The Fatty Acid Synthase Inhibitor Platensimycin Improves Insulin Resistance without Inducing Liver Steatosis in Mice and Monkeys

Abstract

Objectives: Platensimycin (PTM) is a natural antibiotic produced by Streptomyces platensis that selectively inhibits bacterial and mammalian fatty acid synthase (FAS) without affecting synthesis of other lipids. Recently, we reported that oral administration of PTM in mouse models (db/db and db/+) with high de novo lipogenesis (DNL) tone inhibited DNL and enhanced glucose oxidation, which in turn led to net reduction of liver triglycerides (TG), reduced ambient glucose, and improved insulin sensitivity. The present study was conducted to explore translatability and the therapeutic potential of FAS inhibition for the treatment of diabetes in humans.

Methods: We tested PTM in animal models with different DNL tones, i.e. intrinsic synthesis rates, which vary among species and are regulated by nutritional and disease states, and confirmed glucose-lowering efficacy of PTM in lean NHPs with quantitation of liver lipid by MRS imaging. To understand the direct effect of PTM on liver metabolism, we performed ex vivo liver perfusion study to compare FAS inhibitor and carnitine palmitoyltransferase 1 (CPT1) inhibitor.

Results: The efficacy of PTM is generally reproduced in preclinical models with DNL tones comparable to humans, including lean and established diet-induced obese (eDIO) mice as well as non-human primates (NHPs). Similar effects of PTM on DNL reduction were observed in lean and type 2 diabetic rhesus and lean cynomolgus monkeys after acute and chronic treatment of PTM. Mechanistically, PTM lowers plasma glucose in part by enhancing hepatic glucose uptake and glycolysis. Teglicar, a CPT1 inhibitor, has similar effects on glucose uptake and glycolysis. In sharp contrast, Teglicar but not PTM significantly increased hepatic TG production, thus caused liver steatosis in eDIO mice.

Conclusions: These findings demonstrate unique properties of PTM and provide proof-of-concept of FAS inhibition having potential utility for the treatment of diabetes and related metabolic disorders.

Fig 1. Platensimycin inhibits enzymatic activity and expression of fatty acid synthase (FAS) and lipogenic genes.

A: Effect of PTM on FAS enzymatic activity in the liver of eDIO mice treated with PTM (100 mpk, BID, p.o. for 4 days) determined by Malonyl-CoA dependent consumption of NADPH (n = 7–8). B–C: Protein level of FAS in the liver determined by western blot and quantitated by Li-COR. D: Relative RNA levels of FAS, sterol regulatory element-binding transcription factor 1 (SREBP-1c), liver X receptor alpha (LXRα), and carbohydrate-responsive element-binding protein (ChREBP) in the liver determined by quantitative RT-PCR. Bars represent means ± SEM. Asterisks denote statistical significance of treatment group compared to vehicle group. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Fig 2. Platensimycin reduces lipogenesis, liver lipid content but has no effect on markers of fatty acid oxidation (FAO) in species with a low de novo lipogenesis (DNL) tone.

A–B: Dose-dependent inhibition of PTM on DNL of high fructose-fed db/+ (A) and chow diet-fed lean C57BL/6 mice (B) (n = 7–8). Concentrations of PTM in liver and plasma are listed under the panels. C: Effect of PTM on tissue TG contents of liver and skeletal muscle of chow diet-fed mice after 17 days of treatment (30 mpk, QD). D–H: Effect of chronic treatment of PTM (50 mpk, BID), CPT1 inhibitor (50 mpk, BID), and T0901317 (10 mpk, QD) on DNL (D), liver TG content, and plasma β-hydroxybutyrate (β-HBA) in diet-induced obese mice. Liver TG contents were analyzed by MRS imaging on day 10 (E) and biochemical methods on day 16 (F). Plasma levels of β-HBA were analyzed at 4 hours post dosing on day 1 after single dose (G) and day 16 after chronic treatment (H) (n = 7). Bars represent means ± SEM. Asterisks denote statistical significance of treatment group compared to vehicle group. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Fig 3. Effect of Platensimycin on body weight gain and food intake in different mouse models.

A–C: Subchronic treatment of PTM in eDIO mice (40, 125, and 400 μg/h, minipump for 10 days) (n = 8). D–E: Chronic treatment of PTM in db/+ mice on high fructose diet (3, 10, 30, and 100 mpk in drinking water for 29 days) (n = 8). F–H: Subchronic treatment of PTM in db/db mice for 16 days (3, 10, and 30 mpk, BID, p.o. for 16 days). Bars represent means ± SEM. Asterisks denote statistical significance of treatment group compared to vehicle group. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Fig 4. Parallel comparison of PTM and CPT1 inhibitor on glucose and lipid metabolism in perfused liver of lean C57BL/6 mice.

PTM and CPT1 inhibitor are used at 100 μM in the perfusion media (n = 3–5 per group). Bars represent means ± SEM. Asterisks denote statistical significance of treatment group compared to control. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Fig 5. Efficacy of PTM in non-human primates (NHPs).

A–C: Inhibition of de novo lipogenesis (DNL) by PTM in lean cynomolgus monkeys (A) (n = 5 for vehicle and 4 for PTM) and rhesus (B) (n = 4), and lean aged rhesus monkey (C) (n = 6). Animals were dosed with PTM (60 mpk BID p.o. in A and B, 20 and 60 mpk BID s.c. in C and blood samples were collected at 24 hrs post dosing. D: Effect of PTM on DNL of lean rhesus monkeys (60 mpk, p.o.). E–J: Effect of chronic treatment of PTM on body weight (E), fasting glucose levels for predose (F) and 2 hrs post dose at baseline, day 9 and day 22 (G) and insulin (H–I). PTM was dosed at 60 mpk mixed with yogurt for 28 days. J: Liver fat fraction was determined by MRS imaging at baseline and after 28 days of s.c. dosing of PTM (100 mpk, BID) in lean rhesus monkeys. Bars represent means ± SEM. Asterisks denote statistical significance of treatment group compared to vehicle or baseline. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.