Effects of calorie, protein, and branched chain amino acid restriction on ovarian aging in mice

Abstract

Calorie restriction (CR) is an intervention that promotes longevity and preserves the ovarian reserve. Some studies have observed that the positive impacts of CR can be linked to restriction of protein (PR) and branched-chain amino acids (BCAAs) independent of calorie intake. The aim of this study was to compare the effects of protein and BCAA restriction to 30% CR on the ovarian reserve of female mice. For this, 3 month-old C57BL/6 female mice (n=35) were randomized into four groups for four months dietary interventions including: control group (CTL; n = 8), 30% CR (CR; n = 9), protein restriction (PR; n = 9) and BCAA restriction (BCAAR; n = 9). Body mass gain, body composition, food intake, serum levels of BCAAs, ovarian reserve and estrous cyclicity were evaluated. We observed that CR, protein and BCAA restriction prevented weight gain and changed body composition compared to the CTL group. The BCAA restriction did not affect the ovarian reserve, while both PR and CR prevented activation of primordial follicles. This prevention occurred in PR group despite the lack of reduction of calorie intake compared to CTL group, and CR did not reduce protein intake in levels similar to the PR group. BCAA restriction resulted in increased calorie intake compared to CTL and PR mice, but only PR reduced serum BCAA levels compared to the CTL group. Our data indicates that PR has similar effects to CR on the ovarian reserve, whereas BCAA restriction alone did not affect it.

1. INTRODUCTION

The ovarian reserve consists of a finite pool of primordial follicles that are established during intrauterine life [1]. Once activated, primordial follicles develop irreversibly until ovulation or atresia [2]. The progressive decline in the ovarian reserve of women results in fertility decline during midlife, and later in life, once the ovarian reserve is depleted, in menopause [2, 3]. Genetic and environmental factors such as body composition, physical activity, dietary habits, and a variety of diseases can influence the rate of decline in the ovarian reserve in women [4]. Activation of mammalian target of rapamycin (mTOR) in granulosa cells and phosphorylation of Forkhead box O3 (FOXO3) in the oocyte are key steps in primordial follicular activation in female mammals, and major targets for ovarian reserve preservation [5–7]. Therefore, strategies that preserve the ovarian reserve are essential for maintaining female fertility and delaying the onset of menopause [8, 9].

Calorie restriction (CR) is a well-known strategy where the reduced food intake promotes weight and fat loss, resulting in increased health and lifespan in mice [10]. CR also reduces the activation of primordial follicles, preserving the ovarian reserve and extending fertility of female mice, through modulation of mTOR and FOXO3 pathways [11–13]. Some studies suggest that beneficial effects of CR are due the reduced protein intake [14, 15] because protein-restricted diets also reduce body mass and composition, improve insulin sensitivity, and increase lifespan [15, 16]. More specifically, mice on branched-chain amino acids (BCAAs) restricted diets also live longer [17], despite protein and calorie intake being similar to control mice. BCAAs play important roles in regulating energy homeostasis, nutrient-sensing pathways, and insulin sensitivity [18].

In the ovary, activation of primordial follicles is reduced by protein restriction (9% of total calories from protein) and accelerated by increased protein intake (56% of total calories from protein) when compared to mice receiving a standard chow diet (22% of total calories from protein) [19]. The effects of protein restriction in ovarian reserve of mice have been linked to mammalian target of rapamycin complex 1 (mTORC1) signaling and endocrine signals, such as fibroblast growth factor 21 (FGF21) and adiponectin [19]. However, the literature about BCAA restriction and ovarian aging is still scarce and no direct comparison between calorie, protein and BCAA restriction has been performed to this point. Therefore, our study aimed to compare the effects of protein and BCAAs restriction on the preservation of the ovarian reserve compared to a classical, most studied, 30% CR approach.

