1 Introduction

Early life dietary restrictions have an impact on growth and postnatal development. The long lasting consequences are known as ‘fetal programming’, or ‘fetal origins hypothesis’ [1, 2]. Such effects are manifested on prepubertal rats whose mothers were undernourished during pregnancy and lactation. Early dietary restriction results in testes with fewer Sertoli cells (SC) in male adult rats, both in absolute numbers and per seminiferous tubules cross-section [3].

Daily sperm production highly correlates with the number of SC per testis due to SC limited capability of metabolically supporting germ cells [4]. Proliferation of SC in the rat normally happens during fetal and postnatal life, starting on gestation days 12 to 13.5 [5, 6] and stopping not long before puberty initiates, on days 16 [7] to 17.5 [8].

Testicular myoid cells (MC) differentiate on gestation day 14 [9, 10] from cells located in the testicular interstitium [11]. SC are essential to induce MC differentiation and migration to the seminiferous cords periphery. Furthermore, Leydig cells (LC) are also under SC influence which stimulate both, LC differentiation and development in normal numbers. Adult males which were subnutritionally restricted during pregnancy and lactation have fewer MC and higher rates of MC apoptosis [12].

Insulin-like growth factor 1 (IGF-1) is a protein mainly secreted in the liver that plays a fundamental role as a growth and development stimulating factor [13]. It acts by binding itself to its cell membrane receptor (IGF-1R). Such receptor relates structurally and functionally to the insulin receptor [14]. The presence of IGF-1R in several rat tissues has been demonstrated, including the testicular parenchyma [15, 16]. The IGF-1 acts as a paracrine/autocrine factor stimulating LC differentiation and steroidogenic activity [17]. Moreover, IGF-1 stimulates SC protein production, and MC production of proteoglycans that constitute the seminiferous tubule basement membrane [18]. IGF-1 production is sensitive to different factors. Physical activity increases IGF-1 maternal serum levels during pregnancy [19] and intrauterine growth restriction (IUGR) decreases IGF-1 serum levels in the rat [20].

In reference to fetal programming and IGF-1, an isocaloric and low-protein diet during rat pregnancy is associated with diminished fetal hepatic growth. Such effect prolongs into postnatal life [21]. Both prenatal subnutrition and a combination of prenatal and prepuberal subnutrition affect testicular development as well as IGF-1R transcription and expression in adult rats [22]. Furthermore, IGF-1 use for the experimental treatment of prepuberal testicular ischemia, provokes improvements in testicular morphology and spermatogenesis [23]. Therefore, both IGF-1 and its receptor are important for the normal development and function of the testis. However, to our knowledge, the effects of subnutrition on IGF-1 and its receptor during early or late pregnancy have not been studied yet. Consequently, we aimed to study the effect of subnutrition during different pregnancy stages (first or second half) on testicular development and both IGF-1 hormone production and IGF-1R expression in puberal rats.

2 Methods

2.1 Animals

Ethics approval: All procedures involving animals were in compliance with national law 18.611 (which constitutes the National Commission of Animal Experimentation). (https://www.cnea.gub.uy). The evaluation and approval of the experimental protocol (111,900–000716-20, Montevideo, Uruguay) was responsibility of the Ethics Committee in the use of animals (CEUA) from Universidad de la República, Uruguay.

Wistar rats were housed in a room illuminated for 12 h, with 50% humidity and 22 °C temperature. Standard rat chow, with balanced composition (23% protein; 8% fibre; 2.5% calcium; 1.25% phosphorus) was always used.

2.2 Experimental design

2.2.1 Preliminary experiment

It was aimed at obtaining an accurate measurement of the ad libitum food intake for every day along a rats’ pregnancy. Six female Wistar rats (317.8 ± 14.6 g body weight), were used and their daily food intake was measured to calculate their average daily food intake afterwards.

