Anat Ben-Zvi

Acquiring new cellular states entails metabolic reprogramming driven by changes in the expression of cytosolic and mitochondrial metabolic enzymes. Most mitochondrial proteins are synthesized in the cytosol and imported into the mitochondria in a linear form, after which they are folded by a network of mitochondrial chaperones and co-chaperones. Which mitochondrial protein is dependent upon which chaperone for its folding is largely unknown. HSPD1/HSPE1 (HSP60/HSP10) are evolutionarily conserved mammalian homologues of the bacterial proteins GroEL/GroES, forming a chamber-and-lid chaperonin to facilitate the folding of client proteins. The endogenous clients of HSP60 in mammalian cells are not fully known, nor are the HSP60 clients that require HSP10 for folding. We used gene knockdown and stable isotope labelling of amino acids in cell culture (SILAC)-based proteomics to identify HSPD1 client proteins. We found that HSPD1 supports the expression of methylenetetrahydrofolate dehydrogenase 2 (MTHFD2), a key mitochondrial one-carbon (1C) pathway enzyme, in cells and tumours. In addition, HSPD1 directly folds MTHFD2 independently of its co-chaperone HSPE1. HSPD1 interacts with MTHFD2 in mitochondria, and MTHFD2 is degraded by LONP1 in HSPD1 knockdown cells. Our data show that HSPD1 is an MTHFD2 chaperone and can fold an endogenous client protein independently of HSPE1, providing a link between mitochondrial protein folding and the 1C pathway. This article is part of the Theo Murphy meeting issue ‘ProteostaSys: a systems view of proteostasis’.

The mitochondrial chaperone and co-chaperone HSP60/HSP10 (HSPD1/ HSPE1), the mammalian homologues of bacterial GroEL/GroES, form a barrel-and-lid complex to fold newly imported or misfolded mitochondrial proteins. Several lines of evidence suggest that HSPD1 and HSPE1 have diverged in function. HSPD1 and HSPE1 are found in different clusters in the ecological network of gene expression in cancer. MTHFD2 was demonstrated to be an endogenous client protein of HSPD1 that does not require HSPE1 for its folding. Newly imported proteins show differential dependencies on HSP60 and HSP10 in isolated yeast mitochondria. Finally, HSPE1 was found to regulate a mitochondrial GTPase and modulate amyloid fibril formation independently of HSPD1. Nevertheless, the extent to which HSPD1 and HSPE1 have diverged in function has not been explored. Here, we show divergent transcriptomic and metabolic responses to the depletion of HSP60 versus HSP10 in cancer cells, and divergent activation of the mitochondrial unfolded protein response across tissues in Caenorhabditis elegans. Importantly, responses to HSP10 depletion were not nested within those of HSP60 depletion. We speculate that HSPE1 has diverged from HSPD1 and may have additional new cellular functions. This article is part of the Theo Murphy meeting issue ‘ProteostaSys: a systems view of proteostasis’.

An important hallmark of aging—and particularly of neurodegeneration—is the loss of proteostasis, leading to cellular stress. However, the causal mechanisms driving this loss are unclear. We show that SIRT6 has a critical role in maintaining proteostasis. Mechanistically, SIRT6 negatively regulates global translation by controlling ribosomal genes, nucleolar function and TIP5 chromatin localization. SIRT6 deletion increases nucleolar size, rRNA production and protein translation. However, the expression of chaperones remains unchanged, failing to compensate for the excessive translation, leading to reduced folding capacity and production of aggregates. In vivo, we establish a C. elegans model ( sir‐2.4 KO) that shows reduced heat shock resistance and an accelerated age‐dependent reduction in motility. Sir‐2.4 depletion crossed with a neuron‐specific polyQ strain led to premature motility loss and premature death. These results point to proteostasis‐stress intolerance in the absence of SIRT6, that can be rescued by pharmacologically reducing protein translation rates. Our data suggest that SIRT6 deficiency results in proteostasis loss through nucleolar dysfunction. These results highlight that deficient proteostasis begins with chromatin dysregulation resulting in neurodegeneration.

Acquiring new cellular states entails metabolic reprogramming driven by changes in the expression of cytosolic and mitochondrial metabolic enzymes. Most mitochondrial proteins are synthesized in the cytosol and imported into the mitochondria in a linear form, after which they are folded by a network of mitochondrial chaperones and co-chaperones. Which mitochondrial protein is dependent upon which chaperone for its folding is largely unknown. HSPD1/HSPE1 (HSP60/HSP10) are evolutionarily conserved mammalian homologues of the bacterial proteins GroEL/GroES, forming a chamber-and-lid chaperonin to facilitate the folding of client proteins. We used gene knockdown and SILAC-based proteomics to identify HSPD1 client proteins. We found that HSPD1 supports the expression of Methylenetetrahydrofolate Dehydrogenase 2 (MTHFD2), a key essential protein in the mitochondrial one-carbon (1C) pathway, in cells and tumors, and directly folds MTHFD2, independently of its co-chaperone HSPE1. HSPD1 interacts with MTHFD2 in mitochondria, and MTHFD2 is degraded by LONP1 in HSPD1 knockdown cells. Consequently, we observed reduced nucleotide and S-adenosylmethionine (SAM) levels in HSPD1 knockdown and found minimal overlap in the transcriptional and metabolic cellular responses to HSPD1 vs. HSPE1 depletion. In C. elegans, knockout of HSP60 triggers the mitochondrial stress response in the gut, while HSP10 knockout triggers the mitochondrial stress response in muscle tissue. Our data support that HSPD1 is an MTHFD2 chaperone and that, in addition to working together, HSPD1 and HSPE1 have distinct biological functions.

