Skip Navigation


Human Molecular Genetics Advance Access originally published online on June 25, 2007
Human Molecular Genetics 2007 16(17):2135-2148; doi:10.1093/hmg/ddm164
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrowOA All Versions of this Article:
16/17/2135    most recent
ddm164v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Google Scholar
Right arrow Articles by Thompson, C. L.
Right arrow Articles by Iyengar, S. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thompson, C. L.
Right arrow Articles by Iyengar, S. K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2007 The Author(s)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Complement factor H and hemicentin-1 in age-related macular degeneration and renal phenotypes

Cheryl L. Thompson1, Barbara E.K. Klein4, Ronald Klein4, Zhiying Xu1, Jennifer Capriotti1, Tripti Joshi1, Dmitry Leontiev1, Kristine E. Lee4, Robert C. Elston1 and Sudha K. Iyengar1,2,3,*

1 Department of Epidemiology and Biostatistics, 2 Department of Genetics and 3 Department of Ophthalmology, Case Western Reserve University, Wolstein Research Building, Room 1315, 2103 Cornell Road, Cleveland, OH 44106, USA and 4 Department of Ophthalmology and Visual Sciences, University of Wisconsin Medical School, Madison, WI 53706, USA

* To whom correspondence should be addressed. Tel: +1 2163685630; Fax: +1 2163684880; Email: ski{at}case.edu

Received May 6, 2007; Accepted June 21, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
In this study, we investigated the associations of complement factor H (CFH) and hemicentin-1 (HMCN1) with age-related macular degeneration (AMD) and renal function. Three scales, measuring the course of AMD and drusen development, were examined in two samples: the Family Age-Related Macular degeneration Study (FARMS), consisting of families ascertained through a single individual with severe AMD, and an unascertained population-based family cohort, the Beaver Dam Eye Study (BDES), which was also used to assess longitudinal changes in AMD and associations with renal function. Associations were performed by a regression accounting for known risk factors as well as familial and sibling effects. Strong evidence of the association of rs1061170 (Y402H) variation with AMD was confirmed (P = 9.15 x 10–5 in BDES, P = 0.016 in FARMS). This association was observed in multiple AMD scales, suggesting that its role is not phenotype-specific. Polymorphisms in both CFH and HMCN1 appeared to influence the longitudinal rate of change of AMD. The rs1061170 polymorphism was also associated with a reduction in estimated glomerular filtration rate (eGFR) (P = 0.046). Another CFH polymorphism, rs800292, was similarly associated with eGFR [ß = –0.90 (P = 0.022)]. Associations between rs743137 (P = 0.05) and rs680638 (P = 0.022) in HMCN1 with calculated creatinine clearance progression were also observed. Both genes appear to play a role in both AMD and renal pathophysiology. These findings support evidence for common pathways influencing ocular and renal function and suggest that further work is required on their common determinants.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
In recent years, a multitude of studies have reported that variants in the complement factor H (CFH) gene are associated with a significant increase in age-related macular degeneration (AMD) risk (1–6). The majority of these conclusions are based on case–control studies from a variety of populations, but many were follow-ups to a linkage signal obtained from several earlier genome scans (7–12), which showed strong linkage peaks in the 1q region (13). A common polymorphism in CFH, a tyrosine to histidine change (Y402H), shows significant association with AMD in most of these studies, although there are some data that suggest other variants may also be important (14–16). Based on these ascertained data, it has been estimated that the population attributable risk (PAR) of AMD due to CFH is between 25 and 50%, and it has been hypothesized that this gene may fit the common disease common variant paradigm for AMD.

In an earlier report, Schultz et al. (17) found a mutation in the hemicentin-1 (HMCN1) gene on chromosome 1 that predicted AMD in their sample; single nucleotide polymorphisms (SNPs) in HMCN1 also showed an association with AMD in the Family Age-Related Macular degeneration Study (FARMS) sample (12). After discovery of CFH, HMCN1 was discarded by most investigators as a major cause of AMD; however, its role in AMD risk was never fully studied until a recent whole genome association study showed evidence of association (18). Subsequently, several other candidate genes were also suggested to cause AMD (19–23). The most notable of these was the region of PLEKHA1/LOC387715/HTRA1 on chromosome 10 discovered by Jakobsdottir et al. (22) and confirmed independently by others (21,24,25). This location on 10q26 showed the best evidence for linkage in a consortium meta analysis (13), and it is likely that one of these genes plays a very important role in AMD pathophysiology. Jakobsdottir et al. (22) suggested that a locus on chromosome 10q26 accounts for a PAR as high as 57%, but acted independently of CFH. Given the number of genes associated with AMD, upwards of six at last count, attributable risk for disease due to a particular gene will be difficult to assess in ascertained populations unless the controls were selected using a strategy agnostic to disease status. In order to improve upon these estimates, we use the population-based cohort Beaver Dam Eye Study (BDES), which has cases but also controls unascertained for disease, to determine PAR for CFH based upon a threshold of severe affection status.

Independently of AMD, variations in CFH have been associated with different renal diseases. Specifically, CFH has been associated with membranoproliferative glomerulonephritis II (MPGN II), a source of chronic renal dysfunction that progresses to end-stage renal disease (26,27), as well as atypical hemolytic uremic syndrome (26,28–30), a rare renal disorder that often results in chronic kidney failure and high blood pressure. In animal models, administration of CFH was recently shown to confer protection against glomerular scarring in mice with chronic serum sickness (31), and have a strong effect on kidney function (32). The sum total of the evidence from these data suggest that, in contrast to AMD pathogenesis, rarer CFH variants may play a role in renal disease [see review by Saunders et al. (33)]. While the role of highly penetrant, but infrequent, CFH variants in rare renal diseases is indisputable, the question that arises is whether common CFH variants, such as Y402H, have a more global role in renal function. A review of the literature showed a cluster of renal abnormalities, e.g. familial focal segmental glomerulosclerosis (34), glomerulopathy with fibronectin deposits (GFND) (35), renal failure (36), hemolytic uremic syndrome (37,38) and MPGN (39) linking to 1q32. In contrast, two genome scans for quantitative measures of renal function, including estimated glomerular filtration rate (eGFR) and creatinine clearance, did not show similar evidence of linkage to this region on chromosome 1 (40,41). The majority of the renal diseases that show linkage to this region of chromosome 1 are characterized by nephrotic range proteinuria and a decline in renal function, as measured by creatinine clearance or eGFR, along with other characteristic pathologic features. Our hypothesis was that variants in genes such as CFH, that predispose individuals to rare types of chronic kidney disease and reduction of creatinine clearance, may also play a role in reduction of creatinine clearance in the general population, and that association of CFH variants is not exclusive to disease states such as MPGN or hemolytic uremic syndrome.

In light of these findings, we investigate the association of CFH and HMCN1 with AMD severity and AMD-related subphenotypes (i.e. drusen type and drusen size) in two populations, a sample of 34 families ascertained through a severely affected proband (FARMS) and a sample of 2307 unascertained families for which DNA had been collected as part of a population-based cohort, the BDES. The latter sample allows us to more accurately estimate population level risks. Since renal data was also available in the BDES sample, variants in CFH and HMCN1 were also evaluated for an association with calculated creatinine clearance and proteinuria in this sample, including longitudinal changes in these variables. Analyses of intermediate traits, e.g. calculated creatinine clearance and proteinuria, rather than end-stage renal disease, such as MPGN, show that both CFH and HMCN1 may have a more global role in the decline of age-related renal function.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Cross-sectional analyses: FARMS sample
All SNPs were found to be in Hardy–Weinberg proportions. The rs1061170 (Y402H) SNP in CFH was significantly associated with the severity level of AMD, increasing size of drusen and more severe type of drusen (Table 1). For drusen size, the rs1061170 SNP in CFH and the rs721153 SNP in HMCN1 were both significant individually. In the multivariable model including multiple SNPs, both the rs1061170 SNP in CFH and the rs743137 SNP in HMCN1 remained significant for drusen size (rs1061170 ß = 0.31, SE = 0.097, P = 0.0015, rs743137 ß = –0.36, SE = 0.16, P = 0.023 in the multiple SNP model). CFH and HMCN1 are far enough apart that the linkage disequilibrium (LD) between them is negligible and hence we postulate that these genes may have independent effects in this sample. Correlations were calculated between the individual SNPs in BDES. These correlations suggest that SNPs within the same gene are significantly correlated, but no SNP in CFH is strongly correlated with any SNP in HMCN1 (Supplementary Material, Table S1). For the drusen-type scale, only the rs1061170 SNP in CFH showed statistically significant association with more severe type of drusen at the 0.01 level (Table 1). In summary, the Y402H variant showed the most significant effect in severity of AMD and with size and type of drusen, yet associations with HMCN1 cannot be ruled out.


