Protein restriction during puberty alters nutritional parameters and affects ovarian and uterine histomorphometry in adulthood in rats

Abstract

In a large part of the population inefficient ingestion of proteins, whether for cultural, aesthetic or economic reasons, is a global concern. Low‐protein diets can cause severe functional complications, mainly during the development and maturation of organs and systems, including the female reproductive system. The present study investigated the effect of nutritional protein restriction during puberty on the oestrous cycle and expression of sex steroid receptors (AR, ERα e ERβ) in ovarian and uterine tissues of adult rats. Protein restriction promoted lower body weight gain, feed efficiency and higher caloric intake. There was an increase in the oestrus phase arrest without changing the total length of the oestrous cycle. The consumption of low‐protein diet also reduced the thickness of the uterine endometrium (uterine epithelium and endometrial stroma) in addition to increasing the number of primary and atretic follicles in the ovaries. Furthermore, the low‐protein diet reduced the levels of androgen receptor (AR) and increased the oestrogen receptor β (ERβ) in the ovary, while no significant changes were observed in the uterus. Our study reinforces the importance of adequate protein intake during puberty, since physiological changes in this developmental period interfere with the histomorphometry of the ovaries and uteri, possibly resulting in impaired folliculogenesis and fertility in the reproductive period.

1. INTRODUCTION

Nutritional status in prenatal and early postnatal life plays a critical role in the development and homeostasis of the organism. Early malnutrition seems to alter the original programming of the organs, especially those dependent on specific developmental stages, which may result in long‐term metabolic changes. ¹ , ² Proteins are substances found in higher amounts in the body, constituting approximately 10%‐20% of the cell mass. Their actions are divided between structural components such as the formation of the cytoskeleton, collagen fibres and elastin and functional components such as participating in the composition of cellular enzymes, hormones and trophic factors which are essential for cell survival. ³

Studies described in the 1970s and the 1980s have associated nutritional restriction, especially protein restriction, not only with the delay in the intrauterine growth of the offspring, but also with the predisposition for the development of chronic diseases in adulthood, particularly hypertension, obesity and type 2 diabetes. ⁴ , ⁵ Overall, due to cultural, aesthetic or economic reasons, insufficient protein intake is considered an extremely concerning issue.

Puberty is a period of rapid morphological, endocrine and behavioural changes. ⁶ It is a complex process that involves the maturation of hypothalamic‐pituitary‐gonadal (HPG) and hypothalamic‐pituitary‐adrenal (HPA) axes. The physiological events associated with these systems are continuous processes starting in intrauterine life, remaining latent during childhood and extending later. ⁷ Therefore, changes in these periods may have negative or even irreversible consequences in adulthood.

Early protein restriction models are well characterized and widely studied. ⁸ , ⁹ Previous studies have shown that female rats from mothers who consumed a low‐protein diet exhibited a delay in the onset of puberty accompanied by a regular oestrous cycle and atrophied uterine glands. ¹⁰ Additional studies have revealed that changes in maternal nutrition, both hypercaloric and hypocaloric, can lead to definitive consequences on the reproductive system of female offspring, thereby promoting an advance in puberty. ¹¹ Maternal protein restriction during pregnancy and/or lactation causes changes in the female offspring, thus affecting the number of developing and antral follicles and overall ovarian function. ¹² In addition, it delays sexual maturation and induces premature ageing of the entire reproductive function. ¹³ Nutritional deficiency in the gestational and postnatal periods was significantly associated with failures in both size at birth and in growth rate and, particularly, an underdevelopment of the ovaries was observed. ¹⁴ Considering this scenario, a high number of smaller antral follicles and a reduction in the number of Graafian follicles are documented, unravelling the potential damage to the ovulatory process. ¹⁵

The animals subjected to food restriction did not show any variations in the follicle number, but instead, they had a dramatic change in follicular maturation. ¹⁶ These findings indicate that certain periods of sexual development are detrimental to the rodent's life and that malnutrition can permanently affect ovarian development. The diet plays a fundamental role in the regulation of steroidogenesis. Molecular changes in critical periods of development can affect ovarian development in several ways with irreversible damage. ¹⁶ Differential signalling regulated by the androgen receptor (AR) and oestrogen receptor (ER) are extremely important to orchestrate many functions in ovarian and uterine tissues, and any change in hormone receptor regulation may compromise the function of these organs, resulting in severe pathophysiological alterations. ¹⁷

