All the procedures used in this study were approved by the Comité Institucional de Cuidado y Uso de Animales de Laboratorio (CICUAL) from the Hospital de Niños Ricardo Gutiérrez (Res #2018-002) and performed in accordance with the principles and procedures outlined in the Guide for the Care and Use of Laboratory Animals issued by the National Institute of Health of USA. 3-month-old pregnant Sprague Dawley rats (250–300 g) were purchased from the bioterium of the Facultad de Ciencias Veterinarias, Universidad de Buenos Aires (FCV-UBA). In that facility, nulliparous female rats were mated with a stud male. The animals were separated after confirmed copulation either by detection of vaginal plug or spermatozoa in vaginal smear samples. Five days prior to pup delivery, rats were transported to our bioterium in transportation crates and were housed singly in free exchange cages (30 cm x 40 cm x 23 cm) with stainless steel tray containing pine wood shavings as bedding. Animals were maintained under controlled conditions of temperature (20 ± 2°C), relative humidity and lighting (12 h light–12 h dark cycle) with free access to water and pellet laboratory chow (Rat-Mouse Diet, Asociación de Cooperativas Argentinas, Buenos Aires, Argentina).

At delivery, pups were sexed according to the anogenital distance. Litters were adjusted to 8 pups, prioritizing a maximum of 8 male pups per litter when possible. Male pups were randomly assigned to one of the following treatment groups: control group (C), receiving water; G2 and G50 groups, receiving 2 and 50 mg/kg/day G, respectively; and R2 and R50 groups receiving 2 and 50 mg/kg/day R, respectively. G was provided by Sigma-Aldrich (St Louis, USA). R formulation was a liquid water-soluble formulation containing 66.2% of G potassium salt, as its active ingredient. Treatments were given orally from PND 14 to 30. Pups were weaned on PND 21 and euthanized on PND 31. A set of animals were treated from PND 14 to 30 with water or 50 mg/kg/day R and kept without further treatment until PND 90. Tissues and blood were collected from euthanized animals in PND 31 or PND 90 ( Figure 1 ).

Rats were treated from PND 14 to 30 to comprise the full period of maturation of the BTB. This exposure period corresponds to Sertoli cell maturation associated with BTB formation, germ cell meiosis, and the appearance of the first spermatids. The dose of 2 mg/kg/day was selected because it is in order of magnitude of the reference dose (RfD) of 1 mg/kg/day recently reassigned for glyphosate by the US EPA (2017). It is worth mentioning that this dose is used in several reports to analyze reproductive toxicity of the herbicide (21–23). The dose of 50 mg/kg/day was selected based on the no observed adverse effect level (NOAEL) for G (24) and on a previous report from Romano et al (10) who proposed this dose of R as appropriate for future toxicological analyses. No alterations in maternal care and nursing among the experimental groups were detected. Treatments caused no overt signs of toxicity.

Blood was obtained by intracardiac puncture. The samples were allowed to clot at room temperature for 15 min, and then they were centrifuged at 950g for 5 min in order to obtain the serum. Supernatants were immediately frozen at −20°C for subsequent analysis. Biochemical parameters were determined with an automated Cobas c 501 analyzer (Roche Diagnostics, Mannheim, Germany).

At PND 31, animals were euthanized by CO₂ asphyxiation, and testes were dissected, weighed, and used for histological analysis and BTB assay. Also, testes were dissected and snap frozen for reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) and Thiobarbituric Acid Reactive Substances (TBARS) determination. Tissue samples (liver, kidney, stomach, and intestine) were also collected at that time.

At PND 90, animals were anesthetized with a mixture of ketamine-xylazine and tissues were sampled and weighed. Testes were used for histological analysis and BTB assay and to determine daily sperm production (DSP). Epididymides were used to evaluate sperm parameters.

For histological analysis, testes were removed and fixed in Bouin solution, and the other tissue samples were fixed in 10% formalin. Dehydration was carried out at room temperature using ascending concentrations of ethanol and shifting to xylene. After clearing, tissues were embedded in paraffin wax and 3- to 5-μm-thick sections were cut using a microtome (Thermo Fisher Scientific, UK). Sections were transferred to albumenized slides that were preheated to 37°C. Tissues were rehydrated in descending concentrations of ethanol, stained with hematoxylin/eosin and covered with a coverslip. The prepared slides were observed under an Eclipse 50i microscope (Nikon Instruments Inc., Tokyo, Japan) equipped with a digital camera (Canon, Japan). For TUNEL assay, testicular sections were revealed with In Situ Cell Death Kit (Roche Applied Science, Indiana, USA) as previously described (25).

Testosterone was extracted from PND 31 testis homogenates with diethyl ether followed by evaporation of the organic phase and reconstitution of extracted testosterone in 0.1% PBS. Serum testosterone of PND 90 animals were also evaluated. Testosterone concentration was measured with an electrochemiluminescence assay (Roche Diagnostics, Mannheim, Germany) using a Cobas e 411 analyzer according to the manufacturer’s instructions. Testosterone assay sensitivity was 10 ng/dl, and intra and interassay CV were 2.4% and 2.6%, respectively.

