Article Text

Download PDFPDF
Original article
KIF1C mutations in two families with hereditary spastic paraparesis and cerebellar dysfunction
  1. Talya Dor1,
  2. Yuval Cinnamon2,
  3. Laure Raymond3,4,
  4. Avraham Shaag2,
  5. Naima Bouslam3,5,
  6. Ahmed Bouhouche5,
  7. Marion Gaussen3,4,
  8. Vincent Meyer6,
  9. Alexandra Durr3,7,
  10. Alexis Brice3,7,
  11. Ali Benomar5,8,
  12. Giovanni Stevanin3,4,7,
  13. Markus Schuelke9,
  14. Simon Edvardson1
  1. 1Department of Pediatrics, Neuropediatric Unit, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
  2. 2Department of Genetic Research, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
  3. 3Centre de Recherche de l'Institut du Cerveau et de la Moelle épinière, UPMC Univ Paris VI UMR_S975; CNRS UMR 7225; INSERM U975, Hôpital de la Pitié-Salpêtrière, Paris, France
  4. 4Groupe de Neurogénétique, Ecole Pratique des Hautes Etudes, Institut du Cerveau et de la Moelle épinière, Paris, France
  5. 5Faculté de Médecine et de Pharmacie de Rabat, Université Mohammed V Souissi, Equipe de recherche des maladies neurodégéneratives (ERMN), Rabat, Morocco
  6. 6Genoscope, Evry, France
  7. 7APHP, Fédération de Génétique, Hôpital de la Pitié-Salpêtrière, Paris, France
  8. 8Faculté de Médecine et de Pharmacie de Rabat, Université Mohammed V Souissi, Centre de recherche en épidémiologie clinique et essai thérapeutique (CRECET), Rabat, Morocco
  9. 9Department of Neuropediatrics and NeuroCure Clinical Research Center, Charité Universitätsmedizin Berlin, Augustenburger Platz 1, Berlin, Germany
  1. Correspondence to Professor Markus Schuelke, Department of Neuropediatrics, Charité Universitäts medizin Berlin, Augustenburger Platz 1, Berlin D-13353, Germany; markus.schuelke{at} Dr. Simon Edvardson, Department of Pediatrics, Neuropediatric unit, Hebrew University Medical Center, 91120 Jerusalem, Israel; simon{at}


Background Hereditary spastic paraparesis (HSP) (syn. Hereditary spastic paraplegia, SPG) are a group of genetic disorders characterised by spasticity of the lower limbs due to pyramidal tract dysfunction. Nearly 60 disease loci have been identified, which include mutations in two genes (KIF5A and KIF1A) that encode motor proteins of the kinesin superfamily. Here we report a novel genetic defect in KIF1C of patients with spastic paraparesis and cerebellar dysfunction in two consanguineous families of Palestinian and Moroccan ancestry.

Methods and results We performed autozygosity mapping in a Palestinian and classic linkage analysis in a Moroccan family and found a locus on chromosome 17 that had previously been associated with spastic ataxia type 2 (SPAX2, OMIM %611302). Whole-exome sequencing revealed two homozygous mutations in KIF1C that were absent among controls: a nonsense mutation (c.2191C>T, p.Arg731*) that segregated with the disease phenotype in the Palestinian kindred resulted in the entire absence of KIF1C protein from the patient's fibroblasts, and a missense variant (c.505C>T, p.Arg169Trp) affecting a conserved amino acid of the motor domain that was found in the Moroccan kindred.

Conclusions Kinesin genes encode a family of cargo/motor proteins and are known to cause HSP if mutated. Here we identified nonsense and missense mutations in a further member of this protein family. The KIF1C mutation is associated with a HSP subtype (SPAX2/SAX2) that combines spastic paraplegia and weakness with cerebellar dysfunction.

