TGFBI and CHST6 gene analysis in Chinese stromal corneal dystrophies

Abstract

AIM

To investigate whether mutations in TGFBI gene or CHST6 gene correlated with stromal corneal dystrophies (CD) in 8 Chinese probands.

METHODS

Eight unrelated patients with stromal corneal dystrophies were recruited in this study; all affected members were assessed by completely ophthalmologic examinations. Genomic DNA was extracted from peripheral leukocytes, 17 exons of TGFBI gene and the exon of CHST6 gene were amplified by polymerase chain reaction (PCR), sequenced directly and compared with the reference database.

RESULTS

Three heterozygous mutations in TGFBI gene were identified in six patients: c. 370C>T (p.Arg124Cys) was found in exon 4 of TGFBI gene in three members, c. 371G>A (p.Arg124His) was found in one patient; c. 1663C>T (p.Arg555Trp) was found in exon 12 in other two members. In addition, four polymorphisms with the nucleotide changes rs1442, rs1054124, rs4669, and rs35151677 were found in TGFBI gene. Mutations were not identified in the rest of 2 affected individuals in TGFBI gene or CHST6 gene.

CONCLUSION

Within these patients, R124C, R124H and R555W mutations were co-segregated with the disease phenotypes and were specific mutations for lattice corneal dystrophy type I (LCD I), Avellino corneal dystrophy (ACD, GCD II), granular corneal dystrophy type I (GCD I), respectively. Our study highlights the prevalence of codon 124 and codon 555 mutations in the TGFBI gene among the Chinese stromal corneal dystrophies patients.

INTRODUCTION

Corneal dystrophies (CDs) define a heterogeneous group of inherited, bilateral, primary alterations of the cornea that are not associated with prior inflammation or systemic diseases. Based on the affected anatomical layer, corneal dystrophy can be classified as epithelial and subepithelial or Bowman layer and the superficial corneal stroma (anterior corneal dystrophies), stroma layer (stromal corneal dystrophies), Descemet membrane and the endothelium (posterior corneal dystrophies). TGFBI (BIGH3, OMIM: 601692) and CHST6 (OMIM: 605294) are more likely two candidate virulence genes for stromal corneal dystrophies^[1].

Human TGFBI gene locates in chromosome 5q31^[2] and encodes a protein named TGFBIp or keratoepithelin (KE), which is constitutively expressed in numerous tissues of human. However, mutations in the TGFBI gene so far reported are correlating with corneal epithelial and stromal diseases exclusively^[3].

Furthermore, point mutations of TGFBI bring about five phenotype-specific corneal dystrophies: p.R555W in granular CD type I (GCD I,OMIM: 121900), p. R124C in Lattice CD type I (LCD I, OMIM: 122200), p. R124H in Avellino CD, (ACD,GCD II, OMIM: 607541), p.R555Q in Thiel-Behnke CD(CDTB, OMIM: 602082), p.R124L in Reis-Bucklers CD (CDRB,OMIM:608470)^[4]. Particularly noteworthy is the fact that specific mutations in TGFBI consistently cause specific types of corneal deposits. The nature of corneal deposits associated with TGFBI mutations is of amyloid (R124C), non-amyloid (R555W, R555Q and R124L) or mixture of both (R124H)^[5]. A progressive alteration of the cornea resulting in loss of transparency and finally blind. KE is found not only in primary amyloid deposits of hereditary cornea dystrophies, but also in secondary amyloidosis of the cornea of diverse ethiologies^[6]. These mutations were reported independently in various ethnic populations around world including Chinese^[7]-^[12]. Among them, mutations at two positions (R124 and R555) are responsible for approximately 50% of all TGFBI related corneal dystrophies that are diagnosed^[13], indicating the amino acids R124 and R555 represent hot spots for mutations in the TGFBI gene-linked corneal dystrophies^[14].

Carbohydrate sulfotransferase 6 gene (CHST6), encodes the enzyme N-acetylglucosamine-6- sulfotransferase (GlcNAc6ST). CHST6 gene mutation causes the corneal keratin sulfate (KS) metabolism, results the deposition of an unsulfated KS^[15] and macular corneal dystrophy (MCD, OMIM: 217800). It is characterized by the presence of grayish and diffuse glycosaminoglycan deposits in keratocytes and stromal lamellae, causing severe visual impairment^[16].

In this study, corneal lesions of 8 subjects were located in different depth of stromal layer. So we selected TGFBI and CHST6 genes for research. Of the eight cases, mutations were identified in 6 patients in TGFBI gene, which corresponded to three forms of corneal dystrophies:p. R124C in LCD I, p. R124H in GCDII, and p.R555W in GCD I.