View this table:
[in this window]
[in a new window]

  		Table 1. Summary of results of single-SNP association analysis for AMD-related phenotypes in FARMS sample

 
Cross-sectional analyses: BDES sample
All SNPs were found to be in Hardy–Weinberg proportions in the BDES sample. The rs1061170 SNP in CFH was also significantly associated with severity level of AMD, increasing size of drusen and more severe type of drusen (Table 2), again in the additive model. While other SNPs were significant for one or two traits, no other SNP was significant for all three scales of AMD traits, and, in contrast to the FARMS sample, when multiple SNPs were added simultaneously to the model, the rs1061170 SNP in CFH remained the only statistically significant SNP. As with the FARMS sample, the rs743137 SNP in HMCN1 was significant at the P = 0.05 level, but in this sample, it was significant only for the 15-step AMD scale, whereas in FARMS it was significant on the drusen size scale. Since this SNP was not significant in the multiple SNP model, we conclude that HMCN1 may have rare variants that play a role in AMD. In our analysis HMCN1 was not completely examined as the gene is large and the SNPs selected in our study did not cover the entire LD structure.


View this table:
[in this window]
[in a new window]

  		Table 2. Summary of results of single-SNP association analysis of AMD-related phenotypes in BDES sample

 
Because the BDES sample is an unascertained, population-based sample, it gives a more accurate estimate of the PAR, or population attributable fraction (PAF), of CFH to AMD than an ascertained sample permits. Using a threshold of 12 on our 15-step AMD severity scale, which corresponds to either choroidal neovascularization (CNV) or geographic atrophy, i.e. late-stage disease, the PAF equals 24.0%. This can be interpreted as approximately 24% of the end-stage AMD cases in the BDES population being attributable to the histidine variant at the rs1061170 SNP in CFH. Using a threshold of 13 or greater, which corresponds to varying degrees of severity of the CNV lesion, the PAF equals 16.1%. This can be interpreted as approximately 16% of CNV cases being due to the histidine variant of the rs1061170 SNP in CFH.

In the analysis of the renal phenotypes in the BDES sample (Table 3), we found marginally significant association of the rs1061170 (ß = –0.99, SE = 0.50, P = 0.046) and rs800292 (ß = –0.90, SE = 0.39, P = 0.022) SNPs in CFH to renal function, as measured by eGFR. The negative regression coefficient (ß) implies that those with at least one rs1061170 risk variant (dominant model) have a statistically significantly lower eGFR. Because having a high eGFR is desirable and is associated with better kidney function, opposite sign regression coefficients for AMD and renal function risk can be interpreted as the same allele showing both a risk of AMD and diminished renal function. No SNP was significant at the 5% level for association with proteinuria or calculated creatinine clearance.


View this table:
[in this window]
[in a new window]

  		Table 3. Summary of results of single-SNP association analysis of renal phenotypes in BDES sample

 
An investigation into the relationship between renal function and AMD status found no difference between persons with late AMD (AMD scale score greater than or equal to 12) and controls without AMD (AMD scale score of less than or equal to 4) with respect to mean calculated creatinine clearance or presence of proteinuria (data not shown); instead, diabetes was a much stronger predictor of renal function (ß = 3.17, SE = 0.81, P = 8.6 x 10–5) for calculated creatinine clearance and proteinuria (ß = 0.25, SE = 0.036, P < 1 x 10–7). To confirm that our lack of association of AMD with renal status was not due to survival bias, we used the International Classification of Disease 9 codes to examine the death records for any mention of renal disease in BDES. Our data shows that CFH status is associated with excess mortality due to renal failure ({chi}2=4.9, P = 0.03) in an additive model (data not shown).

Longitudinal association: BDES sample
In the analysis of the progression of AMD, the rs1061170 SNP in CFH also provides the strongest association with progression (Table 4), with the highest ß-coefficient for each step of the 15-point AMD severity scale per 5-year period, but the rs800292 SNP is more statistically significant. In addition, when accounting for both significant SNPs, the rs800292 SNP in CFH (ß = 0.11, SE = 0.049, P = 0.024, multivariable model) remains significant in these analyses even when the rs1061170 SNP in CFH (ß = 0.14, SE = 0.066, P = 0.037, multivariable model) is added to the model.


View this table:
[in this window]
[in a new window]

  		Table 4. Summary of results of single-SNP association analysis of longitudinal measures of AMD-related phenotypes in BDES sample

 
The longitudinal analysis of the renal data (Table 5) was quite different, with the two HMCN1 SNPs, rs743137 and rs680638, showing significant (P = 0.05 and 0.022, respectively) and strong (ß = –1.25 and 3.92, respectively) associations with calculated creatinine clearance progression. It is important to consider the measurement of the traits when interpreting the sign of the regression coefficients. In all three AMD severity scales, the higher the value, the more severe the disease. However, the renal traits are not severity scales, but rather continuous measurements; for calculated creatinine clearance, the lower the number (that is, the more negative the regression coefficient), the greater the decrease in renal function. The negative regression coefficient for the rs743137 SNP indicates that what was defined as the risk allele for AMD is also associated with a decrease in renal function. However, what we had defined as the risk allele for rs680628 (the minor allele) was actually protective for calculated creatinine clearance progression. The effect of the rs743147 SNP in HMCN1 (ß = –0.92, SE = 0.59, P = 0.11) appeared to be dominated by the effect of the rs680638 SNP (ß = 3.92, SE = 1.87, P = 0.033) when coupled in the multiple SNP model, and no longer remained significant. This may occur because the C allele at the rs680638 SNP is present on both a risk haplotype and a protective haplotype (see discussion of haplotypes below), confounding the interpretation of single SNP effects. No SNPs were statistically significant for proteinuria or eGFR progression (Table 5).


View this table:
[in this window]
[in a new window]

  		Table 5. Summary of results of single-SNP association analysis of longitudinal measures of renal phenotypes in BDES sample

 
Haplotyping in BDES and FARMS samples
Six of the eight possible CFH haplotypes were found to have a population frequency > 1% in the BDES sample. When individual haplotype probabilities were included as covariates in the model, accounting for age, several haplotypes showed statistically significant association with cross-sectional AMD-related phenotypes (Table 6). Specifically, the G-C-T haplotype, which contains the risk variant of rs1061170, shows a significant association with AMD (P = 1.37 x 10–4), drusen size (P = 1.22 x 10–5) and drusen type (P = 0.032), whereas the T-T-T haplotype, containing the wild-type allele of rs1061170, shows a protective effect on AMD (P = 6.79 x 10–4) and drusen size (P = 0.021). The G-T-T haplotype, which also contains the wild-type allele for rs1061170, was not statistically significantly associated with the AMD scale, but did show significance in the association with drusen size (P = 1.47 x 10–3).


View this table:
[in this window]
[in a new window]

  		Table 6. Summary of results of haplotype association analysis in BDES sample for AMD traits

 
For the renal phenotypes (in the BDES sample), none of the CFH haplotypes showed any significant association with calculated creatinine clearance or proteinuria. However, one CFH haplotype (T-T-T) showed marginally significant association with eGFR (Table 7), with a modest increase in eGFR.


View this table:
[in this window]
[in a new window]

  		Table 7. Summary of results of haplotype association analysis in BDES sample for renal traits

 
For HMCN1, 14 of the 16 possible haplotypes were observed in BDES to have a population frequency > 1% (Table 6). None of the HMCN1 haplotypes showed significant association with any AMD scale in the BDES population. These results suggest that CFH is a stronger predictor of AMD than HMCN1 in the population-based BDES sample. However, a limitation of this analysis is that the sentinel HCMN1 SNP(s) may not have been adequately represented in this haplotype.