The type of diet in the early phase is of great importance in the development and regulation of ovarian and uterine functions, and the presence of low‐protein levels causes irreversible damage to different organs and tissues. The present study aimed to assess the impact of protein restriction during puberty on the oestrous cycle and on the levels of sex steroid hormone receptors (AR, ERα and ERβ) in the ovarian and uterine tissues of rats in adulthood. The endocrine events that occur in rodents are similar to other mammals, including humans. Therefore, the effects of diet on hormonal regulation can provide important information for the understanding of reproductive damage in humans.

2. MATERIALS AND METHODS

2.1. Animals and experimental design

Twenty adult Fischer 344 female rats (25 days old, 60 g body weight) were used in this study. The animals were bred and raised in the vivarium of the Department of Anatomy of the Institute of Biosciences, UNESP, Botucatu. They were kept in polypropylene boxes (43 × 30 × 15 cm), with self‐clad shavings as substrate, in a suitable environment under controlled conditions (12‐h light/dark cycle and temperature 23 ± 1°C). All animals were distributed according to the diet protocol: control group (CG, n = 10) received a solid diet consisting of a standard diet (AIN‐76A containing 17% protein) and water ad libitum. The protein restriction group (RG, n = 10) was submitted to severe protein restriction receiving a diet containing low‐protein content (modified AIN‐93 containing 6% casein, Table 1) and water ad libitum. After weaning (at 25 days of age), all animals received diets for 35 consecutive days, both of which were isocaloric (422 kcal/100 g of food) containing mineral salts and vitamins at the same concentration. ¹⁸

TABLE 1.

Composition of the normoprotein and low‐protein (AIN‐93) diets offered to the animals

Ingredients a	Normoprotein (17% protein) g/Kg	Low protein (6% protein) g/Kg
Starch	397	480
Casein (84% protein) b	202	71.5
Dextrin (90%‐94%)	130.5	159
Sucrose	100	121
Soy oil	70	70
Fibres	50	50
Salt mixtures AIN93G c	35	35
Vitamin mixtures AIN93G c	10	10
L‐Cystine	3	1
Choline hydrochloride	2.5	2.5

^aDiet for gestational phase in rodents—AIN‐93G.

^bCorrected values according to the protein content in casein.

^cAccording to AIN‐93G. To know the detailed composition and utilization of diet, see Cavariani et al. ¹⁹ and Santos et al. ²⁰

After the experimental period, the females were anaesthetized with ketamine and xylazine and were then euthanized in oestrus stage in the morning (9:00 am) for subsequent dissection, harvesting and processing of the biological samples.

2.2. Assessment of the oestrous cycle

The animals were monitored throughout the experimental period using the vaginal wash technique previously described by Marcondes et al. ²¹ and Belardin et al. ²² Thus, the cells detached from the vaginal epithelium were individually collected with the aid of a LabMate 0.5‐ to 10‐µL pipette (International LabMate Ltd, St Albans, UK), containing 10 µL of 0.9% (v/v) saline, and transferred for numbered clean slides. The analysis schedule was always held at 9:00 am All slides were analysed under a microscope coupled with a Zeiss Axiophot II digital camera (Carl Zeiss).

2.3. Monitoring of food and water consumption and body and organ weight

Throughout the experiment, the consumption of water and feed and the respective weights of the animals were measured. After euthanasia, each animal was subsequently weighed and submitted to abdomen‐pelvic laparotomy to remove the organs (uteri, ovaries and oviducts), liver and visceral fat that were weighed separately. The determination of body weight and organ weight was performed using an Owalabor analytical balance (OwaLabor).

2.4. Histopathological analysis

After the animals were euthanized, the ovaries and uterine horns (n = 8) of each group were collected and fixed in buffered formalin solution (10% (v/v)) for 24 h. After the fixation period, the organs were washed in running water, dehydrated in an increasing series of ethanol, diaphanized in xylol and included in paraplastic (Oxford Labware). For the analysis of ovarian and uterine tissues, 4‐µm sections were obtained using a Leica RM2165 microtome and transferred to silanized slides, deparaffinized in xylol and then stained with haematoxylin & eosin (H.E). All analyses were performed using the entire sample and were digitalized under a Zeiss Axiophot II microscope (Carl Zeiss).