2. METHODS

2.1. Animals and diets

This study was approved by the Animal Experimentation Ethics Committee of the Universidade Federal de Pelotas under the number 024998/2021-29. For this study, C57/BL6J mice (n=35) were obtained from the Animal Facility of the Universidade Federal de Pelotas and kept under controlled temperature and light conditions (22± 2°C, 12-hour light cycle from 7AM-7PM) with water ad libitum in groups of 2–3 mice/cage. At the age of 3-month-old, mice were randomized into four groups: control (CTL, n = 8), 30% calorie restriction (CR; n = 9), protein restriction (PR; 68% reduction in all AAs, n = 9) and branched-chain amino acid restriction (BCAAR; 68% reduction in BCAA only, n = 9). The diets were continued for a period of four months. Previous studies from our group observed a significant decline in the number or primordial and total follicles between 3 to 6 months of age, therefore this window of intervention was selected [20].

Mice were monitored for weight gain and food consumption throughout the study. For this, each mouse was weighed individually on a precision scale weekly. Food consumption was verified from the weekly difference between food offered and remaining. The food was offered daily (5–6 PM) for CR mice at the same time according to the weekly average consumption of mice in the control group, with a reduction of 10% in the first week, 20% in the second week and 30% from the third week onwards [21]. Meanwhile, CTL, PR and BCAAR mice had ad libitum access to diets, as shown in Table 1. Importantly, CTL, PR and BCAAR diets were isoenergetic and CTL and BCAAR diets were also isoproteic. As shown in Table 1, CTL and PR diets were made isoenergetic by increasing the amount of carbohydrates in the diet. The CTL and BCAAR diets were made isocaloric and isoproteic by increasing the amount of non-essential AAs in the diet. At the end of the experiment, estrous cycle was synchronized by injection of eCG (2.5 U) and hCG (5 U), 48 and 12 hs before euthanasia, respectively. After a 4-hour fast, mice were euthanized by exsanguination via cardiac puncture under isofluorane, followed by cervical dislocation. Mice were dissected and the ovaries collected and stored, one ovary frozen at −80°C and the other in a 4% formaldehyde solution for histological analysis. The blood was centrifuged and the serum separated for storage at −80°C.

Table 1.

Diet composition for the control/calorie restriction (CR) groups, protein restriction (PR) and branched-chain amino acids (BCAA) restriction (BCAAR) groups.

Ingredients	Control/CR (g/kg)	PR (g/kg)	BCAAR (g/kg)
Branched-chain aminoacids
L-isoleucine	7.8	2.54	2.54
L-leucine	25.4	8.27	8.27
L-valine	8.4	2.73	2.73
Other aminoacids
L-alanine	9.38	3.05	11.81
L-arginine	6.3	2.05	6.3
L-asparagine	20.58	6.7	22.38
L-aspartic Acid	20.58	6.7	24.22
L-cysteine	7.2	2.34	7.2
L-glucaminic acid	28.97	9.43	32.99
L-glutamine	33.77	11	35.78
Glycine	2.96	0.96	5.01
L-histidine	4.6	1.5	4.6
L-lysine	20.38	6.64	20.38
L-methionine	6.7	2.18	6.7
L-phenylalanine	6.6	2.15	6.6
L-proline	7.41	2.41	10.55
L-serine	7.41	2.41	10.12
L-theonine	9.7	3.16	9.70
L-tryptophan	3.4	1.1	3.40
L-tyrosine	6.9	2.25	6.90
Other ingredients
Sucrose	291.0	291.2	291.2
Corn Starch	150.0	232.4	153.0
Matodextrin	150.0	232.4	153.0
Corn oil	52.0	52.0	52.0
Olive oil	29.0	29.0	29.0
Cellulose	30.0	30.0	30.0
Mineral mix (AIN-93M-MX)	35.0	35.0	35.0
Calcium phosphate	8.2	8.0	8.2
Vitamin mix	10.0	10.0	10.0
TBHQ antioxidant	0.012	0.012	0.012
Food coloring	0.1	0.1	0.1
% energy from
Protein	22.0	7.0	21.9
Carbohydrates	59.4	74.0	59.6
Fats	18.6	19.0	18.5
Kcal/g	3.9	3.9	3.9

2.2. Serum concentrations of BCAAs

The serum BCAA levels were measured using a Leucine Standard BCAA Assay Kit (MAK003; Sigma-Aldrich, U.K.). The reaction product results in a colorimetric change proportional to the amount of leucine present in the sample [22].

