Mutation in TECPR2 reveals a role for autophagy in hereditary spastic paraparesis

Abstract

We studied five individuals from three Jewish Bukharian families affected by an apparently autosomal-recessive form of hereditary spastic paraparesis accompanied by severe intellectual disability, fluctuating central hypoventilation, gastresophageal reflux disease, wake apnea, areflexia, and unique dysmorphic features. Exome sequencing identified one homozygous variant shared among all affected individuals and absent in controls: a 1 bp frameshift TECPR2 deletion leading to a premature stop codon and predicting significant degradation of the protein. TECPR2 has been reported as a positive regulator of autophagy. We thus examined the autophagy-related fate of two key autophagic proteins, SQSTM1 (p62) and MAP1LC3B (LC3), in skin fibroblasts of an affected individual, as compared to a healthy control, and found that both protein levels were decreased and that there was a more pronounced decrease in the lipidated form of LC3 (LC3II). siRNA knockdown of TECPR2 showed similar changes, consistent with aberrant autophagy. Our results are strengthened by the fact that autophagy dysfunction has been implicated in a number of other neurodegenerative diseases. The discovered TECPR2 mutation implicates autophagy, a central intracellular mechanism, in spastic paraparesis.

Figure 1

Pedigrees and Neuroimaging Phenotypes (A) Pedigrees of the three Jewish Bukharian families. Filled symbols represent affected individuals, roman numerals indicate generations, and numbers are serial within the family. The slash represents the deceased individual with DNA available from a remaining muscle biopsy. TECPR2 genotype status for the c.3416delT variant is indicated by ΔT/ΔT for a homozygous mutation, ΔT/T for a heterozygous mutation, and T/T for a wild-type form. Circles indicate individuals who underwent exome sequencing. (B) A T1 sagittal MRI from individual II-1 in family A shows a thin corpus callosum at the age of 3 years. (C) A T1 sagittal MRI from individual II-1 in family A shows progressive cerebellar vermian atrophy at the age of 10 years. (D) A T2 axial MRI from individual II-2 in family C at the age of 7 years shows enlarged lateral ventricles and deep sulci, indicating cerebral atrophy. (E) A T2 sagittal MRI from individual II-2 in family C at the age of 7 years shows deep cerebral sulci (indicating cerebral atrophy), a thin corpus callosum, and vermian atrophy.

Figure 2

Protein Attributes of TECPR2 Domain structure and mutation according to UniProt and NCBI conserved domains. Top: Conserved domains in the wild-type TECPR2 according to UniProt. Yellow represents WD (tryptophan-aspartic-acid dipeptide) repeats, purple indicates TECPR domains, and blue represents the polylysin tract. Bottom: The effect of the mutation. Orange indicates the modified protein sequence resulting from a frameshift (at position 1,139) leading to a premature stop codon (at position 1,212).

Figure 3

Effect of TECPR2 Mutation (A) Semiquantatative RT-PCR analysis of the fate of TECPR2 mRNA expression in COS-7 transfectants for the wild-type and c.3416delT (“del”) forms of the transcript. “Mock” indicates the empty vector, “RT(+)” is with reverse transcriptase, and “RT(−)” is without reverse transcriptase. The data reveal no effect of the deletion on PCR-amplified mRNA levels. (B) Immunoblotting of TECPR2 levels with FLAG monoclonal antibody. Notations are the same as in (A). GAPDH was used as a loading control. The data reveal a major disappearance of p. Leu1139Argfs^∗75 TECPR2. Wild-type TECPR2 is seen at the expected molecular weight (154 kDa). (C) Immunoblotting with the TECPR2 antibody. 293T and HeLa are unmodified cell lines; endogenous TECPR2 is not detectable. Labels are the same as in (A) and (B). Wild-type TECPR2, but not the p.Leu1139Argfs^∗75 form, is observed in transfected COS-7 cells. (D) Effect of proteasome inhibition with MG132 and lactacystin. Immunoblotting with FLAG monoclonal antibody for COS-7 transfectants of C-terminal FLAG-tagged TECPR2 is shown. Rescue of the altered TECPR2 and enhancement of wild-type TECPR2 are revealed upon proteasome inhibition. The following abbreviations are used: (−), mock transfected; WT, wild-type; del, variant; and GAPDH, loading control. (E) Effect of proteasome inhibition with MG132 and lactacystin. Immunoblotting with FLAG monoclonal antibody for COS-7 transfectants of N-terminal FLAG-tagged TECPR2 is shown. Notations are the same as in (D).

Figure 4

Effect of the Mutation on Autophagy Markers (A) Immunoblotting with p62 antibody and LC3 antibody for skin-fibroblast lysates of both the affected individual and the control. Cell lines were used at an early passage, and there were a maximum of seven passages. LC3I is the cytoplasmic form, and LC3II is the phosphatidylethanolamine-conjugated form. GAPDH was used as a loading control. The following abbreviations are used: basal, rich medium; str, starvation; and baf, with the lysosomal inhibitor bafilomycin A. (B) Summary of three biological replicates (Figure S3), each in duplicate, for experiments as shown in (A) (with the same notations) with the use of immunoblot scan quantitation and normalization by GAPDH control. Error bars are the SD for the four replicates. The data show across-the-board diminution of both LC3II and p62 amounts in the affected individual as compared to the control. (C) Skin fibroblasts of an affected individual (right) and a healthy control (left) were incubated under starvation conditions in the presence of bafilomycin A for 6 hr, and thin sections were visualized by transmission electron microscopy as described. Arrows show representative autophagosomes or autolysosomes and a possible slight diminution of organelle accumulation within autophagic bodies in the affected fibroblasts.

Figure 5

TECPR2 Knockdown in HeLa Cells (A) Real-time PCR amplification of cDNA from transfected HeLa cells with either nontargeting siRNA (Ctrl-siRNA) or a pool of four TECPR2 siRNAs (TP-siRNA) with the use of primers specific to TECPR2 (exons 16 and 17). (B) Immunoblotting with p62 antibody and LC3 antibody for HeLa cell lysates transfected with either nontargeting siRNA (Ctrl-siRNA) or a pool of four TECPR2 siRNAs (TP-siRNA). Notation is the same as in Figure 4A. LC3II levels are mostly affected by the knockdown, whereas the effect is less pronounced for p62. (C) Basal levels of LC3II and p62 in HeLa cells in which TECPR2 is knocked down by siRNA are viewed by immunofluorescence confocal microscopy. Notation is the same as in (B). The following colors are used: green, LC3 antibody; red, p62 antibody; and yellow, merger of both signals. (D) LC3II and p62 levels in the same cells under the conditions of starvation + bafilomycin. For nontarget siRNA, punctate perinuclear structures, stained for both p62 and LC3II, are most likely autophagosomes. TECPR2 siRNA knockdown shows strongly attenuated LC3II signal, whereas the p62 signal is only slightly diminished.