  • Movement Disorders (other than Parkinsons)
  • Neurology
  • Clinical Genetics
  • Genetics

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Hereditary spastic paraplegia (HSP) is a heterogeneous neurological condition, which had already been described at the end of the nineteenth century by Adolf von Strümpell (1880) and Sigmund Freud (1893).1 ,2 Clinically, the condition can be classified into ‘pure’ and ‘complicated’ forms of HSP.3 ,4 In pure HSP, pyramidal tract signs are caused by corticospinal axonal degeneration and are the only clinical manifestation. In complicated HSP, the pyramidal tract signs are accompanied by other findings, including cerebellar signs, extrapyramidal signs, muscle weakness, sensory deficit, retinal involvement and cognitive decline.5 The age of onset of HSP ranges from early childhood to the third decade of life.

The mode of inheritance of HSP is variable, including autosomal dominant, recessive, as well as X-linked subtypes.5–7 So far, 57 distinct loci have been associated with the disease (SPG1–57), and mutations in specific genes were found in nearly half of them. Genes linked to HSPs are often involved in organelle and microtubule dynamics,6 and are thought to affect transport of cargo in neurons,8 ,9 homeostasis of the endoplasmic reticulum6 ,10 or neuronal signal transduction.11 ,12 Among the genes affecting microtubule transport, mutations in the kinesin family members KIF1A and KIF5A have been shown to cause HSP.9 ,13–15

Here we report the result of the molecular investigation of two consanguineous families with childhood or juvenile onset, autosomal-recessive HSP with spastic paraparesis and cerebellar dysfunction. We identified in KIF1C a nonsense mutation that entirely abolished protein expression and a missense mutation in a highly conserved domain of the protein.

Patients and methods


Subjects from two families were recruited for this study. The parents provided written informed consent for all aspects of the study according to the Declaration of Helsinki. The study was approved by the ethical review boards of Paris-Necker Hospital, the Hadassah Medical Center and the Israeli Ministry of Health.

Four patients originated from two branches of a consanguineous Palestinian family as depicted in the corresponding pedigree (figure 1A). Three are siblings, two males (V_1, V_2) and a female (V_4), and the fourth (V_6) is their male cousin. The four patients had similar clinical manifestations. Pregnancy and birth were normal. Following an initial normal development, the disease manifested between 6 and 10 years of age with tremor, dysarthria and frequent falls. Initial examination revealed pyramidal signs, including lower limb spasticity, hyperreflexia, extensor plantar reflexes and clonus, as well as cerebellar signs comprising gaze nystagmus, ataxia, head titubation, dysmetria and tremor. The lower limb spasticity did not progress and allowed independent walking, although at slow pace with frequent falls. In contrast, the cerebellar symptoms progressed during the 5 years of follow-up. Dysarthria was noted in all patients, with different severity. Speech remained intelligible up to the latest evaluation (patient V_4 at 12 years of age and patient V_1 at 24 years of age). Cognitive function remained intact. Extrapyramidal symptoms and sphincter disturbances were absent. Upper limbs were unaffected, and the patients remained independent in their daily activities. The salient features are summarised in table 1. No evidence of a systemic metabolic disease was evident through analysis of plasma amino acids, lactate, acylcarnitines and urinary organic acids. Brain MRI of patient V_1 at the age of 18 years ruled out atrophy of the cerebellum, the Corpus callosum and the brainstem. However, in the posterior limb of the Capsula interna and the occipital white matter, a mild degree of demyelination was present (figure 2). We conclude that the patients have an autosomal-recessive form of complicated HSP with lower limb spasticity and cerebellar dysfunction.

Table 1

Clinical phenotypes and disease manifesting age of the patients

Figure 1

Molecular genetic findings in the patients. (A) Pedigree and (B) Sanger sequencing results of the Palestinian family. (C) Western blot analysis of protein extracts from cultured control skin fibroblasts and from HEK293 cells reveals an expected single band at ≈150 kDa, which is entirely absent in the patients’ skin fibroblasts. The same membrane was re-probed with anti-α-tubulin antibodies to verify comparable loading. (D) Pedigree and (E) Sanger sequencing results of the Moroccan family. (F) Multispecies alignment of the protein domain around the p.Arg169Trp mutation of the Moroccan family demonstrates the high evolutionary conservation of Arg169 and its neighbouring amino acids. Access the article online to view this figure in colour.

