-Related Eye Disease Study) and CHS (Cardiovascular Health Study) cohorts. ¹⁸ The combined additive effect of the rs1061170 (CFH) and rs10490924 SNPs exhibited a high population attributable risk percentage (PAR %) in AMD. ^{20 27}  

Very recently, another SNP (rs11200638) located 512 bp upstream of the transcription site of HTRA1 in the same 10q26 cluster was implicated in several independent reports on Caucasian, ^{28 29 30} Chinese, ³¹ and Japanese ^{24 32} AMD subjects. It was also demonstrated that this SNP in the promoter region was in LD with rs10490924 that was a further 6.6 kb upstream of HTRA1. ³¹  

Although the rs1061170 SNP (CFH) has been widely replicated across different ethnic groups worldwide, there is a need to replicate the variants in genes in the 10q26 cluster in these populations, to gain a better appreciation of the variants’ role in disease pathogenesis. Earlier, we demonstrated the association of the rs1061170 SNP in an AMD cohort from India and generated haplotypes that indicated similarity in risk and protection to those observed in Caucasians. ³³ Herein, we provide an independent validation of associations of the 10q26 SNPs based on extensive screening of an AMD cohort from India. In addition, we provide risk estimates based on the combined effects of these SNPs with the rs1061170 SNP (CFH) in this cohort. We have also performed a meta-analysis of the associated SNPs across different studies. 

Methods

Subjects and Clinical Evaluation

The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the institutional review board. The cohort comprised 250 unrelated patients with AMD from seven states of India, along with 250 ethnically matched normal controls presenting at the L. V. Prasad Eye Institute, Hyderabad, India, between August 2004 and May 2007. As this was an extension of our previous cohort, diagnosis and enrollment were based on inclusion and exclusion criteria published earlier. ³³ All the cases and controls were examined independently by two retina specialists, and interobserver agreement was determined by calculating the κ statistic. Blood samples were collected from the patients and normal volunteers by venipuncture, with prior informed consent. 

Screening of the 10q26 Region

Genomic DNA was extracted from the peripheral blood leukocytes according to standard protocols. ³⁴ The genomic region containing the first coding exon of the hypothetical LOC387715 (Entrez ID: NC_000010.9; chromosomal region, nucleotides 124204133-124204548 nucleotides; www.ncbi.nlm.nih.gov/sites/entrez/ Entrez Database, provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD) and the promoter region of the HTRA1 (Entrez ID: NC_000010.9; chromosomal region, nucleotides 124210211-124210782) genes were PCR amplified using specific primers published earlier. ^{17 28} The PCR amplicons were purified with PCR purification columns (MO BIO Laboratories, Inc., Carlsbad, CA) and subjected to resequencing (Genetic Analyzer 3100; Applied Biosystems, Inc. [ABI], Foster City, CA) using dye-termination chemistry (BigDye Terminator; ABI), according to the manufacturer’s protocol. Of the 500 subjects enrolled, the actual number of individuals whose genotype data were available for analysis is indicated in Table 1 . 

Generation of Haplotypes at the 10q26 Region

A total of six SNPs in the 10q26 region were used to generate haplotypes from the sequence data in patients and controls. The SNPs were in the following order: rs10490923, rs2136911, rs10490924, rs11200638, −502C>T, and rs2672598. The first three SNPs were in exon 1 of the LOC387715 gene and the remaining two in the promoter of HTRA1.  

Statistical Analysis

Allele and genotype frequencies were estimated by an allele-counting method. Hardy-Weinberg equilibrium calculations were made, and the estimated haplotype frequencies were obtained with Haploview software, that uses the EM algorithm. ³⁵ LD between the three SNPs was analyzed by using the LD plot function of the software. Odds ratios were computed for estimating the risk of AMD with respect to different genotypes. The combined effects of the LOC387715 and HTRA1 to the CFH genotypes were calculated. 

