Early treatment with metformin induces resistance against tumor growth in adult rats

Abstract

It is known that antidiabetic drug metformin, which is used worldwide, has anti-cancer effects and can be used to prevent cancer growth. We tested the hypothesis that tumor cell growth can be inhibited by early treatment with metformin. For this purpose, adult rats chronically treated with metformin in adolescence or in adulthood were inoculated with Walker 256 carcinoma cells. Adult rats that were treated with metformin during adolescence presented inhibition of tumor growth, and animals that were treated during adult life did not demonstrate any changes in tumor growth. Although we do not have data to disclose a molecular mechanism to the preventive metformin effect, we present, for the first time, results showing that cancer growth in adult life is dependent on early life intervention, thus supporting a new therapeutic prevention for cancer.

Abbreviations

AUC

area under the curve

HOMA-IR

homeostasis model assessment of insulin resistance

HOMA-β

homeostasis model assessment of β-cell function

LBW

body weight loss

MHC-I

major histocompatibility complex class I

NAL

nasal anal length

VEGF

vascular endothelial growth factor

Introduction

It has been shown that nutritional stress during early life induces, in adulthood, serious health risks such as obesity, type 2 diabetes, cardiovascular diseases, among others non-communicable diseases, such as cancer.^1,2 Nutritional dysfunction during adolescence programmed adult rats to obesity, indicating adolescence as window to metabolic programming, like other early life phases, such as pregnancy and lactation.^3,4

High birth weight is associated with an elevated risk of developing breast cancer in adult life compared to babies born with normal weight.^1,2,5 These observations suggest that some components of cancer etiology relate to early phases of life.

It has been suggested that certain functional foods, specifically genistein-rich soy products, folic acid, cruciferous vegetable sulforaphane and green tea polyphenols, might prevent cancer.⁶ During adolescence, soy food intake was able to prevent breast cancer in the Asian-American population.⁷

One hypoglycemic drug used worldwide, metformin, has been strongly considered for the prevention and treatment of cancer.^8-10 The pathologic complete response rate to neoadjuvant chemotherapy in diabetic patients with breast cancer was higher in patients taking metformin than in patients who had not been treated.^4,11,12

Various strategies have been reported to prevent cancer and metabolic syndrome, such as exercise training, functional food and metformin treatment, by modulating metabolism.^13,14 These strategies may support the hypothesis that some agents, such as metformin, could display some direct or indirect action on metabolism to resist further deleterious effects of tumor cell growth.

The goal of the present work was to evaluate whether adult rats that were chronically treated with metformin at weaning or from the beginning of adulthood present resistance to a rodent breast carcinoma (Walker 256 tumor) growth.

Results

Metformin treatment, either initiated at weaning or in adult life, did not reduce the final body weight, food intake, fasting glycemia and insulinemia, or HOMA-IR and HOMA-β in animals without tumor cells grafted, as shown in Table 1. Table 2 shows that weanling and adult metformin-treated animals with grafted tumor cells grafted did not present changes in final body weight, food intake, Lee Index and biochemical parameters.

Table 1.

Biometric and biochemical parameters of animals without tumor cells graft treated with or without metformin during 2 different ages

	21 to 100-day-old		60 to 139-day-old
Parameters	Water Metformin		Water Metformin
Final Body Weight (g)	384.0 ± 5.27	365.1 ± 5.82*	429.9 ± 5.63	405.2 ± 6.84
AUC [Body Weight (g)]	2649 ± 32.86	2499 ± 34.86*	3909 ± 38.73	3691 ± 50.06 Ω
AUC [Food Intake (g)]	264.5 ±1.45	249.2 ± 2.77*	286.4 ± 10.65	264.6 ± 8.37
AUC [Food Intake (g/100 g of BW)]	130.1 ± 0.65	127.2 ± 1.89	82.39 ± 1.09	80.12 ± 1.22
Glycemia (mg/dL)	99.31 ± 3.35	97.17 ± 4.93	98.60 ± 4.26	89.25 ± 4.25
Insulinemia (ng/mL)	0.731 ± 0.08	0.672 ± 0.07	0.731 ± 0.13	0.674 ± 0.15
HOMA-IR	3.90 ± 0.42	3.81 ± 0.45	4.20 ± 0.87	3.47 ± 0.87
HOMA-β	57.15 ± 6.27	60.42 ± 7.13	57.52 ± 10.49	59.55 ± 13.26

The data represent the mean ± SEM obtained from each experimental group.* ^Ω P < 0.05, compared to control, by Student's t-test. AUC: area under the curve.