Figure 2

Brain MRI of the Palestinian patient (V_1) at the age of 18 years. (A) Sagittal T1-weighted MRI rules out atrophy of the Corpus callosum, cerebellum and brainstem. Areas of mild demyelination (arrowheads) can be seen in the (B) T2-weighted and (C) FLAIR axial images in the posterior limb of the Capsula interna and the posterior white matter.

Three patients were born from third degree cousins of Moroccan origin. They presented with childhood or juvenile onset (1–16 years of age) cerebellar ataxia with various degrees of pyramidal signs. The initial symptoms were frequent falls, gait instability and/or head tremor. The disease was slowly progressive without affecting the ability to walk after 15–24 years of disease duration. Cervical dystonia, nystagmus, neuropathy and reduced visual acuity were each observed in only one patient. Pyramidal signs included a positive Babinski sign and hyperreflexia in one patient. Cranial CT identified cerebral and cerebellar atrophy in one patient at the age of 22 years. None presented with intellectual deterioration, sphincter disturbances or deep sensory loss.


Autozygosity mapping

DNA of all four affected individuals of the Palestinian family was subjected to autozygosity mapping using the Affymetrix Genechip Human Mapping 250k NspI single nucleotide polymorphism (SNP) array, as previously described,16 and delineating the regions that were homozygous in all four patients using the HomozygosityMapper2012 ( software.17 DNA samples of three patients, two healthy parents and one unaffected sibling from the Moroccan family, were genotyped using the whole genome Illumina Linkage_12 microarray, and regions co-segregating with the disease were identified using the MERLIN software, as described.18

Whole exome analysis

Exonic sequences were enriched from the DNA sample of the Palestinian patient V_2 using the Agilent SureSelect V4 Human All Exon 51 Mb Kit (Agilent Technologies, Santa Clara, California, USA). V.3 of this kit was used in two patients from the Moroccan kindred (II_1, II_7). Sequencing was performed on a HiSeq2000 (Illumina, San Diego, California, USA), which produced 133, 68 and 38 million 100 bp paired-end reads for patients V_2, II_1 and II_7. The FASTQ files were aligned to the human GRCh37.p5 (hg19/Ensembl 67) genome sequence using the Geneious v6.0.5,19 or the BWA software.20 The alignment was first performed as an ungapped alignment allowing a maximum of five mismatches per paired-end read. For the alignment of the remaining sequences, a maximum of three gaps with a maximum length of 1000 bp were allowed. Finally, variants were called in regions with >3× coverage and >25% variant reads within the coding exons ±50 bp flanking regions using the Genome Analysis Toolkit (GATK) software package.21 ,22

Western blot

Skin fibroblasts were obtained from a biopsy taken from the Palestinian patient V_6, and 15 or 30 µg of total protein extract were separated by SDS-PAGE in parallel to samples from control cells and HEK293 protein extract. Western blot was performed using rabbit antihuman KIF1C polyclonal antibody (Abcam ab125903, dilution 1:1000).

Results and discussion

In order to identify the disease-causing gene in the Palestinian family, we genotyped 250k SNPs in the DNA samples of the four patients and searched for autozygous regions. This analysis revealed two regions with homozygous stretches of >100 SNPs—chr8:33938017–34843712 and chr17:4375102–6524298 (hg19)—spanning a total of 3.1 Mb containing 59 genes, 9 of them pseudogenes. We therefore opted for whole exome sequencing of DNA from patient V_2.

SNP calling revealed 47 267 variants in total, 127 of them located in the autozygous regions. The average coverage of the exons within the linked regions was 97×, 97.6% of them were covered >10×. A *.vcf-file containing all 127 variants (single-nucleotide exchanges, deletions, insertions, and InDels) was submitted to the variant evaluating software MutationTaster (,23 which discovered a single homozygous ‘disease-causing’ variant at chr17:4925567C>T. This variant affects exon 22 of the kinesin family member 1C gene (KIF1C NM_006612 or ENST00000320785, c.2191C>T), is not listed in dbSNP build 132, does not occur in the 1000 genome project24 and causes a premature termination codon (p.Arg731*). The variant was represented by 23 reads, 100% of which contained the mutant T allele. Segregation of this mutation with the phenotype was validated in the entire family by Sanger sequencing using the BigDye Terminator protocol on an ABI capillary sequencer (Life Technologies) (figure 1B).

