Abstract
Purpose.:
We tested the association between two intronic polymorphisms (CTG18.1 and rs613872) in TCF4 and Fuchs' endothelial corneal dystrophy (FECD), and analyzed their segregation patterns in families.
Methods.:
We recruited 120 unrelated Caucasian subjects with FECD and 100 controls. Available family members of probands were recruited. Genotyping of the single nucleotide polymorphism (SNP) rs613872 was performed using Sanger sequencing or real-time allelic discrimination assay. The trinucleotide repeat polymorphism, CTG18.1, was genotyped using a combination of short tandem repeat assay and triplet repeat primed PCR assay. The cytosine-thymine-guanine (CTG) repeat length of ≥40 was classified as an expanded CTG18.1 allele. Association of the two loci with FECD was evaluated. Segregation in 29 families was examined.
Results.:
The two polymorphisms are in linkage disequilibrium (r 2 = 0.65 in cases and 0.31 in controls). Significant associations were found between FECD and rs613872 (P = 3.1 × 10−17), expanded CTG18.1 allele (P = 6.5 × 10−25), and their haplotypes (P = 5.9 × 10−19 ). The odds ratio (OR) of each copy of the rs613872 G allele for FECD was estimated to be 9.5 (95% confidence interval [CI], 5.1–17.5). The OR of each copy of the CTG18.1 expanded allele was estimated to be 32.3 (95% CI, 13.4–77.6). The expanded CTG 18.1 allele cosegregated with the trait in 52% (15/29) of families with complete penetrance and 10% (3/29) with incomplete penetrance.
Conclusions.:
We report, to our knowledge, the first independent replication of the expanded CTG 18.1 allele conferring significant risk for FECD (>30-fold increase). The expanded allele cosegregates with the trait with complete penetrance in a majority of families, but we also document cases of incomplete penetrance.
Fuchs' endothelial corneal dystrophy (FECD) is a bilateral, progressive disorder of the endothelium that can result in corneal edema and loss of vision. The condition of FECD is the leading indication for corneal transplantation in the United States and is estimated to affect 4% of the population over the age of 40 in this country.
1,2 The basement membrane of the endothelium, Descemet's membrane, becomes markedly thickened with focal collagenous excrescences, called guttae, that are apparent on slit-lamp biomicroscopy.
3 Premature senescence of the endothelium is a hallmark of the disorder, marked by loss of their normal hexagonal pattern, increased variation in cell size, and decreased cell density.
4,5 Canonical molecular pathways of apoptosis, oxidative stress, and unfolded protein response have been implicated in the pathophysiology of the disorder.
6–11
In 2001, studies on a three-generation family in the United Kingdom with early-onset FECD led to the discovery that heterozygous missense mutations in
COL8A2 (Q455K)
, which encodes for the α2 chain of type VIII collagen found in basement membrane of endothelium, can cause this disorder, but do not appear to underlie the more common, late-onset form of FECD.
12–15 Mutations in a few other genes, including
SLC4A11,
TCF8, and
LOXHD1, have been implicated in a small number of FECD cases, but further studies are required to determine their roles in the disorder.
16–20
In 2006, Sundin et al.
21 reported a common locus for late-onset FECD on 18q21.2-q21.31 (FCD2) in a linkage study with three large families with the disorder. In 2010, Baratz et al.
22 published their genome-wide association study (GWAS) findings of a significant association between single nucleotide polymorphisms (SNPs) spanning a 1 Mb region of chromosome 18q (including
TCF4 and
FLJ45743) and late-onset FECD, most notably the intronic SNP rs613872 in
TCF4. Haplotype analysis of SNPs in this region indicated that more than one variant may contribute independently to the risk of the disease.
Recently, Wieben et al.
23 reported a strong association of a cytosine-thymine-guanine (CTG) repeat expansion in intron 3 of
TCF4 and FECD with greater specificity than rs613872 in identifying the disorder. These two loci are approximately 43 kilobases apart (
Fig. 1).
24 In their report, 52 of 66 subjects with FECD had the expanded CTG alleles compared to 2 of 63 controls.
23 This intronic trinucleotide repeat locus originally was identified and named as CTG18.1 by Breschel et al.
25 This earlier group had found alleles of >37 CTG repeats at this locus to be unstable and present in 3% of their Caucasian population without any known associated phenotype.
