Triacylglycerol (TG) is the main form of lipid storage and is synthesized by a condensation reaction between nonesterified fatty acid (NEFA) and glycerol-3-phosphate (G3P). NEFA is generated by de novo lipogenesis from dietary glucose in the liver or the lipolysis of white adipose tissue (WAT) [1]. Because G3P is converted from glycerol, the glycerol transportation and metabolism are important for TG synthesis [2]. In a physiological fasted state, the WAT constitutes the most important source of glycerol, which flows into the liver to be used as a substrate for gluconeogenesis. In contrast, lipogenic conditions, such as excessive calorie intake and low energy consumption, promote the incorporation of glucose and glycerol from the bloodstream into adipocytes to be used as substrates for TG synthesis [3]. The control of glycerol transport into cells may be relevant to regulate fat accumulation and glucose metabolism.

Aquaporin (AQP) is an integral membrane protein that acts as a channel for the movement of water and small solutes. Thirteen AQPs have been identified and classified into subfamilies based on their function. AQP3, AQP6, AQP7, AQP8, AQP9, and AQP10 are known as aquaglyceroporins that transport glycerol across the cell membrane [4]. AQP7 is the main glycerol transporter in adipocytes and is important for releasing glycerol from adipocytes for uptake by the liver under physiological conditions, which is involved in the development of obesity and type 2 diabetes. During the fed state, AQP7 also contributes to adipose glycerol uptake for lipogenesis [5]. AQP9 is expressed in multiple organs and is the major isoform involved in glycerol influx in hepatocytes. AQP9-KO mice have elevated plasma glycerol levels and reduced plasma glucose levels under starvation conditions, suggesting that a pharmacological blockade of AQP9 may represent a treatment for type 2 diabetes [6]. In adipocytes, AQP9 was suggested to play a role in glycerol transport in rodents and humans. In mouse 3T3-L1 adipocytes, AQP9 contributed 12% of total glycerol permeability across the cell membrane [7]. The protein expression of AQP9 was upregulated by insulin in human adipocytes and was higher in the visceral adipose tissue of obese and type 2 diabetes patients [8]; however, other modulators of adipocyte AQP9 remain to be identified in contrast to hepatocytes [9].

The n-3 polyunsaturated fatty acids (PUFAs), such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are abundant in fish oil and known as functional food ingredients for preventing dyslipidemia [10], improving fatty liver [11], and lowering the risk of metabolic syndrome [12]. Several mechanisms have been proposed from in vivo studies. EPA and DHA reduce hepatic lipid content through the inhibition of lipogenesis and stimulation of fatty acid β-oxidation in the liver [13,14]. In adipose tissue, fish oil promotes energy consumption in the form of heat by activating thermogenic brown adipocytes and the differentiation of beige adipocytes that is inducible thermogenic adipocytes [15]. With respect to glycerol metabolism, including AQPs, n-3 PUFA-depleted rats exhibit decreased protein expression of AQP7 and AQP9 in WAT and liver, respectively [16,17]. Studies using hamsters revealed that fish oil upregulated AQP7 gene expression in subcutaneous and visceral WAT under non-obese conditions [18]. Conversely, little is known regarding the effects of EPA and DHA on protein expression of AQP9 in adipocyte of obesity and type 2 diabetes.

In this study, we focused on the relationship between gluconeogenesis, hepatic glycerol uptake, and adipose glycerol export and determined the effect of fish oil on AQP9 protein expression in liver and WATs using KK mice as an obesity and diabetes model under fasted state. In addition, we elucidated the direct action of EPA and DHA on AQP9 protein expression in mouse 3T3-L1 adipocytes.