Mutational Analysis of TCOF1, GSC, and HOXA2 in Patients With Treacher Collins Syndrome.

Treacher Collins syndrome is an autosomal dominant craniofacial malformation mainly caused by mutations in the TCOF1 gene. Few cases have been observed in the Chinese population. Herein, the authors report the mutational analysis of TCOF1, GSC, and HOXA2 to determine the mutational features of the 3 genes in Chinese patients with Treacher Collins syndrome. Genomic DNA of the patients and their parents was extracted from peripheral blood following a standard protocol. DNA sequencing analysis was performed on all exons and the exon-intron borders of TCOF1, GSC, and HOXA2 in addition to the 1200-bp upstream of TCOF1. Four novel single nucleotide polymorphisms were detected in TCOF1, one of which was in the promoter region. Mutations in GSC and HOXA2 were not found in the 3 patients. Our results suggest the possibility of genetic heterogeneity or different mechanisms leading to the disease. Further functional study of the alteration is necessary to obtain more definitive information.

Treacher Collins syndrome (TCS, OMIM 154500) is a craniofacial development disorder. Although an autosomal recessive form of the syndrome has been reported,[1,2] most cases are inherited as an autosomal dominant trait. High intra- and interfamilial phenotypic variations have been identified, and no genotype-phenotype correlation in the syndrome has been found based on the evaluation of the clinical variability in TCS.[3-5] Anticipation is often observed in TCS families due to ascertainment bias.[4,5] TCS is likely in patients who are symmetrically affected, usually characterized by downward slanting palpebral fissures with lower eyelid coloboma, hypoplasia of the mandible and zygomatic, malformed ears, and conductive hearing loss due to atresia of the external ear canal.

The gene associated with the syndrome is TCOF1, which is located in 5q32-q33.1.[6] The gene contains 28 exons and several alternative splice sites.[3,6,7] To date, >150 different pathogenic mutations have been reported in the coding region of TCOF1, most of which are novel deletions or duplications leading to a premature stop codon.[8] Mutations in the promoter region have been proven to be functional,[9] representing a different mechanism causing the syndrome. In mice, haploinsufficiency of TCOF1 results in the depletion of neural crest cell precursors as the result of high levels of cell death in the neuroepithelium, which can lead to a reduced number of neural crest cells migrating into the developing craniofacial complex.[10] Recent research on TCOF1 mutant mice has demonstrated that effective cavitation of the middle ear is intimately linked to the growth of the auditory bulla, a neural crest cell-derived structure that encapsulates all middle ear components, and that defects in those processes have a profoundly detrimental effect on hearing.[11]

TCS is thought to represent defective structures derived from the embryonic first and second branchial arches, which are involved in a number of syndromic microtia,[12] including Goldenhar syndrome, Nager syndrome, and Miller syndrome. The overlap of patients with these syndromes may suggest a common genetic cause. Among the genes implicated in these syndromic microtia are GSC and HOXA2. A missense mutation in HOXA2 has been shown to cause autosomal recessive microtia with cleft palate.[13] Similarly, a same-sense mutation and a missense mutation have been documented in 8 patients with isolated microtia.[14] We therefore report a mutational analysis of TCOF1, GSC, and HOXA2 in 3 Chinese patients with TCS.

PATIENTS AND METHODS

Subjects

All patients with a clinical diagnosis of TCS were recruited through the Eye and ENT Hospital of Fudan University in China. The study protocols were previously approved by the Ethics Committee of the hospital, and blood samples were collected after informed consent was obtained from patients and their legal guardians. The parents also provided written permission to publish the family photos shown in Figure 1. The parents of these patients were all nonconsanguineous, and none of the patients had any family history of TCS.

FIGURE 1

Patient 1 had slanting palpebral fissures, coloboma of the eyelid, bilateral microtia, atresia of the external ear canal, hearing loss, and hypoplastic zygomatic arches.

