Introduction

Spastic paraplegia (SPG) is a highly heterogeneous group of inherited progressive neurodegenerative disorders, and so far more than 80 genes are known to cause SPGs. SPGs are broadly classified as pure and complicated SPG based on presence or absence of additional clinical manifestations (other than SPG) respectively. SPGs are inherited as autosomal dominant, autosomal recessive, and X-linked disorders. So far 44 genes are described to cause autosomal recessive SPG, majority of these being childhood onset and complicated type of SPG (OMIM).

Spastic Paraplegia 79 (SPG79) is an autosomal recessive SPG, characterized by early onset of SPG and optic atrophy. Additional variable features include peripheral neuropathy, cerebellar signs, seizures, borderline cognitive impairment, and abnormal neuroimaging findings [1]. Recently Rydning et al. [1] described a Norwegian family with three affected siblings (monozygotic twins and their sister) with childhood onset progressive visual loss and spasticity, harboring a compound heterozygous variant (NM_004181.4:c.533G>A;c.647C>A) in UCHL1 gene. In 2013, Bilguvar et al. [2] reported a Turkish family with three siblings affected with this complex form of neurodegenerative disorder (childhood-onset neurodegeneration with optic atrophy, ataxia, and SPG) and a homozygous missense UCHL1 variant (NM_004181.4:c.20A>C). We describe a third family of Indian origin with two affected siblings harboring a novel homozygous, likely pathogenic, splice-site variant, NM_004181.4:c.459+2T>C in UCHL1, which was identified by whole-exome sequencing and validated by Sanger sequencing in the family.

Patients and methods

Patients (P1 and P2)

Both siblings (P1 and P2) presented with history of motor developmental delay and gait imbalance. Progressive spasticity was associated with generalized decrease in the muscle bulk, chest deformity, and joint contractures. Gradually both of them became non-ambulatory by age 5–7 years and there was associated poor vision. Other clinical findings included—tremors, other cerebellar signs, facial dysmorphism, myopathic face, microcephaly, and fasciculations (both tongue and limb muscles). Cognitive functions were normal for age although IQ test at age 9 years for elder sibling (P1) and at age 6 years for younger sibling (P2) revealed scores of 59 and 56 respectively. Sensory system was normal in both siblings. Only elder sibling (P1) had recurrent seizures (age of onset 1.5 years), and was on anti-epileptic drugs. Detailed clinical history and examinations of the patients are given in Fig. 1 and Table 1 and also provided in Supplementary Note.

Fig. 1
figure 1

a, b (Patient 1)—Down slant palpebral fissures, epicanthic folds, broad nasal base, thick full lips, open mouth, micrognathia, relatively large looking ears, and overall appearance of myopathic face. c (Patient 2 at age 6 years 9 months) and d (at age 13 years)—Down slanting palpebral fissures, epicanthic folds, broad nasal base, open mouth, thick full lips, micrognathia, relatively large looking ears, frontal upsweep, anterior unusual hair whorl, and with age overall myopathic facial appearance. e Pedigree of the family. f, g MRI findings for Patient 2—T1 and T2W MRI sagittal images sagittal and h axial image demonstrates large cysterna magna (arrow). Cerebellum and brain stem are well developed. i On FLAIR coronal MRI, there are areas of gliosis in both occipital white matter (open arrow). j, k T2W axial MR images of optic nerves of the patient show near symmetric, relatively small optic nerves (mean 24 mm) on both sides (open arrow). l Age-matched control shows a mean optic nerve measurement of 38 mm. m Coronal MR image of the patient also show near symmetric small sized, optic nerves (arrow)

Full size image
Table 1 Clinical and molecular findings in patients with SPG79 in the patients described in this study and in the patients described by Bilguvar et al. [2] and Rydning et al. [1]
Full size table

Whole-exome sequencing

Whole-exome sequencing was carried out on peripheral leukocyte DNA of the younger sibling (P2) to ascertain coding sequence variants. Exome sequencing and data analysis pipelines were described earlier [3]. The study protocol was approved by the Institutional Ethics Committees of respective institutes. PCR was performed using specific primers F-5′-GGCTAGGGGAAAAGGTACAGC-3′ and R-5′-TTGGCCTTGGGTGGATAACAT-3′, and Sanger sequencing was done using ABI 3130 Genetic analyzer (Life Technologies, Carlsbad, CA) following the manufacturer’s protocol in control, parents, and patient samples to validate the variant. Pathogenicity of the variant was tested using online mutation prediction tools like MutationTaster2 [4], CADD [5], and Human Splicing Finder version 3 [6].

