Association and Familial Segregation of CTG18.1 Trinucleotide Repeat Expansion of TCF4 Gene in Fuchs' Endothelial Corneal Dystrophy

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 (r² = 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.

The expanded CTG18.1 allele in TCF4 gene is strongly associated with Fuchs' dystrophy and cosegregates with the trait in a majority of Caucasian families in our cohort with documented cases of incomplete penetrance.

Introduction

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.

Figure 1.

The intronic TCF4 polymorphisms SNP rs613872 and CTG18.1 trinucleotide repeat alleles are approximately 43 kilobases apart. The TCF4 transcript model is courtesy of Ensembl 2013 (ENST00000565018.2).²⁴ Reprinted with permission from Flicek P, Ahmed I, Amode MR, et al. Ensembl 2013. Nucleic Acids Res. 2013;41:D48–D55.

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

Methods

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.

Study Participants

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.

Genotyping

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 MgCl₂, 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.).

Figure 2.

Overview of TP-PCR to genotype CTG18.1. Adapted by permission from BMJ Publishing Group Limited. Warner JP, Barron LH, Goudie D, et al. A general method for the detection of large CAG repeat expansions by fluorescent PCR. J Med Genet. 1996,33:1022–1026.²⁶ (A) The STR assay uses primers P1 and P2 that flank the CTG repeat, but the assay can fail to genotype expanded alleles. (B) The trinucleotide repeat specific 3′ end of P4 binds at numerous sites within the CTG repeat in the early rounds of amplification, resulting in a mixture of products. Primer P1 is a 5′-6-FAM locus-specific primer that determines the specificity of the reaction. In the early rounds of amplification, the P4 primer is consumed quickly due to the 33:1 molar ratio of P3 to P4. (C) P3 primer amplifies from end of the mixture of products of the prior cycles. The PCR parameters include a long extension time to ensure complete extension of the longer products within the PCR product mixture and allow for the identification of the expanded CTG18.1 allele(s).

Table 1.

Primers for STR Analysis and TR-PCR Assay

Primer	Sequence, 5′–3′
P1	AATCCAAACCGCCTTCCAAGT
P2	CAAAACTTCCGAAAGCCATTTCT
P3	TACGCATCCCAGTTTGAGACG
P4	TACGCATCCCAGTTTGAGACGCAGCAGCAGCAGCAG

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.

Statistical Analysis

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 r² 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 MERLIN³⁶ 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.

Figure 3.

Segregation of SNP rs613872 and CTG18.1 alleles in families with Fuchs' dystrophy. In family F10, all four affected family members are homozygous for rs613872 risk allele G and have the expanded CTG18.1 allele, including two individuals being homozygous for the expanded CTG18.1 allele. Note lack of penetrance of the expanded CTG18.1 allele in individual VVM193 (38-year-old male without any central guttae). Two individuals VVM125 and VVM183 are homozygous and heterozygous for rs613872, respectively. Both lack the expanded CTG18.1 allele and have grade 1 guttae. Family F33 shows cosegregation of the trait with rs613872 and the expanded CTG18.1 allele. Family F14 shows cosegregation of the expanded CTG18.1 allele with the FECD trait with complete penetrance. The SNP rs613872 cosegregates with disease but with incomplete penetrance. Subject VVM209 has grade 1 guttae. Family F32 shows cosegregation of the expanded CTG18.1 allele and rs613872 with the FECD trait, but with incomplete penetrance of both risk alleles. Subject VVM122 is a 95-year-old woman homozygous for the expanded CTG18.1 and rs613872 alleles, but only has grade 1 guttae. Subject VVM175 is heterozygous for both risk alleles and has grade 1 guttae. Subject VVM176 lacks either risk allele and has grade 1 guttae. Family F36 is an example of a family where the trait fails to cosegregate with either rs613872 or expanded CTG18.1 allele. Note VVM208 has one copy of expanded CTG18.1 allele and has only grade 1 guttae. Family F39 is an example of a family where the trait cosegregates with the expanded CTG18.1 and rs613872 alleles, but with incomplete penetrance. Interestingly, VVM149 (52-year-old female) has the expanded allele, but has only grade 1 guttae.