2.5. Morphometric analysis of ovarian follicles, corpus luteum and uterine endometrium

For morphometry, the samples prepared for light microscopy were used, analysing one section and neglecting the five subsequent sections, until the 61st section, resulting in 13 repetitions analysed by ovary ²³ and for each uterine horn. The identification of follicles was based on the classification proposed by Camargo et al. ²⁴ The primordial, primary, growing follicles (with more than three layers of follicular cells), types 6 (antral areas) and 7 (oocyte displaced in the antrum) together, the mature, the atretic and the corpus luteum were counted and analysed. The endometrium was analysed based on the lining epithelium and endometrial stroma. The analysed structures included the number of follicles and corpus luteum, and thickness of the epithelium and endometrial stroma. The data were evaluated using the computerized image analysis system KS‐300 ‐ Zeiss.

2.6. Immunohistochemical analysis for AR, ERα and ERβ in the ovary and uterus

Samples of ovarian and uterine tissue (n = 8 per group) were dewaxed in xylol and hydrated in a series of ethanol. Antigen recovery was performed in a microwave oven using 0.01M sodium citrate buffer, pH 6.0. Next, the sections were subjected to reaction with specific primary antibodies (AR, ERα and ERβ, 1:100; Santa Cruz Biotechnology, Inc) and incubated in a humid chamber overnight at 4°C. After incubation, the slides were washed in TBS‐T buffer and then incubated with a secondary antibody (Biotinylated Mouse Anti‐Rabbit Immunoglobulins—DAKO^® CYT) at room temperature for 1h. Then, the slides were washed in TBS‐T and subjected to reaction with diaminobenzidine chromogen (DAB; Sigma) for 5 min. Finally, the slides were counterstained with haematoxylin. Positive and negative control slides were used. Slides were analysed and the images were scanned using a camera coupled with a Zeiss Axiophot II microscope (Carl Zeiss).

2.7. Western blotting analysis and quantification of AR, ERα and ERβ in the ovary and uterus

Samples of ovarian and uterine tissues (n = 6 per group) were frozen at −80°C. After morphological characterization, tissues were homogenized in RIPA lysis buffer (Pierce Biotechnology) containing protease inhibitors. The homogenates were centrifuged, and an aliquot was used for protein quantification. 70 micrograms of proteins was solubilized and applied onto 4%‐20% polyacrylamide gel (SDS‐PAGE). After electrophoresis, the proteins were transferred to a nitrocellulose membrane. After the membranes were blocked with 3% BSA (w / v), they were incubated with monoclonal antibodies anti‐AR (sc‐7305), anti‐ERα (sc‐8005) and anti‐ERβ (sc‐53494) (Santa Cruz Biotechnology, Inc), diluted 1:500 in 1% BSA overnight. The membranes were washed three times in basal solution (TBS‐Tween 1% [v / v]) and incubated with secondary antibody (1:10 000) in 1% BSA (w / v), conjugated to peroxidase‐HRP (Sigma) for 1 hours. After removing the reactions, the membranes were covered with chemiluminescent substrate for Western blotting. Results were determined by checking for positive reactions or the absence of bands. The band intensity was quantified using the optical densitometry index (IOD) corrected by the endogenous β‐actin (housekeeping protein).

2.8. Statistical analysis

The results were studied using parametric tests (Student's t test) or non‐parametric tests (Mann‐Whitney U test) depending on the data distribution. The results are expressed as the mean ± SD and are presented in tables and graphs, considering a P value < 0.05 as statistically significant. GraphPad Prism® software (version 5) was used for graphical analyses.

3. RESULTS

3.1. Low‐protein diet at puberty affects nutritional parameters and oestrous cycle in adulthood

During the experimental period, the animals were monitored daily for the oestrous cycle and assessment of food, liquid and body weight gain based on the same assessment criteria. Protein restriction at puberty reduced body weight from the 20th to the end of the dietary intake period (day 20 to day 35; Figure 1A). While food consumption was significantly higher in animals that consumed a low‐protein diet after the 20th day (P < 0.05, Figure 1B), the amount of total protein intake was lower (about 50% reduced) during all the periods (Figure 1C). Water consumption was similar between the groups, except on days 5 and 15 of the experiment (Figure 1D). Interestingly, energy consumption was higher in animals under protein restriction after day 20; however, feed efficiency was reduced by approximately 22% in these animals, thereby providing evidence for the changes in the use of calories for weight gain. These findings show the nutritional impact of protein restriction and are aligned with previous findings involving other experimental models.