Lipid oxidation was determined by the colorimetric assay of TBARS (26). Testis homogenates were performed in PBS containing 0.4% w/v butylated hydroxytoluene on ice and then disrupted by ultrasonic irradiation. An aliquot (25 µl, corresponding to 450 µg protein) was added to 175 µl mixed reaction solution (0.15% w/v SDS, 0.5 N HCl, 0.75% w/v phosphotungstic acid, and 0.175% w/v 2-thiobarbituric acid). The mixture was heated in a boiling water bath for 45 min. TBARS were extracted with 200 µl of n-butanol. After a centrifugation at 10,000g for 5 min at 4°C, the absorbance at 532 nm of the butanolic phase was measured. A calibration curve was performed using malondialdehyde (MDA), generated from 1,1,3,3-tetramethoxypropane (0.4–8 µM), as standard to express the absorbance changes as nmol MDA/µg protein.

The permeability of the BTB was assessed with a biotin tracer as described previously by Perez et al (27). Immediately after testes isolation, a solution of 10 mg/ml EZ-Link Sulfo-NHS-LC-Biotin (Pierce) dissolved in PBS containing 1 mM CaCl₂ was injected into the testis. The administered volume represented 10% of testis weight. Testes were then incubated at 34°C for 30 min, immersed in 4% paraformaldehyde and embedded in paraffin. For localization of the biotin tracer, testis sections (5 µm thick) obtained from different levels were deparaffinized and hydrated. To avoid nonspecific staining, sections were blocked with 5% nonfat dry milk in PBS containing 0.01% Triton X-100 for 15 min prior to incubation with streptavidin-rhodamine (1:300; Invitrogen, USA) for 45 min at room temperature. After nuclear staining with DAPI, sections were mounted in buffered glycerin and observed by Ahiophot fluorescent microscope with epi-illumination. At least 50 seminiferous tubules from 3 nonconsecutive testis sections from each rat were examined. Results are expressed as % of permeable tubules.

Total RNA was isolated from testis homogenates with TRI Reagent (Sigma-Aldrich). The amount of RNA was estimated by spectrophotometry at 260 nm. RT was performed on 2 µg RNA at 42°C for 50 min with a mixture containing 200 U MMLV reverse transcriptase enzyme, 125 ng random primers and 0.5 mM dNTP Mix (Invitrogen, Argentina).

qPCR was performed by a Step One Real Time PCR System (Applied Biosystems, Warrington, UK). The specific primers for RT-qPCR are shown in Table 1 . Amplification was carried out as recommended by the manufacturer: 25 µl reaction mixture containing 12.5 µl of SYBR Green PCR Master mix (Applied Biosystems), the appropriate primer concentration and 1 µl of cDNA. The relative cDNA concentrations were established by a standard curve using sequential dilutions of a cDNA sample. The amplification program included the initial denaturation step at 95°C for 10 min, 40 cycles of denaturation at 95°C for 15 s, and annealing and extension at 60°C for 1 min. Fluorescence was measured at the end of each extension step. After amplification, melting curves were acquired and used to determine the specificity of PCR products. The relative standard curve method was used to calculate relative gene expression. Relative mRNA levels were normalized to the reference genes HPRT1 and βactin.

At PND 90, testes were collected and processed following the experimental procedure described in Fernandes et al (28). The tunica albuginea was removed from one testis and the parenchyma was homogenized in 0.9% w/v NaCl containing 0.5% w/v Triton X100. Then, the homogenates were ultrasonic disrupted for 30 s. Samples were diluted at 1:10 and transferred to a Neubauer chamber and counted in quadruplicate. Elongated spermatid nuclei with a shape characteristic of step 19 spermatids and resistant to homogenization were counted to determine the number of elongated spermatid nuclei. To calculate the DSP, the number of spermatids was divided by 6.1, which is the number of days of the seminiferous cycle in which these spermatids are present in the seminiferous epithelium (29).

At PND 90, sperm were recovered from cauda epididymis. The epididymides were placed in a conical tube, covered with 750 µl of fresh medium (30) and sperm were allowed to swim up at 37°C. After 10 min, aliquots of the upper sperm layer were recovered for motility and morphology evaluation. To evaluate total motility (progressive + nonprogressive), sperm suspensions were placed on pre-warmed slides and analyzed subjectively under a light microscope (400× magnification). To assess sperm morphology, a 10 µl aliquot of the sperm suspension was smeared over a microscope slide. After drying in air, the smear was fixed with methanol for 5 min, washed with distilled water, stained with Harris hematoxylin for 15 min, and finally washed with tap water. Sperm morphology was evaluated using a Nikon microscope (Nikon Instruments Inc., Melville, NY, USA) at 1000× magnification. In all cases, at least 200 spermatozoa from each sample were assessed.