Table 2.

Biometric and biochemical parameters of animals with tumor cells graft treated with or without metformin during 2 different ages

	21 to 100-day-old		60 to 139-day-old
Parameters	Water Metformin		Water Metformin
Final Body Weight (g)	374.6 ± 6.43	348.5 ± 6.41*	415.1 ± 10.51	385.0 ± 8.41Ω
Length (cm)	22.16 ± 0.19	22.19 ± 0.13	22.90 ± 0.22	22.58 ± 0.16
Lee Index	32.53 ± 0.22	32.02 ± 0.17	32.15 ± 0.28	32.15 ± 0.17
AUC [Body Weight (g)]	2731 ± 35.43	2577 ± 45.00*	3058 ± 67.00	2802 ± 61.65Ω
AUC [Food Intake (g)]	120.3 ± 4.15	105.8 ± 1.94*	117.1 ± 0.20	113.1 ± 6.25
AUC [Food Intake (g/100 g of BW)]	30.48 ± 0.78	28.90 ± 0.85	26.74 ± 0.85	28.22 ± 1.12
Glycemia (mg/dL)	82.20 ± 4.57	85.53 ± 3.71	84.50 ± 4.94	79.18 ± 3.74
HOMA-IR	0.86 ± 0.14	0.72 ± 0.11	0.96 ± 0.17	0.71 ± 0.10
HOMA-β	15.91 ± 3.22	9.90 ± 1.45	18.51 ± 1.96	10.32 ± 2.79

The data represent the mean ± SEM obtained from each experimental group.* ^Ω P < 0.05, compared to control, by Student's t-test. AUC: area under the curve.

A 32% inhibition of Walker 256 tumor growth was detected in rats that were treated with metformin from the time they were weaned. At 21 d old, no interference on tumor growth was observed in animals that were treated with metformin initiated later in life, at 60 d old, as shown in Figure 2.

Figure 2.

The effect of chronic metformin treatment initiated during 2 different ages on tumor growth of rats. The bars represent the mean ± SEM of the tumor mass related to body weight that were obtained from 12 animals for each group. In the insert of figure, the bars represent the mean ± SEM of tumor mass (g). The letters over the bars represent statistically significant differences among the groups based on one-way ANOVA (P <0.0001); the treatment groups were as follows: (A) Water treatment from 21 to 100 d old, (B) Metformin treatment from 21 to 100-day-old, (C) Water treatment from 60 to 139 d old and (D) Metformin treatment from 60 to 139 d old.

Metformin treatment that was started at weaning was also able to inhibit cachexia by 19%; however, any change was detected in rats treated with metformin later in life as shown in Figure 3.

Figure 3.

Percentage of Cachexia. The bars represent the mean ± SEM of% of rats that lost more than 10% of body weight, which were obtained from 12 animals for each group. The letters over the bars represent statistically significant differences among the groups based on one-way ANOVA (P < 0.0001); where: (A) Water treatment from 21 to 100-day-old, (B) Metformin treatment from 21 to 100 d old, (C) Water treatment from 60 to 139 d old and (D) Metformin treatment from 60 to 139 d old.

Fasting blood insulin levels were decreased by 75% when metformin-treated rats were inoculated with Walker 256 tumor cells, independent of the age when treatment was initiated, as showed in Figure 4A and B.

Figure 4.