As with CFH, none of the HMCN1 haplotypes showed statistically significant association with calculated creatinine clearance or proteinuria (Table 7). However, two (A-C-G-C and G-A-T-C) showed borderline significance, with strong effects in opposite directions on eGFR. This does not rule out the possibility of an effect of CFH or HMCN1, or another gene in this region, on renal function, in particular on eGFR.

Only five of the possible CFH haplotypes were found in the FARMS sample. In this sample, the G-C-T haplotype also showed the most evidence of directly influencing the risk of drusen development (Table 8). Surprisingly, no CFH haplotype was statistically significantly associated with the AMD severity scale.


View this table:
[in this window]
[in a new window]

  		Table 8. Summary of results of haplotype association analysis in FARMS sample

 
Twelve of the possible 16 haplotypes for HMCN1 were found in the FARMS sample. Interestingly, three of the haplotypes showed statistically significant P-values to atleast one scale (Table 8). The A-A-G-T and G-A-G-T haplotypes showed association with the AMD severity scale, the G-A-T-C and G-A-G-T haplotypes showed association with the drusen size scale.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
The complement system has been hypothesized to play a strong role in both kidney and eye diseases through its involvement with innate immunity (42,43). With the identification of a protective risk haplotype for AMD in complement factor B, there is now convincing evidence that perturbance of the complement pathway may have a causal role in these diseases (see also 20). One aim of this study was to investigate whether only CFH played a role or if HMCN1 was also important in the pathogenesis of AMD. The results shown in this study confirm the strong association of the rs1061170 SNP in CFH with AMD. While other CFH and HMCN1 SNPs were associated with AMD or AMD-related phenotypes at the individual level, the rs1061170 SNP was the most consistently associated SNP evaluated and is likely to have the largest effect at a population level. The results of the association of this SNP with several different AMD severity scales show that this SNP affects various AMD-related phenotypes [e.g. soft drusen, pigmentary abnormalities, geographic atrophy (GA) and CNV] that characterize both early and late AMD. This is consistent with recent conclusions of the rs1061170 SNP conferring a similar risk of soft drusen, GA and CNV (44). However, our results indicate that variations in HMCN1 cannot be ruled out as predictors of AMD risk or progression. These results are consistent with the recent release of the data from the Age-Related Eye Disease Study (AREDS) study published electronically in dbGAP (18), that variants in HMCN1 also influence AMD risk. The variants reported as showing the best P-values in AREDS are rs10465495 and rs7546152, but were not genotyped in our sample. The gene is very large (456 kb) and a more extensive canvas of this gene is required to fully comprehend which variants cause the disease. This analysis also supports our previous work in this sample (12) showing that the HMCN1 gene has an important role to play in AMD.

When calculating a PAR or PAF, it is important to consider the sample. PAR and PAF can be calculated without bias if the control individuals have been sampled without regard to disease status. Therefore, the BDES sample was used to compute PAF. We estimated the PAF for severe disease at 24.0%. This is considerably lower than the estimates previously reported as the attributable risks for CFH derived from ascertained samples, as well as the 41% PAR estimated via the Cardiovascular Health Study (45), which was not ascertained with respect to eye disease, but consistent with the 25% PAR estimate reported in the unascertained Physicians' Health Study (16). Our estimates come from an unascertained sample and are supported by data from other epidemiologic and genetic studies which suggest that there are other multiple genetic (Fig. 1), as well as environmental factors contributing to AMD in a population, although the PAR estimates for environmental factors are less known.


Figure 1
View larger version (53K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. A model for the genetic factors of AMD.

 
We also found that the risk of development and progression in the severity of the AMD phenotype could also be predicted by SNPs in both CFH and HMCN1. Finding an association between the progression of AMD over time and individual SNPs in CFH suggest that the same genes may predispose an individual to AMD, that is, increase the odds that an individual has a high score on any or all of the AMD scales, as well as influence its rate of change, that is, how much progression on the scale was observed during the follow-up. The interesting result from the longitudinal analysis is that there was so much statistical significance for the rs1061170 and rs800292 SNPs in CFH on the progression of the AMD severity scale, but no evidence for association with the rate of progression on either of the drusen scales. These results with the AMD severity scale also fit well with association results previously reported using these SNPs as predictors of progression of pigmentary abnormalities in AMD (46). On the other hand, the HMCN1 SNPs show statistical significance only in the longitudinal analysis, suggesting that they may not contribute to AMD initiation (or an individual having a higher score) as much as the progression of AMD in individuals already predisposed owing to other genetic or environmental factors. When considering the biological implications, both genes have an independent impact on AMD. The role for CFH in the kidney is supported by CFH knockout mice that do not develop AMD (47), but rather develop glomerulonephritis. The lack of an eye phenotype is surprising to a certain degree, but these observations suggest that either the genetic background of the mouse is important in determining this phenotype, or the environment or lack of macula in the mouse is not conducive to generating features of AMD. In many animal models where there is redundancy in the physiological system, knock-outs do not lead to anticipated outcomes (48–51). Instead, it may be necessary to develop a knock-in or conditional knock-in mouse model that will develop features of AMD, either on its own or under the right genetic background (genetic interactions). If neither of these experiments succeeds, it may be necessary to manipulate the environmental conditions to promote the manifestation of the phenotype.

While much molecular work has established a link between CFH and rare renal disorders (33), very little work has been done to investigate the association of polymorphisms in CFH with quantitative measures of renal function such as creatinine clearance and urinary protein excretion in the general population. We extended the prior hypothesis based on a knock-out mouse model, as well as that rare variants cause renal disease, to one in which common polymorphisms in CFH affect glomerular function and protein excretion in the general population. It is of interest to note that both CFH and HMCN1 were associated with calculated creatinine clearance, with HMCN1 having stronger effects on the progression of calculated creatinine clearance. It should also be noted that our dataset was population-based and unascertained with respect to renal phenotypes. A review of the literature suggests that several linkages cluster in the 1q32 region and another renal failure gene besides CFH, podocin (NPHS2), which is located approximately 18 Mb from CFH and about 6 Mb from HMCN1, is associated with hereditary/congenital nephrotic syndrome. Saunders et al. (33) noted that the Y402H polymorphism that was previously seen in hemolytic anemia patients was not associated with renal disease. Our data suggest similar trends, although a deeper scan of the CFH gene using additional SNPs will be necessary in populations with focal segmental glomerulosclerosis and other forms of renal disease to confirm this observation; the focus of this study was not renal failure, but rather to look at overall renal function.

Given that there appear to be genetic factors in common between AMD and renal phenotypes, which also share potential environmental factors in common, it may be important to look at the correlation between the two in order to assess causality. We found only slight correlation between AMD and renal function, but a survival bias in CFH due to renal failure. Thus, we may well lack the ability to detect any association between AMD and renal failure because of a survival bias caused by attrition of particular CFH variants.

In several of the phenotypes, including the 15-step AMD severity scale in BDES and the drusen size severity scale in FARMS, multiple SNPs were shown to be significantly associated with the phenotype, including those in HMCN1. The significance of the other SNPs may be postulated to be due to LD with the now well-established rs1061170 CFH SNP. However, the region covered by the seven SNPs studied here is quite large and these SNPs do not appear to be in tight LD with each other. The LD observed in our sample was quite small (see Supplementary Material, Table S1). In addition, the two SNPs studied are 9.1 Mb apart (Table 9), and the LD blocks in neither CFH nor HMCN1 extend past the coding region of the gene (Figs 2 and 3). Finally, SNPs in both CFH and HMCN1 also show association in the AREDS genome-wide association study (18). It seems that, given the large size of both of these genes and the distance between them, the significance may be due to independent small effects that may interact with other mutations in the region. There has also been evidence that multiple mutations in CFH contribute to AMD (14,15,52), and the three CFH SNPs that we evaluated are in separate LD blocks within CFH. Li et al. (15) found that CFH haplotypes not containing the rs1061170 risk variant still showed an increase in AMD risk. We did not find any statistically significant association between any haplotype not containing the rs1061170 risk variant in either sample. In summary, although the rs1061170 polymorphism showed quite a significant effect in our sample, the data also suggest that a deeper canvas of the area is warranted and that other variants may also show disease risk. Our best conclusion is that HMCN1 may bear variants that affect progression of both AMD and renal outcomes independent of CFH status. Multiple genes in close proximity causing disease is often seen when sub-congenic strains bearing partial segments of quantitative trait loci are made for fine-mapping in animal models (52–54). Although less frequently reported in human studies, the incidence of such reports will increase, as additional whole genome association studies are published.