Results

The quality information of exome-sequencing raw data and the variant filtering strategy are outlined in Table S1 and Table S2 respectively. After filtering steps, a total of three variants were shortlisted, among which a likely pathogenic variant was identified after detailed clinical correlation with the patient’s phenotype (Table S3). The variant was a novel homozygous splice-site variation in intron 6 of the UCHL1 gene (NM_004181.4:c.459 + 2T>C) which affects the invariant 5′ donor splice-site of exon 6. The variant was not present in 1000 Genomes project, Exome Variant Server, ExAC, gnomAD, and in our in-house exome database of 96 unrelated individuals from all over India. The in silico splice-site effect assessment of this variant by online tools scored low for the mutant splice-site and predicted possible splicing aberrations in the transcript due to this variant (Figs. S1 and S2). The variant was confirmed in the elder sibling (P1) and validated in the family using Sanger sequencing (Fig. 2). The variant has been submitted to ClinVar (ClinVar accession ID: SCV000746265.1)

Fig. 2
figure 2

Sanger validation of the UCHL1 variant in family. The variation c.459 + 2T>C of UCHL1 is found in homozygous state in probands (a, b) and is heterozygous in their mother (c). Normal control has the homozygous wild-type allele (d). e A schematic representation of UCHL1 gene structure (exons 1–9) showing the earlier variants reported by Bilguvar et al. [2] (c.20A>C) and Rydning et al. [1] (c.533G>A;c.647C>A) and our variant (c.459 + 2T>C) in association with SPG79

Full size image

Discussion

We report on two siblings sharing a common phenotype of an early onset neurodegenerative disorder characterized by SPG, optic atrophy, seizures, and facial dysmorphism. Whole-exome sequencing identified a homozygous deleterious splice-site variant in UCHL1 gene confirming the diagnosis of SPG79 (MIM#615491). This is the third report of a family with SPG79 and mutations in UCHL1 gene.

The UCHL1 encodes a thiol protease enzyme Ubiquitin carboxyl-terminal hydrolase-L1 (UCHL1) which is one of the most abundant proteins in the brain, comprising almost 1–2% of the total soluble protein [7]. Expression of UCHL1 is seen in most neuronal cells, neuroendocrince cells, podocytes, and various other cells (due to its role in cancer pathology) [8]. The protein is mainly a deubiquitinating enzyme that has hydrolase activities in the ubiquitin–proteasome pathway. In vitro studies have also shown ubiquitin ligase activity in this pathway [9]. These studies suggest that UCHL1 enzyme may play an important role in maintaining the ubiquitin stability within the neurons [10].

Variants in UCHL1 have been implicated in the late onset common neurodegenerative disorders like Parkinson’s disease and Alzheimer’s disease [11,12,13,14]. In 2013, Bilguvar et al. reported a family with a early onset form of neurodegenerative disorder and a homozygous UCHL1 missense variant [2]. They described three siblings born to consanguineous parents of Turkish origin with childhood-onset neurodegeneration, optic atrophy, ataxia, and SPG. All patients had the homozygous c.20A>C;p.Glu7Ala variant causing a loss of function of UCHL1 as observed in UCHL1 knockout mice [15]. Recently, Rydning et al. [1] reported a second family with similar phenotype and a compound heterozygous variant in UCHL1 gene (c.533G>A;p.Arg178Gln and c.647C>A;p.Ala216Asp). However, further studies showed contradictory functional consequences as the c.647C>A variant led to loss of function (insoluble protein), whereas the c.533G>A led to increased enzyme (hydrolytic) activity. The authors proposed that while the non-soluble c.647C>A variant results in reduction of functional UCHL1 protein and contributes to neurodegeneration, the increased enzymatic activity of the c.533G>A variant might offer a protective effect on cognitive function as their patients showed conspicuously well-preserved cognitive abilities than the patients described by Bilguvar et al. [2] despite similar neurological dysfunctions.

In the present study, we describe a third family of Indian origin with two siblings similarly affected with childhood-onset optic atrophy and spasticity harboring a homozygous splice-site variant c.459 + 2T>C in UCHL1. Additional features observed were recurrent childhood seizures in one of the probands (P1), facial dysmorphisms, and microcephaly in both the siblings. Cognitive functions appeared to be normal in both the siblings (during their home-visit interaction), similar to that observed by Rydning et al. group in their patients, although measured IQ scores were low (at young age). Apparent discrepancy of observed cognitive abilities (home-visit interaction and formal IQ testing) might be subjective and confounded by the presence of dysarthria, optic atrophy, and method used to measure IQ. The clinical features of the previously reported patients are comparable to that of the probands described in our study. However, progression seems to be more severe relative to previously reported two families, as both siblings were non-ambulatory at an early age (first decade). Both patients showed similar facial dysmorphic features, myopathic face and microcephaly, which have not been discussed in two previously reported families. Further studies are needed to understand facial dysmorphic features as a clinical feature of UCHL1 gene mutations. Additional neuroimaging findings in our patient family (P2) and those unreported in previous two families, include large cisterna magna and bilateral areas of gliosis in occipital white matter (Fig. 1f–m). The hyperintense signals in occipital region do not appear to be due to hypomyelination. A comparison of clinical and molecular features of the probands with the previously described cases with mutations in UCHL1 is given in Table 1. The splice-site variant identified by exome sequencing in intron 6 of the UCHL1 gene, is likely to affect the invariant 5′ donor splice-site of exon 6.

One of the limitation of this study involves the inability to ascertain further functional assessment of the variant to see the actual effect of the variant in splicing aberrations, however, a distinct phenotype–genotype correlation and variation in the consensus splice-site region ensures a possible deleterious effect of the variant in the disease etiology. Another discrepancy is that we could not show complete familial co-segregation as father and normal sibling did not give their consent to participate in the study.

Finally, we report a novel UCHL1 splice-site variant in multiple members of a family of Indian origin affected with SPG79. This finding substantiates that mutations in UCHL1 gene are the cause of this rare phenotype, and should be considered in the diagnosis of patients with early-onset optic atrophy and subsequent progressive neurodegenerative symptoms.