2.3. Histology and follicle counts

For histological evaluation, ovaries were removed from 4% paraformaldehyde and subjected to dehydration in alcohol, cleared in xylol and embedded in Paraplast Plus (Sigma Chemical Company^®, St. Louis, MO, USA). Subsequently, the ovaries were sequentially cut in a Leica automatic microtome model RM2245 (Leica Biosystems Newcastle Ltd, Newcastle Upon Tyne, UK) at a thickness of 5 μm, and 1 out of 6 cuts was removed and placed on slides standard histology slides. After drying the slides in an oven at 55°C, the slides were stained with hematoxylin-eosin, mounted with coverslips and synthetic resin (Sigma Chemical Company^®, St. Louis, MO, USA). Images of ovarian sections were evaluated in a microscope (Nikon Eclipse E200, Nikon Corporation, Japan) using 10 and 40X magnification. The classification of ovarian follicles was determined as follows: primordial when surrounded by a layer of flattened granulosa cells; transition whey surrounded by at least one cuboidal granulosa cell; primaries when surrounded by a layer of cuboidal granulosa cells; secondaries when surrounded by two or more layers of cuboidal granulosa cells; tertiary when the presence of antrum and cumulus oophorus complex is observed, only nucleated follicles were considered [23]. The total number of follicles was divided by the total number of sections. The size of the oocytes was evaluated based on the average of the vertical and horizontal length by measurement in a microscope (Nikon Ecplipse E200, Nikon Corporation, Japan) using three oocytes/mice/follicle type [24].

2.4. Vaginal cytology

For cytological analysis, vaginal cells were collected with a micropipette, by washing the vaginal canal with saline. After that, the material was smeared in a slide and stained with a fast panoptic (Laborclin, Pinhais, PR, Brazil) for visualization under a microscope (Nikon Eclipse E200, Nikon Corporation, Japan). The stages of the estrous cycle were defined by the cell types present in the slide as described previously [25]. In addition, the duration between cycles was defined from the interval between two estrous phases and the relative percentage of each phase adjusted for the total period of 12 days evaluated.

2.5. Statistical analysis

Statistical analyses were performed on GraphPad Prism 6.0 software. Normality was tested using the Shapiro-Wilk test. One-Way ANOVA test was used followed by a Tukey post-hoc to evaluate body mass gain, food consumption, body composition, BCAA levels, follicle numbers and estrous cycle. Weight gain between months was analyzed by Two-way ANOVA for repeated measures followed by a Tukey post-hoc test. P values <0.05 were considered statistically significant.

3. RESULTS

3.1. Body mass, body composition and food intake

We observed that calorie, protein and BCAA restriction prevented body mass gain compared to control mice (Figure 1A, p <0.05). The CR group had the lowest body mass throughout the experiment when comparing with control, PR and BCAA groups (Figure 1B, p<0.05). The normalized weight of intraabdominal adipose tissue indicated a reduction of fat mass in CR mice, while an increase in PR mice was observed compared to CTL mice (Figure 1C, p<0,05).

Figure 1.

A) Body mass during treatment; B) body mass change from 3 to 7 months of age; C) Intraabdominal fat mass; D) Total calorie intake/day/gram of body mass; E) Calorie intake from protein/day/gram of body mass; F) Intake of branched chain amino acids (BCAA)/day/gram of body mass in the control (CTL), calorie restriction (CR), protein restriction (PR) and branched-chain amino acid restriction (BCAAR) groups. Different letters indicate statistical differences between groups at p<0.05. A) Two-way ANOVA repeated measures, B,D,E,F) One-Way ANOVA parametric test and C) One-Way ANOVA non-parametric test.