Figure 2
View larger version (42K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Linkage disequilibrium (LD) plot for the complement factor H gene.

 


Figure 3
View larger version (64K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Linkage disequilibrium (LD) plot for the hemicentin-1 gene.

 


View this table:
[in this window]
[in a new window]

  		Table 9. Details of genotyped SNPs

 
As with all studies, our study is not without limitations. Measurements made on a severity scale are more informative than a purely dichotomous measure (such as affected or unaffected with AMD). However, using a severity scale will require a different interpretation of the regression coefficient obtained compared with the odds ratio of a study using a discrete trait. Instead of representing a risk of developing a disease, it measures the risk of progression due to one step of the scale, and makes the assumption that each step of the scale is equal to the other steps.

When looking at alternate phenotypes in a study, it is important to consider possible ascertainment biases. In the analysis of the renal data, it would be important to consider adjusting for AMD status when looking at CFH and HMCN1 in a sample ascertained on AMD. However, BDES is a population-based sample, and was not subject to ascertainment by AMD status. In addition, we did a correlation analysis of AMD and eGFR. While these two phenotypes are correlated (r = –0.14), the correlation appears to be due almost exclusively to the very strong correlation of both traits with age. After accounting for age, the correlation disappears. Thus, because we accounted for age in our models, it was not necessary to adjust for AMD status in the analyses of the renal data.

Our study is the first to address the issue of renal function and proteinuria with both genes. While our results with both CFH and HMCN1 are significant, this investigation was conducted in a sample with few cases of end-stage renal disease. Therefore, replication of these results in an independent sample is warranted. In addition, the HMCN1 gene is quite large, and there are some regions of the gene that were not covered by the four SNPs we evaluated. Mutations in other parts of the gene may have more of an impact on function and thus be more likely to be associated with renal or AMD-related phenotypes. While we only found modest evidence of CFH mutations associated with renal function or proteinuria, our power to detect an association may have been weakened by the fact that we did not ascertain our sample with respect to renal function, and those individuals who may have had severe renal failure due to mutations in CFH may have not survived long enough to be ascertained into this cohort of older individuals. Finally, we have not yet evaluated the interaction of CFH or HMCN1 with potential covariates, as has previously been shown to exist for the chromosome 10 locus (LOC387715) with smoking (55).

The Cockcroft-Gault function for creatinine clearance is an imperfect measure for estimating renal function. There are many other measures of renal function, including the Modification of Diet in Renal Disease Study Group (MDRD) equation (56,57) for eGFR. We also investigated the association of CFH and HMCN1 with eGFR values, as estimated by the MDRD equation, which have been reported to be similar to the Cockcroft-Gault values (58). However, we did not observe similar influences of CFH and HMCN1 on creatinine clearance and eGFR (Tables 3 and 5).

Another important consideration for the interpretation of our results is the issue of multiple testing. We evaluated seven SNPs in two genes for three related AMD scales and two, related, renal measures. However, since these SNPs were in two genes with prior evidence of association with AMD, and one, CFH, showing strong evidence of linkage to other renal phenotypes, our analyses for CFH may be considered an independent replication. The novel discovery of the role of HMCN1 with eGFR is supported by a very small P-value in the longitudinal analysis, which would remain significant even after correction for multiple testing.

In conclusion, our research shows a strong association of CFH with AMD and evidence of an association with creatinine clearance, providing evidence that these phenotypes are linked through the complement system. It would be interesting to further extend these analyses with additional genetic components of the complement system. There also appear to be significant haplotypes in HMCN1 that may be too rare to detect in unascertained, population-based samples. Careful examination of HMCN1 also supports a role for this molecule in AMD pathogenesis and renal progression, which is also worthy of further pursuit.

The association of the same gene with ocular and renal function is not a new theory. For example, individuals with MPGN II are known to frequently have abnormalities in Bruch's membrane present in the eye (59) as well as peripheral drusen (60). In addition, Van Agtmael et al. (61) showed that a mutation in Col4a1 results in ocular and renal abnormalities. This study offers additional insight into common pathways influencing renal and ocular function.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 REFERENCES
 
Subjects
Two samples were used for this study. A brief summary of the two samples is given (Table 10), but each has been previously described (9,12,62). The first study sample, FARMS, consists of 34 persons with advanced AMD and their extended families, including a total of 297 individuals with available genotypic and phenotypic information. These cases were recruited from the Madison, Wisconsin area and were ascertained based on having advanced AMD. Details of this study have been described earlier (12).


View this table:
[in this window]
[in a new window]

  		Table 10. Comparison of FARMS and BDES samples

 
The second sample involves a population from the BDES. The BDES study population comprises a cohort of 4926 individuals between the ages of 43 and 86 years identified in a census from 1987 to 1988 of the city and township of Beaver Dam, Wisconsin and, because of the relationships within the community, included 602 pedigrees (2307 individuals). The related individuals in these 602 pedigrees were genotyped and used for the analyses presented here. There is little difference in the genotyped individuals compared with the remainder of the sample (63). Data from this population were obtained at baseline, at 5-year and 10-year follow-up. Details of this study are also provided elsewhere (64,65). The data collected in this study included measures of renal function and AMD phenotypic information. The tenets of the Declaration of Helsinki were followed and informed consents were obtained for all participants of both studies. The Institutional Review Boards of the University of Wisconsin and Case Western Reserve University approved both of these studies, and all participants were required to sign a consent form.

Phenotypic evaluation
Each participant in both studies was interviewed and had stereoscopic 30° color fundus photographs of each eye taken. Two gradings were performed on photographs for each eye according to the standardized protocols developed to classify and detect AMD (66,67). Severity of AMD for each person was determined using a 15-step AMD severity scale, which was designed to represent the progression from no disease to early and then late-stage AMD, as described in Schick et al. (9).

Two additional phenotypic measures of drusen were used for this study, one describing the size and the other the type of drusen. The drusen severity scales are described in detail elsewhere (62). The differential weighting of the phenotypic features of AMD was performed to appraise the entire clinical spectrum of the disorder.

The filtering capacity of the kidney is frequently assessed using indirect measures, because direct measurement of kidney function is cumbersome to assess in large numbers of patients, nor it is standard clinical practice. Creatinine clearance, the most widely used measure of renal function, estimates the amount of endogenous creatinine that is filtered through the kidneys. Decrease in creatinine clearance, in the absence of confounding factors, is directly proportional to reduction in the number of functioning nephrons in the kidney. Thus, creatinine clearance is a good surrogate of kidney function. In practice, creatinine clearance can be directly calculated from a 24-h timed urine sample, after adjustment for confounding factors. Alternatively, it can be estimated from serum creatinine using predictive equations such as the Cockroft-Gault (68). Because estimation of creatinine clearance is affected by other factors such as age, race, gender, nutritional status and kidney disease, other predictive equations, such as the MDRD have been developed as measures of kidney function (56,57). The MDRD equation measures GFR, whereas the Cockroft-Gault measures estimated creatinine clearance. Urine creatinine is secreted in the proximal tubule (of the kidney), and hence the Cockroft-Gault overestimates GFR and the two measures are not equivalent. Further, the MDRD formula more accurately measures GFR among those with kidney disease.