FIGURE 1.

General nutritional features in animals consuming low‐protein or normoprotein diets. A, Evolution of body weight, B, food consumption, C, dietary protein consumption, D, liquid consumption, E, energy consumption and (F) feed efficiency. * P < .05; Student's t test. The values are expressed as mean ± SD

The final body weight, ovary weight, oviduct weight and liver weight did not vary with protein restriction during puberty (Table 2). On the other hand, the weight of the uterus and the weight of the kidneys were significantly lower in animals that consumed the low‐protein diet. The total visceral fat content was also reduced in restricted animals.

TABLE 2.

Evaluation of body weight and reproductive organs (ovaries, oviducts and uteri), liver and visceral fat weights between the experimental groups (N = 10 / group)

Parameters	Groups
	Control	Restricted
Final body weight (g)	174.38 ± 26.0	161.88 ± 6.69
Ovary weight (mg)	68.2 ± 6.5	68.8 ± 5.4
Oviduct weight (mg)	27.7 ± 5.8	29.0 ± 4.2
Uterine weight (mg)	418.43 ± 95.8	349.03 ± 63.9*
Liver weight (g)	12.82 ± 2.32	11.47 ± 0.92
Kidney weight (g)	3.16 ± 0.5	2.64 ± 0.2*
Visceral fat weight (g)	6.87 ± 1.54	5.15 ± 0.77*

^* P < .05 indicates a significant difference. Student's t test. Values are expressed as mean ± SD.

The analysis of the oestrous cycle did not show definitive change in the oestrous cycle length; that is, there was no exacerbated prolongation of the cycles in the restricted animals (Table 3). The percentage of permanence in the oestrus phase was significantly higher in restricted animals (about 30% oestrus phase arrest), without, however, causing chronic anovulation. The percentage of permanence in the metaoestrus and dioestrus stages was unaltered by the low‐protein diet (Table 3).

TABLE 3.

Duration of oestrous cycle (in days) and frequency (%) of interruption in oestrus, metaoestrus and dioestrus after protein restriction (n = 10 animals/group)

Groups	Cycle duration (days)	Oestrus persistent/cycle (%)	metaoestrus persistent/cycle (%)	Dioestrus persistent/cycle (%)
Control	5.37 ± 1.41	21.2 (6;26.74)	16.02 (0;21.1)	32.41 (0;42.4)
Restricted	6.12 ± 1.28	30.3 (6;33.2)*	18.91 (6;27.4)	36.22 (8.3;41.5)

^* P < .01 indicates significant statistical difference from the control group. The evaluation of the cycle duration was performed using the Student t test, and the results are expressed as mean ± SD. Fisher's test was used to compare the frequencies.

3.2. Histomorphometrical changes in ovarian and uterine tissues are related to low‐protein diet consumption at puberty

Table 4 depicts the number of ovarian structures (primordial, primary, secondary, pre‐antral, antral 6 and 7, and atretic follicles and haemorrhagic and regressing corpus luteum). The number of primary and atretic follicles was significantly higher in the group that received a low‐protein diet compared to the control group (P < 0.05). It seems that the increase in the number of primary follicles reveals a protective mechanism for follicle reserve considering the scarcity of circulating proteins. On the other hand, the primordial, secondary, pre‐antral and antral follicles and the corpus luteum did not undergo significant quantitative changes (P > 0.05).

TABLE 4.

Follicular counts in the ovaries of animals consuming low‐protein diet (N = 10 animals/group)

Parameters	Groups
	Control	Restricted
Primordial follicle	4.5 ± 2.89	5.75 ± 1.65
Primary follicle	9.75 ± 4.61	15.38 ± 3.28*
Secondary follicle	4.5 ± 1.61	5.63 ± 0.89
Pre‐antral follicle	8.25 ± 3.43	10.5 ± 2.10
Antral follicle	4.38 ± 2.19	4.50 ± 1.73
Atretic follicle	1.11 ± 0.7	2.82 ± 0.9*
Haemorrhagic CL	3.3 ± 1.60	4.0 ± 1.73
Regression CL	3.88 ± 2.07	4.52 ± 2.28

Student's t test. Values are expressed as mean ± SD.