Patient 1

The patient was a 10-year-old boy. His mother became pregnant at 19 years of age, and the pregnancy was complicated by trauma at day 40, which was treated with antibiotics. After birth, slanting palpebral fissures with coloboma of the eyelid, hypoplastic zygomatic arches, bilateral severe microtia and atresia of the external ear canal were observed. Objective audiometry confirmed bilateral conductive hearing loss with a hearing threshold of approximately 70 dB; there were no further abnormal clinical findings (Fig. 1).

Patient 2

This patient was an 11-year-old boy, and there was no history of exposure to teratogenic substances or the presence of illness of the mother. He was born with slanting palpebral fissures, coloboma of the lid, hypoplastic zygomatic arches, mild deformity of the left ear, atresia of the left external ear canal, and stenosis of the right ear canal. His hearing loss of approximately 50 dB was successfully treated by surgery, and no further physical abnormalities were apparent (Fig. 2).

FIGURE 2

Patient 2 with slanting palpebral fissures, coloboma of the lid, mild deformity of the left ear, atresia of the left external ear canal, stenosis of the right external ear canal, hearing loss, hypoplastic zygomatic arches, and deformation of both forearm.

Patient 3

Patient 3 was a 14-year-old girl born to a healthy mother without remarkable pregnancy history. She had slanting palpebral fissures, coloboma of the lid, hypoplastic zygomatic arches and mandible, and stenosis of bilateral external ear canal with conductive hearing loss of 55 dB (Fig. 3).

FIGURE 3

Patient 3 had slanting palpebral fissures, coloboma of the lid, stenosis of bilateral external ear canal, hearing loss, hypoplastic zygomatic arches, hypoplasia of the mandible, and deformation of both forearm.

DNA Sequencing

Genomic DNA of the patients and their parents was extracted from peripheral blood following a standard protocol. All exons and the exon-intron borders of TCOF1, GSC, and HOXA2 were amplified by PCR under optimal conditions using specific primers (Table 1); the 1200-bp upstream of TCOF1 were also amplified.

TABLE 1

PCR Primers for the Amplification of TCOF1, GSC, and HOXA2

PCR ID	Forward Primer	Reverse Primer
TCOF1-Promotor1	gagggcaactgccatgtatt	tggtggtagatcaggggaag
TCOF1-Promotor2	agactcatgcagtgccctct	gatcaacaataccgcccatt
TCOF1-exon2	gcgccaatgggcggtattgttg	ggaaggcagagattgcggctcctc
TCOF1-exon3	catgagtttggggagatctgg	atgtgagctggctttctggag
TCOF1-exon4	gcatgggtcagctcctatcac	tctcctccccagggtctttta
TCOF1-exon5	aagaatgtgggccagtccttt	gcagcaggcatcatccttatc
TCOF1-exon6	catgtcccaagaactgggatt	ctgaccctccctcgtctaggt
TCOF1-exon7	gcctcagagatgtgtgggaac	ctcctggtcaccctaccacag
TCOF1-exon8	ccaggtgggagatgaacgtaa	agaggtgctcatggcagagtg
TCOF1-exon9	ggaggctccaggagtgagagt	atagggcaaatgatgccacac
TCOF1-exon10–11	acgtggtgtcctgtgtctcct	caaaaccacaggaggtcttttga
TCOF1-exon12–13	ctctcctcccctcactcacat	ggggtgctgactgtggtgt
TCOF1-exon14–15	cagaacagatgggggactctg	caacactagcccccagtcaag
TCOF1-exon16	gaaggcacgcacaatgagttt	tccccactatggcacaactct
TCOF1-exon17–18	gttgtgccatagtggggagtg	accacgcccagccctatac
TCOF1-exon19	agtgctgtgctgggtcttagc	ccacaaaaaagctctggcaac
TCOF1-exon20	caggccggtaaattgggttat	aggtgatttggggggttgtag
TCOF1-exon21	accagttttgcccctttgact	acttgttgcagggagtgttcc
TCOF1-exon22	tgtgtgccccatctaacacag	aaacatggccctggagttttt
TCOF1-exon23	aatagaaatggggcctcagga	gacctgagggatcgggtagac
TCOF1-exon24	tgatagggcagggtgatccta	tctcacttggagaggctctgg
TCOF1-exon25–26	gcaccctcttcgctcttaggt	gaggaatgagaccaggtgctg
TCOF1-exon27	gttgtgatggcttctggtggt	ttccccaacacccttctacct
TCOF1-exon28	ggtagaagggtgttggggaag	tttccccttagagcccaccta
GSC-exon1	tggtctgagctccgtcctac	aattaaccaaccggctccat
GSC-exon2	caattctcagcatccccttg	ttcaacttcctgggcctaaa
GSC-exon3	gcccaggaagttgaatgaaa	cttggctccagactgatggt
HOXA2-exon1–1	tcccaggcacacacactaga	cgtaattcatggccttctcc
HOXA2-exon1–2	ccccatacggctgtaatcag	gaaggaagagggtcccagag
HOXA2-exon2	ccaactgtgcgtgtgtctct	ttggtgatgctttgttttgc