2.3 Main experiment

In utero nutritional deprivation of animals in their fetal and embryonic stages before and after male rat sexual differentiation was performed by restricting their mothers’ food intake at 50% of the ad libitum volume estimated in the preliminary experiment. Twenty-four primiparous female Wistar rats (331.9 ± 12.4 g body weight), all mated by the same male rat, were used. Pregnancy was confirmed by presence of spermatozoa in a vaginal smear [24]. Pregnant females were randomly distributed into 4 experimental groups:

Control group (CC) mothers fed ad libitum during pregnancy and lactation.

Restricted group, days 0–10 of pregnancy (RC) pregnant rats were nutritionally restricted by feeding them 50% of the ad libitum food intake. From day 11 till parturition, they were fed ad libitum.

Restricted group, day 11 of pregnancy till parturition (CR) pregnant rats were nutritionally restricted by feeding with 50% of the ad libitum food intake only from day 11 of pregnancy till parturition.

Restricted group, the entire pregnancy (RR)pregnant rats were nutritionally restricted during the entire pregnancy period by feeding with 50% of the ad libitum food intake.

During lactation period (until day 21 of postnatal life) all mothers were fed ad libitum and the litter size was maintained at 8. Postweaning, male pups were housed in cages of 5 animals till puberty (day 40 of life) with ad libitum access to food and water until they were slaughtered and sampled (Scheme 1).

Scheme 1
scheme 1

The scheme summarises the experimental stages whereby the following 4 experimental groups were obtained: Control group (CC), RC (restricted group during the first half of pregnancy), CR (restricted group during the second half of pregnancy) and RR (restricted during the entire pregnancy). In green, stages where there was no in uterus or postnatal restriction. In red, food maternal restrictions either in the first or second half of gestation

Full size image

2.4 Measurements and sampling

2.4.1 Testicle

On day 40 of postnatal life all animals were weighed, measured from their nose to their tail tip, anesthetised with IP Xilacine 4% (20 mg/kg) and Ketamine (60 mg/kg), and decapitated with a guillotine to obtain enough blood volume for further processing.

Their testes were dissected and weighed. The left gonads were immersion-fixed in Bouin’s solution during 15 h. Subsequently, they were histologically processed. Slides were treated with a rabbit primary antibody anti IGF-1R (ab39675, Abcam, UK) and an amplifier kit Mach 2 Double (Mouse-HRP + Rabbit-AP, Polymerdetection kit, Biocare medical, USA) with DAB chromogen. A slide without primary antibody was used as the negative control. The distribution and relative abundance of IGF-1R in SC, LC and MC was described as relative intensity. 300 cells of each type were evaluated using a semi-quantitative intensity scoring (IC) 0–3 scale, where 0 = negative and 3 = strong positivity. Then, a positivity index (PI) for each cell type and for each animal was calculated as follows (Fig. 1):

$${\text{Positivity }} = {1 } \times {\text{ n }}\left( {{\text{IC1}}} \right) \, + { 2 } \times {\text{ n }}\left( {{\text{IC2}}} \right) \, + { 3 } \times {\text{ n }}\left( {{\text{IC3}}} \right)$$

where n = number of cells exhibiting IC = 1, 2 o 3, expressed in % [25].

Fig. 1
figure 1

Histological micrographs of animals’ testes on day 40 of postnatal life from the different groups: Control group (CC), RC (restricted group during the first half of pregnancy), CR (restricted group during the second half of pregnancy) and RR (restricted during the entire pregnancy). The images were immunohistochemically stained against IGF-1R. A Positive myoid cell red arrow; positive Leydig cells blue arrow; B Positive and negative Leydig cells blue arrow. C Negative Sertoli cell nuclei black arrow. Square present in the images A, B and C corresponds to the negative control in the run. Scale bar: 50 µm

Full size image

2.5 Blood serum

1.5 ml of blood without anticoagulant was stored. Subsequently, blood was centrifuged to obtain the serum from each animal. Sera were stored at—20 °C until assayed. A quantitative sandwich ELISA kit (ab231924, ABCAM, USA) with 15,43 pg/ml sensitivity and a measurement range of 31,25–2000 pg/ml was used in order to determine serum IGF-1 concentration.