MATERIALS AND METHEDS

Subjects

Eight sporadic probands with corneal dystrophies from 8 Chinese families were recruited in this study. All the patients were performed with complete ophthalmologic examination by the experienced ophthalmologists (Q.Z.). They were initial treatment, but none underwent surgery. 3 patients were diagnosed as lattice CD type I (LCDI) which was characterized by refractile lattice lines that were associated with white dots and faint haze in the anterior stroma. Twe patients presented with granular CD type I (GCDI) which was characterized by discrete deposits of gray-white, rounded, crumb-shaped opacities in the anterior central stroma.One presented with Avellino CD (GCDII), combined features of both granular, stellate opacities and lattice lesions. Two exhibited ‘atypical phenotypes’ CD with irregular gray-white haze in the subepithelium and superficial stroma of central corneal. Informed consent conforming to the tenets of the Declaration of Helsinki was obtained from each of participants. This study was approved by a local institutional medical ethics committee.

Methods

DNA extraction and polymerase chain reaction

Peripheral blood samples (5mL) were collected from all participators and genomic DNA was extracted with phenol chloroform isoamyl alcohol method^[17]. TGFBI (NCBI Reference Sequence: NC_000005.9) gene was amplified in these 8 patients and CHST6 (NCBI Reference Sequence: NC_000016.9) gene was amplified in the 2 ‘atypical phenotypes’ patients, using polymerphrase chain reaction. The primers and conditions for PCR are listed in Table 1 and Table 2. All primers were synthesized by Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. (Shanghai, China).

Table 1. Primers and annealing temperatures for PCR of TGFBI gene and CHST6 gene.

Exon	Primer sequence (5′-3′)	Amplicon size (bp)	Annealing temp (°C)
TGFBI-E1-F	GCGCTCTCACTTCCCTGGAG	252[18]	62
TGFBI-E1-R	GACTACCTGACCTTCCGCAG
TGFBI-E2-F	AGGCAAACACGATGGGAGTCA	279[19]	62
TGFBI-E2-R	TAGCACGCAGGTCCCAGACA
TGFBI-E3-F	ACCTGTGAGGAACAGTGAAG	200[2]	64
TGFBI-E3-R	GCCTTTTATGTGGGTACTCC
TGFBI-E4-F	TTCCCACATGCCTCCTCGTC	390	58
TGFBI-E4-R	AGTTCCACAGCCTTTTCTAA
TGFBI-E5-F	TCACGAGGGCTGAGAACA	351	60
TGFBI-E5-R	AGGAGAAGCAGGCACAAA
TGFBI-E6-F	CATTTTGCTACTGTGTTTGAA	306	57
TGFBI-E6-R	AGCCATCCTCTACCTCTCC
TGFBI-E7-F	GTGAGCTTGGGTTTGGCTTC	347[19]	56
TGFBI-E7-R	ACCTCATGGCAGGTGGTATG
TGFBI-E8-F	TGAGGTTATCGTGGAGTG	399[19]	55
TGFBI-E8-R	CACATCAGTCTGGTCACA
TGFBI-E9-F	GTGTTTAGAGGGGTTGTTGACT	372	62
TGFBI-E9-R	TCTCTGCACTGCCTAATGAA
TGFBI-E10-F	TAGAAGATACCAGATGTTAAGG	384[19]	54
TGFBI-E10-R	TGTCAGCAACCAGTTCTCAT
TGFBI-E11-F	CTTAATAACCCATCCCAGTGT	332	58
TGFBI-E11-R	TGAATCCCCAAGGTAGAAGA
TGFBI-E12-F	GTGGCCTGGACTCTACTATC	384	64
TGFBI-E12-R	CCCTGGTTGGCCTCATCCTT
TGFBI-E13-F	CATTAGACAGATTGTGGGTCA	378[19]	60
TGFBI-E13-R	GGGCTGCAACTTGAAGGTT
TGFBI-E14-F	CAGCCTGGGCGACAAGAT	396	64
TGFBI-E14-R	GTGCATTCAAAACCAACAG
TGFBI-E15-F	CCCTCAGTCACGGTTGTT	312[19]	56
TGFBI-E15-R	GGAGTTGCCTTGGTTCTT
TGFBI-E16-F	CTTGCACAACTTATGTCTGC	279[19]	56
TGFBI-E16-R	TGCACCATGATGTTCTTATC
TGFBI-E17-F	CAGGAGAGCATGGCAGAAGGA	178	58
TGFBI-E17-R	AGAGAAATTGGCGGAGAG

Table 2. Summary of the annealing temperatures and PCR products length of CHST6 gene, which were divided into 3 overlapping amplicons.