To confirm that KIF1C protein would be truncated or absent in the patients, we performed Western blot analysis on cultured skin fibroblasts that had been isolated from Palestinian patient V_6. As shown in figure 1C, while in control cells it was clearly present, patient fibroblasts entirely lacked KIF1C protein. Of interest, KIF1C has 1103 amino acids with a predicted molecular size of ≈123 kDa. The predicted size of the p.Arg731* mutated protein would be ≈81 kDa. As the antibody was raised against a peptide encoding AA#452–718 of KIF1C (see ‘Methods’ section), it should have been able to bind to the truncated protein. Absence of the protein band in the patients indicates that either nonsense-mediated mRNA decay had taken place, the truncated protein was degraded or could not be detected by the antibody any more due to structural changes of remaining N-terminal epitopes.

The whole genome linkage analysis performed in the Moroccan kindred using 6090 SNPs identified 10 homozygous regions segregating with the disease, including a region on chromosome 17 that was confirmed using additional microsatellite markers and was the only one associated with a previously reported recessive CNS condition (SPAX2 , OMIM %611302; or SAX, HGNC).25 A maximum and significant multipoint LOD score of 3.13 was obtained in this 1.32 Mb region between the markers D17S1876 and D17S1854 containing 46 genes. Exon sequencing of individuals II_1 and II_7 resulted in a mean coverage of 47 and 44× with 58 and 55% of the regions covered at least 30×. A total of 45 648 and 45 255 SNPs were identified in the patients, but only 89 were homozygous, rare (<1%) in HapMap and the 1000 genome cohorts and shared by both patients. Four of those variants were located in coding sequences of the autozygous regions, but three variants in DEFB126 and C21orf49 were frequent in controls in our local exome database of North African individuals, even in the homozygous state, which makes them unlikely candidates for disease mutations. The candidate variant remaining was a missense variant in exon 7 of KIF1C, c.505C>T (p.Arg169Trp), in the significantly linked interval on chromosome 17. The amino acid change was (i) predicted to be deleterious by MutationTaster and SIFT, (ii) segregated with the disease in the pedigree (figure 1D) and (iii) was absent in 8503 exomes (2000 local exomes and 6503 exomes at and in 120 North African control chromosomes. The mutation was located in the motor domain of the protein and affected a highly conserved amino acid motif (figure 1F).

The previous description of missense mutations in the kinesin family of motor proteins in two forms of HSP supported the idea that a kinesin mutation might be responsible for the present disease as well. We have recently identified missense mutations in KIF1A in association with a form of recessive, pure HSP,13 ,26 and missense mutations in KIF5A were described in an autosomal-dominant form of a complicated HSP.9 ,14 ,15 The biochemical defect in kinesin-related HSP is thought to involve protein trafficking along axons, although the scope of involvement of this large gene family in axonal degeneration or secondary demyelination is only partly understood.

Interestingly, no phenotype was observed when the murine Kif1c was deleted.27 We speculate that in the central nervous system of mice, redundant expression of other kinesins might have masked the requirement for Kif1c or the shortness of the axons in mice might render these small animals less susceptible to defects involving long-haul axonal transport. This might also explain why in humans the lower limbs are preferably affected. Alternatively, fine motor defects such as tremor might have been missed in the analysis of mutant mice.

Recent studies in different model systems have described two functions of KIF1C in neuronal development and homeostasis. First, KIF1C was associated with migration of cells during development, including neurons.28 We believe that this function of KIF1C is unlikely to be the cause of the present disease because embryonic and early postnatal development in the patients was normal in all but one patient (II_7 from Morocco). Rather, degeneration occurred during later stages of development when developmental neuronal migration is presumably complete and the axons subsequently only elongate.