Meta-analysis

To understand the significance of the observed associations across the rs10490924 (LOC387715) and rs11200638 (HTRA1) variants in different studies, a meta-analysis was undertaken with the estimated odds ratios under a fixed-effect model. The included studies were based on a literature search in PubMed in October 2007 with the phrases “LOC387715,” “HTRA1,” and “age-related macular degeneration” and their combinations. The articles were restricted to the English language. Only those studies in which the genotype counts (or frequencies) in cases and controls were available, were included in the analysis. Meta-analysis was performed with NCSS-PASS-GESS software (windows XP version) according to the manufacturer’s guidelines. ³⁶  

Results

Among the AMD cases, only 9% had a family history of the disease; the remaining were sporadic cases. There was good interobserver agreement in assignment of AMD status (κ = 0.94 ± 0.06). There was an equal distribution of cases of dry (49.7%) and wet (51.3%) AMD. The mean age of patients with wet AMD (68.8 ± 3.1 years) was slightly higher than that of those with dry AMD (64.4 ± 4.8 years). 

Association of SNPs in the 10q26 Gene Cluster with AMD

Resequencing of the previously connoted LOC387715 gene revealed three SNPs, rs10490923, rs2736911, and rs10490924, which were also seen in other populations. ^{17 19} A novel change resulting in a two-base insertion of TG 155 bp downstream of rs10490924 was also observed. However, as this change was in complete LD with the rs10490924 SNP, it was excluded from further analysis. Resequencing of the HTRA1 revealed three SNPs that included a novel SNP at −502C>T that lies between the previously reported ³¹ SNPs rs11200638 and rs2672598 in the promoter region. Other than these SNPs, no other DNA sequence variants were observed. 

There was no significant deviation from Hardy-Weinberg equilibrium among the controls with respect to the six SNPs: rs10490923 (P = .99), rs2736911 (P = .99), rs10490924 (P = 0.622), rs11200638 (P = 0.926), −502C>T (P = 0.961), and rs2672598 (P = 0.319). The frequency distributions of the minor allele of these SNPs along with their genotypes are provided in Table 1 . There was a significant association of the risk alleles for the two SNPs of LOC387715 and all the three SNPs of HTRA1 among the cases (Table 1) . Further analyses were largely restricted to rs10490924 (LOC387715) and rs11200638 and rs2672598 (HTRA1), as these three SNPs exhibited very high disease odds and PAR% for the AMD risk genotypes in our cohort (Table 1) . 

Subjects homozygous for the risk genotypes in rs10490924 (LOC387715) and rs11200638 and rs2672598 (HTRA1) had a significantly higher risk of AMD, as was evident from their respective disease odds ratios and PAR% than those carrying a single copy of the risk allele (Table 1) , similar to other populations. ^{18 19 20 21 22 23 24 26 28 29 30 31 32}  

LD and Haplotype Analysis at the 10q26 Loci

Pair-wise LD analysis between the five SNPs revealed tight LD between the rs10490924 and rs11200638 SNPs (D′, 0.90; 95% CI, 0.84–0.93). The measure of LD was relatively lower between rs10490924 and rs2672598 (D′, 0.80; 95% CI, 0.74–0.87; Fig. 1 ). 

Four different haplotypes (with haplotype frequencies of >5%) with all six SNPs in the 10q26 region were observed among cases and controls (Table 2) . The estimated frequency of the G-C-T-A-C-C haplotype was almost twofold higher in the cases and was deemed a risk haplotype (P = 8.04 × 10⁻¹⁵), whereas the frequency of the G-C-G-G-C-T haplotype was close to twofold higher in the controls and could be protective (P = 2.01 × 10⁻⁴). Similarly, the frequency of the A-C-G-G-C-T haplotype was significantly higher in the controls (P = 0.0066). The data were reanalyzed with respect to the three major SNPs (rs10490924, rs11200638, and rs2672598) that exhibited significant associations in our cohort (Table 1) . It was observed that the frequencies of the risk haplotypes among patients with AMD were consistent, even when two- or three-locus haplotypes comprising the risk alleles of LOC387715 and either or both of the HTRA1 SNPs were considered (data not shown). Similar observations were noted for the protective haplotypes with respect to the wild-type alleles for these loci among the controls. 

Meta-analysis of the Associated SNPs

Twelve independent studies were identified through the search strategy for the rs10490924 (LOC387715) variant. Of these, four studies did not meet the inclusion criteria because of the unavailability of genotype counts. ^{20 21 27 37} Thus, the final analysis for rs10490924 was based on eight studies (including the present one), ^{17 18 19 22 23 24 26} which included 3629 cases and 4292 controls. Likewise, seven studies were found for the rs11200638 (HTRA1) variant, ^{22 24 28 29 30 31 32} of which only three studies ^{22 24 30} met the inclusion criteria, providing 994 cases and 781 controls. 