Nephrogenic damage erodes the filtration barrier in the glomerulus, such that molecules that would ordinarily have been retained by the barrier leak into the urine. Clinical assessment of the amount of protein that leaks into the urine is conducted by determining the amount of albumin (or total protein) present, after adjusting for the individual's protein intake and turnover. Thus, urine albumin to creatinine ratio or total protein to creatinine ratio is used as a clinical indicator of kidney damage. The presence of a high albumin to creatinine ratio is referred to as albuminuria and an increase in the protein to creatinine ratio is termed proteinuria. While clinical chemistry laboratories routinely measure urine albumin, protein and creatinine, an approximate quantitation of the amount of protein in the urine can be determined using a dipstick (69).

In our sample, renal phenotypes evaluated include calculated creatinine clearance and eGFR, as well as proteinuria [derived from dipstick estimate of amount of protein in the urine corrected for specific gravity (69)]. Calculated creatinine clearance was defined as per the Cockcroft-Gault equation (68). eGFR was calculated by the MDRD equation (56,57).

Deaths and causes of death were confirmed in BDES with the Wisconsin Department of Health and Family Services or the National Death Index for all known deaths and for persons who we were unable to contact. Mortality between the baseline examination and 31 December 2002 was used.

Genotyping
LD plots were generated for both HMCN1 and CFH using Haploview (70) and the HapMap (71) data for the Centre d'Etude du Polymorphisme Humain (CEPH) population, since a majority of our samples were of European descent, with the blocks defined by confidence intervals (72).

Three SNPs in CFH and four SNPs in HMCN1 were genotyped in both samples using TaqMan assays. Two hundred and ninety seven individuals in the FARMS sample were genotyped and 2307 individuals in the BDES population were genotyped. The details of the SNPs are given in Table 9. The three SNPs in CFH cover the three LD blocks in the CFH gene (Fig. 2). The HMCN1 gene is large and the entire LD structure was not examined in this analysis. The four SNPs that were genotyped cover a majority of the gene, but there are regions of the gene that are not well-covered by our SNPs (Fig. 3). Genomic DNA was subjected to PCR amplification in a volume of 25 µl including 1 x TaqMan Universal PCR Master Mix [PE Biosystems; 8% glycerol and 1 x TaqMan buffer (100 mM Tris, pH 8.0, 500 mM KCl)] with 7.5 mM MgCl2, dNTPs (200 µM dATP, 200 µM dCTP, 200 µM dGTP, 400 µM dUTP) and 0.5 U of AmpliTaq Gold DNA polymerase. Assays-on-Demand SNP Genotyping Assay Mix containing the two specific TaqMan MGB probes, forward and reverse primers, was also added. The 96-well or 384-well plate containing the reaction mixture was then run on the ABI 7900 Sequence Detection system. The overall genotyping error rate was estimated at < 1% based on 609 replicates of all seven SNPs. No individual SNP showed error rates significantly higher than any other.

Statistical methods
Testing for Hardy–Weinberg proportions was done by a {chi}2 goodness-of-fit test using only the founders to eliminate the non-independence of family data.

Association analysis was performed by a linear regression accounting for sibling and family effects with the program ASSOC in the S.A.G.E. software package (73). Analyses for AMD were performed on each of the three scales (1): 15-level AMD severity scale with scores from 1 to 15 (2), drusen size severity scale with scores from 0 to 6 and (3) drusen type severity scale with scores from 0 to 3. Analyses were done using the average score over both eyes at baseline as well as scores in the worse eye at baseline. Because average and worse eye results were very similar, only the results from the average scores are reported here. Age (years) was accounted for in all analyses.

Initial analysis was done including each SNP alone in the regression model and evaluating significance. To estimate a more comprehensive model for each phenotype including multiple SNPs, each SNP was added in a stepwise fashion to obtain the most parsimonious model. For each regression, three different genotypic models (dominant, additive and recessive) were tested. The best fitting model, defined as the lowest P-value in the regression, is reported in the results.

Because longitudinal ocular and renal data were available for the BDES sample, we repeated the association analysis on this sample using the rate of change in score during follow-up, of AMD, drusen size and drusen type scale, as surrogates for progression in the respective areas. Ocular phenotypes were measured at the baseline, 5- and 10-year follow-up. For those genotyped individuals in which at least two scores were available (n = 1323), the rate of change per 5-year time period in the score was calculated for each eye using the ordinary least squares (OLS) method, which would allow for both disease advances and improvement, and the eye showing the greatest deterioration along the scale of interest was used for that analysis. Those without scores for at least two time points were excluded from this analysis. Of those individuals genotyped, 882 individuals had data available at time points 1, 2 and 3, 67 individuals had data available at time points 1 and 3, but missing for time point 2. For the latter individuals, the rate of change was assessed using time points 1 and 3. Of the remaining individuals, 367 had data available at time points 1 and 2, and 7 had data available at time points 2 and 3 and the data for those two time points were used.

Renal phenotypes were measured at baseline and the 5-year follow-up. Two thousand three hundred and thirty-six individuals had renal data available at both baseline and 5-year follow-up and were thus included in the longitudinal renal analyses. Renal phenotypes that were evaluated include calculated creatinine clearance and eGFR as well as proteinuria. Age (in years), diabetes status, mean arterial blood pressure (in mmHg) and smoking status were controlled for in the regressions on renal phenotypes.

All ocular-dependent variables are on an ordinal scale, calculated creatinine clearance and eGFR were evaluated as continuous variables. Increases along the AMD scales are associated with disease progression, while decreases in creatinine clearance or eGFR are associated with worse renal function. These are reflected in the models through the sign of the regression coefficient, ß.

The PAR was calculated using a threshold model given that scalar measurements were used. PAR can be calculated as the probability that an exposed individual has a score greater than a given threshold T, minus the probability that an individual has a score greater than T if not exposed to the risk factor of interest, which here is the risk genotype. The PAF can then be calculated as the PAR divided by the probability that an individual has a score greater than T.


Formula 164UM1

This formula assumes a dominant genetic model. The PAF was calculated using the BDES sample, which was a population-based sample, using a threshold of 12 (or greater), which is defined by a person having either neovascular disease or geographic atrophy, clinically considered late AMD end-points.

Given that there appear to be genetic factors in common between AMD and renal phenotypes, which also share potential environmental factors in common, it is important to look at the correlation between the two in order to assess causality. Correlations were performed using the standard correlation coefficient.

Haplotyping
Haplotype likelihoods were reconstructed for both CFH and HMCN1 using DECIPHER, which is part of the S.A.G.E. software package (73). DECIPHER uses an EM algorithm to calculate likelihoods of possible haplotypes for each individual. Most likely haplotypes were also inferred via MERLIN (74). Since the haplotype results agreed very well, we chose to use the probability of having a given haplotype from DECIPHER as a covariate in the regression. Haplotyping was done for both the BDES and FARMS samples, for CFH and HMCN1 independently.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at HMG Online.


   ACKNOWLEDGEMENTS
 
This study was supported, in part, by training grant HL07567, from the National Heart, Lung, and Blood Institute; U.S. Public Health Service research grants GM28356, from the National Institute of General Medical Sciences; U10EY06594, EY015286, EY13438, EY10605 and EY015810 from the National Eye Institute; U01DK057292 from the National Institute of Diabetes and Digestive and Kidney Diseases; resource grant RR03655, from the National Center for Research Resources; and a grant from the Retina Research Foundation. The results of this paper were obtained by using the software package S.A.G.E., which is supported by a US Public Health Service Resource Grant (RR03655) from the National Center for Research Resources. This research was supported by the Gene Expression and Genotyping Facility of the Comprehensive Cancer Center of Case Western Reserve University and University Hospitals of Cleveland (P30CA43703).

