A pseudo‐homozygous missense variant and Alu‐mediated exon 5 deletion in FARS2 causing spastic paraplegia 77

Abstract

FARS2‐associated hereditary spastic paraplegia, later onset spastic paraplegia type 77, is a rarely neurodegenerative disease. Here, we reported two affected siblings in an autosomal recessive spastic paraplegia family with a pseudo‐homozygous missense variant and Alu‐mediated exon 5 deletion in FARS2. Both patients gradually developed altered gaits and weakness in both lower limbs. In our literature review, spastic paraplegia type 77 shows high heterogeneity in clinical manifestations. Our study broadens the scope of pathogenic mechanisms of SPG77 resulting from compound heterozygous mutations in FARS2 and provides strong evidence that deletion in FARS2 due to recombination event mediated by Alu element.

Introduction

Hereditary spastic paraplegia (HSP) is a group of heterogeneous diseases characterized by progressive bilateral spasticity and weakness of lower extremities. ¹ HSP is clinically classified as either pure or complicated forms, with the pure form generally limited to spasticity and weakness of lower extremities and, in some cases, associated with bladder disorders. The complicated form is accompanied by different neurological symptoms including epileptic seizure, cognitive deficits, and developmental delay, among others. ² The hereditary modes of HSP include autosomal dominant, autosomal recessive, X‐linked recessive, and mitochondrial inheritance. ³ At present, more than 90 disease‐causing genes have been reported to contribute to HSP. ⁴

FARS2 encodes human mitochondrial phenylalanyl‐tRNA synthetase (Hsmt^PheRS), which ensures accuracy of in mitochondrial protein translation. ⁵ Patients carrying loss of function variants of FARS2 manifest as one of three distinct phenotypes: early‐onset epileptic encephalopathy, later‐onset spastic paraplegia type 77 (SPG77), and juvenile‐onset epilepsy. ⁶ To date, FARS2 mutations have been identified in 14 cases of SPG77. ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ Deletion mutations are the most common mutations besides missense mutations in SPG77 patients. The pathogenic mechanism of FARS2 gene deletion leading to SPG77 remains unclear.

Herein, we report a pedigree with SPG77, transmitted by compound heterozygous inheritance, arising from a novel Alu‐mediated deletion in exon 5 of FARS2 together with a c.1013G>A (p.A338H) missense variant. We also reviewed possible genotype–phenotype correlations between our two patients and previously reported individuals with FARS2 deletion variants.

Materials and Methods

All participants involved in this study from the HSP cohorts (NCT04010188) assessed at the First Affiliated Hospital of Fujian Medical University by our previously reported pipeline. ¹⁵ , ¹⁶ All participants underwent complete neurological examination including physical examinations, magnetic resonance imaging (MRI), electromyography (EMG), and the estimation of Spastic Paraplegia Rating Scale (SPRS), ¹⁷ Mini–mental state examination (MMSE), and Montreal Cognitive Assessment (MoCA). This study was approved by the Ethics Committee of the First Affiliated Hospital of Fujian Medical University (No.: FYYY2006‐01‐19‐01). Written informed consents were obtained from all family members.

Genomic DNA of all participants was extracted from peripheral blood leukocytes. The causative gene in two affected siblings was explored using whole‐exome sequencing (WES) and whole‐genome sequencing (WGS). Sequencing fragments were aligned to the human genome consensus sequence (UCSC hg38). Variant calling was performed using the Genome Analysis Toolkit (GATK) ¹⁸ and annotated with ANNOVAR. ¹⁹ Co‐segregation of the identified FARS2 variants was investigated by Sanger sequencing with specific primers (see Supporting Information 1: Table S1). Copy number variation (CNV) detection and calling were performed using the Sprinkle tool kit. ²⁰ Total RNA of all participants was extracted from peripheral blood leukocytes and were used for reverse transcription. RT‐PCR and long‐range PCR assays were performed with specific primers (see Supporting Information 1: Table S1).