As expected, we observed that CR mice consumed less calories adjusted to body mass (Figure 1D, p=0.002). In addition, CR mice had lower energy intake from protein (Figure 1E, p=0.0003) and lower consumption of BCAA (Figure 1F, p<0.0001). Mice in the PR group had similar energy intake to the control group, but less energy intake from protein (Figure 1E, p<0.0001). In addition, BCAAR mice consumed more calories than CR and PR groups (Figure 1D, p=0.03), including more energy intake from protein (Figure 1E, p=0.008), despite the reduction in BCAA intake (Figure 1F, p<0.0001). However, BCAA intake relative to body mass was lower in the PR and BCAAR groups compared to the control and CR groups (Figure 1F). When not adjusting for body weight, total calorie intake was reduced in CR and PR mice compared to CTL and BCAAR mice (Suppl. Figure 1A). Protein energy intake was the lowest in PR mice, and not different between CTL and BCAAR mice (Suppl. Figure 1B). Calorie intake from carbohydrates was the highest in PR mice, and not different between CTL and BCAAR mice (Suppl. Figure 1C). Calorie intake from fat followed the same pattern as from total calorie intake (Suppl. Figure 1D).

3.2. Serum concentrations of BCAAs

The analysis of serum BCAA levels, using a leucine standard, indicates that PR mice had lower concentrations of BCAAs compared to the CTL group (Figure 2, P<0.05). Serum levels of BCAA were not different between CTL, CR and BCAAR groups.

Figure 2.

Serum concentrations of branched chain amino acids (BCAA) using a leucine standard in control (CTL), calorie restriction (CR), protein restriction (PR) and branched chain amino acid restriction (BCAAR) groups. Different letters indicate statistical differences between groups at p<0.05 using a One-Way ANOVA non-parametric test.

3.3. Vaginal cytology

Cyclicity was evaluated through vaginal cytology after 3.5 months of treatment. We observed that mice from CR and PR groups had less cycles compared to CTL group in the period evaluated (Figure 3A and B). In addition, CR and PR mice remained longer in the diestrus stage compared to the CTL group (Figure 3C). BCAAR mice had less cycles than the CTL group, however more than CR mice (p>0.05).

Figure 3.

A) Number of complete cycles, as defined by the interval between two separate estrus events; B) number of estrus, as defined by the detection of the cellular pattern indicating estrus; C), percentage of time in each stage for control (CTL), calorie restriction (CR), protein restriction (PR) and branched-chain amino acid restriction (BCAAR) groups. Different letters indicate statistical differences between groups at p<0.05 using a One-Way ANOVA parametric test.

3.4. Ovarian reserve

We observed that CR and PR mice had more primordial follicles compared to CTL mice (Figure 4A; p <0.05). However, BCAAR mice were not different from CTL mice (Figure 4A). Furthermore, PR mice had more transition follicles and a lower number of secondary follicles when compared to CTL group (Figure 4B, D; p<0.05). Regarding to other follicle classes, no differences were observed (Figure 4B–C, E–F).

Figure 4.

Number of A) primordial; B) transition; C) primary; D) secondary; E) tertiary; F) total follicles between control (CTL), calorie restriction (RC), protein restriction (PR) and branched-chain amino acid restriction (BCAAR) groups. Different letters indicate statistical differences between groups at p<0.05 using a One-Way ANOVA non-parametric test.

Oocytes diameter is associated to primordial follicle activation as follicles have to reach a certain diameter in order to be activated [26]. We observed that PR mice had smaller oocytes in primordial and primary follicles compared to CTL mice (Figure 5, p<0.05).

Figure 5.

Oocyte size in A) primordial; B) primary; C) secondary follicles between control (CTL), calorie restriction (RC), protein restriction (PR) and branched-chain amino acid restriction (BCAAR). Different letters indicate statistical differences between groups at p<0.05 using a One-Way ANOVA parametric test.

4. DISCUSSION

In this study, we performed the first direct comparison between the effects of restriction in the intake of calories, protein and BCAA on the ovarian reserve in female mice. We observed that BCAAs restriction prevented body mass gain but did not affect the ovarian reserve. Conversely, CR and PR promoted a more intense reduction in body mass gain and preserved the ovarian reserve. Interestingly, PR promoted its benefits even though energy intake was similar to the CTL group. This points to a direct effect of AAs restriction in primordial follicle activation, which is not mimicked by BCAA restriction alone. This is interesting as PR may be a better strategy than CR for translational studies manipulating human diets to preserve fertility.