Abbreviation: CL, corpus luteum.

^* P < .05 shows significant statistical difference.

The low‐protein diet reduced the thickness of the uterine endometrium of the animals (Table 5). The uterine epithelium of the analysed uterine horns was significantly thinner in the proximal, middle and distal parts in animals subjected to a low‐protein diet compared to the epithelium of normoprotein animals (P < 0.05). Similarly, the endometrial stroma showed reduced thickness in the animals that consumed a low‐protein diet compared to the stroma of the control animals (P < 0.01). The layers of the myometrium and perimetrium did not show any morphological or thickness variation (P > 0.05).

TABLE 5.

Morphometric analysis of the uterine epithelium and endometrial stroma in the uterine horns of animals subjected to protein restriction

Parameters	Groups
	Control	Restricted
Uterine epithelium (μm)
PP	47.78 ± 17.45	26.12 ± 12.19*
MP	54.24 ± 14.47	28.24 ± 14.28*
DP	41.32 ± 10.30	20.63 ± 8.85*
Endometrial stroma (μm)
PP	372.44 ± 39.04	123.14 ± 36.11**
MP	451.42 ± 41.56	175.76 ± 72.91**
DP	411.50 ± 45.37	188.80 ± 77.03**

Student's t test (N = 8 animals per group).

Abbreviations: DP, distal part; MP, middle part; PP, proximal part.

Values are mean ± SD.

^* P < .05,

^** P < .01.

In the restricted group, some changes in the ovarian tissue were observed mostly associated with ovarian stroma. Figure 2A,B shows normal folliculogenesis and luteogenesis. The restricted animals presented dilation of the ducts of the rete ovarii, through which the structures related to ovarian vascularization take place (Figure 2C,D). The appearance of vascular congestion in the entire stroma was also frequent along with the infiltration of lipid droplets in the ovarian parenchyma.

FIGURE 2.

Photomicrographs of the ovarian tissue of animals underwent protein restriction. A,B, Control ovary showing folliculogenesis and normal corpus luteum formation. C,D, Dilated area of the rete ovarii in restricted animals. HE. Bar = 100 µm. (*) indicates atretic follicles; arrow = primary follicles

In normal uterine tissue, the high columnar uterine epithelium lines the uterine stroma, which contains endometrial glands with a secretory aspect (Figure 3A,B). The changes observed in the restricted animals were mostly associated with the endometrium, in which the lumen was reduced and presented a slit papillary‐like projection (Figure 3C). In the epithelium, numerous sites of vacuolization with cellular debris appeared surrounding the endometrial projections (Figure 3D,E). The endometrial glands had irregular epithelium and leucocyte infiltration was observed in the stroma and inside the endometrial glands (Figure 3E‐G).

FIGURE 3.

Photomicrographs of the uterine tissue of animals underwent protein restriction. A,B, Normal endometrium. C, Endometrium‐containing slit papillary projections. D, Papillae lined by low columnar epithelium. E, Epithelial vacuolization; arrow shows cellular vacuolization. F, Inflammatory infiltration in the stroma; arrow shows narrow glands with epithelial cells and leucocytes. G, Glandular leucocyte infiltration; (*) polymorphonuclear cells invading the lumen of the gland. HE. Bar = 100 µm

3.3. AR and ERβ were changed in the ovaries but not in the uteri after low‐protein consumption at puberty

The intensity of AR, ERα and ERβ immunoreactivities in the ovarian follicles and in the endometrium was analysed depending on its location and presence at the cytoplasmic and nuclear levels (Table 6).

TABLE 6.

