Screening for GARS Variants in A Cohort of Chinese Patients With Inherited Peripheral Neuropathy

Background CMT2D is a rare subtype of axonal CMT, caused by the mutation of glycyl-tRNA synthetase (GARS) gene which is also a disease-causing gene of distal spinal muscular atrophy type V (dSMA-V) or hereditary motor neuropathy 5A (HMN5A). There were only several case reports in China, and no epidemiological study of CMT2D/ HMN5A yet. Methods We recruited the patients of Chinese Han descent clinically diagnosed with inherited peripheral neuropathy (IPN) from the Department of Neurology at Chinese PLA General Hospital (Beijing, China) from December 20, 2012 to July 31, 2019. All patients underwent a detailed medical history, neurological examination, laboratory examination, electrophysiological studies, and genetic testing. Results A total of 206 unrelated patients underwent genetic analysis, and we found four mutations of GARS from four different families, including c.794C>T (p.S265F), c.374A>G (p.E125G), c.1000A>T (p.I334F) and c.781T>G (p.Y261D), the rst three of them were considered pathogenic. As for the three pathogenic mutation carriers, one patient was diagnosed as CMT2D, two patients were diagnosed as HMN5A. Conclusion GARS mutation is a rare cause of inherited peripheral neuropathy and the phenotype tends to be CMT2D or HMN5A. There might be a relatively higher mutation frequency in Asian population compared with Caucasians. Combination of clinical phenotype, auxiliary tests and genetic evidence to assess the pathogenicity of genetic variants in patients suspected as IPN is of vital importance. mutation reported in a patient with HMN5A [16], and in our study, we found Patient 3, a 21-year-old female also carried c.1000A > T variant. Unlike c.794C > T (p.S265F) and c.374A > G (p.E125G) which occupied in the core catalytic domain of GARS, c.1000A > T (p.I334F) was located in the anticodon binding domain.


Introduction
Inherited peripheral neuropathies (IPN) include a large heterogenous group of hereditary diseases with more than 100 causative genes reported to date.
Charcot-Marie-Tooth disease (CMT), the most common IPN with a worldwide incidence of 1 in 2500, comprises a group of clinically and genetically heterogeneous peripheral neuropathies [2,3], and is roughly classi ed into Type 1 (CMT1; demyelinating) and Type 2 (CMT2; axonal) according to median nerve motor conduction velocity. More than 80 genes have been reported as being associated with CMT [4]. Owing to the development of molecular genetics, the classi cation of CMT is re ned.
The mutation of glycyl-tRNA synthetase (GARS) gene causes Charcot-Marie-Tooth disease type 2D (CMT2D) and distal spinal muscular atrophy type V (dSMA-V), also called hereditary motor neuropathy 5A (HMN5A) [4] . In Chinese Mainland, one family with CMT2D caused by mutation of c.999G > T (p.E333D) and the other family with HMN5A caused by mutation of c.383T > G (p.L128R) have been reported [5,6], and no epidemiological study of CMT2D/HMN5A was reported yet. This study was approved by the Chinese PLA General Hospital Ethics Committee, in accordance with the principles stated in the Declaration of Helsinki.
Informed written consent was obtained from each patient enrolled in this study.

Electrophysiological examination
All patients underwent nerve conduction study (NCS) in which their skin temperature was maintained at 32°C or above during the examination. NCS were performed on the median, ulnar, tibial, peroneal, and sural nerves using the Keypoint electromyography (EMG) system (Medoc Ltd, Israel). The results were measured according to the normal reference values utilized by the EMG laboratory of Chinese PLA General Hospital (median motor nerve: amplitude ≥5.0 mV, velocity ≥50.0 m/s; median sensory nerve: amplitude ≥5.0 µV, velocity ≥50.0 m/s; ulnar motor nerve: amplitude ≥5.0 mV, velocity ≥50.0 m/s; ulnar sensory nerve: amplitude ≥5.0 µV, velocity ≥50.0 m/s; tibial motor nerve: amplitude ≥5.0 mV, velocity ≥40.0 m/s; peroneal motor nerve: amplitude ≥3.0 mV, velocity ≥45.0 m/s; and sural sensory nerve: amplitude ≥6.0μV, velocity ≥50.0 m/s). NCS were considered abnormal if any of the studied parameters was found to be abnormal [7,8].

