The lab continues to use in vitro systems and model organisms to study the function of PKD1, PKD2 and PKHD1, the genes responsible for the most common forms of autosomal dominant and autosomal recessive polycystic kidney disease. In last year's report, we had described finding additional evidence indicating that mutant cells had altered cellular metabolism, and this was associated with abnormal mitochondrial structure in both immortalized and freshly isolated renal epithelial cells from Pkd1 mutant mice and in human ADPKD kidneys. We further found that Pkd1 mutant cells had a small but significant increase in mitochondria membrane potential as assayed using tetramethyl-rhodamine, methyl ester (TMRM) and increased fragmentation of their mitochondrial network. To gain insights into how polycystin-1 (PC1), the protein encoded by PKD1 (the gene most commonly mutated in autosomal dominant polycystic kidney disease), could regulate mitochondrial and metabolic functions, we analyzed PC1 subcellular localization using live cell imaging and biochemical methods. Since PC1 is a large protein that undergoes complicated cleavage, we generated constructs with distinct N-terminal and C-terminal fluorescent tags for these studies. By live cell imaging of cells transfected with the dual fluorescently-tagged PC1 construct, we observed three patterns. In the majority of cells, only the N-terminal tag was visible and it was present predominantly in the ER. This result was consistent with immunoblot analyses that showed that the cleaved N-terminus is much more abundant than either full length PC1 or its cleaved C-terminus. A smaller number of cells was found to express both the N-terminal and C-terminal tags. In these cells, the tags mostly co-localized to the ER but sometimes could be seen segregating to distinct subcellular regions. Finally, there was a rare set of cells with the NTF localized to the ER and a distinct mitochondrial pattern for the C-terminal tag. Using a set of truncation constructs, we identified a putative mitochondrial targeting sequence that was both necessary and sufficient for the mitochondrial localization of the C-terminal cleavage product of polycystin-1 (CTT). We used a split GFP assay to show that the CTT localized to the mitochondrial matrix, and we found that transient expression of CTT partially rescued mitochondrial network structure in three Pkd1 mutant renal epithelial cell lines. Using Drosophila to model in vivo effects, we found that transgenic expression of mouse CTT resulted in decreased viability and exercise endurance but increased CO2 production, consistent with altered mitochondrial function. We published a summary of these findings this year. While our work was in review, another group also reported mitochondrial abnormalities in ADPKD, independently validating at least part of our study. Our results, however, also differ in important ways with other published reports. Two groups had previously described a nuclear localization pattern for the CTT, something that we did not observe, even when we used constructs identical to those published. Another group had previously localized PC1 to ER-mitochondrial-attachment sites based on cell fractionation studies. We cannot exclude this possibility, particularly given the high level of expression of PC1 in the ER, but in contrast to their findings we have unambiguously shown that a fragment of PC1 enters the mitochondria. Our current working model is that PC1 may play a direct role in regulating mitochondrial function and cellular metabolism but rather than serving to govern cellular energetics, its function may be to transduce signals, perhaps mechanical, by releasing a PC1-CTT fragment that fine-tunes mitochondrial function and acetyl-CoA levels. These outputs mediate epigenetic or cytoskeletal changes that ultimately determine cell shape and tubule diameter. In ongoing work, we have collaborated with the NHLBI mouse transgenic core to make a mouse line with an EGFP knocked into the C-terminus of PC1 by CRISPR. While our initial founder failed to successfully transmit the tag, a later set of mice have been successfully produced. Studies are ongoing to determine if the added sequence disrupts PC1 function and to see if we can now detect PC1 by either live cell imaging or antibody-based methods. In other studies, we have begun characterization of a putative PC1 binding partner identified by mass spectrometry. The gene encoding it was among the most highly differentially expressed in our transcriptomic studies of both early and late onset PKD mouse models. Characterization of the protein has been complicated by problems with antibodies (variable sensitivity and specificity), discordance between the proteins predicted and apparent size on immunoblot for both native and recombinant protein, and confusion about its nomenclature as several distinct genes appear to share a similar name. Despite these challenges, in preliminary studies, some of the proteins properties appear to change with cell density and knock-down of the gene in cell culture results in a cellular phenotype that appears to synergize with Pkd1 knock-down. These studies are ongoing. We had previously generated a Pkhd1 mouse model that had lox P sites flanking exons 3 and 4 and used it to show that homozygous deletion of the exons resulted in renal cystic disease and hepatic fibrocystic disease,
Showing the most recent 10 out of 33 publications
