We have found that feeding activates previously unrecognized transcriptional programs in liver and muscle designed to maintain protein folding homeostasis, or proteostasis. Heat shock factor 1 (HSF1), the master transcription factor that controls the heat shock response (HSR), regulates and is required for these normal physiological responses. The programs are inhibited by fasting, and hyperstimulated by refeeding after fasting. The setting and pattern of regulation, inhibition by fasting and activation in liver and muscle by feeding closely resembles mTOR responses. We further found that the HSF1 transcriptional program is in fact downstream of the canonical Tsc1/2-Rheb-mTORC1 axis and is required for mTOR-dependent protein synthesis. Regulation appears to go both ways. Inhibition of mTOR suppresses the HSF1 transcriptional response, and HSF1 loss- of-function in liver suppresses both mTOR activation and protein synthesis. mTOR regulation clearly distinguishes the HSF1 feeding response from a classical heat shock response, which is not regulated by mTOR. Transcriptome-wide RNA-seq results further distinguish the hepatic feeding response from the HSR, as fewer than 10% of genes are common to both responses. The RNA-seq results also show that feeding induces an unfolded protein response (UPR) in the ER, which is distinct from the cytoplasmic protein folding response (cPFR) we describe. While cytoplasmic and ER protein folding responses are distinguished by both subcellular distribution and proteins/pathways involved, they appear to be mechanistically linked, as perturbations in one affect the other (e.g. the XBP1s transcriptional program is suppressed in Hsf1 null liver). Based on our preliminary findings we hypothesize: 1) Feeding acutely increases protein synthesis and therefore the protein folding burden in liver and muscle. 2) mTOR simultaneously promotes protein synthesis and the cellular machinery for maintaining proteostasis. 3) HSF1 loss of function increases the cytoplasmic protein folding burden, which 4) suppresses mTOR-dependent protein synthesis and thereby 5) reduces the protein folding burden in both cytoplasm and ER. Similarly, 6) XBP1 loss of function induces ER stress, which suppresses mTOR dependent protein synthesis and the protein folding burden in both cytoplasm and ER, which suggest 7) cross-talk between cytoplasmic and ER protein folding responses. We contrast the physiological feeding responses in muscle and liver with the classical HSR at all levels, regulation, transcriptional programs, and effects of HSF1 on both mTOR and the ER protein folding response.
Aims 1 -3 test hypotheses related to feeding, whereas Aim 4 asks whether the proposed mechanisms extend more broadly to other settings of mTOR-dependent protein synthesis.
Aim 4 thus hypothesizes that like feeding, 1) exercise coordinately drives mTOR-dependent protein synthesis and an HSF1 transcriptional program required for proteostasis and muscle growth, and 2) cross-talk between cytoplasmic (HSF1) and ER (XBP1s) protein folding responses occurs with exercise as we had seen with feeding. Proposed studies test these hypotheses in detail.
We have found that feeding activates previously unrecognized transcriptional programs in liver and muscle required for protein synthesis and the metabolic adaptations associated with feeding. Proposed studies assess molecular mechanisms for both tphysiological feeding and exercise responses.