Sural nerve biopsy
Sural nerve biopsy was performed on Patient 1 with GARS variant with informed consent. A segment of nerve was xed in 3% glutaraldehyde buffered to pH 7.4 with 0.1 M phosphate buffer. Cross-sections of 1 mm thickness were post-xed in 0.1 M osmic tetroxide for 2 h, dehydrated in a series of graded ethanols and propylene oxide, and embedded in epoxy resin (LX-112). Semithin sections were stained with toluidine blue or paragon. Thin sections were stained with lead citrate and uranyl acetate, and examined under an electron microscope [9].

Genetic analysis
All patients underwent genetic analysis via NGS (high throughput target sequencing). We examined IPN-associated genes, especially CMT-associated genes ( Table 1). Genomic DNA was extracted from the peripheral leukocytes of fresh blood samples obtained from patients with a clinical diagnosis of IPN. Target genes were captured by GenCap target region probe (MyGenostics Inc, Medford, MA, USA) and ampli ed by polymerase chain reaction. The eluted DNA was nally ampli ed for 15 cycles according to the following procedure: 98°C for 30 s (1 cycle), 98°C for 25 s, 65°C for 30 s, 72°C for 30 s (15 cycles), and 72°C for 5 min (1 cycle) [7,8]. The ampli ed product was puri ed using SPRI beads (Beckman Coulter, Brea, CA, USA) according to manufacturer's protocol. Enriched libraries were sequenced using a HiSeq 2000 sequencer (Illumina, San Diego, CA, USA), which generated 100 bp paired reads [7,8].
Depth reading of NGS identi ed PMP22 duplications/deletions, and multiplex ligation-dependent probe analysis (MLPA) was applied to con rm the results. Sanger direct sequencing was used to con rm and detect variants in the patients and their family members [7,8].

Bioinformatics analysis
Polymorphism Phenotyping 2 (PolyPhen-2) (http://genetics.bwh.harvard.edu/pph 2/), sorting intolerant from tolerant (SIFT) (http://sift.jcvi.org/), and Mutation Taster (http://www.mutationtaster.org/) were used to predict potential functional effects of GARS mutations [7,8]. PolyPhen-2 classi ed the predicted effects of amino acid substitutions on the function of human proteins as "benign,""possibly damaging," "probably damaging," or "unknown." The functional impact of the mutation was predicted as"tolerated" or "damaging" by SIFT and as "polymorphism" or "disease-causing" by Mutation Taster [7,8]. The pathogenicity was determined by the ACMG guideline. Table 1 List of examined Charcot-Marie-Tooth disease-associated genes  Table 2.   [10]. He presented with bilateral distal lower limbs weakness and atrophy at the age of 13 and subsequently presented with weakness of bilateral distal upper limbs. No subjective sensory abnormality was found. Physiological examination showed muscle atrophy in bilateral thenar eminence, interosseous muscle, tibialis anterior muscle and calf muscle. Weakness of extremities, more severe in distal lower limbs, were found. Deep tendon re exes were decreased, and bilateral Babinski signs were negative. Super cial sensation was lost in distal extremities with intact deep sensation ( Table 2). NCS showed axonal motor neuropathy (Table 3). EMG showed active and chronic denervation potentials in both upper and lower limbs. Sural nerve biopsy showed unclear laminar structure in myelin sheath, axonal degeneration and necrosis of peripheral nerve myelin sheath and axon (Fig. 2). Patient 1 had positive family history and some of the family members had similar presentation ( Table 2, Fig. 1a). III 5, 27-year-old male, son of the proband, presented with distal lower limbs weakness at the age of 12, then upper limbs weakness at the age of 14. Physiological examination showed weakness of distal extremities and super cial sensory loss in gloving and socking pattern ( Table 2). NCS of III 5 showed axonal motor neuropathy (Table 3) and EMG showed active and chronic denervation potentials in both upper and lower limbs. Sanger test con rmed that affected family members (II1, II3, II4, II6, III1 and III3) carried c.794C > T mutation and unaffected members (III2 and III4) did not. The variant was predicted to be pathogenic by SIFT, PolyPhen-2, and Mutation Taster. It was also not present in the 1000 Genomes Project database. According to the standards and guidelines of ACMG, the variant was considered to be pathogenic.  [4,7] . He presented with bilateral upper limbs weakness at the age of 12 and presented with bilateral lower limbs weakness later. No subjective sensory abnormality was reported. Some family members had similar presentation,indicating autosomal dominant inheritance (Fig. 1b). Sanger test was performed on his mother (II 8) and son (IV 2), and the c.374A > G mutation was found. Physiological examination showed prominent weakness and muscle atrophy of bilateral hands. Mild weakness of bilateral distal lower limbs was also found. Deep tendon re exes were decreased, and bilateral Babinski signs were negative ( Table 2). Super cial and deep sensation were intact. NCS showed axonal motor neuropathy (Table 3). EMG showed active and chronic denervation potentials in upper and lower limbs. The variant was predicted to be pathogenic by SIFT, PolyPhen-2, and Mutation Taster. It was also not present in the 1000 Genomes Project database. According to the standards and guidelines of ACMG, the variant was considered to be pathogenic.