Conflict of Interest statement. No author has any conflict of interest with any results presented in this manuscript.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Edwards A.O., Ritter R., Abel K.J., Manning A., Panhuysen C., Farrer L.A. Complement factor H polymorphism and age-related macular degeneration. Science (2005) 308:421–424.[Abstract/Free Full Text]

  2. Haines J.L., Hauser M.A., Schmidt S., Scott W.K., Olson L.M., Gallins P., Spencer K.L., Kwan S.Y., Noureddine M., Gilbert J.R., et al. Complement factor H variant increases the risk of age-related macular degeneration. Science (2005) 308:419–421.[Abstract/Free Full Text]

  3. Klein R.J., Zeiss C., Chew E.Y., Tsai J.Y., Sackler R.S., Haynes C., Henning A.K., SanGiovanni J.P., Mane S.M., Mayne S.T., et al. Complement factor H polymorphism in age-related macular degeneration. Science (2005) 308:385–389.[Abstract/Free Full Text]

  4. Zareparsi S., Branham K.E., Li M., Shah S., Klein R.J., Ott J., Hoh J., Abecasis G.R., Swaroop A. Strong association of Y402H variation in complement factor h at 1q32 to susceptibility for age-related macular degeneration. Am. J. Hum. Genet. (2005) 77:149–153.[CrossRef][Web of Science][Medline]

  5. Sepp T., Khan J.C., Thurlby D.A., Shahid H., Clayton D.G., Moore A.T., Bird A.C., Yates J.R. Complement factor H variant Y402H is a major risk determinant for geographic atrophy and choroidal neovascularization in smokers and nonsmokers. Invest. Ophthalmol. Vis. Sci (2006) 47:536–540.[Abstract/Free Full Text]

  6. Souied E.H., Leveziel N., Richard F., Dragon-Durey M.A., Coscas G., Soubrane G., Benlian P., Fremeaux-Bacchi V. Y402H complement factor H polymorphism associated with exudative age-related macular degeneration in the French population. Mol. Vis. (2005) 11:1135–1140.[Web of Science][Medline]

  7. Weeks D.E., Conley Y.P., Tsai H.J., Mah T.S., Rosenfeld P.J., Paul T.O., Eller A.W., Morse L.S., Dailey J.P., Ferrell R.E., Gorin M.B. Age-related maculopathy: an expanded genome-wide scan with evidence of susceptibility loci within the 1q31 and 17q25 regions. Am. J. Opthalmol. (2001) 132:682–692.[CrossRef][Web of Science][Medline]

  8. Majewski J., Schultz D.W., Weleber R.G., Schain M.B., Edwards A.O., Matise T.C., Acott T.S., Ott J., Klein M.L. Age-related macular degeration—a genome scan in extended families. Am. J. Hum. Genet. (2003) 73:540–550.[CrossRef][Web of Science][Medline]

  9. Schick J.H., Iyengar S.K., Klein B.E., Klein R., Reading K., Liptak R., Millard C., Lee K.E., Tomany S.C., Moore E.L., et al. A whole-genome screen of a quantitative trait of age-related maculopathy in sibships from the Beaver Dam Eye Study. Am. J. Hum. Genet. (2003) 72:1412–1424.[CrossRef][Web of Science][Medline]

  10. Seddon J.M., Santagelo S.L., Book K., Chong S., Cote J. A genomewide scan for age-related macular degeneration provides evidence for linkage to several chromosomal regions. Am. J. Hum. Genet. (2003) 73:780–790.[CrossRef][Web of Science][Medline]

  11. Abecasis G.R., Yashar B.M., Zhao Y., Ghiasvand N.M., Zareparsi S., Branham K.E., Reddick A.C., Trager E.H., Yoshida S., Bahling J., et al. Age-related macular degeneration: a high-resolution genome scan for susceptibility loci in a population enriched for late-stage disease. Am. J. Hum. Genet. (2004) 74:482–494.[CrossRef][Web of Science][Medline]

  12. Iyengar S.K., Song D., Klein B.E., Klein R., Schick J.H., Humphrey J., Millard C., Liptak R., Russo K., Jun G., et al. Dissection of genomewide-scan data in extended families reveals a major locus and oligogenic susceptibility for age-related macular degeneration. Am. J. Hum. Genet. (2004) 74:20–39.[CrossRef][Web of Science][Medline]

  13. Fisher S.A., Abecasis G.R., Yashar B.M., Zareparsi S., Swaroop A., Iyengar S.K., Klein B.E., Klein R., Lee K.E., Majewski J., et al. Meta-analysis of genome scans of age-related macular degeneration. Hum. Mol. Genet (2005) 14:2257–2264.[Abstract/Free Full Text]

  14. Hageman G.S., Anderson D.H., Johnson L.V., Hancox L.S., Taiber A.J., Hardisty L.I., Hageman J.L., Stockman H.A., Borchardt J.D., Gehrs K.M., et al. A common haplotype in the complement regulatory gene factor H (HF1/CFH) predisposes individuals to age-related macular degeneration. Proc. Natl Acad. Sci. USA (2005) 102:7227–7232.[Abstract/Free Full Text]

  15. Li M., Atmaca-Sonmez P., Othman M., Branham K.E., Khanna R., Wade M.S., Li Y., Liang L., Zareparsi S., Swaroop A., et al. CFH haplotypes without the Y402H coding variant show strong association with susceptibility to age-related macular degeneration. Nat. Genet. (2006) 38:1049–1054.[CrossRef][Web of Science][Medline]

  16. Schaumberg D.A., Christen W.G., Kozlowski P., Miller D.T., Ridker P.M., Zee R.Y. A prospective assessment of the Y402H variant in complement factor H, genetic variants in C-reactive protein, and risk of age-related macular degeneration. Invest. Ophthalmol. Vis. Sci. (2006) 47:2336–2340.[Abstract/Free Full Text]

  17. Schultz D.W., Klein M.L., Humpert A.J., Luzier C.W., Persun V., Schain M., Mahan A., Runckel C., Cassera M., Vittal V., et al. Analysis of the ARMD1 locus: evidence that a mutation in HEMICENTIN-1 is associated with age-related macular degeneration in a large family. Hum. Mol. Genet. (2003) 12:3315–3323.[Abstract/Free Full Text]

  18. bGAP National Eye Institute (NEI). Age-Related Eye Disease Study (AREDS). (2007) http://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?id=phs000001.

  19. Zareparsi S., Buracynska M., Branham K.E., Shah S., Eng D., Li M., Pawar H., Yashar B.M., Moroi S.E., Lichter P.R., Petty H.R., Richards J.E., et al. Toll-like receptor 4 variant D299G associated with susceptibility to age-related macular degeneration. Hum. Mol. Genet. (2005) 14:1449–1455.[Abstract/Free Full Text]

  20. Gold B., Merriam J.E., Zernant J., Hancox L.S., Taiber A.J., Gehrs K., Cramer K., Neel J., Bergeron J., Barile G.R., Smith R.T., AMD Genetics Clinical Study Group. Variation in factor B (BF) and complement component 2 (C2) genes is associated with age-related macular degeneration. Nat. Genet. (2006) 38:458–462.[CrossRef][Web of Science][Medline]

  21. Rivera A., Fisher S.A., Fritsche L.G., Keilhauer C.N., Lichtner P., Meitinger T., Weber B.H. Hypothetical LOC387715 is a second major susceptibility gene for age-related macular degeneration, contributing independently of complement factor H to disease risk. Hum. Mol. Genet. (2005) 14:3227–3236.[Abstract/Free Full Text]

  22. Jakobsdottir J., Conley Y.P., Weeks D.E., Mah T.S., Ferrell R.E., Gorin M.B. Susceptibility genes for age-related maculopathy on chromosome 10q26. Am. J. Hum. Genet. (2005) 77:389–407.[CrossRef][Web of Science][Medline]

  23. Haines J.L., Schnetz-Boutaud N., Schmidt S., Scott W.K., Agarwal A., Postel E.A., Olson L., Kenealy S.J., Hauser M., Gilbert J.R., Pericak-Vance M.A. Functional candidate genes in age-related macular degeneration: significant association with VEGF, VLDLR, and LRP6. Invest. Ophthalmol. Vis. Sci. (2006) 47:329–335.[Abstract/Free Full Text]

  24. Yang Z., Camp N.J., Sun H., Tong Z., Gibbs D., Cameron D.J., Chen H., Zhao Y., Pearson E., Li X., et al. A variant of the HTRA1 gene increases susceptibility to age-related macular degeneration. Science (2006) 314:992–993.[Abstract/Free Full Text]

  25. Dewan A., Liu M., Hartman S., Zhang S.S., Liu D.T., Zhao C., Tam P.O., Chan W.M., Lam D.S., Snyder M., et al. HTRA1 promoter polymorphism in wet age-related macular degeneration. Science (2006) 314:989–992.[Abstract/Free Full Text]