Results

Demographics of Study Subjects

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

Table 2.

Demographic Information, and TCF4 CTG18.1 and rs613872 Genotyping Results of Caucasian FECD Cases and Controls

Characteristic	Cases, n = 120	Controls, n = 100	P Value*
Men/women	41/79	40/60	4.0 × 10−1
Age ± SD, y	70.6 ± 10.7	67.3 ± 11.6	3.1 × 10−2
rs613872
GG	17	3	3.1 × 10−17
GT	78	18
TT	25	79
CTG18.1†
XX	7	1	6.5 × 10−25
SX	81	6
SS	32	93
Haplotype
G-X	0.382	0.040	5.9 × 10−19
T-X	0.014	0.000
G-S	0.085	0.080
T-S	0.519	0.880

Counts of measured genotype at each locus and estimated haplotype frequencies between the two loci.

^* Potential risk factors age and sex were adjusted for in genetic association tests.

^† Alleles were dichotomized with X denoting CTG_≥40 and S denoting CTG_<40.

STR Assay and TR-PCR Assay Results

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, CTG₁₃,CTG₁₈, and one homozygote, CTG₁₂,CTG₁₂), one FECD subject with one stable CTG18.1 allele of 12 CTG repeats and another expanded CTG18.1 allele (CTG₁₂,X), and one FECD subject homozygous for expanded CTG18.1 allele (X,X) are shown in Figure 4.

Figure 4.

Representative STR analysis and TP-PCR electropherogram tracings. The STR analysis results are shown above the corresponding TP-PCR tracings. The CTG allele sizes are shown with arrows. The TP-PCR tracings have a characteristic ladder with a 3 base pair periodicity. (A) The STR and TP-PCR tracings of unaffected subject heterozygous for two stable CTG18.1 alleles of less than 40 CTG repeats (CTG₁₃, CTG₁₈). On the TP-PCR tracings, the CTG₁₃ and CTG₁₈ alleles give peaks as well as all the intermediate priming sites. Note the slight continuation of the pattern past the CTG₁₈ allele peak due to mispriming at the repeat end. (B) The STR and TP-PCR tracings of unaffected subject homozygous for stable CTG18.1 allele (CTG₁₂, CTG₁₂). (C) The STR and TP-PCR tracings of FECD subject with one stable CTG18.1 allele of 12 CTG repeats and another expanded CTG18.1 allele (CTG₁₂, X). Note STR results revealed only the CTG₁₂ allele and failed to detect the expanded allele. Persistence of the ladder pattern on RP-PCR tracing indicates presence of an expanded allele. (D) The TP-PCR tracing of FECD subject homozygous for expanded CTG18.1 allele (X, X). The STR analysis failed to detect any CTG18.1 allele. This characteristic ladder pattern on the TP-PCR tracing indicates the presence of two expanded CTG18.1 alleles. Note the continuation of the ladder pattern on the TP-PCR tracing.

Association Analysis of TCF4 Polymorphisms

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 r² 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).

Segregation of Expanded CTG18.1 and rs613872 Alleles in FECD Families

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).

Table 3.

Segregation and Penetrance of TCF4 Expanded CTG18.1 and rs613872 Alleles in 29 FECD Kindreds