Here, we conducted an independent validation study to reassess
TCF4 polymorphisms CTG18.1 and rs613872, and FECD in 120 unrelated Caucasian FECD patients, and confirm the significant association. One copy of the expanded CTG18.1 allele confers a >32-fold increase for disease. Additionally, segregation of the expanded CTG18.1 allele in a majority of the pedigrees in our cohort suggests that
TCF4 CTG18.1 is an important locus for the common, late-onset form of FECD. Finally, we demonstrated the feasibility of efficiently genotyping the CTG18.1 trinucleotide repeat allele using a combination of short tandem repeat (STR) analysis and a triplet repeat primed polymerase chain reaction assay (TP-PCR) originally described by Warner et al.
26
The study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center and was performed in compliance with the tenets of the Declaration of Helsinki. Study participants were enrolled after written informed consent.
We recruited subjects from tertiary care cornea clinics who underwent an ophthalmic examination with slit-lamp biomicroscopy. The FECD severity was graded by a cornea fellowship–trained ophthalmologist (VVM) using a modified Krachmer scale (0, no central guttae; 1, up to 12 central guttae; 2, ≥12 nonconfluent central guttae; 3, 1–2 mm confluent central guttae; 4, 2–5 mm confluent central guttae; 5, 5 mm confluent central guttae without edema; 6, 5 mm confluent central guttae with edema).
27 Subjects selected as cases had a grade of 2 or greater by slit-lamp examination or by histopathologic diagnosis after corneal transplantation. Control subjects recruited from our clinics had no central guttae (grade 0) by slit-lamp examination. A detailed medical and family history was obtained from each study subject. Self-reported ethnicity information was collected. Family members were recruited and examined when possible.
Genomic DNA was extracted from leukocytes of peripheral blood using AutoGen FlexiGene by Qiagen (Valencia, CA).
We genotyped SNP rs613872 using Sanger sequencing or real-time allelic discrimination assay (TaqMan SNP Genoptyping Assay; Life Technologies, Carlsbad, CA).
The trinucleotide repeat CTG18.1 polymorphism was genotyped using a combination of an STR assay and TP-PCR assay. The STR assay was performed using genomic DNA on all study subjects. On samples where STR assay detected only one allele or failed to detect any alleles, repeat primed PCR assay was performed to confirm the presence of an expanded allele(s).
For STR assay, a 5′ FAM primer was used for the PCR. After PCR, 5 μL of DNA were mixed with 10 μL of Promega Internal Standard 600 (ILS600; Promega, Madison, WI). Sequencing was done using an ABI 3730XL DNA Analyzer (Applied Biosystems, Inc., Foster City, CA) and data analysis was performed using the ABI GeneMapper 4.0 (Applied Biosystems, Inc.).
The TP-PCR based on the technique originally described by Warner et al.
26 (
Fig. 2) was used to detect expanded CTG18.1 allele(s). The P1 is a fluorescent primer designed to a region upstream from the CTG18.1 allele. The companion reverse primer P4 on the complementary strand is comprised of 5 units of the CTG repeat and a 5′ tail to serve as an anchor for a second reverse primer P3, which prevents progressive shortening of the PCR products during subsequent cycles. The 5′ tail of primer P4 and the “common” flag primer P3 share no homology with human sequence. The PCR was performed with the following conditions: 200 ng of genomic DNA, 1 μmol/L of locus-specific (forward) primer P1, and 0.03 μmol/L of repeat-specific (reverse) primer P4, and 1 μmol/L of primer P3, 200 μmol/L dNTPs, 1.5 mmol/L MgCl
2, and 1U of Taq DNA polymerase by 5 Prime (Gaithersburg, MD). Primer sequences are provided in
Table 1. The cycling parameters were an initial denaturation of 9 minutes at 95°C, followed by 10 cycles of 95°C for 30 seconds, 62°C for 30 seconds, and 72°C for 4 minutes, and then 30 cycles of 95°C for 45 seconds, 62°C for 45 seconds, and 72°C for 4 minutes with a 15-second extension at each cycle. The final extension step was 72°C for 10 minutes. The TP-PCR products were analyzed on the ABI 3730XL DNA Analyzer (Applied Biosystems, Inc.).