  26. Dragon-Durey M.A., Fremeaux-Bacchi V., Loirat C., Blouin J., Niaudet P., Deschenes G., Coppo P., Herman Fridman W., Weiss L. Heterozygous and homozygous factor H deficiencies associated with hemolytic uremic syndrome or membranoproliferative glomerulonephritis: report and genetic analysis of 16 cases. J. Am. Soc. Nephrol. (2004) 15:787–795.[Abstract/Free Full Text]

  27. Abrera-Abeleda M.A., Nichimura C.A., Smith J.L.H., Sethi S., McRae J.L., Murphy B.F., Silvestri G., Skerka C., Jozsi M., Zipfel P.F., et al. Variations in the complement regulatory genes factor H (CFH) and factor H related 5 (CFHR5) are associated with membranoproliferative glomerulonephritis type II (dense deposit disease). J. Med. Genet. (2005) 42:582–589.[Free Full Text]

  28. Ying L., Katz Y., Schlesinger M., Carmi R., Shalev H., Haider N., Beck G., Sheffield V.C., Landau D. Complement factor H gene mutation associated with autosomal recessive atypical hemolytic uremic syndrome. Am. J. Hum. Genet. (1999) 65:1538–1546.[CrossRef][Web of Science][Medline]

  29. Buddles M.R., Donne R.L., Richards A., Goodship J., Goodship T.H. Complement factor H gene mutation associated with autosomal recessive atypical hemolytic uremic syndrome. Am. J. Hum. Genet. (2000) 66:1721–1722.[CrossRef][Web of Science][Medline]

  30. Perez-Caballero D., Gonzalez-Rubio C., Gallardo M.E., Vera M., Lopez-Trascasa M., Rodriguez de Cordoba S., Sanchez-Corral P. Clustering of missense mutations in the C-terminal region of factor H in atypical hemolytic uremic syndrome. Am. J. Hum. Genet. (2001) 68:478–484.[CrossRef][Web of Science][Medline]

  31. Alexander J.J., Pickering M.C., Hass M., Osawe I., Quigg R.J. Complement factor H limits immune complex deposition and prevents inflammation and scarring in glomeruli of mice with chronic serum sickness. Am. Soc. Nephrol. (2005) 16:52–57.[Abstract/Free Full Text]

  32. Alexander J.J., Wang Y., Chang A., Jacob A., Minto A.W., Karmegam M., Haas M., Quigg R.J. Mouse podocyte complement factor H: the functional analog to human complement receptor 1. J. Am. Soc. Nephrol. (2007) 18:1157–1166.[Abstract/Free Full Text]

  33. Saunders R.E., Goodship T.H.J., Zipfel P.F., Perkins S.J. An interactive web database of factor H-associated hemolytic uremic syndrome mutations: insights into the structural consequences of disease-associated mutations. Hum. Mutat. (2006) 27:21–29.[CrossRef][Web of Science][Medline]

  34. Tsukaguchi H., Yager H., Dawborn J., Jost L., Cohlmia J., Abreu P.F., Pereira A.B., Pollak M.R. A locus for adolescent and adult onset familial focal segmental glomerulosclerosis on chromosome 1q25-31. J. Am. Soc. Nephrol. (2000) 11:1674–1680.[Abstract/Free Full Text]

  35. Vollmer M., Jung M., Ruschendorf F., Wienker T., Reis A., Krapf R., Hildebrandt F. The gene for human fibronectin glomerulopathy maps to 1q32, in the region of the regulation of complement activation gene cluster. Am. J. Hum. Genet. (1998) 32:1724–1731.

  36. Cohn D.H., Shohat T., Yahav M., Ilan T., Rechavi G., King L., Shohat M. A locus for an autosomal dominant form of progressive renal failure and hypertension at chromosome 1q21. Am. J. Hum. Genet. (2000) 67:647–651.[CrossRef][Web of Science][Medline]

  37. Caprioli J., Castelletti F., Bucchioni S., Bettinaglio P., Bresin E., Pianetti G., Gamba S., Brioschi S., Daina E., Remuzzi G., Noris M. International registry of recurrent and familial HUS/TTP. Complement factor H mutations and gene polymorphisms in haemolytic uraemic syndrome: the C-257T, the A2089G and the G2881T polymorphisms are strongly associated with the disease. Hum. Mol. Genet. (2003) 12:3385–3395.[Abstract/Free Full Text]

  38. Neumann H.P., Salzmann M., Bohnert-Iwan B., Mannuelian T., Skerka C., Lenk D., Bender B.U., Cybulla M., Riegler P., Konigsrainer A., Neyer U., Bock A., et al. Haemolytic uraemic syndrome and mutations of the factor H gene: a registry-based study of German speaking countries. J. Med. Genet. (2003) 40:676–681.[Abstract/Free Full Text]

  39. Neary J.J., Conlon P.J., Croke D., Dorman A., Keogan M., Zhang F.Y., Vance J.M., Pericak-Vance M.A., Scott W.K., Winn M.P. Linkage of a gene causing familial membranoproliferative glomerulonephritis type III to chromosome 1. J. Am. Soc. Nephrol. (2002) 13:2052–2057.[Abstract/Free Full Text]

  40. Fox C.S., Yang Q., Cupples L.A., Guo C-Y., Larson M.G., Leip E.P., Wilson P.W.F., Levy D. Genomewide linkage analysis to serum creatinine, eGFR, and creatinine clearance in a community-based population: The Framingham Heart Study. J. Am. Soc. Nephrol. (2004) 15:2457–2461.[Abstract/Free Full Text]

  41. Hunt S.C., Coon H., Hasstedt S.J., Cawthon R.M., Camp N.J., Wu L.L., Hopkins P.N. Linkage of serum creatinine and glomerular filtration rate to chromosome 2 in Utah pedigrees. Am. J. Hypertens. (2004) 17:511–515.[CrossRef][Web of Science][Medline]

  42. Nozaki M., Raisler B.J., Sakuai E., Sarma J.V., Barnum S.R., Lambris J.D., Chen Y., Zhang K., Ambati B.K., Baffi J.Z., Ambati J. Drusen complement components C3a and C5a promote chorodial neovascularization. Proc. Natl Acad. Sci. USA (2006) 103:2328–2333.[Abstract/Free Full Text]

  43. Zipfel P.F., Heinen S., Józsi M., Skerka C. Complement and diseases: defective alternative pathway control results in kidney and eye diseases. Mol. Immunol. (2005) 43:97–106.[CrossRef][Web of Science]

  44. Magnusson K.P., Duan S., Sigurdsson H., Petursson H., Yang Z., Zhao Y., Bernstein P.S., Ge J., Jonasson F., Stefansson E., et al. CFH Y402H confers similar risk of soft drusen and both forms of advanced AMD. PLoS Med. (2006) 3:e5.[CrossRef][Medline]

  45. Conley Y.P., Jakobsdottir J., Mah T., Weeks D.E., Klein R., Kuller L., Ferrell R.E., Gorin M.B. CFH, ELOVL4, PLEKHA1 and LOC387715 genes and susceptibility to age-related maculopathy: AREDS and CHS cohorts and meta-analyses. Hum. Mol. Genet. (2006) 15:3206–3218.[Abstract/Free Full Text]

  46. Thompson C., Jun G., Klein B.E.K., Klein R., Capriotti J., Lee K.E., Iyengar S.K. Genetics of pigment changes and geographic atrophy. Invest. Ophthalmol. Vis. Sci. (2007) 48:3005–3013.[Abstract/Free Full Text]

  47. Pickering M.C., Cook H.T., Warren J., Bygrave A.E., Moss J., Walport M.J., Botto M. Uncontrolled C3 activation causes membranoproliferative glomerulonephritis in mice deficient in complement factor H. Nat. Genet. (2002) 31:424–428.[Web of Science][Medline]

  48. Bielsky I.F., Hu S.B., Young L.J. Sexual dimorphism in the vasopressin system: lack of an altered behavioral phenotype in female V1a receptor knockout mice. Behav. Brain Res. (2005) 164:132–136.[CrossRef][Web of Science][Medline]

