Dietary Protein Deciency in Early Life of F1 and F2 Generations of Sprague Dawley Rats Disrupts the Reproductive Function

Background: The ability to reproduce eciently is an important characteristic that has evolved through natural selection. Nutrition can modulate reproductive activities at different levels, its effect on nutrition is therefore complex and less predictable. This study aims at investigating the underlying effect of persistent dietary protein deciency during early life on reproductive parameters of subsequent (F 1 and F 2 ) generations. Method: Rats in group of four (4) were fed daily, with different ration of protein diet (PD) formulated as: 21% protein diet, 10%protein diet, 5%protein diet and control diet (rat chow, containing 16-18% protein). They were fed ad libitum before mating, throughout gestation and lactation, and next generations were weaned to the maternal diet. Reproductive function analysis (which include; gestation and pubertal hormonal proling, onset of puberty, oestrus cyclicity, sexual response) and morphometric analysis of the ovarian structure were carried out to assess associated consequences. Results: showed signicant reduction in the fertility index as a consequence of altered reproductive function in the protein decient models at P ≤ 0.05. Low protein diet posed suboptimal intrauterine condition, which was linked to increased prenatal morbidity and mortality, lowered birthweight delayed onset of puberty, induced cycle irregularity, altered follicular maturation and endocrine dysfunction in the protein decient groups. Reproductive status of an individual female organism critically depends on the maintenance of ovarian structure and function that has been associated with the hypothalamic pituitary-gonadal axis, hormonal events and sexual maturity. Conclusion: There is therefore an association between persistent early life protein deciency and reproductive response which mechanistically involves life-long changes in key ovarian cytoarchitecture and function.


Introduction
Reproductive health encompasses the reproductive processes, functions and system at all stages of life. It is an important component of general health (WHO, 2008). Nutrients when detected by the cellular sensor systems as dietary signals can in uence certain phenotypic changes which may alter activities and homeostatic control process (Elizondo et al., 2019). Maternal malnutrition during gestation and lactation impairs embryonic fetal development which results in deleterious outcomes and imprints. With the growing human population, approximately one billion people suffer from protein de ciency one-third of which are stunted under ve, because the global food system is currently failing to meet its nutritional needs (Wu et al., 2014) Undernutrition occurs in adolescent pregnancy because of the competition that exists between fetus and mother for nutrients (Vega et al., 2016;Micheal and Sandel., 2003). Early-life exposures trigger processes that sets individual ready for particular circumstances that are anticipated in the postnatal and adulthood environment (Chan et al., 2015). Protein undernutrition can cause stunted growth, anemia, physical weakness, and edema (Abey et al., 2019). The ability to reproduce is central to the life history of all sexually reproducing organisms. The primordial follicle pool is built up at the early stage of development, and is therefore vulnerable to exposures at this stage of development. Therefore, reproductive maturation and capabilities depends on early-life events, (Chan et al;. Previous reports from clinical and experimental studies have indicated that early-life problem is associated with a decline in ovarian follicular reserve, changes in ovulation rates, and altered age at onset of puberty. However, mechanisms underlying the regulation of the relationship between the early-life developmental environment and postnatal reproductive function are not clear (Somchit et al;2007).
Nutrition has pivotal effects on reproduction as it can modulate reproductive activity at multiple levels (Vega et al., 2017). The success of reproduction in all animals depends on the function of the hypothalamus-pituitary-gonad axis. The interaction between nutrition and reproduction has been established to have important implications for the reproductive performance of ewes (Somchit-Assavacheep, 2011). Also, Nutritional status of individuals in uences virtually all aspects of female reproductive performance starting at the fertilization to their oocytes and embryo quality (Silvestris et al., 2019). More than 10% of the world's population are affected by infertility (Rouchou, 2013). In developing countries, there are severe social, psychological and economic consequences for infertile men and women, especially in the low-resource settings where it may be associated with a signi cant risk of further impoverishing the health reserves and sustenance in the community, threatening survival and worsening poverty (Gerrits et al., 2012). This present work was undertaken with the aim to investigate the perinatal dietary protein de ciency effects on reproductive health of two (2) different generation (F 1 and F 2 -generations) of rat model. As such, the pubertal attainment, estrus cycle, fertility response, hormone pro le and morphology of the ovary were used to validate this, the established consequence can inform some therapeutic interventions in cases of reproductive health issues resulting from persistent perinatal dietary de ciency.