We observed that BCAA restriction had no effect on the ovarian reserve of young female mice. Despite BCAA restriction being effective in reducing body mass gain compared to CTL mice, this decrease was less evident than in CR and PR mice. Also, adjusted calorie and protein intake was higher in the BCAA restricted mice compared to CTL mice, which could blunt beneficial effects of this intervention. We also observed that CR and PR attenuated age-related decline in the ovarian reserve as previously reported [27, 28]. Activation of mTOR in granulosa cells and phosphorylation of FOXO3 in the oocyte regulate primordial follicular activation [5–7]. CR [11–13] and PR [19] modulate mTOR activation in ovarian tissues, therefore, reduced primordial follicle activation can be mediated though reduced mTOR activation in those diets. However, no previous study directly compared CR, PR and BCAA restriction. Our findings are aligned with prior reports that show that reducing calorie intake in female rodents results in greater maintenance of fertility [11]. We have previously shown that 30% CR for three months prevented primordial follicle decline in mice [27]. Similarly, 10% CR was enough to positively impact the ovarian reserve after a six-month intervention [29]. Interestingly, we observed that the effects of PR in the ovarian reserve are very similar to CR. Previous studies showed that activation of primordial follicles was reduced by PR (9% of total energy intake) and accelerated by increasing protein intake (56% of energy from protein) when compared to a control diet (22% of energy from protein) [28]. Despite a similar effect in the preservation of the ovarian reserve, the reduction of protein intake imposed by CR is small compared to PR, and calorie intake in PR mice is greater than in CR mice. Therefore, CR and PR can promote similar effects on ovarian aging, despite divergent macronutrient intake profiles. A previous study showed that a 75% BCAA restriction (compared to 67% in our study) reduced the number of primordial follicles in young female C57BL/6J mice [30]. However, the diets low on BCAA, also had lower AA content, and lead to reduced serum BCAAs levels [30]. In our study, reduced BCAA in the diet was compensated by increasing other AAs, and we observed that serum BCAAs concentration was reduced only in PR compared to CTL mice, when total AAs availability was reduced. This suggests that divergent pathways may be regulating the effect of restrictive diets on the ovarian reserve.

We observed that BCAA, protein and calorie restriction prevented body mass gain throughout the study. Interestingly, protein and BCAA restriction reduced body mass gain even though these mice had a higher adjusted energy intake. These results are similar to previous observations, where even with higher adjusted calorie intake, reducing proteins and BCAAs intake prevented body mass gain in mice [17, 31]. Similarly, studies have also shown in humans that moderate protein restriction can reduce body and fat mass gain [15]. Low AA diets increase food intake, however, adding back BCAAs in these low AA diets restored food intake to control mice levels [32], suggesting regulation of food intake is mediated by BCAA intake. Very low-protein diets decrease food intake and promote weight loss, associated to hypothalamic inhibition of mTOR signaling in mice [33]. Our results are similar, as PR mice had a similar adjusted calorie intake to CTL mice, however BCAAR mice had a 10% higher adjusted calorie intake compared to CTL mice. BCAAR mice had higher calorie and protein intake than CTL and PR mice, resulting in less pronounced body mass decrease compared to PR mice. Previous studies have shown that BCAA restriction has different effects in male and female mice, as lifelong restriction of BCAAs increased longevity in males only [17]. Additionally, female mice fed lifelong low BCAA diets had lower body mass only after 10 months of age, while male mice showed significant differences already at 5 months of age [17]. Thus, our findings indicate that BCAA restriction in an isoproteic diet has moderate effects in body mass gain in young female mice compared to PR and CR, which may impair any beneficial effects in the ovarian reserve.

