A novel TFG c.793C>G mutation in a Chinese pedigree with Charcot‐Marie‐Tooth disease 2

Abstract Introduction Mutations within TFG gene were recently reported to cause Charcot‐Marie‐Tooth disease 2 (CMT2). However, only few pedigrees were documented so far. Here, we reported a Chinese CMT2 pedigree with 8 affected cases and a novel TFG mutation. Methods Clinical evaluation and electrophysiological study were performed in all the affected individuals. Whole‐exome sequencing was conducted, followed by the Sanger sequencing and co‐segregation analysis to verify the variants. Results All cases presented with a phenotype of CMT2, including slowly progressive symmetrical muscle atrophy and weakness predominantly in the distal limbs. Sensory loss in the distal limbs was present in the proband and his father. Age at onset ranged from 37 to 44 years, and was younger in male cases, compared with female cases. Nerve conduction study revealed normal motor nerve conduction velocity but decreased compound muscle action potential. Electromyography test revealed fibrillation potential and positive sharp waves. The creatine kinase activity was increased in all cases. After genetic investigations, we identified a novel TFG c.793C>G (p.Pro265Ala) mutation in the family. This mutation alters the conserved amino acid residue and is absent in 1000G, ExAC, dbSNP, EP6500, and 200 in‐house controls. It co‐segregated with the disease in the family. Conclusions Our report provided additional evidence that the heterozygous TFG mutations were associated with CMT2.

According to the nerve conduction velocity (NCV) of median motor, CMT is divided into three types: demyelinating (CMT1) with NCV < 35 m/s, axonal (CMT2) with NCV > 45 m/s, and intermediate CMT (ICMT) with NCV 35-45 m/s (Mathis, Goizet, & Tazir, 2015;Thomas, Guergueltcheva, & Gondim, 2016). Patients with CMT usually suffer from slowly progressive muscle weakness and atrophy of distal extremities and sensory loss of distal end. The symptoms usually begin in the first decade to the third decade but can be great variable. The inherited pattern of CMT includes autosomal dominant, autosomal recessive, and X-linked. To date, more than 80 genes have been associated with CMT (Pareyson, Saveri, & Pisciotta, 2017).
Among these identified genes, PMP22 duplication is the most common cause, accounting for 40%-50% of all CMT and about 70% CMT1 (Li, Parker, Martyn, Natarajan, & Guo, 2013;van Paassen et al., 2014). The second most common cause of CMT is GJB1 mutations, which are transmitted with the X-linked model (Murphy, Laura, & Fawcett, 2012). In the CMT2 subtype, at least 17 genes have been identified, and mutations in MFN2, MPZ, and NEFL are more common than others (Boerkoel, Takashima, & Garcia, 2002). However, the genetic causes of about 60% CMT2 patients remain elusive.

| Subjects
This study was approved by the Ethics Committees of Second Affiliated Hospital of Zhejiang University School of Medicine.
Written informed consents were obtained by all participants. Clinical evaluation was performed by at least two senior neurologists. A total of 8 patients were involved in this family. Nerve conduction study and electromyography (EMG) were performed in available individuals. PMP22 duplication/deletion was excluded before the wholeexome sequencing (WES). Besides, 200 normal individuals with no history of neurological disorders were collected as controls.

| Genetic investigations
Genomic DNA was extracted from peripheral blood, using QIAamp genomic DNA kits (Qiagen). WES was performed in the proband, followed by the Sanger sequencing to verify the variant. Co-segregation analysis was conducted in all available familial members. Minor allele frequency (MAF) was searched on ExAC, dbSNP, and 1,000 Genomes Browser. In silico algorithms SIFT, PolyPhen-2, Mutation Taster, and CADD were used for functional prediction analysis.
Variants were classified according to American College of Medical Genetics and Genomics (ACMG) guideline.

| Clinical features
A total of 8 affected individuals were included in this four-generation pedigree (Figure 1a). The inheritance pattern was autosomal dominant. Age at onset ranged from 37 to 44 years. All affected individuals presented with CMT2 phenotypes, including slowly progressive symmetrical muscle atrophy and weakness predominantly in the distal limbs. Sensory loss in the distal limbs was present in the proband and his father. No sensory impairment was detected in other affected individuals. The median motor NCV ranged from 54.6 to 62.1 m/s (Table 1). Compound muscle action potential (cMAP) was decreased. EMG test revealed fibrillation potential and positive sharp waves in the proband and his older sister (III-12). The creatine kinase (CK) was increased in all cases.
The proband (III-14) was a 44-year-old man who had a 6-year history of muscle weakness and atrophy in four extremities. At the age of 38, he felt muscle weakness in his lower limbs after exercise.
The weakness slowly progressed in the following months. Half a year after the onset, he noticed muscle atrophy at the distal end of lower limbs. There was no sensory dysfunction at early stage. At the age of 41, he exhibited muscle atrophy in his hands and his fine movement was impaired. In addition, he presented with muscle cramps The other familial members II-1, III-1, III-2, III-3, III-4, and III-5 did not exhibit muscle weakness or atrophy at present. II-2 had passed away and was reported not to have muscle atrophy. NCV study and EMG test were not performed in these members.