Immunolocalization (cytoplasmic and nuclear scores) and intensity of AR, ERα and ERβ staining in the experimental groups (n = 8 animals per group)

Targets	Control	Restricted
Ovary
AR
Primordial follicle	C+,N0	C++,N0
Primary follicle	C+,N0	C++,N0
Growing follicle	C++,N++	C+,N+
Antral follicle (6 and 7)	C+,N+	C0,N+
Mature follicle	C++,N+	C+,N+
Atretic follicle	C+,N0	C+,N0
Corpus luteum	C++,N+	C++,N+
Stroma	C+,N0	C++,N0
ERα
Primordial follicle	C+,N0	C+,N0
Primary follicle	C0/+,N0/+	C+,N0
Growing follicle	C0/+,N0/+	C0,N0
Antral follicle (6 and 7)	C0/+,N0/+	C0,N0/+
Mature follicle	C+,N0/+	C+,N+
Atretic follicle	C+,N0/+	C0/+,N0/+
Corpus luteum	C+,N+	C+,N0/+
Stroma	C0/+,N0/+	C+,N0/+
ERβ
Primordial follicle	C0/+,N0/+	C+,N0
Primary follicle	C+,N0/+	C+,N+
Growing follicle	C0/+,N0/+	C0,N0
Antral follicle (6 and 7)	C0/+,N0/+	C+,N0/+
Mature follicle	C+,N0/+	C+,N+
Atretic follicle	C+,N0/+	C0/+,N0/+
Corpus luteum	C+,N+	C+,N0/+
Stroma	C0/+,N0/+	C+,N+
Uterus
AR
Epithelium	C++,N+/++	C++,N+
Stroma	C++,N+/++	C+/++,N+
Myometrium	C++,N+	C++,N+
Perimetrium	C0,N0	C0,N0
ERα
Epithelium	C+,N++	C+,N++
Stroma	C+,N+/++	C0/+,N0/+
Myometrium	C+,N+/++	C0/+,N0/+
Perimetrium	C0,N0	C0,N0
ERβ
Epithelium	C+,N++	C+,N++
Stroma	C+,N++	C0/+,N++
Myometrium	C+,N++	C+,N0/+
Perimetrium	C0,N0	C0,N0

Immunoreaction intensities were classified as strong (+++), moderate (++), weak (+) or absent (0). ^{C, N} Difference in staining intensity between cytoplasm (C) and nucleus (N). Antibody dilution for all targets was 1/100.

Table 6 and Figures 4A and 5A allow us to assess the variation in the staining intensity (cytoplasmic versus nuclear staining) of AR, ERα and ERβ receptors in the ovary and uterine tissues. The immunostaining of AR in the ovary varied from weak to moderate, with an increase in primordial, primary and stromal follicles and a reduction in antral follicles in the restricted group. The ovarian ERα and ERβ staining showed a low variation between the groups and appeared weak in the different types of follicles and corpus luteum. Differences in the intensity of nuclear labelling were also observed in the uterus of both animal groups, mainly regarding the presence of ERα and ERβ receptors, with a notable reduction in the uterine myometrium. Although the endometrium of the restricted animals showed a reduction in nuclear immunostaining, the levels of AR seemed to be higher in these animals. While no changes in ERβ immunoreactivity were observed in the uterus of control and restricted animals, the endometrial stroma and myometrium of the restricted animals had a reduction in ERα nuclear staining. There were no significant differences for AR and ER staining in the uterine perimetrium between the control and restricted groups (Table 6).

FIGURE 4.

Profile of AR, ERα and ERβ levels in the ovarian tissue. A, Photomicrographs of the AR and ERs in the ovarian tissue of control and restricted animals; n = 7 animals/group. Images are representative of ovarian follicles (primordial, primary, growing, antral and mature follicles when present). B, Representative profile of the AR (90 kDa), ERα (66 kDa) and ERβ (56 kDa) in extracts of 40 μg proteins from 6 samples/group. C, Optical densitometry analysis of AR, ERα and ERβ after normalization with β‐actin. Data are expressed as mean ± SD

FIGURE 5.

