Preservation of Proteomic Stability and Promotion of Protein Lipidation by HSF1

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

One of our aims is to examine the role of HSF1 in repressing proteomic instability in cancer. Our preliminary results show that HSF1 knockdown induced global protein ubiquitination, aggregation, and even amyloid formation in various cancer cells, particularly NF1-deficient malignant peripheral nerve sheath tumor (MPNST) cells. Accompanied with this proteomic instability is impairment in cell viability. Importantly, Congo red (CR) treatment, which can block amyloidogenesis, partially rescued this impaired viability. In stark contrast, HSF1 is dispensable for the proteome of non-transformed human Schwann cells. Mechanistically, in MPNST cells HSF1 defends the essential mitochondrial chaperone HSP60 against the direct assault from soluble amyloid oligomers. To survive and adapt to impaired protein quality, owing to HSF1 deficiency, MPNST cells mobilized JNK to repress mTORC1 and protein translation, thereby attenuating protein quantity to alleviate proteomic imbalance. mTORC1 stimulation, via either pharmacological JNK blockade, genetic TSC2 depletion, or supplement with the leucine analog NV-5138, markedly aggravated the proteomic imbalance elicited by HSF1 deficiency. This profound proteomic imbalance instigated cell death, which can be largely blocked by either translation inhibition or CR treatment. Together, our study highlights driving proteomic imbalance as a new therapeutic concept to combat malignancies. Based on these results, we are currently interrogating: 1) whether combined HSF1 inhibition and mTORC1 stimulation can effectively counter in vivo tumor growth in MPNST xenograft models. In the last fiscal year, we have successfully established MPNST orthotopic xenograft model by transplanting human MPNST cells into the sciatic nerves of NIH-III nude mice and monitored tumor growth using bioluminescence imaging. To generalize our notion, we also established a syngeneic melanoma model using B16F10 cells in C57BL/6J mice. We have conducted pilot experiments to determine the doses of the recently developed HSF1 inhibitor DTHIB in vivo. 2) the precise mechanisms underlying this amyloid-dependent cell death. Interestingly, our series of experiments have confirmed that this cell death is synchronized and characterized by cell membrane permeabilization. Our further experiments exclude other common cell death mechanisms, including apoptosis, necroptosis, ferroptosis, pyroptosis, as well as autophagic cell death. Our second aim is to examine the role of HSF1 in regulating gene expression beyond the heat shock response. Our studies revealed that HSF1 acts as a prime modifier of the c-MYC-mediated transcription. HSF1 deficiency diminishes c-MYC DNA binding and dampens its transcriptional activity genome-widely. Mechanistically, c-MYC, MAX, and HSF1 assemble into a transcription factor complex on genomic DNAs. Within this complex, HSF1 physically recruits the histone acetyltransferase general control nonderepressible 5 (GCN5), promoting histone acetylation and augmenting c-MYC transcriptional activity. Thus, our studies reveal that HSF1 specifically potentiates the c-MYC-mediated transcription, discrete from its canonical role in countering proteotoxic stress. Based on these findings, we are currently investigating: 1) how heat shock and other proteotoxic stressors affect this HSF1-mediated c-MYC activation; and 2) whether c-MYC can reciprocally affect the transcriptional activity of HSF1. In the last fiscal year, we have found that upon heat shock c-MYC and HSF1 interactions are disrupted, accompanied by a genome-wide reduction in the expression of c-MYC target genes. Intriguingly, heat shock resulted in c-MYC protein aggregation, becoming detergent-insoluble. By stark contrast, its canonical binding partner MAX remains detergent soluble. Our studies further reveal that recombinant c-MYC proteins, but not MAX, become aggregates in vitro. Importantly, recombinant c-MYC/MAX dimers do not form aggregates in vitro under the same condition. Moreover, by using a synthetic peptide library we have identified two aggregation-prone regions on c-MYC proteins.