  49. Bontekoe C.J., McIlwain K.L., Nieuwenhuizen I.M., Yuva-Paylor L.A., Nellis A., Willemsen R., Fang Z., Kirkpatrick L., Bakker C.E., McAninch R., Cheng N.C., Merriweather M., et al. Knockout mouse model for Fxr2: a model for mental retardation. Hum. Mol. Genet. (2002) 11:487–498.[Abstract/Free Full Text]

  50. Fukamauchi F., Wang Y.J., Mataga N., Kusakabe M. Paradoxical behavioral response to apomorphine in tenascin-gene knockout mouse. Eur. J. Pharmacol. (1997) 338:7–10.[CrossRef][Web of Science][Medline]

  51. Shi W., Wang X., Shih D.M., Lauback V.E., Navab M., Lusis A.J. Paradoxical reduction of fatty streak formation in mice lacking endothelial nitric oxide synthase. Circulation (2002) 105:2078–2082.[Abstract/Free Full Text]

  52. Ennis S., Goverdhan S., Cree A., Hoh J., Collins A., Lotery A. Fine scale linkage disequilibrium mapping of age related macular degeneration in the complement factor H gene region. Br. J. Ophthalmol. (2007) 91:966–970.[Abstract/Free Full Text]

  53. Frantz S., Clemitson J.R., Bihoreau M.T., Gauguier D., Samain N.J. Genetic dissection of region around the Sa gene on rat chromosome 1: evidence for multiple loci affecting blood pressure. Hypertension (2001) 38:216–221.[Abstract/Free Full Text]

  54. Podolin P.L., Lord D.P., Lord C.J., Hill N.J., Todd J.A., Peterson L.B., Wicker L.S., Lyons P.A. Congenic mapping of the insulin-dependent diabetes (Idd) gene, Idd10, localizes two genes mediating the Idd10 effect and eliminates the candidate Fcgr1. J. Immunol. (1997) 159:1835–1843.[Abstract]

  55. Schmidt S., Hauser M.A., Scott W.A., Postel E.A., Agarwal A., Gallins P., Wong F., Chen Y.S. Cigarette smoking strongly modifies the association of LOC387715 and age-related macular degeneration. Am. J. Hum. Genet. (2006) 78:852–864.[CrossRef][Web of Science][Medline]

  56. Levey A.S., Coresh J., Balk E., Kausz A.T., Levin A., Steffes M.W., Hogg R.J., Perrone R.D., Lau J., Eknoyan G., et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann. Intern. Med. (2003) 139:137–147.[Abstract/Free Full Text]

  57. Manjunath G., Sarnak M.J., Levey A.S. Prediction equations to estimate glomerular filtration rate: an update. Curr. Opin. Nephrol. Hypertens. (2001) 10:785–792.[CrossRef][Web of Science][Medline]

  58. Saleem M., Florkowski C.M., George P.M., Woltersdoft W.W. Comparison of two prediction equations with radionuclide glomerular filtration rate: validation in routine use. Ann. Clin. Biochem. (2006) 43:209–313.

  59. Appel G.B., Cook H.T., Hageman G., Jennette J.C., Kashgarian M., Kirschfink M., Lambris J.D., Lanning L., Lutz H.U., Meri S., Rose N.R., Salant D.J., et al. Membranoproliferative glomerulonephritis type II (dense deposit disease): an update. J. Am. Soc. Nephrol. (2005) 16:1392–1403.[Abstract/Free Full Text]

  60. Huang S.J., Costa D.L., Gross N.E., Yannuzzi L.A. Peripheral drusen in membranoproliferative glomerulonephritis type II. Retina (2003) 23:429–431.[CrossRef][Web of Science][Medline]

  61. Van Agtmael T., Schlotzer-Schrehardt U., McKie L., Brownstein D.G., Lee A.W., Cross S.H., Sado Y., Mullins J.J., Poschl E., Jackson I.J. Dominant mutations of Col4a1 result in basement membrane defects which lead to anterior segment dysgenesis and glomerulopathy. Hum. Mol. Genet. (2005) 14:3161–3168.[Abstract/Free Full Text]

  62. Jun G., Klein B.E., Klein R., Fox K., Millard C., Capriotti J., Russo K., Lee K.E., Elson R.C., Iyengar S.K. Genome-wide analyses demonstrate novel loci that predispose to drusen formation. Invest. Ophthalmol. Vis. Sci. (2005) 46:3081–3088.[Abstract/Free Full Text]

  63. Klein B.E., Klein R., Lee K.E., Moore E.L. Do multiple families alter estimates of risk for age-related cataract in a population-based study? The Beaver Dam Eye Study. Ophthalmic Epidemiol. (2003) 10:97–106.[CrossRef][Web of Science][Medline]

  64. Klein R., Klein B.D.K., Linton K.L.P. Prevalence of age-related maculopathy: the Beaver Dam Eye Study. Ophthalmology (1992) 99:933–943.[Web of Science][Medline]

  65. Klein R., Klein B.E., Jensen S.C., Meuer S.M. The five-year incidence and progression of age-related maculopathy: the Beaver Dam Eye Study. Ophthalmology (1997) 104:7–21.[Web of Science][Medline]

  66. Klein R., Davis M.D., Magli Y.L., Klein B.E.K. Wisconsin Age-Related Maculopathy Grading System. VA, USA: National Technical Information Service, Springfield. Accession number PB 91–184267/AS.

  67. Klein R., Davis M.D., Magli Y.L., Segal P., Klein B.E., Hubbard L. The Wisconsin age-related maculopathy grading system. Ophthalmology (1991) 98:1128–1134.[Web of Science][Medline]

  68. Cockcroft D.W., Gault M.H. Prediction of creatinine clearance from serum creatinine. Nephron (1976) 16:31–41.[Web of Science][Medline]

  69. Constantiner M., Sehgal A.R., Humbert L., Constantiner D., Arce L., Sedor J.R., Schelling J.R. A dipstick protein and specific gravity algorithm accurately predicts pathological proteinuria. Am. J. Kidney Dis. (2005) 45:833–841.[CrossRef][Web of Science][Medline]

  70. Barrett J.C., Fry B., Maller J., Daly M.J. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics (2005) 21:263–265.[Abstract/Free Full Text]

  71. The International Hapmap Consortium. The international hapmap project. Nature (2003) 426:789–796.[CrossRef][Medline]

  72. Gabriel S.B., Schaffner S.F., Nguyen H., Moore J.M., Roy J., Blumensteil B., Higgins J., DeFelice M., Lochner A., Faggard M., Liu-Cordero S.N., Rotimi C., et al. The structure of haplotype blocks in the human genome. Science. 296:2225–2229.

  73. Statistical analysis for genetic epidemiology. http://darwin.cwru.edu/sage/://darwin.cwru.edu/sage/. (2005) S.A.G.E. 5.0.

  74. Abecasis G.R., Cherny S.S., Cookson W.O., Cardon L.R. Merlin—rapid analysis of dense genetic maps using sparse gene flow trees. Nat. Genet. (2002) 30:97–101.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Proc. Natl. Acad. Sci. USAHome page
H. C. Bravo, K. E. Lee, B. E. K. Klein, R. Klein, S. K. Iyengar, and G. Wahba
Examining the relative influence of familial, genetic, and environmental covariate information in flexible risk models
PNAS, May 19, 2009; 106(20): 8128 - 8133.
[Abstract] [Full Text] [PDF]

Home page
 Arch OphthalmolHome page
R. Klein, M. D. Knudtson, K. E. Lee, and B. E. K. Klein
Serum Cystatin C Level, Kidney Disease Markers, and Incidence of Age-Related Macular Degeneration: The Beaver Dam Eye Study
Arch Ophthalmol, February 1, 2009; 127(2): 193 - 199.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrowOA All Versions of this Article:
16/17/2135    most recent
ddm164v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Google Scholar
Right arrow Articles by Thompson, C. L.
Right arrow Articles by Iyengar, S. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thompson, C. L.
Right arrow Articles by Iyengar, S. K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?