Profile of AR, ERα and ERβ levels in the uterine tissue. A, Photomicrographs of the AR and ERs in the uterine tissue of control and restricted animals; n = 7 animals/group. Images are representative of endometrial tissue (uterine epithelium and stroma). B, Representative profile of the AR (90 kDa), ERα (66 kDa) and ERβ (56 kDa) in extracts of 40 μg proteins from 6 samples/group. C, Optical densitometry analysis of AR, ERα and ERβ after normalization with β‐actin. Data are expressed as mean ± SD

Analysis of the sex steroid receptor expression revealed a significant reduction in AR levels and a significant increase in ERβ in the ovaries of animals consuming a low‐protein diet (Figure 4B,C). However, there were no significant changes in the expression of ERα; there was only a slight increase in the restricted group (P = 0.068). Regarding the uterine tissue, there was a slight increase in AR levels, but without significant difference (P = 0.240). The expression of ERα and ERβ remained unchanged in the uterus of restricted animals compared to controls (Figure 5B,C).

4. DISCUSSION

The maintenance of body parameters encompasses a series of influences on crucial targets for weight regulation, such as satiety, thermogenesis, feeding efficiency and body composition. ²⁵ Protein restriction at puberty led to a reduction in body weight, visceral fat, feed efficiency and protein consumption, while food and energy consumption was higher in animals that received a low‐protein diet. Similar results were observed by Oliveira et al., ²⁶ who demonstrated that consumption of a low‐protein diet at puberty is associated with reduced body weight, consumption of protein and visceral fat; however, they observed a reduction in food and energy intake, diverging from our present findings. Previous studies by França et al. ²⁷ and Dos Santos et al. ²⁸ also reported an increase in food and energy intake and lower body weight, with a significant increase in body fat content. These differences may be due to differences in animal gender and models. Notably, a recent study by our group using maternal protein restriction showed a reduction in body mass index, energy intake, feed efficiency and visceral fat by the offsprings. Regarding the ovarian tissue, a significant reduction in the amount of primordial and primary follicles associated with an increase in the number of atretic follicles was also observed in adulthood. ²⁹ Thus, we believe that some maternal programming conditions may overlap with the effects on nutrition and ovarian activities found after low‐protein consumption during puberty.

The weight of the ovaries, oviducts and liver was unchanged, while the weight of the uterus and kidneys was reduced in restricted animals. Because the ovaries and oviducts are tiny organs and have a primordial function, there may be some protective mechanism to prevent morphological damage. He et al. ³⁰ carried out studies in goats, with a maternal protein restriction diet and found no change in liver and kidney weights. Otherwise, Ramadan et al. ³¹ documented a significant reduction in foetuses liver weight in rats subjected to a low‐protein diet. The reduced weight of the uterus is possibly linked to the reduction of the endometrial layers (uterine epithelium: proximal part: 26.12 ± 12.19, middle part: 28.24 ± 14.28 and distal part: 20.63 ± 8, 85 and endometrial stroma: proximal part: 123.14 ± 36.11, middle part: 175.76 ± 72.91 and distal part: 188.80 ± 77.03) observed in our study; this may have a negative impact during the implantation of blastocyst. In a study on maternal energy restriction, Hoffman et al. ³² reported similar results in lambs, where the weight of the uterus was lower in the restriction group, without, however, changing the weight of the ovaries.

There was no change in the length of the oestrous cycle or anovulation. There was only an increased permanence in the oestrus stage. Mejia‐Guadarrama et al. ³³ analysed the effects of protein restriction in sows and also detected no differences in the length of the oestrous cycle and in the ovulation process. In addition, Hussein et al., ³⁴ in a study with Egyptian buffalo heifers, observed no differences in the length of the oestrous cycle after energy restriction of 50% calories. However, ovarian activity started late when compared to the animals under high feed intake.

The primordial, secondary, pre‐antral, antral follicles and the corpus luteum did not undergo significant quantitative changes. On the other hand, there was an increase in the number of primary and atretic follicles. Guzmán et al. ¹² documented an increase in the number of pre‐antral and antral follicles in rats whose mothers suffered protein restriction during pregnancy and lactation when compared to unrestricted animals or those restricted only during pregnancy. Recently, Winship et al. ³⁵ evaluated the effects of maternal protein restriction in mice and investigated the number of primordial, growing, primary, secondary and antral follicles. There was only a significant reduction in the primordial follicles, with no significant change in the others. We believe that the increase in primary follicles found in our study may indicate a protective mechanism for follicular reserve due to the shortage of circulating protein.

