Exploiting Enhanced Lipid Metabolism in Glioblastoma through Diet Modification

Abstract

Enhanced lipid metabolism has emerged as a central metabolic node in glioblastoma, serving as a 'gain of function' that allows these cells to efficiently adapt to their dynamic tumor microenvironment. Seemingly contradictory to this, pre-clinical studies have demonstrated anti-tumor activity in mice fed a high-fat/low-carbohydrate ketogenic diet (KD), both alone and in combination with radiation therapy (RT). In this study, we sought to identify mechanisms underlying the antitumor activity of a KD in glioblastoma from a metabolic perspective to better understand factors contributing to this apparent disconnect. Immunocompromised and immunocompetent mice were injected orthotopically with human and mouse-derived glioblastoma cell lines and randomized to four treatment arms. Mice were fed ad libitum a standard diet (SD), KD (Bio-Serve), or a modified unsaturated fatty acid (uFA) rich diet (MD; 60/30/10: fat/protein/carb) alone or in combination with hypofractionated RT (6 Gy x 3). Global metabolomic profiling of tumors and serum were carried out using LC/GC-MS. Lipid droplets were analyzed by flow cytometer and confocal microscopy using BODIPY staining and free fatty acids were measured using a commercially available kit. A KD demonstrated independent anti-tumor activity and potent synergy with RT in two aggressive glioblastoma models. Metabolomic profiling of tumors revealed significant changes in tumor metabolism in KD-fed mice when compared to SD, with an accumulation of uFAs being a key finding. We therefore sought to determine if this accumulation of fatty acids in KD mice contributed towards the observed anti-tumor activity. Consistent with in vivo results, in vitro studies using the uFA linoleic acid demonstrated anti-proliferative activity, reduced clonogenic capacity, and potent synergy when combined with RT in glioblastoma cells. Through a series of investigations, we went on to determine that this anti-tumor activity was attributed to the ability of uFA to override lipid storage homeostasis in glioblastoma cells, resulting in lipotoxicity. Based on these findings, we hypothesized high fat concentrations, rather than carbohydrate restriction, contributed to the anti-tumor activity of a KD. To test this, we generated a MD rich in uFA that did not require carbohydrate restriction. Similar to a KD, mice fed a MD demonstrated both independent anti-tumor activity and potent synergy when combined with RT. High concentrations of uFA represents a key factor underlying the anti-tumor activity of a KD in glioblastoma by targeting lipid homeostasis. A MD consisting of high concentrations of uFA without carbohydrate restriction demonstrates promising anti-tumor activity in glioblastoma models. As a major limitation of a KD is tolerability, particularly in glioblastoma patients, a MD represents a promising form of dietary modification that may be translated clinically.