Results

Clinical features in an autosomal recessive spastic paraplegia family

Two affected patients (Fig. 1A), including a woman (age 38; II‐1, the proband) and her younger brother (age 35, II‐2) gradually developed altered gaits, stiffness, and weakness in both lower limbs at 7 and 6 years old respectively, with slowly deteriorating symptoms. Proband began relying on crutches at age 27 and her younger brother at age 29. Detailed neurological examination revealed phenotypes of hypertonia, hyperreflexia, symmetrical distal muscle weakness (MRC grade 4), and positive for Babinski sign in both lower limbs. The phenotype was consistent between the affected siblings. MRI scans of the brain and spinal cord showed normal results (Fig. 1B). Electromyography was found to be normal. SPRS scores for patients II‐1 and II‐2 were 21 points and 26 points, respectively. They exhibited normal cognition and language function (MoCA scores 27 and MMSE scores 30 for patient II‐1; MoCA scores 26 and MMSE scores 29 for patient II‐2). Combined with the above clinical features, we classified both patients as pure form HSP.

Figure 1.

Genetic analysis of families with hereditary spastic paraplegia. (A) Family pedigree of affected siblings with suspected HSP. Filled and empty symbols represent affected and unaffected individuals, respectively. Arrow indicates the proband in this family. (B) T2‐weighted MR image showing no positive changes in II‐1 or II‐2. (C) Sanger sequencing traces of both alleles in affected individuals (top) and unaffected parents (bottom). (D) Kinship analysis by KING‐robust confirming biological paternity of I‐1 in subjects II‐1 and II‐2. (E) Copy number variation analysis of each exon for both alleles of FARS2 in affected and unaffected family members showing the absence of paternal exon 5 in the father and both affected children. (F) Agarose gel of RT‐PCR products of FARS2 for each family member showing a truncated allele (500 bp) in subjects I‐1, II‐1, and II‐2, but not I‐2 (left). Sanger sequencing of both bands showing two populations of mRNA, with the lower band lacking a region between exons 4 and 6 (right). (G) Whole genome sequencing data confirming the deletion at exon 5 of FARS2. Blank spaces indicate reduced sequencing depth and the location of gross deletion. (H) Sequencing traces for the breakpoint junction with microhomology region marked in red. Proximal reference is marked in blue; distal reference is marked in purple. (I) Diagram of the likely Alu‐mediated mechanism of deletion.

Identification of a novel missense mutation and deletion of exon 5 in FARS2

Whole‐exome capture identified a homozygous missense variant in FARS2 (NM_006567: c.1013G>A: p.A338H) that was present in both patients (Fig. 1C). According to the ACMG guidelines, ²¹ this variant was classified as likely pathogenic (PM1 PM2 PM3 PP1 PP3), which was described in detail in Supporting Information 1: Table S2. Interestingly, although WES suggested that the missense variant was homozygous in the patients, genotyping of the parents by Sanger sequencing indicated maternal inheritance, as the paternal FARS2 allele carried no such variant at that position (Fig. 1C). We hypothesized three possible explanations for this phenomenon: (i) biological non‐paternity of subject I‐1, (ii) uniparental diploidy inherited from subject I‐2, or (iii) one inherited allele contained an undetected deletion. Parentage analysis by the KING‐robust estimator supported that I‐1 was indeed the biological father of II‐1 and II‐2 (Fig. 1D). Subsequent comparison of exome DNA sequence between the proband and her parents indicated that the proband carried both the maternal and paternal alleles of FARS2, thereby ruling out the possibility of uniparental diploidy (see Supporting Information 1: Table S3).