| Genetic analysis
Using WES, we identified a heterozygous TFG (NM_001195478) mutation c.793C>G in the proband (Figure 1c). This mutation causes a substitution of proline 265 with alanine (p.Pro265Ala) and resides in an evolutionarily conserved region (Figure 1d). This mutation was also detected in II-6, III-6, III-8, III-10, and III-12, all of whom were affected individuals. We did not find this mutation in II-1, III-1, III-2, III-3, III-4, III-5, and III-7. Co-segregation with the disease was confirmed in the family. This mutation was located in the conserved amino acid residue and was absent in 1000G, ExAC, dbSNP, EP6500, and our 200 in-house controls. It is predicted to be tolerable by SIFT, benign by PolyPhen-2, disease causing by Mutation Taster, and damaging by CADD. According to the ACMG guideline, this mutation should be classified as "likely pathogenic."

| D ISCUSS I ON
Here, we presented a novel TFG mutation in a Chinese CMT2 pedigree. The mutation p.Pro265Ala alters the conserved amino acid residue and is absent in public genomic database and our in-house controls. In addition, the location of this mutation was adjacent to the pathogenic mutation p.Gly269Val. The co-segregation in the family revealed that this mutation might be causative for the disease. Due to the technical and material limitations, the evaluation of functional alternation of the mutant protein was not conducted.
Since loss-of-function mechanism has been reported as molecular pathogenesis of TFG, we believed this novel mutation may be biologically pathogenic.
The patients in this pedigree exhibited symmetrical muscle atrophy and weakness predominantly in the distal limbs. Distal sensory impairment was present in the proband and his father. Motor nerve conduction studies were within normal limits. EMG showed denervation and reinnervation. Based on the clinical and electrophysiological findings, these patients met the diagnosis criteria of CMT2.
In some patients, sensory impairment was not present. It is possible that their sensory loss was mild, similar to the CMT2 patients with TFG p.Gly269Val mutation (Tsai et al., 2014). Interestingly, some patients in our study had onset of muscle atrophy in the upper limbs, which is consistent with the patients described by Fabrizi, Høyer, & Taioli (2020). In addition, four patients in this study had muscle cramps and fasciculations, which were not common in CMT2.
Mutations within TFG have been associated with several distinct phenotypes, including HMSN-P, (Alavi et al., 2015;Ishiura et al., 2012) hereditary spastic paraplegia (HSP), (Tariq & Naz, 2017) neuroaxonal dystrophy "plus" syndrome, (Catania, Battini, & Pippucci, 2018) and motor neuron disease with sensory neuropathy (Li, Meng, & Wu, 2019). This implied the clinical heterogeneity in patients with TFG mutations. Even the identical TFG mutation could cause different phenotypes. Different symptoms were also found across the family members. In addition, the presence of fasciculation and cramps is prominent in HMSN-P. The different clinical features among patients with TFG mutations required further explanation.
Other modifying genes or environmental factors are potential contributing causes. Alternatively, age at examination may affect the phenotypic differences.
TFG is ubiquitously expressed in human neurons, including the brain, spinal motor neurons, and dorsal root ganglia (Yagi, Ito, & Suzuki, 2016). Different mutations in TFG cause disparate clinical phenotypes, implying that each mutation has its unique pathogenic mechanism. Alternatively, different expressions of TFG in peripheral axons, lower motor neurons, and upper motor neurons might influence the phenotypes. Although the full scope of TFG functions is still unclear, recent studies revealed that TFG functions at endoplasmic reticulum exit sites and regulates secretory protein vesicle biogenesis and egression from the endoplasmic reticulum (Beetz, Johnson, & Schuh, 2013;Yagi et al., 2016). Therefore, impaired endoplasmic reticulum function might be the major mechanisms of mutant TFG in TFG-related disorders.

| CON CLUS IONS
In summary, we reported a novel TFG c.793C>G mutation in a Chinese pedigree with CMT2. Our results provided additional evidence that heterozygous TFG mutations were associated with CMT2.

ACK N OWLED G M ENTS
The authors sincerely thank the patients for their help and willingness to participate in this study. This work was supported by grants from Zhejiang analytical testing technology program (2018C3706).

CO N FLI C T O F I NTE R E S T
The authors declare no potential conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Ding-Wen Wu contributed to funding, data acquisition, analysis and interpretation, and drafting of the manuscript. Yanfang Li contributed to data acquisition, analysis and interpretation, drafting of the manuscript. Xinzhen Yin contributed to data acquisition. Baorong Zhang contributed to study design, study supervision, data acquisition, analysis and interpretation of data, and critical revision of the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data supporting the results of this study are publicly available.