Transmembrane protein 106B, a risk factor in frontotemporal lobar degeneration, is a lysosomal type II transmembrane protein and affects autophagy

摘要

Frontotemporal lobar degeneration (FTLD) is a fatal neurodegenerative disease with presenile onset. Clinically, it mainly presents with language disorders or personality and behavioural changes whereas pathologically patients Show atrophy of the frontal and temporal lobes of the brain. Like in other neurodegenerative disorders, abnormal protein deposition can be detected in the affected areas of the brain nervous system. However, several different proteins have been identified to be the main component of These inclusions accordingly leading to the differentiation of so far five distinct types of FTLD, namely FTLD-tau, FTLD-TDP (TAR DNA-binding protein 43), FTLD-FUS (Fused in Sarcoma), FTLD-PR (dipeptide repeat protein) and FTLDUPS (ubiquitin-proteasome system). FTLD-TDP comprises 45 % of all FTLD cases and thus represents one of the two main pathological subtypes of FTLD. In the last few years, tremendous progress has been made in the identification of the genetic causes for FTLD-subtypes; among them, the identification of mutations in the progranulin (GRN) gene in FTLD-TDP. Interestingly, even though haploinsufficiency of progranulin was demonstrated to be causative for FTLD-TDP, the same GRN mutation could present with different ages of disease onset in different patients. This argued for additional factors that might modulate disease onset. In order to identify such genetic factors, a genome-wide association study was performed in genetically or pathologically confirmed FTLD-TDP cases. Thereby, twelve single-nucleotide polymorphisms mapped to a 68 kb interval located on chromosome 7p21.3 implicating that this might be a common genetic susceptibility locus for FTLD-TDP. This region only comprised one gene encoding for the transmembrane protein 106B (TMEM106B). udInterestingly, the risk allele of TMEM106B was especially associated with FTLD risk in patients carrying a GRN mutation which suggested a functional relationship between those two proteins. However, TMEM106B was an uncharacterized protein of unknown function. Thus, the Motivation of my study was to investigate the biochemical features of TMEM106B, followed by examining the relationship between TMEM106B and GRN and finally, by investigating TMEM106B function.udIn the first part of this study, Membrane orientation, cellular localization and the glycosylation status of TMEM106B were determined and tools developed. By sequential inactivation of the five predicted N-glycosylation motifs, TMEM106B was demonstrated to be a type II transmembrane protein that is N-glycosylated at the amino acid positions 146 (N1), 152 (N2), 165 (N3), 184 (N4)and 257 (N5). Moreover, only N4 and N5 proved to be complex glycosylated whereas N1, N2 and N3 did not. By immunofluorescence, TMEM106B was determined to be a lysosomal protein.udInterestingly, mutants where one of the two complex glycosylation motifs was deleted showed a different intracellular localization whereas deleting the non-complex glycosylation motifs did not change TMEM106B localization. This indicated that complex glycosylation was essential for correct TMEM106B positioning in the cell. udIn the second part of this study, the influence of TMEM106B expression on GRN levels was analysed in various cell lines. However, neither overexpression nor knockdown of TMEM106B changed intracellular or secreted GRN levels indicating that both proteins probably do not influence each other directly. However, interestingly, bafilomycin A1 (BafA1) treatment which inhibits lysosomal acidification and thus lysosomal function increased both GRN and TMEM106B protein levels suggesting that both proteins might act in a common pathway or might be located in the same compartment. Since treatment with proteasomal inhibitors did not increase TMEM106B levels, this observation further indicated that TMEM106B is mainly degraded by the lysosome.udIn the third part of this study, the endogenous function of TMEM106B was investigated using siRNA-mediated TMEM106B knockdown in a cell culture model. Thereby, TMEM106B knockdown was shown to change lysosomal positioning as lysosomes clustered tightly at the microtubule-organizing center instead of being distributed throughout the cell. A rescue experiment, where endogenous TMEM106B was knocked down first and then, additionally, either a control vector or exogenous TMEM106B was transfected, proved that lysosomal clustering was the result of TMEM106B loss and not a side effect of siRNA transfection.udFurthermore, lysosomal clustering upon TMEM106B knockdown was shown to be dependent on functional retrograde transport and an intact microtubule network. In addition, lysosomes were demonstrated to be still acidic and, in principle, functional upon TMEM106B knockdown.udInterestingly, however, lysosomal and autophagosomal protein levels increased significantly upon TMEM106B knockdown, suggesting that the autophagic pathway might be affected by TMEM106B levels. Since GRN had been implicated in playing an important role for lysosomal function and thus in the autophagic pathway, the finding that TMEM106B also has an impact on this pathway might explain why TMEM106B polymorphisms are especially associated with GRN mutation carriers but also why TMEM106B is a general risk factor for FTLD.udChanges in the autophagic pathway seem to be common in neurodegenerative disorders as for example in Alzheimer’s, Parkinson’s and Huntington’s disease, the autophagic pathway has been reported to be impaired in the course of disease. My findings would support the notion that also in FTLD, autophagy plays an essential part in disease progression.