The HIF complex was initially identified

The HIF complex was initially identified as an important component of the cell machinery involved in the response and pathways cu to hypoxia (Semenza, 2001). Further studies have shown that HIF can also modulate whole-body energy homeostasis by controlling hypothalamic neurons (Varela et al., 2017, Zhang et al., 2011). Specifically, the HIF complex in POMC neurons directs the real-time homeostasis of feeding and energy expenditure (Zhang et al., 2011). In addition, the induction of HIF activity in neuronal cells (neuron-specific FIH null mice) is involved in the regulation of respiration, energy balance, and lipid metabolism (Zhang et al., 2010a). Thus, HIF has emerged as an important component of the hypothalamic neuronal machinery involved in the control of body mass and metabolism (Zhang et al., 2011). With these concepts in mind, we decided to evaluate the potential involvement of the hypothalamic HIF complex in experimental DIO. In order to determine the general impact of changes in the hypothalamic levels of HIF-1 in metabolic phenotype parameters, we used bioinformatics to evaluate datasets from a large number of mouse strains (Andreux et al., 2012). Hypothalamic HIF-1 presented positive correlations with caloric intake, respirometric parameters and lean body mass, whereas a negative correlation was found for blood glucose levels. Using a human dataset, hypothalamic HIF-1 was positively correlated with hypothalamic markers of inflammation, apoptosis, metabolism, autophagy, and the ubiquitin proteasome system. Due to the magnitude of the datasets and the high reproducibility of other bioinformatics analyses performed using similar methods and the same datasets (Andreux et al., 2012, Williams et al., 2016), we propose that hypothalamic HIF-1 can be involved both in sensing and regulating nutrient status in the body and is modulated in parallel with a number of other hypothalamic markers known to be affected in DIO (Cavadas et al., 2016, Jais and Bruning, 2017, Thaler et al., 2013, Velloso and Schwartz, 2011). Next, using immunofluorescence staining, we showed that HIF-1 is highly expressed in the ARC and median eminence, with most labeling present in POMC neurons and glial cells. Previously, studies have shown that activation of the HIF pathway in POMC neurons is an important mechanism of nutrient sensing that modulates caloric intake and energy homeostasis (Varela et al., 2017, Zhang et al., 2011). In the present study we expand this concept by showing that mice fed a HFD (35% lard fat, mainly rich in saturated fatty acids) present increased hypothalamic expression of HIF-1 proteins. This occurred in parallel with the reduction of hypothalamic VHL protein levels, which is an E3 ubiquitin ligase responsible for the targeting of HIF-1α by ubiquitin for subsequent degradation by the ubiquitin/proteasome system. In DIO, ubiquitin and UCHL-1 (ubiquitin carboxy terminal hydrolase) are obesity-related factors in the hypothalamus, likely playing an important role in the genesis of obesity by interfering with the integrated signaling network that controls energy balance and feeding (Wang et al., 2011). In addition, in prolonged DIO, the hypothalamic ubiquitin/proteasome system fails to maintain an adequate rate of protein recycling, leading to the accumulation of ubiquitinated proteins (Ignacio-Souza et al., 2014). This can be one of the important mechanisms involved in the progression of obesity and occurs in a similar way to the defects in the ubiquitin/proteasome system in the hippocampus of ageing rodents (Cavadas et al., 2016). Studies have shown that the HIF-1 complex is involved in the pathophysiology of adipose tissue that leads to the development of obesity, inflammation, and insulin resistance (Halberg et al., 2009, Jiang et al., 2013, Kihira et al., 2014, Lee et al., 2014). In a model of DIO, HIF-1 activation is involved in the impairment of lipid and glucose metabolism, being involved in the biogenesis of hepatic steatosis (Ochiai et al., 2011, Rahtu-Korpela et al., 2014). Conversely, HIF-1 complex activation in the central nervous system has beneficial roles in the regulation of metabolism (Zhang et al., 2010a). A study has shown that neuronal-specific FIH (Factor Inhibiting HIF-1α) KO mice had a hypermetabolic phenotype with increased food intake and energy expenditure, reduced fat mass, improved insulin sensitivity, and reduced hepatic steatosis(Zhang et al., 2010a). However, the authors were not able to demonstrate which brain region was responsible for this metabolic regulation, since FIH was deleted in all neurons. Because many of these metabolic processes are controlled by neuronal regulatory circuits located in the hypothalamus, we hypothesized that in the ARC the HIF-1 complex could have a protective effect against hypothalamic inflammation triggered by dietary SFAs, thus regulating energy expenditure and systemic metabolism. To test this hypothesis, we inhibited the HIF-1 complex in the ARC and evaluated a number of metabolic and neuroinflammatory parameters in mice fed chow and HFD. We inhibited the β-subunit of the HIF-1 complex as this subunit is mandatory for the DNA binding and activation of both HIF-1 and HIF-2 complexes. This approach avoids a compensatory activation of HIF-1 by other isoforms. Our results show that, under HFD feeding, the downregulation of the HIF-1 complex in the ARC exacerbates the metabolic phenotype induced by SFAs, namely increased body mass and decreased energy expenditure. We did not observe changes in caloric intake, as reported elsewhere (Zhang et al., 2011, Zhang et al., 2010a). In our experiments, food intake was measured under baseline conditions (that is every morning at the same hour, without fasting or any other stimuli that could influence feeding). In a study using a POMC/HIFβlox/lox model Zhang and collaborators demonstrated that the fasting (6h or 24 h) induced food intake were the same both in the POMC/HIFβlox/lox and in wild-type mice (Fig. 2F and G). They also showed that food intake in POMC/ HIFblox/lox mice and the matched controls were suppressed by leptin in a similar manner. However, HIF in POMC neurons is required for glucose/nutrient-dependent hypothalamic regulation of feeding behavior (Zhang et al., 2011). Also, in an endothelial-HIF KO mice (HIF-1ec) food intake under baseline conditions was the same as in WT mice (Varela et al., 2017); however, only in fasting condition there was a significantly elevated increase in food intake in HIF-1αEC mice relative to controls (Varela et al., 2017). The difference in the experimental protocols for measurement of food intake can produce different results. Another point is that in our model, HIF-1 was inhibited in the hypothalamus and not only in POMC neurons.