The results reinforced the earlier findings that the rs10490924 (LOC387715) risk genotype TT contributed to an increased risk of AMD (pooled OR, 8.13; 95% CI, 6.82–9.68) compared with a single copy (pooled OR, 2.47; 95% CI, 2.23–2.74) of the risk (T) allele (Fig. 2) . The pooled estimate of the odds ratios for the homozygous and heterozygous risk alleles of rs10490924 had very narrow confidence intervals. There was a marked degree of homogeneity, and most of the studies, including the present one, clustered around the pooled estimate. But unlike the rs10490924 SNP, the present study deviated slightly from the pooled estimate of the rs11200638 SNP (Supplementary Fig. S1). 

Combined Effect of the LOC387715 and the CFH (rs1061170) SNPs in AMD

Since rs10490924 (LOC387715) was the most significantly associated SNP in our patient cohort (Table 1) , we generated two-locus ORs for rs10490924 along with the previously associated CFH (rs1061170) SNP (published earlier), ³³ to obtain their combined effect in AMD pathogenesis. All nine possible genotypes were compared to the baseline wild-type genotypes GG+TT of rs10490924 and rs1061170, respectively (Table 3) . 

As evident from the table, subjects bearing the homozygous risk genotypes TT (rs10490924) and CC (rs1061170) SNPs were more susceptible to AMD (OR, 73.89; 95% CI, 8.69–628.13) with a PAR of 93.7%. The combined effects of the CC homozygotes (rs1061170) to all the genotypes of rs10490924 exhibited a PAR of 58% to 93.7% (Table 3) . The results were almost similar when the combined effects of rs1061170 of CFH and rs11200638 (HTRA1) were calculated (Supplementary Table S1). This finding could again be attributable to the tight LD between the rs10490924 and rs11200638 SNPs (Fig. 1) . 

Discussion

AMD, a complex disease, is attributed to multiple genes and their interactions with varying magnitudes of effect. ² As a continuation of our earlier results on the association of rs1061170 (CFH) with AMD, ³³ we undertook the present study to explore the involvement of other AMD-associated variants in the 10q26 region in an extended cohort. The results from the present study indicate strong associations with the LOC387715 (rs10490924) and HTRA1 (rs11200638 and rs2672598) SNPs with AMD and provide an independent replication of these results in an Indian population. 

Similar to rs1061170, rs10490924 exhibited significant associations across multiple populations, including the Japanese, who did not exhibit any association with CFH. ^{17 18 19 20 21 22 23 24 25 26 27 37} The consistently high disease odds for the risk genotype provide strong evidence for a functional implication of the rs10490924 variant in AMD. ^{17 22 38} A recent study has provided some clues toward understanding the underlying function of the hypothetical LOC387715 gene. ³⁹  

Earlier studies in Chinese, ³¹ Japanese, ^{24 32} and U.S. Caucasian ^{22 28 29} populations have demonstrated the association of rs11200638 (HTRA1) in AMD. It has also been shown that rs11200638 confers a similar risk in both dry and wet AMD cases in Caucasian populations. ²⁹ The present data showed observed almost similar contributions of rs10490924 (as in Caucasian populations), ^{16 17 18 19 20 21 22 26 27} and rs11200638 (as observed in Chinese and Japanese populations) ^{24 28 31 32} to AMD susceptibility. 

Recent studies, however, have provided convincing evidence of the significant involvement of the LOC387715 SNP, but not HTRA1, in the development of AMD. ²² Although the underlying functions of LOC387715 are yet to be unveiled, it was suggested that rs11200638 is a strongly associated marker in the vicinity of rs10490924 because of the strong LD between these SNPs. ^{22 31} In the present study, we also observed a strong LD between these two SNPs (Fig. 1) . It was also noted that the frequencies of the risk alleles and risk genotypes of rs10490924 and rs11200638 were significantly higher in patients (Table 1) . Likewise, haplotypes bearing the risk alleles of these two SNPs were significantly higher among AMD cases and the converse was true for the controls (Table 2) . But nonavailability of haplotype data for these SNPs in other populations prevented us from making any further comparisons. 

