Long-term ritonavir exposure increases fatty acid and glycerol recycling in 3T3-L1 adipocytes as compensatory mechanisms for increased triacylglycerol hydrolysis

Abstract

Lipodystrophy with high nonesterified fatty acid (FA) efflux is reported in humans receiving highly active antiretroviral therapy (HAART) to treat HIV infection. Ritonavir, a common component of HAART, alters adipocyte FA efflux, but the mechanism for this effect is not established. To investigate ritonavir-induced changes in FA flux and recycling through acylglycerols, we exposed differentiated murine 3T3-L1 adipocytes to ritonavir for 14 d. FA efflux, uptake, and incorporation into acylglycerols were measured. To identify a mediator of FA efflux, we measured adipocyte triacylglycerol lipase (ATGL) transcript and protein. To determine whether ritonavir-treated adipocytes increased glycerol backbone synthesis for FA reesterification, we measured labeled glycerol and pyruvate incorporation into triacylglycerol (TAG). Ritonavir-treated cells had increased FA efflux, uptake, and incorporation into TAG (all P < 0.01). Ritonavir increased FA efflux without consistently increasing glycerol release or changing TAG mass, suggesting increased partial TAG hydrolysis. Ritonavir-treated adipocytes expressed significantly more ATGL mRNA (P < 0.05) and protein (P < 0.05). Ritonavir increased glycerol (P < 0.01) but not pyruvate (P = 0.41), utilization for TAG backbone synthesis. Consistent with this substrate utilization, glycerol kinase transcript (required for glycerol incorporation into TAG backbone) was up-regulated (P < 0.01), whereas phosphoenolpyruvate carboxykinase transcript (required for pyruvate utilization) was down-regulated (P < 0.001). In 3T3-L1 adipocytes, long-term ritonavir exposure perturbs FA metabolism by increasing ATGL-mediated partial TAG hydrolysis, thus increasing FA efflux, and leads to compensatory increases in FA reesterification with glycerol and acylglycerols. These changes in FA metabolism may, in part, explain the increased FA efflux observed in ritonavir-associated lipodystrophy.

Figure 1

Effects of triacsin C on fatty acid uptake and disposition after ritonavir exposure. 3T3-L1 cells were exposed to ritonavir (black bars) or vehicle (white bars) during and after differentiation for a total of 14 d. Cells were pretreated with triacsin C or control medium for 10 min before labeling by addition of [9,10-³H]palmitic acid for a 15-min incubation. A, ³H in total cellular lysate; B, ³H in intracellular TAG; C, ³H in DAG; D, ³H in intracellular (IC) nonesterified FA pool; E and F, release of total glycerol (E) and total FA (F) during the 15-min incubation. All data are normalized for total cellular DNA. For all graphs, n = 18 from three independent experiments. **, P < 0.01; ***, P < 0.001 for ritonavir vs. vehicle.

Figure 2

Effects of ritonavir on ATGL and HSL protein expression. 3T3-L1 adipocytes treated with ritonavir or vehicle for 14 d. A, Representative ATGL and HSL Western blots (R, ritonavir; V, vehicle); B, quantitation of signal intensity from Western blots. All data are normalized for total cellular DNA. For B, n = 9, from three independent experiments. *, P < 0.05 for ritonavir (black bars) vs. vehicle (white bars).

Figure 3

Effects of ritonavir in glucose-deprived 3T3-L1 cells on unstimulated and isoproterenol-stimulated lipolysis and glyceroneogenic capacity using glycerol or pyruvate. 3T3-L1 adipocytes were treated with ritonavir (black bars) or control medium (vehicle, white bars) for 14 d. Release of total glycerol (A) and total nonesterified FA (B) in response to 3 h glucose deprivation (Fast; left side of dashed line) followed by loading with either 100 μm unlabeled glycerol plus [2-³H]glycerol or 200 μm unlabeled pyruvate plus [1-¹⁴C]pyruvate for 30 min and subsequently studied after no additions (NA) in the nonstimulated state or after stimulation with 1 μm isoproterenol (Iso) for 2 h (right side of dashed line). C and D, Incorporation of [2-³H]glycerol (C) and [1-¹⁴C]pyruvic acid (D) into TAG. E, Incorporation of glycerol into TAG in picomoles per microgram DNA (label from either glycerol or pyruvate was converted to picomoles using specific activities reported in Materials and Methods). F, Incorporation of pyruvic acid (picomoles per microgram DNA) into TAG. G, Steady-state TAG mass in glycerol-loaded cells. H, Steady-state TAG mass in pyruvate-loaded cells. For all graphs, n = 15 from four independent experiments. **, P < 0.01; ***, P < 0.001 for ritonavir vs. vehicle.

Figure 4

Gyk, PEPCK1, and ATGL mRNA expression in 3T3-L1 adipocytes treated with ritonavir or vehicle for 14 d. A–C, Gyk (A), PEPCK1 (B), and ATGL (C) mRNA expression ratios in response to 3 h glucose deprivation followed by loading with 200 μm pyruvate for 30 min and subsequently not stimulated (NA, hatched bars) or stimulated with 1 μm isoproterenol (Iso, gray bars) for 2 h. Mean expression ratios with sem are reported. For all graphs, n = 8–9 from three independent experiments. See Fig. 3 for other details. *, P < 0.05; **, P < 0.01; ***, P < 0.001 for mean expression ratio different from 1.0.

Figure 5

Ritonavir’s effects on glycerolipid metabolism. Proposed model of ritonavir’s effects on FA cycling includes increased TAG hydrolysis, augmented recycling of both FA and glycerol, and inhibition of glyceroneogenesis from pyruvate. Abbreviations used: Glycerol-P_i, glycerol-3-phosphate; Gyk, glycerol kinase; MAG, monoacylglycerol; NEFA, nonesterified FA; PEPCK1, phosphoenolpyruvate carboxykinase.