Results
Patient 3 was a 21-year-old female who carried c.1000A > T variant reported to cause HMN5A [13] . Weakness of bilateral distal lower limbs appeared about 1 year ago. No subjective sensory abnormality was found. Her father and brother had similar presentation and her brother (II 2) carried the same mutation (Fig. 1c). Physiological examination showed mild weakness and muscle atrophy of bilateral distal upper and lower limbs. Deep tendon re exes were decreased, and bilateral Babinski signs were negative (Table 2). Super cial and deep sensation were intact. NCS showed axonal motor neuropathy (Table 3). EMG showed active and chronic denervation potentials in upper and lower limbs. The variant was predicted to be pathogenic by SIFT, PolyPhen-2, and Mutation Taster. It was also not present in the 1000 Genomes Project database. According to the standards and guidelines of ACMG, the variant was considered to be pathogenic.
Patient 4 was a 38-year-old female who carried c.781T > G variant which was not reported before. She presented with numbness of distal extremities for 2 months at the rst admission to our hospital in 2013. There were no other patients in her family and no genetic cosegregation presented in this family. Physiological examination revealed decreased super cial and deep sensation in bilateral lower limbs, positive Romberg sign, and decreased deep tendon re exes. Strength of extremities were normal, and bilateral Babinski signs were negative (Table 2). NCS showed sensory neuropathy with intact motor nerve conduction ( Table 3). EMG of upper and lower limbs was normal. The variant was predicted to be damaging/probably damaging by SIFT, PolyPhen-2, and Mutation Taster. It was also not present in the 1000 Genomes Project database. According to the standards and guidelines of ACMG, the variant was considered to be of uncertain signi cance. Other auxiliary examinations including blood routine test, blood glucose, liver function, renal function, vitamin B12, homocysteine, routine cerebral spinal uid (CSF) test, lung CT and PET-CT etc, were normal, except for positive anti-Hu antibody in serum and CSF. With 7 years' follow-up, no malignancy was found and symptoms persisted. Reexamination of NCS showed an improvement of sensory nerve action potential (SNAP) and sensory nerve conduction velocity (SNCV) in left median and bilateral sural nerves and decreased SNAP amplitude in bilateral ulnar nerves (Table 3).