The effect of chronic metformin treatment initiated during 2 different ages on fasting insulinemia of rats with or without inoculation with Walker 256 tumor cells. The bars represent the mean ± SEM of fasting plasma insulin levels obtained from 12 rats for group in which treatment was initiated at 21-day-old to 100 day-old (A) and the group in which treatment was initiated at 60-day-old to 139-day-old (B). The letters over the bars represent statistically significant differences among the groups based on one-way ANOVA (P < 0.0001); where: a) Animals that received water treatment from 21 to 100-day-old or from 60 to 139-day-old without tumor cells grafted and b) with tumor cells grafted; (C) Animals that received metformin treatment from 21 to 100-day-old or from 60 to 139-day-old without tumor cells grafted and (D) with tumor cells grafted. The symbols (−) and (+) indicate the absence or presence of tumor, respectively.

Discussion

Metformin treatment started at weaning promoted a 32% inhibition of tumor growth. Clearly, drug treatment initiated at an adult age, at 60 d old, did not cause any reduction of tumor mass. These results note for the first time that the antitumor- effect of metformin may be closely linked to treatment early in life.

It is known that incidence of cancer is higher in older populations.^15-17 The present work used animals at 2 different ages to test the tumor cell growth. It has been shown that 840-day-old rats show a smaller tumor growth than rats grafted at 140 d old or at 100 d old. ^18,19 Animal with different ages when received same stimulus might not respond with the same magnitude because they have differences in how the organism function, such as metabolism; however, in the case of Walker 256 tumor cells growth regarding the age which the tumor cell were grafted the 100-day-old and 140-day-old rats the tumor growth disclosed similar magnitude. The result suggest that the 2 organisms respond to aggression of similar behavior. By other hand, previous metformin treatment, started early, into adolescence, and later, in the begin of adulthood, present different responses. Earliest treatment gives some resistance to growth tumor; while, later one does not confer none resistance. Why we used animals of different ages many question arise, the mechanism are open to be revealed. The current results, at least, indicate that early life, such as adolescence, is a phase that metformin treatment might modulate organism functions. It has been observed that metformin can cross blood brain barrier and provokes effect in the central nervous system (CNS) and in the metabolism.^20,21 Changes in CNS during early phases is one the keys to change CNS structure and function, which induces a permanent changes in organism function showing dysfunctions in adult age, such as in cardiometabolic ones.^22-25

The results presented in the current work suggested that the attenuation of tumor growth may be modulated during adolescent age by an antidiabetic drug, metformin. Several studies have shown that adolescence is a sensitive period for interventions that can program the organism to block or attenuate a possible disease that may appear later in life.²⁶ Our group has shown that a high-fat or low-protein diet administered during adolescence induces more drastic consequences on metabolism than when these diets are administered only during adulthood.^27,28

Reduced tumor cell growth, in vivo, by tumor cell transplantation, or in vitro, in cancer cell lines, was observed both in short or chronic treatment with metformin. Moreover, the anti-tumor effects were attributed to a direct effect of the drug, due to its concentration in the blood or in the cell culture solution.^29-31

One relevant aspect of the metformin treatment protocol used in the current work must be highlighted: the metformin treatment was halted 24 h before tumor cell inoculation to discard the direct effects of metformin on tumor growth. During the 14 d of tumor growth, the rats did not receive any metformin doses. It has been documented that metformin plasma elimination half-life, after multiple oral dosages administration, is approximately 4–9 h.^32,33 Based on that, once again, our data indicate that metformin may provide protection against tumor growth in rats as a result of treatment that started early in life. It has been shown that the immunological system plays a crucial role on cancer immunity via cytotoxic T lymphocytes that lyse tumor cells through the recognition of tumor antigens on the major histocompatibility complex class I (MHC-I). Reduced levels of MHC-I are found in tumors, which may culminate in declined sensitivity to cytotoxic T lymphocytes lysis and consequently, assist tumor cells to escape attack by immune cells.³⁴ Metformin was able to drastically enhance MHC-I expression in tumor cells compared to those not treated with metformin.³⁴ Another study shows that animals treated with metformin had an increase of CD8 T cell memory before tumor inoculation and survived longer than the control group, resulting in greater protective anti-tumor immunity.^35,36 These observations support the hypothesis that metformin's antitumor effect is related to the stimulation of the immunological system to reduce Walker 256 tumor growth. Nevertheless, the effect only appears if treatment starts in early life and not during adulthood.

