Each tissue in the body has its own range of pO2. Compared to pO2 in arterial blood (~ 104 mmHg), the average pO2 at the tissue level is ~ 40–50 mmHg (~ 5–6%). However, pO2 is much lower in the retina (2–15 mmHg; ~0.5-2%), brain (0.4 − 10mmHg; ~0-1.5%) and solid tumors (0–10 mmHg; ~0-1.5%) [30]. The physiological pO2 of AT in humans ranges between ~ 25–85 mmHg (~ 3–11%) [8, 31], whereas the pO2 level in SM ranges from ~ 8–25 mmHg (1–3%) [32,33,34,35]. Due to more variation in oxygen supply (blood flow) and consumption (metabolic rate) throughout the day, pO2 levels may fluctuate more in metabolically active tissues such as SM compared to other tissues [36]. Heterogeneity in pO2 within organs may also be present, dependent on the anatomical location. Importantly, alterations in the oxygenation of metabolic organs have been linked to changes in tissue metabolism, as discussed in detail below.

4.1 The role of altered oxygen tension in adipose tissue

The AT serves as the primary energy storage depot and plays an important role in buffering the daily influx of excess calories. Abdominal fat mass expansion results in enlargement of adipocytes (hypertrophy) [37], leading to impaired clearance of meal-derived triglycerides and lipid spillover into the circulation in individuals with obesity [38]. The excessive flux of lipids that are directed towards other tissues results in ectopic lipid accumulation in SM, pancreas, liver and heart, thereby contributing to insulin resistance, impaired insulin secretion and related cardiometabolic impairments [37, 39, 40].

Notably, when excessive energy influx leads to expansion of the AT, inadequate angiogenesis may hamper sufficient oxygen supply. Several studies have demonstrated that angiogenesis and capillary density are reduced in the AT in individuals with obesity [41,42,43], which may lead to lower oxygen supply to the tissue. Indeed, we have previously demonstrated that alterations in AT blood flow induced by local micro-infusion of vasoactive agents into the AT microenvironment were accompanied by concomitant changes in AT pO2 [43]. Interestingly, although the link between obesity and AT hypoxia has been demonstrated in rodents [44, 45], conflicting findings have been reported in human studies, showing both higher [43, 46, 47] and lower [42, 48] AT pO2 in individuals with obesity compared to normal weight, despite lower AT blood flow (that is, oxygen supply) [41,42,43], as reviewed [31]. Conflicting findings have also been reported regarding the association between AT pO2 and insulin sensitivity in humans [31]. Some cross-sectional studies have shown that AT oxygenation was positively associated with insulin sensitivity in individuals with obesity [43, 46], whereas others found an inverse association between AT oxygenation and insulin sensitivity in men and women, independent of adiposity [49]. To provide a better insight into the relationship between changes in adiposity and AT pO2, we recently performed an intervention study in humans, showing that diet-induced weight loss decreased AT pO2, which was accompanied by improved insulin sensitivity in humans with overweight/obesity [47]. Discrepancies in human studies may relate to differences in the study populations and/or techniques used to measure pO2, as well as differences in the way insulin sensitivity was measured or defined [31, 50].

4.1.1 Effects of altered oxygen tension on adipose tissue lipid metabolism

To better elucidate the effect of hypoxia on lipid metabolism in adipocytes, several in vitro studies have been performed, in which exposure to different oxygen levels has been applied either during or after adipocyte differentiation. A study that examined the effects of severe hypoxia after differentiation found that exposure of 3T3-L1 cells to 1% O2 for 24 h reduced free fatty acid (FFA) uptake and increased lipolysis [45]. In line with the reduction in FFA uptake following hypoxia exposure, there was a decrease in the expression of the fatty acid transport proteins (FATP) and CD36, as well as the transcription factors peroxisome proliferator-activated receptor γ (PPARγ) and CCAAT/enhancer-binding protein-α [45]. In line with this, lipolysis was markedly increased, whereas lipogenesis was reduced when adipocytes were exposed to 1% O2 during differentiation [51, 52].