2.6 Quantitative histology

Images were captured (light microscope Olympus BX50, video camera SSC-C158P; Sony, Tokyo, Japan and Image Pro Plus Media Cybernetics program, Silver Spring, MA, USA) at a final magnification of 2500 × in the computer’s screen from slides treated with Hematoxylin and Eosin. Testicular volume was estimated using the testicular weight, assuming that the testicular density is 1 [26]. The volume density of seminiferous tubules and testicular interstitium was determined by point counting [27], superimposing a grid of 100 points in 30 randomly taken images (200x) of each animal’s testicular parenchyma. The points that fell on each histological structure of interest were counted. The volume density of such structure was calculated as follows:

Vv = Pn/Pt, where Pn is the number of points of the grid that fell on the given structure and Pt is the total number of points per image.

The seminiferous tubules’ diameter was determined by measuring two perpendicular diameters in 30 seminiferous tubule transverse cross-sections per testicle.

The total number per testicle of MC, LC and SC was calculated with the following equation:

$${\text{Total number of cells}}\, = \,{\text{Ns }} \times \, \left( {{\text{L}}/{\text{thickness of the histological section}}} \right)$$

where Ns is the average number of nuclei cells per seminiferous tubule cross-section and L is the total length of the seminiferous tubules of a given testicle.

The seminiferous tubules were assumed as cylindrical and their lengths were estimated from the following equations [28]:

$${\text{L }} = {\text{ Vs}}/\left( {\pi \left[ {{\text{Ds}}/2} \right]^{2} } \right),\,{\mkern 1mu} {\text{and Vs }} = \,{\mkern 1mu} \left( {{\text{Vv seminiferous tubules}}} \right)\,{\mkern 1mu} \times {\text{ absolute testicular vol}},$$

where Vs is the seminiferous tubules total volume and Ds is the diameter of these tubules. The cell number per seminiferous tubule cross section was multiplied by the seminiferous tubules total length in order to obtain the number of cells per testicle.

2.7 Statistical analysis

All variables were expressed as mean ± sd. Normality was verified by Shapiro–Wilk’s test. The variation among rats (within the same litter) and within the same treatment was considered as part of the experimental error. The differences among groups for all variables including the testicular and body weights were compared by ANOVA. Group effect and individual effect within group were studied and considered different if p ≤ 0.05. Post hoc differences among groups were also studied with Tukey tests. Pearson correlations were studied among the different variables. Tendencies were detected if p ≤ 0.10. The program used was Statistica version 6 (Palo Alto CA, U.S.A).

3 Results

All animals presented both, balano-preputial separation and had started spermatogenesis. Both body and testicular weights as well as IGF-1 serum concentration were different among the different experimental groups, according to the pregnancy stages of nutritional restriction with an intra-assay coefficient of variability less than 10%. Both body and testicular weights were higher in CC group and lower in RR group. No differences were found either in animals’ body length (Table 1).

Table 1 Body and testicular weights, body length and IGF-1 serum concentration
Full size table

IGF-1R expression in LC was higher, but MC was lower in CC group than in RR group. The absolute volume of seminiferous tubules as well as the total number of SC were higher in CC group than in RR group (Table 2, Figs. 1, 2). The diameter of seminiferous tubules was also higher in CR than in RR group.

Table 2 IGF-1 receptor positivity index and morphometrical variables
Full size table
Fig. 2
figure 2

Histological micrographs of animals’ testes on day 40 of postnatal life from the different groups: Control group (CC), RC (restricted group during the first half of pregnancy), CR (restricted group during the second half of pregnancy) and RR (restricted during the entire pregnancy). The images were Hematoxylin and Eosin stained. A transverse section of seminiferous tubule at 10× magnification. B seminiferous epithelium and part of the testicular interstitium at 40× magnification, Leydig cell blue arrow; Sertoli cell black arrow; Myoid cell red arrow; elongated spermatid yellow arrow. Scale bar: 50 µm

Full size image

Positivity index in: Leydig (LC PI), myoid (MC PI) and Sertoli (SC PI) cells. Diameters of seminiferous tubules (DST), absolute volume of seminiferous tubules (AVS), and Sertoli cells per tubule section and total SC number per testicle from control animals (CC), RC (restricted group during the first half of pregnancy), CR (restricted group during the second half of pregnancy) and RR (restricted during the entire pregnancy) at 40 days of life. a vs. b vs c, P ≤ 0.05.

