Fructose uptake by brown adipose tissue is independent of carbohydrate response element-binding protein and does not cause elevated de novo lipogenesis

Abstract

Brown adipose tissue (BAT) is a heat-generating organ burning significant amounts of calories from fatty acids and glucose. The importance of glucose metabolism in the context of thermogenic function has been underlined by several studies. However, fructose metabolism and consequences of fructose overfeeding are poorly studied in BAT. Here we provide evidence that brown adipocytes use fructose as a substrate, however to a lesser extent than glucose. Furthermore, our data suggest that carbohydrate response element binding protein (ChREBP) and its target glucose transporter 5 (GLUT5) are not essential for fructose uptake and metabolism in BAT. Notably, we report that high fructose feeding has no effect on ChREBP activity and thus de novo fatty acid synthesis in BAT as opposed to liver and intestine. Instead, excessive carbohydrate loading of brown adipocytes induced by both, high-fructose feeding and impairment of ChREBP-dependent glucose metabolism, causes a massive accumulation of hexosylceramide species, as revealed by mass spectrometry-based lipidomics. Based on our data we hypothesize a reprogramming of fructose utilization upon impaired carbohydrate metabolism from canonical glycolysis and pentose phosphate pathway towards glycosphingolipid synthesis.

1. Introduction

Fructose consumption in the form of added sugar in refined foods or sugar-sweetened drinks has strongly increased in many countries worldwide since the 1970s [1–3]. Excessive fructose intake is associated with overweight and cardiometabolic disorders including hypertriglyceridemia, fatty liver disease, type 2 diabetes and cardiovascular diseases [3–7]. The development of fatty liver and hypertriglyceridemia in response to fructose overfeeding is mechanistically well understood, in particular how it is linked to de novo lipogenesis (DNL), the metabolic pathway converting carbohydrates into long-chain fatty acids [8]. In human and rodent hepatocytes, fructose is rapidly metabolized to C3 intermediates of glycolysis via ketohexokinase (KHK), aldolase B (ALDOB) and triokinase (TKFC) in a relatively unregulated fashion [9], thereby providing acetyl-CoA as substrate for DNL. When in excess, fructose can thus promote liver steatosis and cause hypertriglyceridemia [9–12].

In parallel to metabolically supplying the DNL pathway, fructose metabolism in hepatocytes efficiently stimulates the transcription of DNL enzyme genes, predominantly through activation of the transcription factor carbohydrate response element-binding protein (ChREBP) [9,13]. As shown by studies in liver-specific ChREBP-knockout mice, activation of ChREBP is essential for the development of obesity and disturbed glucose homeostasis in mice fed with a high-fructose diet [14]. Several members of the hexose transporter family, SLC2A2 (GLUT2), SLC2A5 (GLUT5) and SLC2A8 (GLUT8), have been implicated in hepatocyte fructose uptake [2,9,15]. Of note, the selective fructose transporter SLC2A5 [16] is activated by ChREBP in both the small intestine and the liver, a regulation that establishes a feed-forward mechanism linking fructose uptake, fructose metabolism and DNL [14,17]. Quantitative postprandial studies have determined that a significant proportion of fructose is metabolized already in enterocytes of the small intestine and converted into glucose via gluconeogenesis [18]. In addition, microbiota in the colon convert fructose into short-chain fatty acids [18–20]. Intestinal metabolism may be the primary route for ingested fructose when moderate amounts of the sugar are consumed, at least in mice [9,18]. If fructose intake is high, a larger portion of the monosaccharide is directed to the liver via the portal vein and is cleared by hepatocytes [11,18]. In consequence of the splanchnic processing, only a minor proportion of ingested fructose reaches the systemic circulation [21] and the concentration of fructose in peripheral blood is low compared to glucose [21–23]. Nevertheless, systemic fructose concentrations rise postprandially [21] and fructose use in peripheral organs can be of relevance. For example, fructose metabolism has been shown to promote the development of cardiomyopathy in mice [24]. The significance of fructose utilization in adipose tissue as another major site of energy metabolism is not well researched. Slc2a5 is expressed in white adipocytes [25] and Slc2a5-KO mice have reduced epididymal white fat mass [26]. In line with this observation, fructose promotes the differentiation of 3T3-L1 preadipocytes [26,27]. Even less is known about the functional significance of fructose metabolism in brown adipose tissue (BAT). BAT is a heat-generating organ that burns significant amounts of calories from fatty acids and glucose, and its importance in energy metabolism is increasingly recognized.