Notably, studies investigating the effect of mild hypoxia exposure, either during or after differentiation, showed conflicting results. Whereas one study found that differentiation of 3T3-L1 adipocytes under mild hypoxia markedly increased lipogenesis and resulted in increased formation of large lipid droplets [52], another study reported that 7-days mild hypoxia exposure after differentiation resulted in a significant reduction in lipid droplet size and triglyceride content [53]. The latter study also found enhanced basal but blunted isoproterenol-induced lipolysis following hypoxia exposure [53]. Furthermore, the expression of lipogenic proteins, including fatty acid synthase, lipin-1, and PPARγ, were found to be decreased following hypoxia exposure [53]. Taken together, although conflicting results have been reported depending on the severity, duration and timing of hypoxia exposure, it is evident that altered pO2 has effects on both lipid breakdown and lipid synthesis in adipocytes (Fig. 3a).

4.1.2 Effects of altered oxygen tension on adipose tissue glucose metabolism

Hypoxia induces a switch from fat oxidation towards glucose utilization in adipocytes [54]. Many studies have demonstrated that acute exposure of adipocytes to hypoxic conditions leads to increased glucose uptake, glycolysis and upregulation of anaerobic enzymes to maintain intracellular energy supply, as reviewed [31]. Indeed, acute and severe exposure to 1% O2 has been found to increase basal glucose uptake in both human and rodent adipocytes [45, 55, 56]. However, exposure to 1% O2 for 24 h resulted in the development of insulin resistance in human and 3T3-L1 adipocytes [56].

Fig. 3

The effects of hypoxia exposure on glucose homeostasis in humans. The effects of mild physiological hypoxia exposure on metabolic and endocrine processes in (a) adipose tissue, (b) skeletal muscle and (c) the liver. AMPK, AMP-activated protein kinase. ↑, increased by hypoxia; ↓, decreased by hypoxia; ↔, unchanged by hypoxia; ⇅, conflicting evidence. Figure created with BioRender.com

study found that acute exposure to 1% O2 enhanced insulin-dependent and insulin-independent glucose uptake in 3T3-L1 adipocytes [57]. Interestingly, the latter study also demonstrated that repeated exposures to hypoxia (1% O2, 4 h/d for 4–8 days) improved insulin sensitivity, reflected by enhanced insulin-mediated glucose uptake, and increased triglyceride accumulation through enhanced protein expression of the lipogenic transcription factors PPARγ and sterol regulatory element-binding proteins(SREBP) in 3T3-L1 cells [57]. It is important to note, however, that in many of these experiments the cells were exposed to non-physiological pO2. We have recently demonstrated that prolonged exposure to low physiological oxygen levels (5% O2 for 14 days) increased basal glucose uptake in differentiated human multipotent adipose-derived stem cells [58]. In conclusion, hypoxia exposure increases basal glucose uptake, whereas effects on insulin-mediated glucose uptake in adipocytes seem dependent on the exposure conditions (that is, duration and/or severity of hypoxia) (Fig. 3a).

4.1.3 Effects of altered oxygen tension on adipokine secretion

Although AT has long been recognized as an energy storage depot, it was not until the discovery of leptin in 1994 that the concept of AT as an important endocrine organ gained acceptance [59]. Besides a role in impaired lipid metabolism, multiple lines of evidence have demonstrated that hypertrophic expansion of adipocytes is also associated with perturbations in the secretory function of AT. Increased AT expression and/or secretion of inflammatory markers, as seen in obesity, may contribute to low-grade systemic inflammation and insulin resistance [60, 61]. Of note, we have recently demonstrated that circulating immune cell populations and inflammatory gene expression in AT show distinct associations with liver and muscle insulin sensitivity [62, 63].