Exon	Primer sequence (5′-3′)	Amplicon size (bp)	Annealing temp (°C)
CHST6PF1	AGTGGGGTACGGGATGGG	559	62
CHST6PR1	GCCAGGGTGAAGGACTGC
CHST6PF2	GCAACCTGTCCGACCTCT	531	62
CHST6PR2	CAGTGAAGGCGTAGAGCG
CHST6PF3	AAGCCGCCACCCTTTCTG	500	60
CHST6PR3	GGGACCTGCTTCTCCGTGC

Primer 1 was quoted from Afshari et al^[18]. Primer 3 was quoted from Munier et al^[2]. Primer 2, 7,8,10,13,15,16 were cited from Ma et al^[19]. The rests were designed by DNA Star primerSlect Programme. (DNA Star Inc., Madison, WI).

Mutation screening

The 25µL PCR reaction mixture included 10×PCR buffer(Mg²⁺ plus) (2×GC buffer I was used in exon 1), 80ng DNA, 5.0mmol/L of each of dNTPs, 2.0U rtaq DNA polymerase, and 5.0µmol/L of each of the forward and reverse primers. All reagents used in this procedure were purchased from TaKaRa Biotechnology Co., Ltd. (Dalian, China). Thermocycling included an initial denaturation step at 94°C for 5 minutes, followed by 36 cycles of denaturation at 94°C for 30 seconds, annealing at 53-64°C for 30 seconds, and extension at 72 for 30-40 seconds, a final extension at 72°C for 5 minutes.

The PCR products were purified and terminator cycle sequencing was carried out using the Big Dye v3.1 kit and unidirectional sequencing reactions were analyzed in ABI 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA). Sequencing results from patients as well as the consensus sequences from the NCBI Human Genome Database were imported into the SeqManII program of the Lasergene package (DNAStar Inc., Madison, WI) and aligned to identify variations.

RESULTS

Sequencing Analysis

We identified a heterozygous C to T transition at position c.370 (CGC to TGC) in exon 4 of TGFBI in three patients, which caused the substitution of Arg for Cys at codon 124, can be related to LCDI(Figure 1). We detected a heterozygous G to A transition at nucleotide c.371 (CGC to CAC), located in exon 4 in one GCDII patient. This nucleotide substitution resulted in the missense mutation Arg124His (Figure 1). We found a heterozygous C to T transition at nucleotide c.1663 (CGG to TGG), located in exon 12 in other two GCDI patients (Figure 1). This mutation caused an Arginine to Tryptophan substitution on protein level at codon 555. In addition, we found four inconsequential heterozygous single nucleotide polymorphisms in TGFBI gene,with the nucleotide changes 651C/G (L217L, rs1442 ) in exon 6,981 G/A (V327V, rs1054124 ) in exon 8, 1620 C/T (F540F, rs4669)in exon 12, 1803A/G (L601L, rs35151677) in exon 13. To the best of the authors' knowledge, L601L was rarely reported in research. Mutations were not identified in the rest two atypical individuals either in TGFBI or CHST6 gene. Detail mutations in TGFBI gene were listed in Table 3 and Figure 1.

Figure 1. The experimental maps of three loci mutations and four SNPs.

A: The heterozygous c.370C>T(p.R124C) mutation in exon 4 in one of three CDLI families; B: The heterozygous c.371G>A (p.R124H) mutation in exon 4 in the GCDII family; C: The heterozygous c.1663C>T (p.R555W) mutation in exon 12 in two GCDI families; D: A heterozygous 1803A/G(L601L)SNP in exon 13; E: A homozygous 981G/G (V327V) SNP in exon 8; F: A heterozygous 651C/G(L217L) SNP in exon 6; G:A heterozygous 1620C/T(F540F)SNP in exon 12. Blank arrows highlight the position of nucleotide substitutions. Control electropherograms are shown for comparison purposes (A-G, bottom panel).

Table 3. Details of TGFBI mutations including amino acid change and cSNP in each patient.

Patients	Dystrophy	TGFBI mutations
		Amino Acid Change	cSNP
1	GCDI	c.1663C>T, R555W	651C/G,L217L
			981G/G, V327V
			1620C/T,F540F
2	GCDII	c.371G>A, R124H	—
3	GCDI	c.1663C>T, R555W	981G/G,V327V
			1620C/T,F540F
			1803A/G,L601L
4	LCDI	c.370C> T, R124C	651C/G, L217L
5	LCDI	c.370C> T, R124C	651C/G, L217L
6	Atypical CD	—	651C/G,L217L
			981G/A,V327 V
			1620C/T,F540F
7	LCDI	c.370C>T, R124C	651C/G,L217L
			981G/A,V327 V
			1620C/T,F540F
8	Atypical CD	—	651C/G,L217L
			981G/A, V327V
			1620C/T,F540F
			1803A/G,L601L,

DISCUSSION

In this study, 3 point mutations in TGFBI gene were identified in the 8 individuals with corneal dystrophies. That were, Arg124Cys mutation in three patients with LCDI, Arg555Trp mutation in two patients with GCDI and Arg124His mutation in one patient with GCDII. Remaining two atypical CDs may be caused by some other genes excluding from TGFBI or CHST6 gene. It confirmed a strong correlation between TGFBI gene specific mutations and stromal corneal dystrophies in Chinese.