The TP-PCR was performed on all the samples in which the STR analysis revealed only one apparent allele or no alleles. Characteristic tracing patterns of the triplet repeat primed electropherograms were used to distinguish samples that truly were homozygous for a stable CTG18.1 allele (<40 CTG repeats) from those that harbored an expanded CTG18.1 allele (≥40 CTG repeats) that was undetectable by STR analysis. The TP-PCR also was used to confirm the presence of two expanded CTG18.1 alleles in samples where STR analysis failed to genotype any allele.
Comparisons of demographic features between cases and controls were performed by a 2-sample
t-test for age and by Fisher's exact test for sex. For the trinucleotide repeat CTG18.1 polymorphism, we dichotomized alleles, such that CTG
≥40 was considered an expanded allele, denoted as X, and CTG
<40 was considered a normal allele, denoted as S. Hardy-Weinberg equilibrium (HWE) was examined in cases and controls, separately, by an exact test.
28 The degree of linkage disequilibrium (LD) between rs613872 and the CTG18.1 polymorphism was estimated by
r 2 and
D'.
29,30 Logistic regression models were fit to test association between genotype and FECD affection status by the likelihood ratio test.
31 Risk factors age and sex were included as covariates.
32–34 The genotypic value was coded in an additive manner; that is, 0, 1, and 2 denoted TT, TG, and GG genotypes, respectively, for SNP rs613872, and SS, SX, XX denote genotypes, respectively, for the CTG18.1 polymorphism. Association between FECD and haplotypes between the two loci also was evaluated using haplo.stats.
35
Segregation of these two
TCF4 polymorphisms in 29 families with FECD was examined. Considering the late-onset nature of FECD, we took a conservative approach, such that individuals with FECD were classified as “affected,” those unaffected after age 40 were classified as “unaffected,” and all others were assigned an unknown affection status. In particular, model-based linkage analysis using MERLIN
36 was performed in a multiplex family with four affected individuals (family F10,
Fig. 3), which were genotyped by the Illumina HumanLinkage-24 array (Illumina, Inc., San Diego, CA). A dominant genetic model with penetrance of 0.95, sporadic rate of 0.001, and disease-predisposing allele frequency of 0.001 was assumed.
We included in the studies 120 unrelated Caucasian patients with FECD and 100 unaffected Caucasian control subjects (
Table 2). The age of each control subject is provided in
Supplementary Table S1. There was 1 case with Krachmer grade 3, 18 cases with Krachmer grade 4, 12 cases with Krachmer grade 5, and 89 cases of Krachmer grade 6. The mean Krachmer guttae grade was 5.575 (± 0.77). The cases were on average older than the controls (70.6 ± 10.7 vs. 67.3 ± 11.6 years,
P = 0.03). There were more females in cases than in controls (66% vs. 60%,
P = 0.40). Considering both are important risk factors for FECD, we included them in the logistic regression models in testing association between FECD and genotype.
37
The STR assay was able to detect and precisely size CTG18.1 alleles of up to 43 repeats. Of the FECD subjects, 86/120 (72%) had only one apparent allele compared to 23/100 (23%) of control subjects. The STR assay failed to detect any CTG18.1 alleles in 6/120 (5%) FECD subjects compared to 1/100 (1%) control subjects.
The TP-PCR assay was able to identify the presence of larger, expanded CTG18.1 alleles undetected by STR assay. The TP-PCR assay also was able to distinguish samples that truly were homozygous for a stable CTG18.1 allele from those samples that had one stable CTG18.1 allele and another expanded CTG18.1 allele that was undetectable by STR assay. All seven samples where the STR assay failed to detect any CTG18.1 allele had characteristic TP-PCR tracings indicative of two expanded CTG18.1 alleles.
The STR and TP-PCR electropherogram tracings of two unaffected subjects with stable CTG18.1 alleles of less than 40 CTG repeats (one heterozygote, CTG
13,CTG
18, and one homozygote, CTG
12,CTG
12), one FECD subject with one stable CTG18.1 allele of 12 CTG repeats and another expanded CTG18.1 allele (CTG
12,X), and one FECD subject homozygous for expanded CTG18.1 allele (X,X) are shown in
Figure 4.
The distribution of rs613872 and the expanded CTG18.1 in cases and controls are summarized in
Table 2. In controls, both alleles were in HWE (
P = 0.14 and 0.13, respectively), which guaranteed the genotyping quality, but in cases both were out of HWE (
P = 1.0 × 10
−3 and 7.6 × 10
−6, respectively), which indicated their association with the disease status.