At least five animals per treatment group were used, and data were presented as mean ± SD. The variables under study were first submitted to tests of normal distribution (Shapiro-Wilk’s Test) and for homogeneity of variances (Levene’s Test). Then, one-way ANOVA followed by Tukey-Kramer test for the comparison of multiple groups was performed. To assume normal distribution, percentages were expressed as ratios and subjected to the arcsine square root transformation. Results were compared using the unpaired Student t test. Probabilities <0.05 were considered statistically significant. InfoStat 2016 (Grupo InfoStat, Facultad de Ciencias Agropecuarias, Universidad Nacional de Córdoba, Argentina) was used.

Initially, rats were exposed to low doses (2 and 50 mg/kg/day) of G or R from PND 14 to 30, a period of life that is essential to complete a functional BTB. Serum biomarkers for renal (urea, creatinine) and liver (aspartate and alanine aminotransferases: AST and ALT) function were analyzed. These biomarkers did not change in treated groups compared with the control ( Table 2 ). In addition, treatments did not alter liver, kidney, stomach, and intestine histology (data not shown). To explore the impact of G or R exposure on male reproductive system we assessed body and testicular weights after treatments. No differences in body and testis weight and in the ratio testis/body weight at any G or R dose tested were observed ( Table 2 ).

Histological examination of the testes revealed differences in the seminiferous tubules among control and treated groups ( Figure 2 ). Some seminiferous tubules of G2, G50, and R2 groups showed a disorganized epithelium, with an apparent low cellular adhesion. In R50 treated males, tubules presented severe disorganization and epithelial desquamation of the most differentiated cells (spermatocytes and round spermatids). In this group, the percentage of affected tubules is higher than in the other groups tested. Although epithelium disorganization was observed in treated groups, the histological characteristics of Sertoli cells -nucleus located parallel to basement membrane next to spermatogonia and premeiotic spermatocytes and triangular in shape- remained unaltered. Additionally, TUNEL analysis of rat testis section from control and R50 treated animals were performed. Only a few seminiferous tubules with some TUNEL positive cells located in areas not corresponding to positions occupied by Sertoli cells were observed in both groups ( Supplementary Figure 1 ).

To analyze the effects of G or R treatment on BTB integrity, the permeability of the BTB was evaluated using a biotin tracer which is excluded from the adluminal compartment of intact seminiferous tubules. Figure 3A shows that the tracer entered the adluminal compartment in animals treated with G or R at both doses tested. Figure 3B shows the data obtained after determining the percentage of permeable tubules in the different experimental groups. A significant increase in the percentage of seminiferous tubules with a permeable barrier in G and R groups was observed.

The next set of experiments was performed to evaluate possible mechanisms that would explain the deleterious effects of G or R treatments on BTB permeability. First, as testosterone is the main regulator of BTB formation and integrity, intratesticular testosterone (ITT) levels and androgen receptor (AR) expression were evaluated. Figure 4A shows that G and R treatments did not modify either ITT levels or AR expression. Secondly, as it has been demonstrated that reactive oxygen species (ROS) mediate some of the harmful effects in BTB integrity, we decided to evaluate TBARS levels after G or R treatments. Figure 4B shows that TBARS levels were not modified by G or R exposure. Thirdly, the expression of intercellular junction proteins such as claudin11, occludin, ZO-1, connexin43, 46, and 50 which are components of the BTB, was analyzed. Figure 4C shows that G or R treatments did not modify claudin11, occludin, ZO-1, connexin43, 46, and 50 mRNA levels.

In order to evaluate possible consequences of juvenile herbicide treatment in adulthood, a set of animals was treated with 50 mg/kg/day of R from PND 14 to 30 and then allowed to grow until PND 90. Several parameters were analyzed at this age. As it was observed in PND 31, Table 3 shows that no differences in urea, creatinine, AST, and ALT levels between groups were observed.

Figure 5A shows the histological examination of testis at PND 90 in control and R50 groups. Tubules had normal cellular associations and complete spermatogenesis. Figure 5B shows the analysis of the frequency of the stages of the seminiferous epithelium. No alteration of the presence of the different stages of the cycle of the seminiferous epithelium was found between groups. Figure 5C shows the study with the biotin tracer and Figure 5D shows the data obtained after determining the percentage of permeable tubules. A small increase in the percentage of permeable tubule in R50 group was observed. Despite this small increase in BTB permeability, testicular weight, and DSP were not modified by juvenile treatment with R ( Table 3 and Figure 6A ).

Cauda epididymal sperm were used to analyze sperm motility and morphology. No differences in motility and morphology were observed in sperm of both groups ( Figures 6B, C ). In addition, no changes were observed in epididymal weight and in epididymal/body weight ratio between groups ( Table 3 ).