Regarding serum levels of BCAAs, we observed that only PR mice had a reduction in serum concentrations of BCAAs compared to control mice. These results suggest that BCAA levels are regulated post-ingestion, given that even with the same levels of dietary BCAA restriction for PR and BCAAR, only PR mice had reduced serum BCAAs. Similarly, mice fed diets with double the amount of BCAAs did not show increased serum levels of BCAAs, reinforcing the hypothesis that serum concentrations are regulated post ingestion [34]. Serum BCAA concentration is highly correlated to total protein intake in the diet [14], suggesting a direct effect of protein intake on whole body BCAA catabolism. This is aligned with our observations, as only PR mice had a reduction in serum levels of BCAAs. Since our BCAA diet was compensated by increasing other AAs intake to be isoproteic, total AAs intake was slightly higher in BCAAR mice compared to CTL and much greater than in PR mice. Other studies observing effects of BCAA restriction in the diet did not compensate intake of other AAs to make the diets isoproteic [30]. The metabolism of BCAAs is complex and influenced by several factors, such as body composition, insulin sensitivity, gut microbiota and inflammation [35], therefore future studies should observe the effects of BCAA restriction in different conditions, such as in older and/or obese mice.

The dietary pattern also affected cyclicity of mice. We observed that CR and PR mice spent less time in estrus and had fewer cycles during the evaluated period. This was less evident in BCAAR mice. It is well known that CR affects female fertility via hypothalamic and peripheral mediators associated to energy homeostasis. While CR can preserve the ovarian reserve of preantral follicles, it can negatively affect growth of tertiary follicles, ovulation and fertility while mice are calorie restricted [36]. Therefore, the reduction in cyclicity in CR and PR indicates less ovulatory cycles, and less opportunities to become pregnant, which could have a negative effect in fertility. Rodents fed low-protein diets (4% of total energy intake from protein) also had disruption of estrous cyclicity and inhibition of follicular development, mainly associated with impaired expression of hypothalamic proteins and FGF21 levels [37]. Disturbances related to the estrous cycle in rats fed low-protein diets were reversed after a period of refeeding with normal-protein diets (18% of energy from protein) [37]. However, we observed that dietary restriction of BCAAs only interfered less with cyclicity. This may be related to the higher energy intake and the lower impact of this diet on body mass gain, which seems to be a hallmark effect of these restrictive diets regarding preservation of the ovarian reserve and cyclicity. Thus, our findings highlight the importance of calorie intake for regulation of the estrous cycle, but also macronutrient ratio, helping to optimize diets aiming not only to preserve the ovarian reserve, but also cyclicity and reproductive longevity.

The importance of the pattern of body fat distribution as a modulator of the metabolic profile in humans and rodents is well known [38]. Different dietary patterns have also been shown to change body composition [39, 40]. Rodents fed high-fat diets have increased visceral fat deposition, hyperinsulinemia, and insulin resistance within just four weeks [41]. Conversely, CR can improve body composition, insulin resistance and inflammatory profile in mammals [42, 43]. We have previously shown that 30% CR for three months was able to prevent body mass gain and visceral fat deposition, associated to increased insulin sensitivity in young female mice [27]. We again observed a reduction in body mass and intrabdominal adipose tissue in CR mice. However, while others suggest that diets restricted in protein and BCAAs reduced fat mass and benefited glucose metabolism [15, 44, 45], the same was not observed in our study. PR mice had increased deposition of intrabdominal fat. BCAA restriction did not seem to affect fat deposition compared to CTL mice in our study. Such differences can be explained by the carbohydrate:protein intake ratio in the diet and by the AA distribution pattern. The PR group had a strong reduction in total AA intake which may drive increased deposition of body fat.

In conclusion, our results show that PR and CR have very similar effects in the preservation of the ovarian reserve. This happened even though the two strategies had divergent effects in calorie and protein intake. In turn, BCAA restriction alone was not able to mimic these effects and preserve the ovarian reserve, as it resulted in greater energy and protein intake and less impact on body mass. More studies are needed to understand the mechanisms by which these diets affect metabolism and preserve the ovarian reserve of aging females.