The results of the meta-analysis of the rs10490924 SNP were consistent with those in previous studies, ^{18 26} and there was a lesser degree of heterogeneity across different studies. The data generated from the present study were consistent with respect to the pooled estimate of odds ratios for the homozygous and heterozygous risk alleles of rs10490924 (Fig. 2) , but deviated slightly from that of the rs11200638 SNP. As major studies on this HTRA1 SNP could not be included in the analysis, ^{28 29 31 32} we were not able to interpret anything conclusive. 

The combined effect based on two-locus odds ratios for rs1061170 (CFH) and rs10490924 (LOC387715) indicated an increased risk of AMD for the homozygotes over the heterozygotes (Table 3)similar to the Caucasian populations. ^{17 20 26} A similar situation was observed with respect to the rs1061170 and rs11200638 (HTRA1) SNPs. ^{28 29} The combined effect of the homozygous risk alleles of CFH with both the LOC387715 and HTRA1 SNPs exhibited similar PAR% in the present dataset. Of interest, the joint effect of CFH and LOC387715 (and HTRA1) in the present study has been consistent in most studies worldwide, although the studies were conducted in ethnically and geographically different populations. ^{18 20 28 29}  

Functionally, the presence of LOC387715/ARMS2 mRNA has been demonstrated in the retina and other cell lines; it localizes to the mitochondrial outer membrane in transfected mammalian cells. ²² Although the functional implications of the rs11200638 SNP in AMD pathology has been suggested based on the detection of HTRA1 in drusen of both wet ³¹ and dry AMD ²⁹ eyes and in promoter-based assays, this finding was not replicated in another study. ²² Thus, the precise role of the LOC387715 and HTRA1 SNPs in AMD is as yet unknown, and their interactions with different factors in the complement pathway remains speculative. However, such speculations cannot be addressed unless extensive data on gene–gene interactions in the background of other nongenetic factors leading to the pathophysiology of AMD are elucidated. ³⁸  

In conclusion, the present study highlights the significant association of the 10q26 SNPs in an AMD cohort from India, thereby providing an independent validation of the previously observed results in other populations. ^{16 17 18 19 20 21 22 23 24 25 26 27} Certain similarities in disease risk were noted for the LOC387715 and HTRA1 SNPs, individually and jointly with CFH. Jointly, the homozygous risk genotypes of LOC387715 (rs10490924) and the CFH (rs1061170) exhibited a higher PAR in the present cohort than in the Caucasian populations. ^{20 26} Also, the rs10490924 SNP conferred a higher susceptibility to AMD than did the HTRA1 SNPs, as evident from genotype, haplotype, and meta analyses. Overall, these results underscore the functional importance of these SNPs in AMD pathogenesis and provide risk estimates in the present cohort that may be useful in predictive testing. 

²  Contributed equally to the work and therefore should be considered equivalent authors.

 Supported in part by a grant from the Department of Biotechnology (BT/PR4774/Med/12/181/2004), Government of India (SC).

 Submitted for publication May 11, 2007; revised October 8 and December 14, 2007; accepted March 12, 2008.

 Disclosure: I. Kaur, None; S. Katta, None; A. Hussain, None; N. Hussain, None; A. Mathai, None; R. Narayanan, None; A. Hussain, None; R.K. Reddy, None; A.B. Majji, None; T. Das, None; S. Chakrabarti, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Allele and Genotypes Frequencies and Odds Ratios of the Six SNPs in Cases and Controls