38 The LD measured by
r 2 between the two loci was estimated to be 0.65 and 0.31 in cases and controls, respectively, and LD measured by
D' was estimated to be 0.93 and 1.00 in cases and controls, respectively.
Both rs613872 and the expanded CTG18.1 alleles were significantly associated with FECD (P = 3.1 × 10−17 and 6.5 × 10−25, respectively); so was their haplotypes (global P = 5.9 × 10−19). The odds ratio (OR) of each copy of the rs613872 G allele for FECD was estimated to be 9.5 (95% confidence interval [CI], 5.1–17.5). The OR of each copy of the expanded CTG18.1 allele, X, was estimated to be 32.3 (95% CI, 13.4–77.6).
Details of the segregation and penetrance of the expanded CTG18.1 and rs613872 alleles in each examined family are summarized in
Table 3. The expanded CTG18.1 allele cosegregated with the FECD phenotype in 52% (15/29) of our FECD kindreds with complete penetrance and 10% (3/29) with incomplete penetrance. The rs613872 allele cosegregated with the FECD phenotype in 38% (11/29) of our FECD kindreds with complete penetrance and 31% (9/29) with incomplete penetrance. Representative pedigrees are shown in
Figure 3. In family F10 (
Fig. 3), all four affected people share a haplotype harboring the expanded CTG18.1 allele, resulting in a maximum logarithm of the odds (LOD) score of 0.90 in the region (
Fig. 5).
We replicated the recently reported association of the expanded
TCF4 CTG18.1 allele with FECD in our Caucasian cohort.
23 The
TCF4 gene encodes for the E2-2 protein and belongs to the family of ubiquitously-expressed class 1 basic helix-loop-helix transcription factors involved in cellular proliferation and differentiation.
46 The
TCF4 is a large gene spanning over 360 kilobases, has 20 exons, and encodes a 667-amino acid protein.
Microdeletions or missense mutations resulting in haploinsufficiency of
TCF4 are causal for autosomal dominant Pitt-Hopkins syndrome, characterized by microcephaly, encephalopathy, epilepsy, psychomotor delay, and episodic hyperventilation.
39,40 Although it would be unusual for the same gene to result in such different disease phenotypes, such as Pitt-Hopkins syndrome and Fuchs' dystrophy, some have hypothesized that variants of
TCF4 may impact upon
TCF8 expression.
22 Mutations in
TCF8 cause another endothelial disorder, posterior polymorphous corneal dystrophy, and also have been found in a few individuals with FECD.
18,19
The 1 Mb region of chromosome 18q (including
TCF4 and
FLJ45743), found to be associated with late-onset FECD in the GWAS study published by Baratz et al.,
22 falls within the linkage interval on 18q21.2-q21.31 on a study of three large families with the trait published by Sundin et al.
21 In the FECD family F10 in our cohort (
Fig. 3), all four affected subjects shared a haplotype harboring the expanded CTG18.1 allele, resulting in a maximum LOD score of 0.90 in the region (
Fig. 5). The expanded CTG18.1 allele or another causal variant in LD with this polymorphism may be responsible for the previously reported GWAS and linkage signals on chromosome 18q21.2-q21.31 for FECD.
41,42
The TP-PCR assay originally was described by Warner et al.
26 as a general method for the detection of CAG trinucleotide repeat expansion in a cohort of subjects with myotonic dystrophy. The assay since has been applied to other pathogenic repeat expansions, as in Friedreich ataxia, spinocerebellar ataxia, and fragile X syndrome.
43–45 The combination of STR and TP-PCR analysis is a simple, efficient method to genotype the CTG18.1 polymorphism.
We were able to detect and precisely size the CTG18.1 allele up to 43 repeats using the STR analysis. We were able to detect the presence of larger, expanded alleles with use of the TP-PCR assay. The TP-PCR can help determine the zygosity of the allele by detecting the presence of an expanded allele by its characteristic tracing on the capillary electropherogram. In theory, the ability of the TP-PCR assay to detect the presence of an expanded allele is not limited by the number of trinucleotide repeats in the DNA template.
44 Although TP-PCR helps detect the presence of enlarged alleles, it does not allow for determination of their length. Southern blot analysis would allow for the quantification of the size of the large repeat expansions. Given the possibility for mosaicism, the detection of two alleles on STR analysis would not necessarily exclude the possibility of undetected large expansions without performing TP-PCR or Southern blot analysis.