During tumor growth, increasing vascularization is critical and vascular endothelial growth factor (VEGF) increases. One of the most abundant flavonoids present in fruits, quercetin, exhibited anti-angiogenic properties that interfere with the formation of endothelial cells and was able to inhibit Walker 256 tumor growth in rats.^37,38 Interestingly, metformin treatment inhibited hyperstimulation ovarian syndrome in rats, partially due to a decrease in neovascularization associated with low levels of VEGF. Another study also showed reduced VEGF levels with metformin treatment on an ovarian tumor.^39,40 Combining these previous observations, the “programming” antitumor growth effect of metformin may be based on inhibiting neovascularization in Walker 256 tumor-bearing rats.

It has been shown that the Walker 256 tumor leads to a decrease in insulin blood levels, which is related to a reduced glucose-induced insulin released by the pancreatic β-cell, while tumor growth is decreased by insulin administration.^41,42 The authors concluded that the Walker 256 tumor disrupted insulin secretion control and insulin tissue sensitivity. These observations suggest that metformin action improves the metabolism, most likely by improving insulin secretion or normalizing insulin sensitivity, one of the mechanisms of its antitumor activity. However, our results show that the metformin effect is not related to increasing insulin level or to improving insulin sensitivity.

In summary, data from the present work suggest that metformin has an antitumor growth protective effect, which is developed in early life, i.e., in adolescent rats. Our study implies preventive and therapeutic possibilities that consider tumor growth capacity to be determined during early life.

Materials and Methods

Experimental animals

Male and female Wistar rats used for mating were supplied by the central animal facility of State University of Maringá and maintained in the animal facility of laboratory of secretion cell biology. Pregnant Wistar rats were placed in individual cages until the weaning of litters that were used in the protocol experiments. During all procedures, animals were kept under controlled temperature conditions (21±2°C), a light/dark cycle of 12 h (07:00-19:00 h) with ad libitum access to water and a standard diet (Nuvital®, Curitiba, PR, Brazil). The Ethics Committee for Animal Experimentation of the State University of Maringá approved the protocols.

Animal groups and metformin treatment

To evaluate the effect of metformin as a preventive strategy against tumor growth and to evaluate the more susceptible phase of life of metformin action against the tumor growth, the treatment was administered in 2 different age periods: one starting at the equivalent of childhood, from weaning (21 d old) until 100 d old, and the other starting at adulthood, from 60 d old until 139 d old. Male rats were treated with metformin (Metformin hydrochloride, Medley, Brazil) dissolved in water at a dose of 250 mg/kg of body weight by gavage once a day between 10:00 to 11:00 h. Both groups of rats received metformin treatment for 79 d, from weaning to 100 d old or from 60 to 139 d old, whereas 2 other groups (control groups) received water during the same periods. Figure 1 summarizes the experimental protocols used.

Figure 1.

Schematic representation of the protocols. Metformin treatment was initiated during 2 different age periods, at weaning or during adulthood. For this purpose, 21-day-old male rats were submitted to metformin gavage once a day until they reached 100 d old. Another batch of animals, at 60 d old, were submitted to metformin gavage until the reached 139 d old. Thus, both groups of rats received metformin treatment for 79 d, and the same procedure was performed with the groups that received water instead of the drug. After treatment ended, 101-day-old animals and 140-day-old animals were inoculated subcutaneously with carcinoma cells, and after 14 d of inoculation, the animals were sacrificed and subsequent protocols performed.

The dose of metformin used in the current work has no toxicity to the experimental animals, as observed in other studies,^43,44 complying with the therapeutic range of metformin. Furthermore, regarding the mammalian species,⁴⁵ the dosage utilized is equivalent to a subclinical dosage used in humans during treatment of diabetes, which may range from 500 mg up to 850 mg administered twice or 3 times a day.⁴⁶ After the end of the metformin treatment, half of the rats in each of the experimental groups were randomly used in subsequent protocols, and another batch of animals were inoculated with tumor cells.

