Short-term calorie and protein restriction provide partial protection from chemotoxicity but do not delay glioma progression

Abstract

Short-term starvation (STS) protects normal cells while simultaneously sensitizing malignant cells to high-dose chemotherapeutic drugs in mice and possibly patients. The fasting-dependent protection of normal cells and sensitization of malignant cells depends, in part, on reduced levels of insulin-like growth factor-1 (IGF-1) and glucose. Calorie restricted diets with defined macronutrient (carbohydrate, protein, fat) ratios were evaluated for the effects on stress sensitization markers and protection in mice treated with high-dose chemotherapy. We show that short-term CR significantly reduced both glucose and IGF-1 levels, but when specific macronutrient deficiencies were tested, only the complete lack of proteins reduced IGF-1 levels. Short-term 50% CR combined with either severe protein-deficiency or ketogenic diets improved chemotoxicity resistance similarly to the standard 50% CR, but did not result in the high protection caused by STS. Notably, a high protein diet reversed the beneficial effects of short-term CR. In a subcutaneous mouse model of glioma, feeding a low protein (4% calories from protein) diet for more than 20days did not delay tumor progression once the tumor became palpable. Also, cycles of short-term (3days) 50% CR did not augment the chemotherapy efficacy of cisplatin in a murine breast cancer model. These results indicate that the protection from chemotoxicity and retardation of the progression of certain tumors achieved with fasting is not obtained with short-term calorie and/or macronutrient restriction.

Keywords: Calorie restriction; Chemotherapy; Fasting; Insulin-like growth factor 1; Macronutrients; Protein restriction; Stress resistance; Tumor progression.

Conflict of interest statement

Conflict of interest

VDL and TEM have equity interest in L-Nutra, a company that develops medical food.

Copyright © 2013 Elsevier Inc. All rights reserved.

Figures

Fig. 1

Calories supplied by macronutrients of the experimental diets in %. AIN93G standard chow was the reference diet supplied to mice. The experimental diets modified in the macronutrient composition (fat, protein and carbohydrates) were all based on this diet. The low-carbohydrate LCHP diet had calories from carbohydrates reduced to 20% compared to the AIN93G formulation (13% vs. 63.9%) but contained more protein (45.2%) and fat (41.8%). Diets 20% P-1 (soybean oil as fat source) and 20% P-2 (coconut oil as fat source) had calories from protein sources reduced to 20% compared to the AIN93G formulation; the 0% P diet contained no protein; all these diets were isocaloric to the AIN93G standard chow. The ketogenic high fat diet 60% HF was designed to supply 60% of the consumed calories from fat sources, the calories coming from protein and carbohydrates were reduced proportionally. The 90% HF diet was a ketogenic diet which contains 90% of fat while supplying only minimal carbohydrates (less than 1%) and half of the protein content (9%). Detailed diet composition and calorie content are summarized in Table S2.

Fig. 2

Calorie restriction reduces bodyweight, glucose and IGF-1. A) Female CD-1 mice, age 12–15 weeks were either fed ad lib (gray square) with AIN93G rodent standard chow, exposed to 40%, 60%, 80% and 90% calorie restricted AIN93G diets (triangles) or fasted (STS, green rectangle) until mice lost 20% of their initial bodyweight (dotted line). N = 5 per experimental group. All data presented as mean ± SEM. B) Linear fit for the severity of the CR regimen vs. the duration (days) until 80% bodyweight was reached. C) Blood glucose levels for mice once 80% bodyweight was reached. Red line represents mean; * p < 0.05, *** p < 0.001, ANOVA, Tukey’s multiple comparison. D) Serum IGF-1 levels for mice once 80% bodyweight was reached. Red line represents mean; *** p < 0.001, ANOVA, Tukey’s multiple comparison. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Effects of macronutrient defined diets on bodyweight, food intake, glucose and serum IGF-1. Five female CD-1 mice, age 12–15 weeks were either fed ad lib with AIN93G rodent standard chow (black circle) or with A) two different low protein diets (20% P-1 and 20% P-2), a diet low in carbohydrates but high in protein (LCHP), a protein deficient diet (0% P) or B) a high fat diet (60% HF) and ketogenic diet (90% HF). A detailed overview over the macronutrients is given in Table S1. C) Daily ad lib calorie intake for diets AIN93G, 20% P-1, 20% P-2, LCHP and 0% P. D) Daily ad lib calorie intake for diets 60% HF and 90% HF; AIN93G shown as reference. All data presented as mean ± SEM. E) Serum IGF-1 levels after 9 days of ad lib feeding. Lines represent mean; *p < 0.05,*** p < 0.001, ANOVA, Tukey’s multiple comparison compared to AIN93G control.

Fig. 4

Stress resistance test for calorie restricted macronutrient defined diets. Mice were fed ad lib (AIN93G), were fasted for 60 h (STS) or fed with 50% calorie restricted diets with defined macronutrient compositions (AIN93G, LCHP, 0% P, 60% HF, 90% HF) for 3 days (green box) prior to an intravenous injection of doxorubicin (24 mg/kg, red dashed line). Survival was followed for 25 days post injection, after which the remaining animals were considered survivors. B) Blood glucose levels after 3 days of feeding ad lib and CR diets, as well as after 60 h STS. Lines represent mean. * p < 0.05, *** p <0.001, ANOVA, Tukey’s multiple comparison. Survival data plotted from pair-matched pooled experiments with the statistical software Prism (GraphPad Software). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Tumor progression of GL26 glioma and 4T1 breast cancer in vivo. A) Subcutaneous tumor progression of murine GL26 glioma is shown by total tumor volume in mm3. Tumor measurements were started once the tumors became palpable under the skinat day 10. Animals were fed ad lib with either AIN93G(N = 5)as a control or with the low protein diet 20% P-1 (N = 6). All data presented as mean ± SEM. B) Subcutaneous tumor progression of murine 4T1 breast cancer is shown by total tumor volume inmm3. Tumor measurements were started once the tumors became palpableunder skinatday 12. Control animals(N = 10) received no treatment and tumor progressed rapidly, reaching the endpoint volume of 2000mm3 by day 54 post tumor implantation. Cisplatin (CDDP) animals (N = 9) were injected at days 15, 33 and 44. The first CDDP dose was delivered at 12 mg/kg by intravenous injection, the two subsequent injections were delivered at 8 mg/kg to avoid chemotoxicity. Mice in the 50% ICR + CDDP group (N = 9) were fed in intermittent regimens with the AIN93G diet reduced to 50% of the daily calorie intake of the control group for three days (ICR, green box) prior to cisplatin injection. Injection schedule was identical as for the CDDP group. All data presented as mean ± SEM; *** p < 0.001, ANOVA, Tukey’s multiple comparison, compared to control. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)