In another recent study carried out in zebrafish, kif1c was found to be important for anterograde movement of secretory vesicles in young neurons.8 Deficiency in a related function might be the underlying pathophysiology of the condition we characterise here.

The involvement of the cerebellum in our patients is not surprising since patients with KIF5A and KIF1A mutations also presented with mild cerebellar signs.15 ,26 Vice versa, patients with HSP sometimes exhibit additional signs of cerebellar ataxia.29 ,30

Kif1c is expressed in the cerebellum of mice,27 chimpanzees and humans (GEO profile 37684198), supporting a conserved role in mammalian cerebellar function. The major cerebellar signs comprised severe tremor and dysarthria, typical of lesions of the efferent cerebellar pathways or their connections with the Nucleus ruber and Thalamus, rather than a defect in the cerebellum per se.31 Indeed, the cranial MRI of Palestinian patient V_1 revealed no structural defects in the cerebellum. The cerebellar phenotype is therefore consistent with a predominantly functional defect of axonal transport, as might occur upon kinesin deficiency.

Interestingly, the mutation described in the Palestinian kindred results in complete loss of KIF1C protein, while previously described variants in other kinesins associated with HSP were missense mutations, presumably generating a dysfunctional protein, as it is probably the case for the missense mutation we identified in the Moroccan patients.

A similar clinical phenotype (HSP including cerebellar signs) and an overlapping area of autozygosity were described 5 years ago in another Moroccan family, unrelated to the family reported here, but no mutations were found in the coding region of KIF1C at the time.25 The locus has been designated as SPAX2 (Spastic ataxia 2) by OMIM (%611302) or as SAX2 by the HUGO Gene Nomenclature Committee (HGNC). We speculate that intronic or promoter mutations or structural rearrangements in KIF1C may underlie this disease, but the family could not be contacted to test this hypothesis.

In summary, we have identified a novel disease-causing gene in two families with a phenotype compatible with SPAX2/SAX2 that is characterised by pyramidal and cerebellar signs due to a loss of function of KIF1C. Hence, we propose to clinically classify this disease as complicated HSP.


The authors would like to thank the patients and their parents for participation in the study. They are also grateful to the DNA and cell bank of the Centre de Recherche de l'Institut du Cerveau et de la Moelle épinière (technical coordinator: Sylvie Forlani), the P3S (Technical coordinator: Wassila Carpentier) and the ICM (technical coordinator: Yannick Marie) genotyping/sequencing facilities. The excellent technical assistance of Vivi Zuri and Na'ama Lanxner is much appreciated.



  • Contributors TD and ABe phenotyped patients. YC, LR and AS performed sequencing. LR, GS and MS validated the results of the next generation sequencing. YC performed Western blotting. MG and VM generated the data from the next generation sequencing. NB, GS, ABo and MS performed whole genome linkage analysis. AD, ABr, GS, MS, SE supervised the work and obtained funding support. TD, GS, MS and SE wrote the manuscript.

  • Funding The project was funded by a grant of the Einstein Foundation Berlin (A-2011-63) to MS and SE, The Deutsche Forschungsgemeinschaft—NeuroCure Center of Excellence (Exc 257) to MS, the Joint Research Fund of Hadassah and Hebrew University (to TD), the French National Agency for Research (ANR, grant ‘SPAX’ to AD and ‘LIGENAX’ to GS), the Association Française contre les Myopathies (‘‘LIGENAX’’ to GS), the European Community (7th Framework Program, Omics call, to ABr), the Fondation Roger de Spoelberch (to ABr) and the Verum foundation (to ABr). This study also benefited from funding from the programme ‘Investissements d'avenir’ ANR-10-IAIHU-06 (to the Brain and Spine Institute, Paris). YC was generously supported by the Human Frontier Science Program.

  • Competing interests None.

  • Ethics approval The ethical review boards of the Paris-Necker Hospital, the Hadassah Medical Center and the Israeli Ministry of Health.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement PCR oligonucleotide sequences.