We then conducted CNV analysis to test the third possibility that an unknown deletion resulted in the absence of a paternal allele in exome data, which could appear as a single homozygous allele in WES data. CNV analysis indicated that the copy number of the maternal allele was indeed twice that of other family members (Fig. 1E). Sanger sequencing identified a truncation from exon 4 to exon 6 (Fig. 1F), thus confirming that subject I‐1 harbored a deletion variant that was inherited by the patients and resulted in pseudo‐homozygosity. These collective results indicated that patients II‐1 and II‐2 were, in fact, compound heterozygous and carried two distinct variants in FARS2 that together resulted in pure form HSP.

Deletion breakpoints contain Alu elements in the same orientation

WGS of the proband to detect the deletion range and breakpoints identified a 6733 bp deletion (chr19: 41896116–41941714) between intron 4 and intron 5 of FARS2 (Fig. 1G). Sanger sequencing of the regions flanking the breakpoints revealed the presence of an AluY element adjacent to the breakpoint in intron 4 and an AluSg element at the breakpoint in intron 5 (UCSC database). Interestingly, sequence within the Alu elements at the breakpoints exhibited 30 bp microhomology and followed the same orientation. Further sequence analysis indicated that the boundaries of the deletion shared 81% homology with AluY and AluSg elements, respectively (Fig. 1H). This breakpoint architecture provided evidence supporting Alu‐mediated deletion as the likely mechanism leading to loss of exon 5 (Fig. 1I).

Pathogenic FARS2 deletion variants in other SPG77 cases

Only three pedigrees with SPG77 carrying deletion variants in FARS2 have been reported to date (Table 1). ⁹ , ¹⁰ , ²² All patients developed spastic gait, lower limb weakness, ankle clonus, and Babinski sign. One pedigree (one patient) was classified as SPG77 pure form, while the other two pedigrees (three patients) were classified as complicated form. One case in Italy was classified as the complicated form with cerebral palsy and developmental delay, while the other pedigree included two siblings who presented with dysphonia.

Table 1.

Clinical features of the six affected individuals carrying FARS2 variants.

	Family in this study		Family in previous study
			Family A	Family B	Family C
	Patient II‐1	Patient II‐2	Proband	Proband	Brother	Sister
Deletion type	ex5,chr19:41896116‐41941714	ex5,chr19:41896116‐41941714	ex.2‐4,chr6:5284218‐5442731	ex.1‐2 and ex.1 of LYRM4	ex.6,Chr6:5564777‐5639774	ex.6,Chr6:5564777‐5639774
Second variant	c.1013G>A, p.R338H	c.1013G>A, p.R338H	c.1082C>T, p.P361L	c.1082C>T, p.P361L	c.422G>A, p.G141E	c.422G>A, p.G141E
Sex	F	M	M	M	M	F
Age at onset	7 years	6 years	2 years	5 years	2 years	18 months
Age at evaluation	14 years	14 years	8 years	12 years	13 years	7 years
Initial symptom	Difficulty walking	Difficulty walking	DD and cerebral palsy	Gait difficulties	Delayed walking	Delayed walking
Spasticity at gait	+	+	+	+	+	+
Limb weakness (DUL/DLL)	−/+	−/+	+/+	NA	−/+	−/+
Muscle atrophy (DUL/DLL)	−/−	−/+	NA	−/+	NA	NA
Muscle tonus (DUL/DLL)	−/++	−/++	NA	−/+	NA	NA
Tremor (DUL/DLL)	−/−	−/−	−/+	−	+/+	+/+
Babinski sign	+	+	+	+	+	+
Extensor plantar response	+	+	+	+	+	+
Clonus (ankle)	+	+	+	+	+	+
Cognitive deficits	−	−	+	−	NA	NA
Epilepsy	−	−	−	−	NA	NA
Dysphonia	−	−	+	−	+	−
Developmental delay	−	−	+	−	NA	NA
Foot deformity	−	+	+	+	NA	NA
SPRS score (0–52)	21	26	NA	NA	NA	NA
SPATAX‐EUROSPA score a (0–7)	5	4	NA	NA	NA	NA
MRI	Normal	Normal	NA	Normal	Normal	Normal
EEG	Normal	Normal	NA	Normal	Normal	Normal
electromyography	Normal	Normal	NA	Normal	Normal	Normal

−, absence; +, presence; DD, developmental delay; DUL/DLL, distal upper limbs/distal lower limbs; NA, not available; SPRS, Spastic Paraplegia Rating Scale.