Family ID	Size*	Presence of CTG≥40 Allele†	Co-Segregation of CTG≥40	Penetrance of CTG≥40‡	Co-Segregation of rs613872	Penetrance of rs613872§	Comments
F10	Multigeneration affected members, n = 4 (8 members examined)	+	Yes	Complete (4/4)	Yes	Incomplete (4/8)	2 individuals with grade 1 guttae with rs613872 and without CTG≥40; 38 year-old male with CTG≥40 without guttae
F14	Multigeneration, n = 4 (8)	+	Yes	Complete (4/4)	Yes	Incomplete (4/7)	1 individual with grade 1 guttae with rs613872 and without CTG≥40
F24	Multigeneration, n = 2 (3)	+	No	N/A	Yes	Incomplete (2/3)
F21	Sib pair (3)	+	Yes	Complete (2/2)	Yes	Incomplete (2/3)
F29	Multigeneration, n = 22 (34)	+	No	N/A	No	N/A
F32	Multigeneration, n = 4 (7)	+	Yes	Incomplete (5/6)	Yes	Incomplete (5/6)	1 individual with grade 1 guttae without rs613872 or CTG≥40; 1 individual with CTG≥40 with grade 1 guttae
F33	Multigeneration, n = 3 (3)	+	Yes	Complete (3/3)	Yes	Complete (3/3)
F34	Multigeneration, n = 2 (2)	+	Yes	Complete (2/2)	Yes	Complete (2/2)
F36	Multigeneration, n = 2 (5)	+	No	N/A	No	N/A	1 individual with grade 1 guttae with CTG≥40; 1 individual with grade 1 guttae without CTG≥40 or rs613872
F38	Sib pair (5)	−	No	N/A	No	N/A
F39	Sib pair (4)	+	Yes	Incomplete (2/3)	Yes	Incomplete (2/4)	1 individual with grade 1 guttae with CTG≥40 and rs613872
F45	Sib pair (6)	+	No	N/A	No	N/A
F46	Multigeneration, n = 2 (4)	−	No	N/A	Yes	Complete (2/2)
F49	Sib pair (2)	−	No	N/A	No	N/A
F51	Multigeneration, n = 2 (4)	−	No	N/A	No	N/A
F56	Multigeneration, n = 3 (5)	+	Yes	Complete (3/3)	Yes	Complete (3/3)
F60	Multigeneration, n = 3 (4)	+	Yes	Complete (3/3)	Yes	Complete (3/3)
F71	Multigeneration, n = 2 (3)	+	Yes	Complete (2/2)	Yes	Incomplete (2/3)	1 individual without any guttae with rs613872
F72	Multigeneration, n = 2 (3)	+	Yes	Complete (2/2)	Yes	Complete (2/2)
F82	Multigeneration, n = 2	−	No	N/A	No	N/A
F89	Multigeneration, n = 3 (4)	+	No	N/A	No	N/A	2 individuals with CTG ≥40 with FECD
F112	Single generation, n = 3 (3)	+	Yes	Complete (3/3)	Yes	Complete (3/3)
F113	Multigeneration, n = 3 (3)	+	No	N/A	No	N/A	33-y/o female with FECD without CTG≥40 or rs613872
F116	Sib pair (2)	+	Yes	Complete (2/2)	Yes	Complete (2/2)
F121	Sib pair (2)	+	Yes	Complete (2/2)	Yes	Complete (2/2)
F138	Multigeneration, n = 2 (3)	+	Yes	Incomplete (2/3)	Yes	Incomplete (2/3)	1 individual with grade 1 guttae with CTG≥40 and rs613872
F161	Multigeneration, n = 2 (3)	+	Yes	Complete (2/2)	Yes	Incomplete (2/3)	1 individual without any guttae and homozygous for rs613872
F169	Sib pair (2)	+	Yes	Complete	Yes	Complete (2/2)
F173	Multigeneration, n = 2 (2)	+	Yes	Complete (2/2)	Yes	Complete (2/2)

N/A, not applicable.

^* n, number of family members affected with Fuchs' dystrophy defined as guttae grade ≥ 2 in more severely affected eye; number in adjacent parentheses includes total number of individuals examined, including proband and biological relatives over the age of 40.

^† Presence of CTG_≥40 allele in at least one examined family member indicated by plus sign; minus sign indicates expansion-negative family.

^‡ Penetrance determined by slit-lamp examination findings of individuals ≥ 40 years of age. Numbers in parentheses include the number of affected subjects in the numerator and the number of carriers of at least one copy of the expanded CTG18.1 allele in the denominator.

^§ Parentheses include the number of affected subjects in the numerator and the number of carriers of at least one copy of rs613872 G risk allele in the denominator.

Figure 5.

Linkage analyses on family F10 reveals a signal on chromosome 18q21. A dominant genetic model with penetrance of 0.95, sporadic rate of 0.001, and disease-predisposing allele frequency of 0.001 were assumed.

Discussion

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.