Gene	dbSNP ID (Nucleotide Change)	Allele/Genotypes	Cases (n, %)	Controls (n, %)	P , †	OR (95% CI)	PAR %‡
LOC387715	rs10490923 (G > A)	G allele	206 (95.6)	129 (87.2)		1	—
		A allele	14 (6.4)	19 (12.8)	0.033*	0.46 (0.22–0.95)	—
		GG	97 (88.2)	67 (73.6)	—	1	—
		GA	12 (10.9)	23 (25.3)	0.012*	0.36 (0.17–0.77)	—
		AA	1 (0.9)	1 (1.1)	0.794	0.69 (0.04–11.23)	—
LOC387715	rs2736911 (C > T)	C allele	219 (96.1)	145 (95.4)		1	—
		T allele	9 (3.9)	7 (4.6)	0.754	0.85 (0.31–2.34)	—
		CC	106 (94.6)	86 (91.4)	—	1	—
		CT	6 (5.4)	8 (8.6)	0.537	0.60 (0.20–1.80)	—
		TT	—	—	—	—	—
LOC387715	rs10490924 (G > T)	G allele	147 (38.1)	203 (64.2)		1	—
		T allele	239 (61.9)	113 (35.8)	5.34 × 10−12 *	2.92 (2.17–3.97)	—
		GG	35 (18.1)	89 (43.8)	—	1	—
		GT	77 (39.9)	89 (43.8)	0.002*	2.20 (1.34–3.61)	25.3
		TT	81 (42.0)	25 (12.4)	0.0001*	8.24 (4.54–14.94)	67.1
HTRA1	rs11200638 (G > A)	G allele	189 (41.3)	241 (65.5)		1	—
		A allele	269 (58.7)	127 (34.5)	4.32 × 10−12 *	2.70 (2.03–3.59)	—
		GG	50 (21.8)	78 (42.4)	—	1	—
		GA	89 (38.9)	85 (46.2)	0.049*	1.63 (1.10–2.59)	19.8
		AA	90 (39.3)	21 (11.4)	0.0001*	6.69 (3.69–12.10)	68.9
HTRA1	−502C > T	C allele	445 (97.2)	345 (93.8)		1	—
		T allele	13 (2.8)	23 (6.2)	0.017*	0.44 (0.22–0.88)	—
		CC	217 (94.3)	160 (86.9)	—	1	—
		CT	13 (5.7)	24 (13.1)	0.014*	0.39 (0.20–0.81)	—
		TT	—	—	—	—	—
HTRA1	rs2672598 (T > C)	T allele	122 (26.8)	180 (49.2)		1	—
		C allele	334 (73.2)	186 (50.8)	3.39 × 10−11 *	2.65 (1.98–3.55)	—
		TT	24 (10.5)	48 (26.2)	—	1	—
		TC	73 (32.1)	84 (45.9)	0.084	1.74 (0.97–3.11)	19.7
		CC	131 (57.4)	51 (27.9)	0.0001*	5.14 (2.86–9.24)	57.9

*  Significant.

†  Based on χ² results.

Figure 1.

 

Pair-wise LD analysis across the six 10q26 SNPs. Black: LD (high D′) between the LOC387715 (rs10490924) and HTRA1 (rs11200638) SNPs; dark and light gray: moderate to low LD; white: almost no significant LD between the SNPs.

Table 2.

 

Estimated Haplotype Frequencies at the 10q26 Region between Cases and Controls

Haplotypes	Cases	Controls	P
G-C-T-A-C-C	56.7	29.7	8.04 × 10−15 *
G-C-G-G-C-T	15.7	26.3	2.01 × 10−4 *
A-C-G-G-C-T	6.2	11.6	0.0066*
G-C-G-G-C-C	10.2	13.3	0.1661

 Data are the percentage of cases and controls with each haplotype.

*  Significant.

Figure 2.

 

Meta-analysis of the association between the rs10490924 (LOC387715) SNP and risk of AMD across different studies (Rivera et al., ¹⁷ Conley et al., ¹⁸ Schaumberg et al., ²⁶ Tanimoto et al., ²³ Ross et al., ¹⁹ Kanda et al. ²² and Kondo et al. ²⁴ ). The boxes indicate the estimated ORs for the homozygous (left) and heterozygous (right) risk alleles. Each box plot represents an individual study, and the size of the box is proportional to the respective sample sizes. The odds ratio (OR) corresponds to the center point of the box and the horizontal line indicates the 95% CI for each study. Diamond: the pooled estimate of the OR in the fixed-effect model. Vertical dotted line: the point estimate of the pooled OR.

Table 3.

 

Two-Locus Odds Ratios for the rs1061170 (CFH) and the rs10490924 (LOC387715)

	rs10490924
		GG	GT	TT
rs1061170
TT	1.00	3.06 (1.25–7.50)	5.62 (1.83–17.25)
TC	1.61 (0.60–4.35)	2.46 (1.00–6.00)	11.99 (4.50–31.95)
CC	7.78 (1.62–37.30)	23.33 (5.61–97.01)	73.89 (8.69–628.13)

 Odds ratios with 95% CI were calculated to compare all the genotypes with respect to the baseline wild-type genotypes of rs1061170 and rs10490924 (TT and GG).

Supplementary Materials

Supplementary Figure S1 - (PDF) 

Supplementary Table S1 - (PDF) 

The authors thank the patients and healthy volunteers for participating in the study and Subhadra Jalali and Avinash Pathangay for providing some of the initial patients’ samples.