Discussion
In 2003, Antonellis et al. con rmed that GARS gene was the pathogenic gene of CMT2D/HMN5A for the rst time [11]. Up to now, only 20 mutations of GARS gene have been reported to be associated with CMT2D/HMN5A [5,12,13]. Classic phenotype of CMT2D/HMN5A is weakness and atrophy of upper extremities, especially the thenar eminence and the rst dorsal interosseous muscle groups. The main distinguishing characteristic of the two disorders is sensory de cits in a stocking and gloving pattern in patients with CMT2D. While sensory loss may vary from family members: sensation may be normal in some family members. Patient 2 and Patient 3 in our study presented with classic phenotype of HMN5A, while the Patient 1 who presented with prominent super cial sensory loss in distal extremities was diagnosed as CMT2D. Upper limbs were the most frequent onset sites, and lower limbs onset of Patient 1 could also be seen [9]. Compared with HMN5A, weakness of lower limbs in patients with CMT2D tended to be more prominent [9]. Most patients were adolescent onset, while infant onset was also reported and tended to be more severe [14][15][16]. Severity of phenotype could be various even in one family [14,16].
The results of electrophysiological studies con rmed motor peripheral neuropathy in the rst three patients. NCS of the Patient 1 and Patient 2 showed distinctively lower CMAP in median nerve than in ulnar nerve, and this imbalance involvement argued against a primary length-dependent distal axonopathy and was more in favor of a motor neuronopathy [9]. Sensory loss was the characteristic of CMT2D, and in other studies, sensory nerve action potential amplitude was decreased or diminished in CMT2D patients [6,12]. However, we did not think normal sensory nerve conduction study could deny the diagnosis of CMT2D for Patient 1, for sensory nerve degeneration in CMT2D could involve small-size and middle-size bers, sparing large myelinated bers [9,16], and sural nerve biopsy showed unclear laminar structure in myelin sheath, axonal degeneration and necrosis of peripheral nerve myelin sheath and axon.
A Korean patient with c.794C > T (p.S265F) was diagnosed as HMN5A [17], however our patient with c.794C > T (p.S265F) was diagnosed as CMT2D, because sensory nerve damage was revealed in sural nerve biopsy [10]. Previous studies showed that the mutation c.374A > C (p.E125G) was associated with both CMT2D and HMN5A [9,11]. Coexistence of CMT2D and HMN5A phenotypes caused by same mutation remained unknown etiology. Recently, the GARS c.794C > A (p.S265Y) mutation was reported in a Malian family with CMT2D [12], and c.373G > A (p.E125K) was associated to an infant patient with failure to thrive and severe muscle weakness [18]. These two amino acid sites might be critical to GARS. Previously, c.1000A > T (p.I334F) mutation was reported in a patient with HMN5A [16], and in our study, we found Patient 3, a 21-year-old female also carried c.1000A > T variant. Unlike c.794C > T (p.S265F) and c.374A > G (p.E125G) which occupied in the core catalytic domain of GARS, c.1000A > T (p.I334F) was located in the anticodon binding domain.
An unreported GARS gene mutation c.781T > G (p.Y261D) was found in Patient 4. This mutation was predicted as pathogenic by different prediction tools and was located in the core catalytic domain of GARS, the same as c.794C > T (p.S265F) and c.374A > G (p.E125G). However, no genetic cosegregation presented in this family and no pure sensory neuropathy associated with GARS gene mutation was reported yet. According to the guidelines of ACMG, the pathogenicity of this mutation was not considered. Due to subacute onset sensory neuropathy, long-term positive anti-Hu antibody, and no any other possible reason of sensory neuropathy, the patient was diagnosed as anti-Hu antibody neuropathy. Anti-Hu antibody is a paraneoplastic marker associated with peripheral and central nervous system disturbances, such as subacute sensory neuropathy, cerebellar dysfunction, limbic encephalitis and motor neuron disease etc [19,20]. Considering anti-Hu antibody combined with nervous system lesion had a strong connection with tumor especially small cell lung cancer [19,21,22], and in some cases neurological symptoms presented ahead of tumor diagnosis [19], consecutive follow-up for Patient 4 was necessary. Asymptomatic carriers with pathogenic variation were found in different studies [12,23], so we cannot exclude the possibility that descendants of Patient 4 may show symptoms later. But based on current evidence, we should not tell the Patient 4's family this possibility that may increase unnecessary psychological burden.
An American study that screened 100 patients diagnosed with dSMA, HMN, or motor axonal CMT for mutations in GARS found 3 mutations [16], while a Taiwan study reported two heterozygous mutations found from 54 axonal CMT patients indicating a higher mutation frequency [24]. Another American study showed very low frequency of GARS gene mutation, only 0.4% of 3216 CMT patients [25]. In this study, we found three pathogenic mutations of In general, GARS mutation is a rare cause of CMT and the phenotype tends to be CMT2D or HMN5A. Clinical characteristics, NCS, and even sural nerve biopsy and skin biopsy are essential to distinguish them. As the advance of next-generation sequencing technologies including disease-speci c gene panels, whole-exome sequencing and whole-genome sequencing etc, novel likely pathogenic genes and mutations would be found increasingly. In this study, we suggest the importance on combining clinical phenotype, auxiliary tests and genetic evidence to assess the pathogenicity of genetic variants in patients suspected as IPN.