Acknowledge the structure and function of TGFBIp will help us revealing the disease mechanism.

Humans TGFBIp in corneas from individuals ranging from 6 months to 86 years of age, TGFBIp is more abundant in mature corneas than in the developing cornea and the processing of TGFBIp changes during postnatal development of the cornea^[20]. In mouse embryon, at 18.5 dpc, the expression of TGFBIp in the epithelium appears more than the developing stroma cornea. Nolabeling is observed in the retina^[21]. The origination of TGFBIp was not imported into the cornea from plasma but from local synthesis suggested corneal deposits with inherited corneal diseases caused by mutations in the TGFBI gene^[22].

The human TGFBIp is composed of 683-amino acid protein is highly conserved between species and contains an N-terminal secretory signal peptide, four 140-amino acid tandemly internally domains homologous to one another, known as fasciclin-1, denoted “FAS1” domains, and an arg-gly-asp (RGD) motif at the C terminus^[23]. The FAS1 domain represents an ancient cell adhesion domain homologous to fasciclin I protein in Drosophila^[24]. In humans, there are four proteins containing FAS1 domains, specifically, two secretory proteins, TGFBIp and periostin, and two membrane proteins, stabilin-1 and stabilin-2^[21]. TGFBIp as a secreted protein and plays a “linker” role in the interaction between extracellular matrixc (ECM) and integrins^[25]. The FAS1 domain mediates cell adhesion and migration via interactions with integrins a3β1^[26] and avβ5^[27].

Missense mutations in the FAS1 domains of the human TGFBIp cause corneal dystrophies. Most of TGFBI mutations so far reported are located in the FAS1 domain with two mutational hot spots in R124 and R555. Arg-124 is located in the turn between helicesα1 andα2 of FAS1 domain 1,whereas Arg-555 is located in the helicesα3-α4 of FAS1 domain 4,whichis the most variable region of the FAS1 fold^[28]. Amino acid changes in FAS1-1 did not affect the stability of the intact TGFBIp, mutations in the FAS1-4, in contrast to the FAS1-1, appeared to affect the stability of TGFBIp, the similar behavior of intact TGFBIp and the isolated FAS1- 4, suggests that the stability of FAS1-4 may be the determiner of CD's mechanism^[29].

Moreover, the common mutations of TGFBI at positions 124 and 555 are likely to affect protein-protein interactions directly and reduced TGFBIp stability or solubility as the cause of amyloid deposition, do not apparently affect its binding to type I collagen, fibronectin and laminin or to forming fibrillar structures^[27], whereas the rare mutations that affected FAS1 core residues are likely to cause misfolding and secretion of the protein within the cell^[30]. While, the deposits of mutant TGFBIp are probably present only in the cornea of affected patients, suggesting that mutant forms of TGFBI may require other cornea-specific factors to trigger the deposition of abnormal TGFBI protein^[31].

The general notion is that TGFBI mutations may impair protein secretion, folding, degradation or interaction with other macromolecular components thus promoting the accumulation of corneal deposits.

Clinically, disease diagnoses are mostly based on the silt-lamp biomicroscope examination and the ophthalmologists' experience. The TGFBI gene-linked corneal dystrophies mainly include the lattice and granular types, as well as a lot of subtypes. It is difficult to make the accurate diagnosis and classification just relying on the clinical features. But the corneal tissue for histopathology was difficult to obtain in the initial treatment of patients, repeated grafts and bullous keratopathy with deficient corneal endothelial cells was not apparent in the originally grafted tissue. In our researches, followed by clinical examination, genetic screen was the efficient method to confirm the diagnosis, we establish the specific correlations between phenotype-genotype in TGFBI mutations: Arg124Cys and LCDI, Arg124His and GCDII, Arg555Trp and GCDI. Unraveling the genetic defects of corneal dystrophies will aid efforts directed at treatment.

Overall, our results about TGFBI and CHST6 gene analysis indicating that R124 and R555 codons were also the hot point mutations in Chinese as other populations in TGFBI gene. Genetic analysis will serve as the accurate and rapid adjacent tool to diagnose distinct CD populations. Patients will benefit from genetic counseling.