High positive correlations were found between body weight and testicular weight, total number of Sertoli cells, IGF-1 serum concentration and LC PI (Table 3).

Table 3 Correlation study of body weight
Full size table

Correlations between body weight and: testicular weight, IGF-1 serum concentration, IGF-1 receptor positivity index in Leydig cells (LC PI), myoid cells (MC PI) and total number of Sertoli cells from control animals (CC), RC (restricted group during the first half of pregnancy), CR (restricted group during the second half of pregnancy) and RR (restricted during the entire pregnancy) at 40 days of life. T = a statistical tendency. **: 0.001 ≤ P ≤ 0.01; ***: P ≤ 0.001.

In addition, a high correlation between IGF-1 serum concentration and both, body weight and total number of Sertoli cells was found. Besides, a statistical tendency (p < 0.10) with LC PI was also found (Table 4).

Table 4 Correlation study of IGF-1 serum concentration
Full size table

Correlations study of IGF-1 serum concentration and: testicular weight, IGF-1 receptor positivity index in Leydig cells (LC PI), Myoid cells (MC PI) and total number of Sertoli cells (SCTN) from control animals (CC), RC (restricted group during the first half of pregnancy), CR (restricted group during the second half of pregnancy) and RR (restricted during the entire pregnancy) at 40 days of life. T = a statistical tendency (p ≤ 0.10). **: 0.001 ≤ P ≤ 0.01.

4 Discussion

To our knowledge, this is the first report that demonstrates the impact of subnutrition on either of the rat’s pregnancy halves, affecting body and testes development, IGF-1 production and IGF-1 receptor abundance when male pups reach puberty. Subnutrition during the entire pregnancy [29] and/or lactation affects body weight in later stages [12, 22]. We found that subnutrition during either the entire pregnancy or just its second half similarly affects body weight. This might be due to fetal exponential growth during the last third of gestation that happened within the treatment; however, subnutrition during the first half of pregnancy also affected body weight, so much so that it couldn’t be compensated in its postnatal life until puberty (40 days). In rats, animals finish their body development later in life. It is well known that an average weight of 200 g acts as an initiating factor for puberty. This suggests that not only age but also body weight, could be considered a biomarker of puberty beginning [30]. In addition, balano-preputial separation and spermatogenesis beginning are described as other markers [31]. Even though in our conditions only CC group reached puberty standard weight, all animals presented balano-preputial separation and had initiated spermatogenesis. This indicates that treatment in any pregnancy half affects body development but not the timing of puberty initiation.

In line with this, it has been established that body weight depends on various factors, being one of the most important IGF-1 concentration [32,33,34]. In our study, differences in IGF-1 concentration among the experimental groups have been found. The treatment applied in different pregnancy stages affected body weight, being lighter in all groups as compared to their controls.

Having said that, this finding is not observed in serum IGF-1 concentration, since animals nutritionally restricted in the first half of pregnancy behave as their controls and, the ones nutritionally restricted in the second half behave as the animals which were nutritionally restricted during whole pregnancy. Thus, more studies are needed in order to establish what kind of link could be acting between variations in IGF-1 serum concentration and body weight in our experimental design.

As far as testicular weight is concerned, we confirmed that it is lower in animals which were nutritionally restricted during the entire pregnancy as compared to their controls at birth. This finding had been reported earlier [35]. Nevertheless, in our present work, in this variable nutritionally restricted animals in the first and second half of pregnancy, behaved differently from their controls but were not different among them. Considering that the gonadal differentiation stage in the rat starts on day 13,5 and that the nutritionally restricted group was only treated until day 10, it can be concluded that the pubertal testicular weight is affected by subnutrition applied even before gonadal differentiation. As far as we know, it is the first time the impact of subnutrition during the undifferentiated sexual stage of pregnancy on both testicular and body development in puberty is established. It is worthy noting that body weight was affected by the treatment, but body length was not. This seems in line with findings reported in lambs, whereby body weight was affected but not body length [36], thus suggesting the axial skeleton is prioritized as compared to muscular and adipose tissues.