Several studies have shown that alterations in pO2 influence the endocrine function of adipocytes and/or AT. We have shown that mild intermittent hypoxia exposure in vivo in men with obesity for seven consecutive days (6 h/day) consistently reduced AT pO2 and altered AT protein expression, compared to normoxia exposure [26, 31, 64]. Also, multiple cell culture studies have demonstrated that the expression and secretion of adipokines are regulated by microenvironmental pO2. These studies, however, vary with respect to the severity of hypoxia, the total duration as well as pattern of hypoxic episodes, which seems to impact study outcomes [64]. Exposure to severe hypoxia (1% O2) has been shown to induce pro-inflammatory responses in adipocytes, as illustrated by the upregulation of tumor necrosis factor (TNF-α), interleukin 6 (IL-6), monocyte chemoattractant protein 1 (MCP-1), and plasminogen activator inhibitor-1 (PAI-1) gene expression, as reviewed elsewhere [30, 31]. Similarly, the expression of vascular endothelial growth factor (VEGF), which promotes angiogenesis, was stimulated by severe hypoxia in both human and rodent adipocytes [55]. Again, it is important to emphasize that many in vitro experiments have exposed cells to severe hypoxia, which may not reflect physiological conditions in humans. Application of hypoxic conditions that mimic physiological pO2 levels is important to better understand the metabolic and endocrine abnormalities in adipocytes in obesity. In a study investigating the impact of 24 h exposure to different pO2 levels in human white preadipocytes, it was observed that leptin, VEGF, and IL-6 expression increased with lower pO2, while adiponectin gene expression showed an opposite pattern [65]. Other studies have demonstrated that prolonged exposure to mild physiological oxygen levels decreased pro-inflammatory gene expression (i.e., IL‐6, PAI‐I, TNF-α, MCP‐1 and dipeptidyl‐peptidase‐4) and altered adipokine secretion in differentiated human adipocytes as compared to 21% and/or 10% O2 [58, 66]. We have previously shown that AT pO2 was lower in femoral compared to abdominal subcutaneous AT in postmenopausal women [58]. In addition, we found that exposing differentiated human mesenchymal stem cells from both abdominal and femoral subcutaneous AT to low physiological oxygen levels (5% compared to 21% O2) altered gene expression and adipokine secretion in cells from people with overweight or obesity, but not in cells derived from individuals with normal body weight [58, 67].

Further studies are warranted to examine the interaction between adipokine expression/secretion and microenvironmental pO2 in AT, thus providing a more comprehensive understanding of the role that altered pO2 may play in AT dysfunction and related cardiometabolic complications in obesity (Fig. 3a).

4.2 The role of altered oxygen tension in skeletal muscle

SM plays an important role in whole-body glucose uptake [68]. The regulation of SM glucose disposal involves both insulin-dependent and insulin-independent mechanisms, as extensively reviewed elsewhere [14]. In individuals with obesity, however, lipid spill-over from the AT together with impairments in substrate oxidation result in the accumulation of triglycerides and toxic lipid intermediates such as ceramides and diacylglycerols (DAGs) in SM [68]. This leads to the activation of protein kinase C (PKC), which in turn blunts phosphorylation of the insulin receptor substrate, thereby impairing insulin signaling in muscle cells. As a result, insulin-stimulated translocation of GLUT4 towards the cell membrane and, consequently, glucose uptake into the myocyte will be reduced [68, 69], resulting in elevated postprandial glucose levels in. individuals with obesity if insulin secretion is insufficient [70].

4.2.1 Effects of altered oxygen tension on glucose uptake

Insulin-independent glucose uptake is partly mediated via the AMP-activated protein kinase (AMPK) pathway, which is activated under stress conditions such as exercise and hypoxia [68, 71, 72]. AMPK exerts a central role in the regulation of energy homeostasis and is involved in multiple signaling pathways, including glucose and lipid uptake, and fatty acid oxidation [72]. Once muscle cells are experiencing a low energy status, more ATP will be converted to AMP resulting in the activation of AMPK, which will result in the inhibition of ATP-consuming pathways, such as gluconeogenesis and protein synthesis, and stimulation of processes that contribute to ATP generation such as glucose uptake, glycolysis and fatty acid oxidation [73]. AMPK activation under hypoxic conditions has been demonstrated in various tissues and cell types, representing a prudent strategy to conserve energy during oxygen deprivation [74]. By reducing cellular ATP consumption, the demand for oxygen is also decreased, aligning with an adaptive response to hypoxic conditions [74].

When SM cells are exposed to hypoxia, physiological adaptation will occur to maintain tissue homeostasis (Fig. 3b). Increased glucose disposal following hypoxia exposure has been reported in animal models and primary muscle cells [75]. Available evidence suggests that hypoxia exposure as well as muscle contraction stimulates glucose uptake via insulin-independent pathways in rodent and human myotubes [14]. Indeed, we have recently found that exposure to low physiological pO2 levels increases insulin-independent glucose uptake, at least partly due to activation of AMPK in primary human myotubes but did not affect insulin-mediated glucose uptake [26]. In line, we have demonstrated that mild hypoxia exposure for seven consecutive days did not significantly alter adipose, hepatic and SM insulin sensitivity, assessed by the gold standard two-step hyperinsulinemic-euglycemic clamp [26].

4.2.2 Effects of altered oxygen tension on myokines

Similar to AT, SM is an important endocrine organ that produces and secretes various myokines (muscle-secreted proteins) that can influence metabolism through autocrine, paracrine, and endocrine signaling.