Role of Thioredoxin-Interacting Protein (TXNIP) in Regulating Redox Balance and Mitochondrial Function in Skeletal Muscle

The Muoio lab studies the interplay between lipid whole body energy balance,

mitochondrial function and insulin action in skeletal muscle. Data from our lab suggests that lipid-induced insulin resistance in skeletal muscle may stem from excessive incomplete oxidation of fatty acids, which occurs when high rates of β-­oxidation exceed TCA cycle flux (Koves et al., 2005; Koves et al., 2008). Most notably, we have shown that mice with a genetically engineered decrease in mitochondrial uptake and oxidation of fatty acids are protected against diet-­induced insulin resistance (Koves et al., 2008). This

suggests that an excessive and/or inappropriate metabolic burden on muscle

mitochondria provokes insulin resistance. Our working model predicts that: 1) high rates of incomplete β-oxidation reflect a state of ”mitochondrial stress,” and 2) that energy-overloaded mitochondria generate a yet unidentified signal that mediates insulin

resistance. One possibility is that this putative mitochondrial-derived signal stems from redox imbalance and disruptions in redox sensitive signaling cascades. Therefore, we are interested in identifying molecules that link redox balance, mitochondrial function and insulin action in skeletal muscle. The work described herein identifies thioredoxin-interacting protein (TXNIP) as an attractive candidate that regulates both glucose homeostasis and mitochondrial fuel selection.

TXNIP is a redox sensitive, α-arrestin protein that has been implicated as a negative regulator of glucose control. Mounting evidence suggested that TXNIP might play a key role in regulating mitochondrial function; however, the molecular nature of this relationship was poorly defined. Previous studies in TXNIP knockout mice reported that deficiency of this protein compromises oxidative metabolism, increases glycolytic activity and promotes production of reactive oxygen species (ROS), while also affording protection against insulin resistance. Therefore, we hypothesized that TXNIP might serve as a nutrient sensor that couples cellular redox status to the adjustments in mitochondrial function. We tested this hypothesis by exploiting loss of function models to evaluate the effects of TXNIP deficiency on mitochondrial metabolism and respiratory function.

In chapter 3, we comprehensively evaluated oxidative metabolism, substrate

selection, respiratory kinetics and redox balance in mice with total body and skeletal muscle-­specific TXNIP deficiency. Targeted metabolomics, comprehensive bioenergetics analysis, whole-body respirometry and conventional biochemistry showed that TXNIP deficiency results in reduced exercise tolerance with marked impairments in skeletal muscle oxidative metabolism. The deficits in substrate oxidation were not secondary to decreased mitochondrial mass or increased H₂O₂ emitting potential from the electron transport chain. Instead, the activities of several mitochondrial dehydrogenases involved in branched-chain amino acid and ketone catabolism, the tricarboxylic acid (TCA) cycle and fatty acid β-oxidation were significantly diminished in TXNIP null muscles. These deficits in mitochondrial enzyme activities were accompanied by decreased protein abundance without changes in mRNA expression. Taken together, these results suggest that in skeletal muscle TXNIP plays an essential role in maintaining protein synthesis and/or stability of a subset of mitochondrial dehydrogenase enzymes that permit muscle use of alternate fuels under conditions of glucose deprivation.

Based on these conclusions, we questioned whether additional regulatory

mechanisms could contribute to the reduced oxidative metabolism in the absence ofTXNIP. Several metabolic enzymes of the TCA cycle have been shown to be redox-sensitive protein targets regulated by the thioredoxin (TRX1/TRX2) and glutathione (GSH) redox-mediated circuits. TXNIP has been shown to respond to oxidative stress by shuttling to the mitochondria where it binds to TRX2 and/or other proteins, thus affecting downstream signaling pathways, such as the apoptotic cascade. Therefore, we speculated whether there was a role for redox imbalance in mediating the mitochondrial phenotype of the TXNIP knockout (TKO) mice. In chapter 4, we present preliminary evidence that increased glucose uptake promotes non-mitochondrial ROS production, causing a shift in redox balance, decreased GSH/GSSG, and S-glutathionylation of α-­ketoglutarate dehydrogenase (&alpha-KGD). This post-translational modification protects the protein from permanent oxidative damage, but at the cost of reversible loss of activity and subsequent disruption of TCA cycle flux that contributes, in part, to the diminished oxidative metabolism observed in the TXNIP deficient mice.

In aggregate, this work sheds new light onto the physiological role of TXNIP in

skeletal muscle as it pertains to substrate metabolism and fuel switching in response to nutrient availability. This work has important implications for metabolic diseases such as obesity and type 2 diabetes, which are characterized by marked disruptions in fuel selection.