^a Disability score: 0 (no functional handicap), 1 (no functional handicap but signs at examination), 2 (mild, able to run, walking unlimited), 3 (moderate, unable to run, limited walking without aid), 4 (severe, walking with one stick), 5 (walking with two sticks), 6 (unable to walk, requiring wheelchair), and 7 (confined to bed).

Discussion

The pedigree reported here is an autosomal recessive pure form SPG77 rather than complicated form previously described in other patients with FARS2 deletion. In our literature review, SPG77 shows high heterogeneity in clinical manifestations. Reported phenotypes are mainly characterized by a combination of spastic paraplegia, dysarthria, developmental delay, and a wide range of abnormalities in brain MRI, including diffuse brain atrophy. ²³ To establish a comprehensive perspective of genotype–phenotype correlations in SPG77, further studies, especially those focusing on pathogenic variants, will be necessary in larger HSP cohorts. Recently, FARS2 was identified as a potential pathogenic gene to cause cardiomyopathy. ²⁴ However, no symptoms of myocardial damage symptoms have yet been observed in our study cohort. It will be necessary to check for cardiovascular systems in our patients during follow‐up and as a standard practice in future cases.

All previously reported deletion mutations distributed in exons 1–6 of FARS2 could potentially compromise the structure or function of Hsmt^PheRS. The protein sequence contains four functional domains, including the N‐terminal region (residues 37–83), a catalytic domain (residues 84–325), a linker region (residues 326–358), and an anticodon binding domain (residues 359–451). ²⁵ In this current study, a deletion in exon 5, spanning residues 304–355, resulted in a partial in‐frame deletion of the aminoacylation domain and linker region. The missense variant was also situated in exon 5 and introduced a premature stop codon at R338. It is therefore reasonable to speculate that these structural changes to Hsmt^PheRS could interfere with its function in neurons, impairing neuronal development, which could lead to SPG77.

It is the first report of which we are aware describing an Alu‐mediated deletion in FARS2 associated with SPG77. A new chimeric Alu element formed at the junction depending on microhomology of two Alu repeat elements. Greater focus on Alu elements is warranted in genetic analysis of HSP patients carrying deletions or duplications in future work. Notably, our initial WES analysis suggested the patients were homozygous for the missense mutation, but later found by co‐segregation analysis to be pseudo‐homozygous. It illustrates the need familial co‐segregation analysis to verify homozygous mutations.

In summary, our study broadens the scope of pathogenic mechanisms of pure form SPG77 resulting from compound heterozygous mutations at the same site in FARS2. These results also provide strong evidence supporting a possible role of Alu‐specific microhomology‐mediated recombination in the development of SPG77.

Author Contributions

SHL designed and carried out the study. JHX drafted the manuscript and figures. JYJ coordinated the research process and helped to draft the manuscript. XYY and CYH participated in the data analysis. NW and WJC participated in clinical evaluation of the patients. XL were involved in the study design, performed clinical evaluation of the patients and critical evaluation of the manuscript.

Conflict of Interest

The authors declare that they have no competing interests.

Supporting information

Table S1.

Funding Statement

This work was funded by Excellent Youth Training Program of the First Affiliated Hospital of Fujian Medical University grant 2023FY‐JCQN‐1; Local Science and Technology Development Project guided by the central government grants grant 2022L3011; Fujian Research and Training Grants for Young and Middle‐aged Leaders in Healthcare; Fujian Provincial Health Technology Project grant 2022ZQNZD003; National Natural Science Foundation of China grants 82171403, 82222022, and U2005201.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.