In our current study, the restricted group showed morphological changes in the ovarian tissue, such as dilation of the ducts of the rete ovarii and vascular congestion throughout the stroma. Winship et al. ³⁵ evaluated ovarian morphology after maternal protein restriction in mice and found no changes compared to the control group. In our study, the changes described for the uterine tissue were especially associated with the endometrium, in which the reduced lumen was commonly observed together with the presence of slit papillary projections, vacuolization and irregular glandular epithelium, in addition to leucocyte infiltration in the stroma and uterine glands. Similar changes were previously reported by Brasil et al. ¹⁰ where maternal protein restriction and energy restriction in lactating rats resulted in reduced volumetric density of the uterine epithelium and lumen and also decreased the length of the glands in the female offspring at puberty. These changes allow us to suggest that both protein restriction at puberty and during the lactation period causes damage to the uterine tissue in adulthood.

It is well accepted that oestrogens have a direct effect on follicular growth and maturation, ³⁶ thus increasing the proliferation of granulosa cells, stimulating the expression of gonadotropin receptors, modulating the synthesis of androgen and progesterone by theca cells, in addition to stimulating the development and growth of the endometrium. The effects of oestrogens are often mediated by ERs, with ERα and ERβ subunits belonging to the nuclear receptor superfamily. ³⁷ While ERα regulates certain ovarian functions and its absence alters essential oestrogenic functions during folliculogenesis and fertility, ³⁸ ERβ is the form most predominantly expressed in granulosa cells of the ovaries, being required for antrum formation, maturation of the pre‐ovulatory follicle, expression of genes related to LH receptor, aromatase and follicular rupture, in addition to exhibiting oncostatic properties in cancerous epithelial cells. ³⁹

Molecular changes in critical periods of development may affect ovarian development as a result of irreversible consequences to fertility. ¹⁶ In our study, we observed a reduction in the intensity of ovarian AR, while the ERs did not show significant changes. In the ovaries, differences were found in the cytoplasmic AR immunostaining, with an increase in primary follicles and stroma, in addition to a reduction in antral follicles. Conversely, in uterine tissues, differences in nuclear ER immunostaining were observed with a reduction in the myometrium. There were no nuclear or cytoplasmic differences in the other structures. In the ovaries, protein restriction led to a significant reduction in AR levels and an increase in ERβ levels, while ERα tended to increase, but not significantly. The uterine AR showed a slight elevation following protein restriction, while the ERs remained unchanged. It is well documented that maternal protein and energy restriction during lactation result in differential expression of AR and ER in the ovaries of pubertal rats. Faria et al. ¹⁵ reported a significant reduction in ovarian AR and ERα after a protein‐restricted diet in lactational period in rats. In a study involving caloric restriction in mice, Słuczanowska‐Głąbowska et al. ⁴⁰ observed an increased expression of ER without altering the expression of AR. The decrease in AR may be associated with changes in the ovulatory process, since AR knockout mice show failure or complete abolition of ovulation as reported by Wang et al. ⁴¹ Overall, signals regulated by the AR and ER are extremely important to guide essential events in the ovaries, and changes in the hormone receptor regulation may compromise its normal functioning as reported by others ¹⁷ and in the current study.

5. CONCLUSION

Our results reinforce the importance of adequate intake of protein throughout life, especially during puberty, where protein restriction at this stage profoundly affects nutritional status and reproductive cycle in adulthood. Moreover, protein restriction compromises body weight gain and the histology and morphometry of the ovaries and uteri in adulthood. Functionally, sex steroid receptors are deregulated in the ovarian tissue. Some of these changes may result in impaired folliculogenesis and fertility.

ETHICAL APPROVAL

Animal care and experimental procedures were performed in accordance with the ethical principles of animal research adopted by the Brazilian College of Animal Experimentation (COBEA), approved by the Animal Experimentation Ethics Committee of the Institute of Biosciences of Botucatu (UNESP).

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

DAMO, HSS, LAL and LGAC conceived the hypothesis of the study, collected and analysed the data and drafted the manuscript. The authors have read and approved the final version of this manuscript.

Supporting information

Figure S1

de Morais Oliveira DA, Lupi LA, Silveira HS, de Almeida Chuffa LG. Protein restriction during puberty alters nutritional parameters and affects ovarian and uterine histomorphometry in adulthood in rats. Int J Exp Path. 2021;102:93–104. 10.1111/iep.12388