Metabolic and biometric parameters

Throughout metformin treatment, body weight and diet consumption were measured weekly. Food intake was evaluated by determining the difference between the amount of food remaining and the total amount that was previously placed in the cage, divided by the number of animals in the box and the number of days. The area under the curve (AUC) was calculated for body weight and food intake. Additionally, the values were expressed relative to 100 g of body weight.

At the end of metformin treatment (after 79 d) and 12 h of fasting, a batch of animals from each experimental group (rats that received Water or Metformin treatment from 21 to 100 d old, and rats that received Water or Metformin treatment from 60 to 139 d old) were weighed and anesthetized (Thiopental 45 mg/kg body weight). The nasal anal length (NAL) was measured to calculate the Lee index using the equation: Lee index = (body weight [g]^1/3/NAL [cm]) × 100.⁴⁷ Subsequently, the animals were decapitated, blood samples were collected and centrifuged, and the plasma was stored at -20 ºC for further analysis. Glucose concentration was determined by the glucose oxidase method⁴⁸ using a commercial kit (Gold Analisa®, Belo Horizonte, MG, Brazil), and insulin levels were evaluated using a radioimmunoassay (RIA)⁴⁹ with a gamma counter (2470 Wizard² Automatic Gamma Counter, PerkinElmer).

Fasting glucose and insulin levels were used to calculate the homeostasis model assessment of insulin resistance (HOMA-IR) and the homeostasis model assessment of β-cell function (HOMA-β), using the following equations: [HOMA-IR = fasting glucose (mmol/L) × fasting insulin (mU/L)/22.5] and [HOMA-β = [20 × fasting insulin (mU/L)]/[Fasting glucose (mmol/L) – 3.5].^50,51

Inoculation of Walker 256 tumor cells

Walker 256 cells were injected intraperitoneally into the rats. After one week, the cells were collected, centrifuged and suspended in phosphate buffered saline (PBS; 16.5 mmol/L of phosphate, 137 mmol/L of NaCl and 2.7 mmol/L of KCl), pH 7.4, and cell viability were evaluated using the trypan blue exclusion method. To eliminate direct effects of metformin on tumor growth, treatment was stopped 24 h before tumor inoculation. Thus, a batch of rats treated with water or metformin since weaning, at 101 d old, and another batch of animals treated with water or metformin since adulthood, at 140 d old, were inoculated subcutaneously (8.0 × 10⁷ viable tumor cells/animal) into their right rear flanks.⁵² Animals from 2 different ages groups, with or without metformin treatment, were inoculated at the same time with the same batch of tumor cells.

Tumor growth evaluation and characterization of cachexia

During tumor growth, body weight and food intake [(g) (g/100 g of body weight)] were evaluated every 2 d, and AUC was calculated. After 14 d of tumor cells inoculation and 12 h of fasting, animals from all groups were anesthetized (Thiopental 45 mg/kg of body weight); NAL was measured to calculate the Lee index. Subsequently, rats were sacrificed by decapitation and blood samples were collected. Plasma was used to determine the glucose and insulin concentration, as previously described, and the values were also used to calculate HOMA-IR and HOMA-β.

Tumor mass was carefully dissected and weighed to calculate the percentage of body weight loss (LBW) using the following equation: LBW (%) = 100 × (bwi – bwf + tm + gbw)/(bwi + gbw); Where, bwi: initial body weight (g), bwf: final body weight (g) of rats with tumor; tm: tumor mass (g); and gbw: body weight gain (g) of rat without tumor during the 14 d of the experiment. The rats were considered cachectic when the LBW was higher than 10%.⁵² Tumor mass values were expressed relative to 100 g of body weight.

Statistical analyses

Data are given as the mean ± SEM and were subjected to a variance analysis (one-way ANOVA) followed by a Bonferroni test. P values less than 0.05 were considered statistically significant. Tests were performed using GraphPad Prism version 6.0 for Windows (GraphPad Software, La Jolla CA, USA).

Funding Statement

This research was supported by grants from the Brazilian Research National Foundation: The Conselho National de Desenvolvimento Científico e Tecnológico (CNPq), the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Paraná State Research Foundation (Fundação Araucária).

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.