PCR, polymerase chain reaction.

A mixture with a total volume of 20 μL was prepared for each reaction including 1× HotStarTaq buffer, 2.0 mM Mg2+, 0.2-mM dNTP, 0.2 μM of each primer, 1 U HotStarTaq polymerase (Qiagen Inc), and 1-μL template DNA. The cycling program was 95°C for 15 minutes; 11 cycles of 94°C for 15 seconds, 62°C to 0.5°C per cycle for 40 seconds, and 72°C for 1 minute; 24 cycles of 94°C for 15 seconds, 54°C to 58°C for 30 seconds, and 72°C for 1 minute; and 72°C for 2 minutes. The PCR products were purified using SAP and ExoI. A mixture of 1 U SAP, 6 U ExoI, and 8-μL PCR products was incubated at 37°C for 60 minutes, followed by incubation at 70°C for 10 minutes. The reaction mixture included 2-μL BigDye 3.1 mix, 2-μL sequencing primer (0.4 μM), and 1- to 2-μL purified PCR product. The cycling program was 96°C for 1 minute followed by 28 cycles of 96°C for 10 seconds, 50°C for 5 seconds, and 60°C for 4 minutes. The final products were then analyzed using a capillary sequencer (ABI Prism 3730xl sequencing).

RESULTS

We identified 12 different variations in TCOF1, 1 previously reported SNP in GSC, and no alterations in HOXA2. Among the 12 variations in TCOF1, −26T>A, 17693G>A, 21761–21765delCTCTC and 21968G>T have not been previously reported as a TCOF1 mutation or polymorphism and were not in the dbSNP. All the 4 variations were also identified in healthy unaffected controls in the form of compound heterozygosity.

Analysis of patients’ normal parents showed that there was no parental origin of the −26T→A or 17693G→A mutations. Interestingly, only the father of patient 1 and mother of patient 2 possessed the 21968G→T alteration in the homozygous form, explaining the heterozygous genotype of their children. Analysis of patient 2's normal parents showed that the father possessed the 5-bp deletion, whereas his mother did not, indicating that the 5-bp deletion was inherited from the father. All of the alterations found within our subjects are described in Table 2.

TABLE 2

Polymorphisms Found in This Study

Gene Tested	Exon or Intron	Patient	Position in Cutted Sequence	SNP Property	Functional Change	dbSNP Identifier
TCOF1	5’UTR	1, 2, 3	−89T→G	None	Unkown	rs4565199
	5’UTR	3	−26T→A	None	Unkown	Unkown
	Intron6	3	14036G→A	None	Unkown	rs2255796
	Exon11	2	17681C→T	Synonymous	Pro526Pro	rs2071238
	Exon11	3	17693G→A	Synonymous	Gly530Gly	Unkown
	Exon12	2	18111A→G	Synonymous	Ser614Ser	rs2071239
	Exon13	2	18434G→C	Missense	Ala665Pro	rs2071240
	Intron16	2	21761-21765delCTCTC	None	Unkown	Unkown
	Exon17	2	21786T→C	Missense	Val887Ala	rs7713638
	Exon17	1, 2	21968G→T	Missense	Ala948Ser	Unkown
	Exon24	3	38922C→T	Missense	Ala1390Val	rs15251
	Intron25	3	41054G→C	None	Unkown	rs2569062
GSC	Intron2	1	1244G→T	None	Unkown	rs3905049

SNP, single nucleotide polymorphism.