The persistence of memory reflects not only how it is formed but also how forgetting is delayed. The mechanisms controlling forgetting remain obscure, and in particular it is unclear to what degree this process is actively regulated. We discovered that cold exposure delays forgetting of specific olfactory memories by more than eight-fold and that adaptation to cold abolishes this effect. To study the underlying mechanism, we performed RNA sequencing, mutant analyses and pharmacological assays. Here we show that regulation of membrane properties underlies the presence and absence of cold-induced delayed forgetting. Furthermore, lithium can also delay forgetting in cold-sensitive worms but not in cold-tolerant worms; this effect involves downregulation of the diacylglycerol pathway in AWC neurons and long-term suppression of activity in the downstream AIY interneurons. Thus, the genetic tractability of worms might be harnessed to study the mechanism of action of lithium and cold exposure and, more fundamentally, how memory is stored and lost.

Protein clearance is fundamental to proteome health. In eukaryotes, it is carried out by two highly conserved proteolytic systems, the ubiquitin-proteasome system (UPS) and the autophagy-lysosome pathway (ALP). Despite their pivotal role, the basal organization of the human protein clearance systems across tissues and cell types remains uncharacterized. Here, we interrogated this organization using diverse omics datasets. Relative to other protein-coding genes, UPS and ALP genes were more widely expressed, encoded more housekeeping proteins, and were more essential for growth, in accordance with their fundamental roles. Nevertheless, UPS and ALP subsystems had varied expression patterns. Furthermore, each system showed a layered organization. The larger layer included genes that were differentially expressed across tissues; tissue-specific upregulation was associated with tissue-specific functions, phenotypes, and disease susceptibility. The second smaller layer included genes that were stably expressed across tissues, more highly and widely expressed, had more protein interactions, and were more essential for growth, suggesting that they act as a core. Last, we compared protein clearance to other branches of the proteostasis network. Protein clearance and folding were closely coordinated across tissues and more plastic than protein synthesis. Taken together, we propose that the proteostasis network is organized hierarchically and is tailored to the proteome composition. This organization could contribute to and illuminate tissue-selective phenotypes.

HIKESHI-related hypomyelinating leukodystrophy (HHL) is a life-threatening disorder caused by homozygous pathogenic variants in HIKESHI. Symptoms include infantile onset progressive spastic dystonic quadriplegia, nystagmus, failure to thrive, diffused hypomyelination, and severe morbidity or death following febrile illness. V54L variants in HIKESHI are particularly prevalent within the Ashkenazi Jewish population. Here, we identified a novel P78S disease-causing variant in HIKESHI in a patient of Christian Arab origin, presenting with clinical and radiologic features characteristic of HHL. In silico analysis suggests that the mutated residue may affect the HIKESHI protein’s dimerization domain. We generated a comprehensive set of induced pluripotent stem cells (iPSCs) from the index case and two additional HHL patients. To investigate mechanisms potentially linked to febrile illness in HHL, we used these cells to study the heat shock (HS) response. HHL-iPSCs showed dramatically decreased levels of HIKESHI compared with healthy controls following HS. In addition, they exhibited increased HSP70 mRNA levels in response to HS, suggesting an increased sensitivity. HHL-iPSCs had impaired HSP70 translocation to the nucleus. Our results provide a human-relevant model for HHL.

David H. Koch Institute for Integrative Cancer Research, and a biology professor at MIT for the rest of her career.Susan Lindquist brought together basic and applied scientists, physicians, mathematicians, biologists, and chemists. During her 15-year career at the Whitehead Institute alone, she was a dedicated mentor to over 100 postdoctoral fellows, graduate students, and undergraduates, many of them women, guiding them to productive careers in research. This research topic on Protein Folding, Misfolding, and Degradation is dedicated to the memory of Susan Lindquist and her ground-breaking research.Zhu and Cohen reviewed the cell non-autonomous regulation of the protein homeostasis

Protein clearance is fundamental to proteome health. In eukaryotes, it is carried by two highly conserved proteolytic systems, the ubiquitin-proteasome system (UPS) and the autophagy-lysosome pathway (ALP). Despite their pivotal role, the basal organization of the human protein clearance systems across tissues and cell types remains uncharacterized. Here, we interrogated this organization using diverse omics datasets. Relative to other protein-coding genes, UPS and ALP genes were more widely expressed, encoded more housekeeping proteins, and were more essential for growth, in accordance with their fundamental roles. Most of the UPS and ALP genes were nevertheless differentially expressed across tissues, and their tissue-specific upregulation was associated with tissue-specific functions, phenotypes, and disease susceptibility. The small subset of UPS and ALP genes that was stably expressed across tissues was more highly and widely expressed and more essential for growth than other UPS and ALP genes, suggesting that it acts as a core. Lastly, we compared protein clearance to other branches of the proteostasis network. Protein clearance and folding were closely coordinated across tissues, yet both were less pivotal than protein synthesis. Taken together, we propose that the proteostasis network is organized hierarchically and is tailored to the proteome needs. This organization could contribute to and illuminate tissue-selective phenotypes.