Presence of the expanded
TCF4 CTG18.1 allele confers more than a 32-fold increased risk for the development of the disorder. However, in the association arm of our study, we found the expanded CTG18.1 allele in control subjects without evidence of any central guttae (grade 0), including one individual who was homozygous for this expanded risk allele. In the report by Wieben et al.,
23 5% of controls had an expanded CTG18.1 allele (40 CTG repeats or greater) compared to 7% of controls in our study.
Differences in study design may have impacted upon the much higher OR of 9.5 for one copy of the minor rs613872 G allele in our study than that of 5.5 reported by Baratz et al.
22 Inclusion criteria for a FECD subject in our study was Krachmer grade 2 guttae or higher (as opposed to grade 1 guttae in the study of Baratz et al.
22) Additionally, recruitment from our tertiary care surgical corneal practice resulted in a FECD cohort with more advanced disease in our study.
In our familial studies, the expanded CTG18.1 allele segregated with complete penetrance in a larger proportion of our FECD kindreds than did the SNP rs613872 G risk allele. In the three families (F32, F39, and F138) where the expanded CTG18.1 allele cosegregated with the trait with “incomplete penetrance,” all of the “unaffected” individuals with the expanded CTG18.1 risk allele had grade 1 guttae, and so were not completely normal. However, one subject from family F32 without the expanded CTG18.1 or rs613872 risk alleles had grade 1 guttae despite these risk alleles cosegregating with the other affected members of his family. Grade 1 guttae also were found in various other individuals from different FECD families without the expanded CTG18.1 or rs613872 risk alleles.
There are over 32,000 trinucleotide repeats across the genome and 878 trinucleotide repeat-containing genes.
47 Only some repeats are genetically unstable, and may increase in size during DNA replication from generation to generation and impact on gene function to result in disease.
48 Breschel et al.
25 presented some limited Southern blot data suggesting instability of this trinucleotide repeat in the original description of this locus. Additional familial studies are required to document the extent of instability of this locus. With other established trinucleotide repeat disorders, repeat expansion from generation to generation may result in earlier onset of disease with each successive generation (anticipation). However, anticipation is not an accepted clinical hallmark of Fuchs' dystrophy, but requires more formal study.
Our rationale to define an expanded CTG18.1 allele as one with 40 or more CTG repeats was to choose a conservative cut-off value slightly higher than the 37 repeats reported by Breschel et al.
25 as the threshold of instability. Further studies are required to determine if intermediate expansions in the 40 to 50 repeats range are associated with the disorder or contribute to disease. Breschel et al.
25 observed no transmission in their pedigrees of repeat lengths in this range. Additionally, the study by Wieben et al.
23 observed only one affected subject in their intermediate category of 40 to 50 repeats.
Our data replicated the strong association of the
TCF4 expanded CTG18.1 allele with Fuchs' dystrophy in our Caucasian cohort and, to our knowledge, are the first to formally calculate an OR of conferred risk and include familial segregation data. However, the expanded CTG18.1 allele is neither sufficient nor necessary for development of the trait. The expanded CTG18.1 may be a causal mutation with incomplete penetrance, a phenotype modifying or susceptibility locus requiring the interaction of additional gene(s) and/or environmental factors to result in the FECD phenotype, or simply a polymorphism in LD with another causal variant in the region. Further studies of the CTG18.1 polymorphism in other ethnic groups may help determine if this variant is the causal driver of the GWAS signal.
49 Additional studies are required to determine if the expansion interferes with normal function of
TCF4 and causes disease. Our data are compatible with the growing body of evidence that FECD is an oligogenic disorder in which the
TCF4 polymorphism CTG18.1 may have an important role.
The authors thank all of our patients and their families for their participation in this study, as well as our study coordinators, Mike Molai and Emily Linsenbardt, for their efforts to recruit and enroll the subjects, and Imran Hussain, PhD, for his critical review of the manuscript.
Supported by Grants R01EY022161 (VVM) and P30EY020799 from the National Eye Institute, National Institutes of Health, Bethesda, Maryland, and an unrestricted grant from Research to Prevent Blindness, New York. The authors alone are responsible for the content and writing of the paper.
Disclosure: V.V. Mootha, None; X. Gong, None; H.-C. Ku, None; C. Xing, None