DISCUSSION

The molecular mechanism responsible for TCS is still not defined; however, there has been remarkable work done in this area. First, haploinsufficiency has been proposed as one of the molecular mechanism underlying the disorder because deletion or insertion mutations in TCOF1 were the most likely to result in the creation of a premature termination codon and a truncated protein. Second, as mutations that do not alter canonical splice signals but influence splicing have been recognized as a novel form of mutation,[15] synonymous alterations in TCOF1 should be further investigated with functional assays before excluding pathogenicity. And last, mutations in the promoter region of TCOF1 could impair the DNA-binding to the YY1 transcription factor.[9] It suggests a possibility that changes outside of the coding region might alter expression level of functionally normal protein.

As previous studies mainly focused on the coding region of TCOF1, the current study detected all the exons and the exon-intron borders of TCOF1 in addition to the 1200-bp upstream of TCOF1. We identified 1 novel SNP −26T→A in the promoter region of TCOF1; however, functional study of this abnormality is necessary to obtain more definitive information. It is interesting to note that only the father of patient 1 and mother of patient 2 possessed the 21968G→T alteration in the homozygous form, explaining the heterozygous genotype of their children. Also interestingly, patient 2's father possessed the 5-bp deletion, whereas his mother did not, indicating that the 5-bp deletion was inherited from the father. All of the novel variations could also be found in the control samples. Therefore, no pathogenic mutations were identified in our patients. A lack of mutations in TCOF1 suggests that other novel genes or complex changes in gene regulatory networks might be responsible for TCS in the subjects analyzed.

Mutations in TCOF1 seem to be extremely variable, and TCS demonstrates great phenotypic variability.[3,4] Due to the clinical overlap, TCOF1 was analyzed and excluded in patients with Goldenhar, Nager, and Miller syndromes.[16,17] Therefore, it is meaningful to clarify whether the patients with TCS have alterations in those genes related with other syndromes.[14,18] HoxA2 is a key transcription factor during development of the second branchial arch that has a main contribution in development of the external and middle ear in mouse.[19] GSC is a transcription factor that plays an essential role during the process of gastrulation in early embryonic development.[20] Mice with a homozygous disruption of gsc revealed multiple defects containing the lower mandible as well as components of the inner ear and the external auditory meatus.[21] Both of the genes have been identified responsible for the syndromic microtia.[13,14] We therefore sequenced all the exons of GSC and HOXA2; however, we did not find any mutations in these genes except one previously reported SNP 1244G→T in GSC. Although these syndromes all derived from first and second branchial arches in embryonic period, they might be altered by different genes and regulated by complex gene networks.

In the present study, we performed TCOF1, GSC, and HOXA2 mutation analysis in 3 Chinese patients with TCS. We detected 12 polymorphic changes in TCOF1, 4 of which were novel. And we also excluded mutations of GSC and HOXA2 in the 3 patients. Mutations in the TCOF1 gene are not always found in patients with TCS.[8,17,22] These results suggest the possibility of genetic heterogeneity or the existence of different mechanisms leading to the syndrome. We hypothesized several possibilities to explain the undetected TCOF1 mutations in these TCS patients. First, there may be another gene that might be located near TCOF1 that is responsible for TCS. Second, nonsequential factors that can modulate the expression of TCOF1, for instance, the methylation of the gene or the mi-RNA regulation, may be involved. Further study is needed to explore the potential mechanism of these alterations in the occurrence of TCS.