Materials And Methods
Ethical Statement: The research was carried out following the guidelines of the Act 2004 health research standards for care and use of laboratory animal models. The College of Medicine University of Lagos Health Research Ethics Committee (local HREC; REC 11), gave approval to the protocol to carry out the research (CMUL/HREC/11/18/462).
Animal Grouping/Maintenance Averagely 6-8 weeks old virgin female Wistar rats (n=40) were obtained from the animal house of the College of Medicine University of Lagos, Lagos Nigeria. Rats were grouped into four (4) according to the ration of protein in the diets; All rats were maintained in clean capacious plastic cages (n=10 per cage) under standard laboratory conditions which include; neat environment, good aeration and lighting, with suitable (25 0 C ± 2 0 C) at a 12-hour light/dark cycle and feeding ad libitum.
Diet Formulation: Non-puri ed isocaloric diet was formulated. The diet was formulated using non-puri ed constituents and standard formula, scored to requirements following adaptation from New Non-puri ed Diet (NTP-2000) for Rodents by Ghanto Rao, 1997. The various protein ration diet for rat was formulated by scoring the proximate protein content to percentage in each dietary component and con rming the overall proximate analysis. Contents include: Maize, Fish meal, Soybean meal, cassava, wheat offal, Bone meal, Oyster shell, premix, methionine, lysine, Common salt in varying percentage towards the attainment of the experimental diet protein percentage for each grouping, in the modi ed method of Silas et al., (2014). Feed were fed to animals in pelletized form.
Time-Mating: Rats were maintained on group diet for about 6weeks before mating. Three (3) weeks prebreed, at proestrus/estrus phase, rats in each group were time-mated with certi ed reproductive male, following three (3) weeks pre-breed vaginal cytology, to establish cycle pattern. The presence of sperm plugs and/ spermatozoa in the vagina smear was used to con rm day 0 of conception. the con rmed pregnant dams were separated to produce F 1 generation, this process was repeated for F 1 generation to produce F 2 generation while the different diet group feeding continues. Pregnant rats were weighed every other day, underwent vaginal delivery and pups' weight was also taken at birth. Weaning was done at postnatal day 28 and anogenital distance (AGD) was used to determine and separate sex (≤ or ≥ 2.5mm as the reference value (Zambrano et al., 2006).

Day 6 and 19 hormonal Assay:
The pregnancy hormone was monitored. Maternal blood at 6th and 19th day of gestation between 9 th and 10 th hour of the day, was collected, cold centrifuged at 3500rpm, and prepared for ELISA assay. serum estradiol and progesterone levels were measured by ELISA following manufacturer's instructions on kit. This was also repeated for the F 1 generation.

Implantation Study
At gestation day 4 (GD5), dams were injected at the tail, with Evans blue dye (1 mg/ml) and allowed to circulate for about 15 min, before sacri cing through cervical dislocation and the implantation sites (identi ed as distinct blue bands along the uterine horn) were observed, counted and recorded (Oludare and Iranloye, 2016).

Pregnancy Outcome
Birth weight, Fetal appearance and average fetal outcome and survival rate in each treatment group were taken.

Fertility Index
Fertility response and fertility index calculations were done following slightly modi ed procedures.

Onset of Puberty
Caliper was used to take the anogenital distance (AGD) for determination of sex; using ≤ or ≥ 2.5mm as the reference value (Zambrano et al., 2006). Male and female progeny were separated and the female were maintained ad libitum on the formulated diet until puberty. Onset of Puberty de ned as the age of vaginal opening (Goldman et al., 2007), was assessed by visual inspection every other day. Opening was established as the complete separation of the membrane sheath covering the vaginal ori ce.