In reference to serum IGF-1 concentration, it was affected by which pregnancy half treatment was applied. We know from early reports that subnutrition during the entire pregnancy affects IGF-1 concentration when animals are adults [21]. In our conditions, we saw animals from both CC and RC groups were not different whereas animals from both RR and CR groups were not different among them as well. It is important to take into consideration that the main source of IGF-1 is the liver [37] and that the testicle is a minor source. Moreover, a significant peak of this hormone’s plasma levels takes place at puberty [38]. Even though the testicular production of IGF-1 relies on SC and LC [39], if we bear in mind that the liver is the main source of IGF-1, we cannot rule out that such difference is more related to the treatment effect on liver growth during early stages, which would be another long-lasting effect of fetal programming. It has been demonstrated that subnutrition during the entire pregnancy affects liver weight and IGF-1 production [40]. The high correlation found in our study between serum IGF-1 and body weight might be thus explained.

On the other hand, target cells for IGF-1 express receptors in the presence of such hormone. When studying IGF-1 receptor positivity index we found its expression was different as compared among L, M and S studied cells. Regarding LC we saw that subnutrition in either pregnancy half diminishes IGF-1 receptor production and its expression has a tendency to correlate with serum IGF-1 concentration. No treatment effect on receptor expression in SC was found. MC behaved differently to SC. Higher PI values were present in the MC from RR group in contrast to the other groups. In addition, we saw that PI values expressed by MC have a tendency towards negative correlation with serum IGF-1 concentration and it was precisely the RR group the one who showed lower concentration of the hormone. MC synthesise basement membrane components of the seminiferous tubules [18] when stimulated by various factors, being one of them IGF-1. Therefore, the increase in the expression of RR group might be linked to the compensation for lack of hormone in the RR group. This might indicate a long term effect.

With respect to morphometrical data, we found that both seminiferous tubules diameter as well as the absolute volume of seminiferous tubules were lower if underfed in either pregnancy half. Additionally, the SC number per cross-section of seminiferous tubules was not affected by the treatments. However, although the total number of SC per testicle was lower in both pregnancy halves if underfed (both CR and RC groups), the testicular weight at puberty was lower only in group CR. This is coherent if we take into account that, in order to calculate the SC total number per testicle, the testicular weight is used as a variable. Assuming that seminiferous tubules are cylindrical, the volume they occupy per testicular section was obtained. Then, it was corrected for the total volume of the testicle, considering the testicular density = 1 [26]. Therefore, animals with higher testicular weight presented higher testicular volume and, as such, higher total number of SC since its volume of total seminiferous tubule per testicle was also higher. Bigger testes have more SC and, theoretically, higher daily sperm output.

It is important to highlight that the SC conduct spermatogenesis and that it exists a strong correlation between the SC number and the theoretical daily sperm production of a testicle [41]. Therefore, subnutrition either in the first pregnancy half or during the entire pregnancy, might affect sperm production in puberty. It was previously reported that nutritionally restricted animals during the entire pregnancy (at 40% of the ad libitum food intake) and lactation contain fewer SC per cross-section of seminiferous tubules in adult life [3]. However, SC number per section was not affected either in nutritionally restricted adults with a 50% [22] or in neonatal pups nutritionally restricted during the whole pregnancy at 40% of the ad libitum food intake [35]. The differences found between the previous literature and our present work might be accounted for not only by the different treatments but also by the fact that we are working at a different age than the mentioned reports.

5 Conclusion

In conclusion, subnutrition in the different pregnancy stages affects body development, testicular weight and testicular morphology in pubertal male rats. Specifically, subnutrition during the different pregnancy halves diminishes the SC total number, and IGF-1 receptor expression in LC and MC cells which indicates a differential impact on the testicular function. Although we detected differences in serum IGF-1 concentration, further studies will be necessary to establish how it affects testicular growth as well as IGF-1 production at the testicular level.