Oestrus Cycling
Oestrus cycle monitoring follows vaginal opening. vaginal smears were collected using lavage technique to index the stage of the estrous. To establish the timing of the oestrus. cyclicity was recorded based on the proportion of leucocytes, epithelial cells and corni ed cells, in the smear. Cycle was counted on intervals between proestrus and the next proestrus, and average duration is calculated in each treatment groups. (Marcondes et al., 2002). The cycle regularity was evaluated on; cycle length, percentage time spent in each cycle phase. Longer cycle length, more than 50% time in one stage is considered irregular cycle. Prolonged diestrus also indicated acyclic period or pseudopregnancy .
Pubertal Hormone: F 1 -generation serum estradiol and progesterone, Follicle Stimulating Hormone, Luteinizing Hormone, Testosterone and cortisol levels were measured at puberty (speci cally the Estrus phase) by EIA kit (Monobind Inc. Lake forest USA), using hormone standards and the speci c antibodies following manufacturer's instructions of kit.

Morphometric analysis
Both ovaries were removed for morphological study at diestrus phase. Tissue samples were xed in formalin (10%) solution for 14-16 hours, at 4 0 C, the samples were dehydrated with alcohol solutions with increasing volume; followed by para n embedding. The samples were cut with a microtome 5µm diameter and consecutive staining with hematoxylin and eosin, sections were placed on slides (Ahmadi et al., 2017). In the nal stage, ovarian morphology was assessed; the tissue sections taken from the ovaries were observed under light microscope with a magni cation of 40 and the number of each of the follicular groups were counted, the thickness of theca layer and granulosa were then measured (Roshangar et al., 2014).
Statistical Analysis: Data were analyzed using diets as factors. Results are presented as the mean ± SEM. Statistical analysis was performed using GraphPad Prism 7.0 for analysis of variance (ANOVA) followed by posthoc Turkey's test and/or Dunnette's test, at P ≤ 0.05, statistical difference is considered signi cant. Pregnancy/Pregnancy Outcomes Figure 2 shows the signi cant reduction in the gestation weight of the protein de cient pregnant dam at F 0 and F 1 generation. In Fig 3, circulating level of progesterone and estradiol at gestation day 6 and 19 in the 5%PD and 10%PD (de cient groups) were not signi cantly different from the 21% Protein diet and control group. There were signi cantly distinct features in protein de cient groups for the fertility index (Fig 4). Fetal resorption rate, Pups birthweight, and average fetal outcome at F 1 & F 2 -generations ( Table   2) were statistically different compared to control diet, while the gestation length (Table 2), did not change across the groups.

Reproductive Function
The onset of puberty measured as date of vaginal opening (Fig 5) in control and 21%PD groups were within the normal range (PND 32), while the protein de cient groups (10%PD and 5%PD) had a late date of pubertal onset (PND 40 and 44 respectively). The pattern of oestrus cycle was disrupted in the nutritionally challenged groups (Fig 6a). Within four (4) weeks of vaginal cytology, Cycle irregularity was displayed for each of the protein de cient groups as the percentage of time spent in each oestrus phase (Fig 6a), was of higher variance compared to control's time spent in each cycle phase (Fig 6a) and cycle length (Fig 6b). Irregular cycle is taken as more than 50% time spent in diestrus stage or cycle length above 7days. The 10%PD group in F 1 had persistent estrus phase (i.e vaginal corni cation), while the F 1 and F 2 5%PD displayed prolonged diestrus compared to other groups.
Hormone at Puberty   Fig 7 & 8 shows the pubertal hormone in all the groups, the protein de cient groups had signi cantly decreased level of follicle stimulating hormone, Progesterone (Fig 7), cortisol and testosterone (Fig 8), which are essential for the maturation of the ovarian structure. Although there was variation in the estradiol levels across the group but this was not signi cant. The Luteinizing hormone remained the same across the groups, in rst and second generation.

Ovarian Morphology
The quanti cation of follicle and corpora lutea shows that at F 1, all other follicles were the same except the primary follicles which was signi cantly lowered and the cystic follicle which was higher signi cantly (Fig 9a) in the protein de cient groups. At F 2 ; protein de ciency altered the number of primordial follicle (↑), cystic follicle (↑) and corpora lutea (↓) especially in the severe de cient group (5%PD), compared to control, signi cant at P<0.05. The severely de cient group had signi cantly higher follicular diameter and lower theca and granulosa thickness (Fig 9b).

Discussion
Perinatal exposure to a protein de cient diet and general mother/child malnutrition during critical period of development is known to lead to certain life-long psychopathological changes and increased susceptibility to dysfunction that underlies most origins of health defects and disease. This study is targeted at investigating the reproductive consequences of perinatal dietary protein de ciency in two (2) subsequent generations of rat models, providing associated mechanism. The organ-system in the reproductive setup, in an attempt to make up for the de ciency may have undergone certain physiological and metabolic shifts which may directly or indirectly impart the shift in the hypothalamic pituitarygonadal axis that controls an aspect of reproduction.
Models under nutritional inadequacy were observed to have grown at a very slow pace during gestation, thus corroborating a previous report by Calin et al., (2019), The fetuses are exposed to sub-optimum environment as a result of the competition that exist between the mother and the fetus for the available nutrient, and this may have resulted in the consequential imprints observed in later life. Maternal constraint alters the development of fetus within the uterine which was evident as IUGR conditioning in the severely de cient group (5%PD), and this corresponds with the low birthweight at both F 1 and F 2generations. The sub-optimal intrauterine condition that result from the nutritional stress explains the low birthweight as earlier stated by Zambrano et al., 2005a. The survival, growth and development depended critically on the nutritional status of the mother and extent of protein de ciency, this result therefore emphasize that maternal protein intake directly impart intrauterine growth and survival (Woodall et al., 1996) and could also be explained by fetal programming effects.
Fertility index is key to reproductive health, delayed conception and reduced fertility response are consequences of protein de ciency as reported in this study. Although dietary protein de ciency did not affect pregnancy hormone, litter size and gestation length, across the generations, it is thought that there was a metabolic adjustment to cater for the pregnancy condition, and this must have accounted for the maintenance of the pregnancy till birth, for those dams that could carry pregnancy to term, this also explains why mortality rate was higher in the de cient group. The high rate of fetal resorption and survival in the protein de cient group has been hypothesized to share common etiology and therefore can be linked to the in uence of dietary insu ciency on the maintenance of a viable pregnancy (Lee et al., 2000).
Delayed onset of puberty corresponds with lowered birthweight. We observed that vaginal opening(Onset of puberty) in the protein de cient models did not occur within the normal age range earlier reported by The altered ovarian architecture may directly impart function and this may in part be responsible for some abnormalities in reproductive function and fertility response earlier discussed. Corpus luteum matures during oestrus cycle and then regress. A degenerating corpora luteum is characterized by increased amount of brous tissue and yellow-brown lipofuscin pigment (Ahmadi et al., 2017), and Corpus lutea cyst is one of the functional cysts known to affect ovarian function in female domestic animals, while follicular cysts regress spontaneously with time and are clinically unapparent. Another underlying mechanism for the variation in ovarian morphology could be alteration in the follicular atresia process in the ovary, this process regulates the size and number of follicles in the developing pool. Increase in follicular atresia can be observed following induced condition of stress (which may be nutritional stress), where it continues to cause decrease in corpora lutea and in amed ovary (Chou and Chen, 2018).

Conclusion
Nutrition is an important factor affecting pubertal development. It can therefore be suggested from this study that intragenerational perinatal protein de ciency can consequentially program reproductive developmental process, presenting a sub-optimal reproductive response and function in subsequent generation adults, as a function of normal development of ovary, ovarian follicles, endocrine dysfunction, ovarian degeneration and morphological variation. These therefore underlies the reproductive responses and function in perinatal dietary protein de cient model from one generation to another.   Data are expressed as mean ± SEM, Control (standard rat chow containing 16-18% Protein), 21%PD (upgraded daily recommended intake), 5% (Protein de cient diet), 10% (Mild protein de cient diet). a < signi cantly different from 21% diet group and or control > b<signi cantly different from the 10% diet group> (*P<0.05, **P<0.001, ***P < 0.0001) I & II; Average cycle length at F1 and F2 -generations III & IV;

Declarations
Percentage Time in Diestrus phase.at F1 and F2 -generations.

Figure 8
Concentrations of Cortisol and Testosterone (µg/g) in the F1 -generation in each of the different diet group of rats. Data are expressed as mean ± SEM, Control (standard rat chow containing 16-18% Protein), a < signi cantly different from 21% diet group and or control > b<signi cantly different from the 10% diet group> (*P<0.05, **P<0.001, ***P < 0.0001).