Alterations in the testicular parenchyma of Foxn1 +/- and Foxn1 -/- adult mice

: Nude mice carry an autosomal recessive mutation in the Foxn1 gene and therefore are homozygous recessive animals ( Foxn1 -/- ). The fertility rate of homozygous male ( Foxn1 -/- ) is low, which seems to be related to the delay in the production of gametes at the beginning of sexual maturity. The present study evaluated the structural and organizational aspects of the testicles of homozygous and heterozygous offspring related to the Foxn1 gene in mice, describing its implications on spermatogenesis. Adult males Balb/c , Foxn1 +/- and Foxn1 -/- mice were used. Testes and epididymis were harvested for histological, biochemical, and sperm transit analyses. Gonadal weight was signifi cantly lower in Foxn1 +/- and Foxn1 -/- animals, the same behavior was noticed for the activity of antioxidant enzymes. In addition, tubular parameters such as epithelial proportion, length, and area, as well as germ and Leydig cell’s populations were signifi cantly reduced in the aforementioned groups, leading to lower sperm production. In conclusion, our results indicate the importance of the Foxn1 in Leydig cell’s function, refl ecting in the preservation of spermatogenesis, thus in germ cell’s population and sperm cell production.


INTRODUCTION
The Foxn1 gene, located on chromosome 11 (Flanagan 1966), is responsible for encoding the Foxn1 transcription factor, whose expression is found specifi cally for thymus (Nehls et al. 1996) and skin epithelial cells (Brissette et al. 1996).Loss of function in one or both domains leads to the nude phenotype, commonly used in scientifi c studies.Nude mice carry an autosomal recessive mutation in the Foxn1 gene and therefore are homozygous recessive animals (Foxn1 -/-) (Oliveira et al. 2020).
The fertility rate of homozygous females (Foxn1 -/-) is low, which seems to be related to the delay in the production of gametes at the beginning of sexual maturity (Alten & Groscurth 1975) and the impairment of the development of the mammary glands (Nagasawa & Yanai 1977).Thus, the most effi cient breeding system uses homozygous males (Foxn1 -/-) and heterozygous females (Foxn1 +/-) (Rygaard & Friis 1974).The homozygous neonate can be identifi ed 24 hours postpartum by the lack or deformity of the vibrissae.
N u d e m i ce h av e re d u ce d s e r u m concentrations of gonadotropins and testosterone (Rebar et al. 1982), although their synthesis is normal (Brünner et al. 1986).The same pattern can be observed in thymectomized rodents during the neonatal period.These animals have reduced levels of gonadotropins and testosterone concerning normal ones (Pierpaoli & Besedovsky 1975).Some studies show that Swiss nude mice have an impaired function of the hypothalamic-pituitary-adrenal axis, which could be related to the lack of thymic factors (Daneva et al. 1995).Besides, the thymic factor called thymulin, produced by the thymus epithelial cells shows a physiological role in the communication of the thymus-hypothalamuspituitary.This was found after the observation that gene therapy, to reestablish the production of thymulin in nude female mice, preventing deficits in the serum concentrations of LH and FSH, typically found in these adult animals (Goya et al. 2007).On the other hand, the concentration of androgens modulates the size of the thymus in rodents, since castrated animals have increased thymus.The evidence of such regulation is the expression of androgen receptors in thymus epithelial cells (Olsen et al. 2001).Therefore, the present study aimed to evaluate the structural and organizational aspects of the testicles of homozygous and heterozygous offspring related to the Foxn1 gene in mice, describing its implications for the spermatogenic process.

Animals
The experiment followed the ethics guidelines provided by the National Council for Animal Experimentation Control (CONCEA) and was approved by the Ethics Committee of Animal Use of the Federal University of Viçosa, Brazil (protocol nº 64/2019).All methods agreed with the Brazilian guidelines that involve the use of animals in research (Law 11,794 / 2008).
The fertility rate of homozygous females Foxn1 -/-is low, as they show a delay in the production of gametes at the beginning of sexual maturity and the development of the mammary glands (Alten & Groscurth 1975, Nagasawa & Yanai 1977).In addition, the lack of fur compromises the heating and well-being of the offspring.For these reasons the most efficient form of reproduction for obtaining Foxn1 -/-animals is to use homozygous males Foxn1 -/-and heterozygous females Foxn1 +/- (Rygaard & Friis 1974).The homozygous neonate can be identified 24 hours postpartum by the lack or deformity of the vibrissae.In the present study mating occurred as previously described, and at 21 days of age 8 homozygous males Foxn1 -/-and 8 heterozygous males Foxn1 +/-were separated for the experiment.These animals were compared with homozygous males Foxn1 +/+ .
All animals were distributed in individual cages made of opaque polyethylene and closed with stainless steel lids and adequate ventilation.The animals were maintained in controlled photoperiod (12h light/dark cycles) and temperature (21 ºC), with free access to water and chow, being euthanized at 120 days of age.
Thiopental (i.p. 30mg/kg) was used for anesthesia.For each group (n=8), the testes and epididymis were removed: six testicles from each group were fixed in Karnovsky fixative (Karnovsky 1965) for histological analysis, while six testicles were stored at -80ºC for oxidative stress analysis.Four testicles with their epididymis were frozen and kept freezer at -20ºC for sperm transit analysis.

Histology
The testes were fixed for 24h.The albuginea was removed, weighed, and its weight subtracted from the testicular weight to obtain the testicular parenchyma weight.Tissue fragments were routinely dehydrated in ethanol and embedded in glycolmethacrylate (Historesin®, Leica).Semiserial sections (3µm) were made and stained with toluidine blue/sodium borate, 1%.The  (2022) 94(2) e20211123 3 | 15 software ImageJ® was used for morphometrical analysis.
The volumetric proportions of the tubular and interstitial compartments were obtained by counting 2,660 points placed on 10 digital images per animal as described by Dias et al (2019).The mass of the testicle was considered equal to its volume (Johnson et al. 1981).
The tubular diameter was calculated as the average of 20 cross-sections of seminiferous tubule for each animal.The height of the seminiferous epithelium was calculated using the same tubules cross-sections, from the tunica propria to the tubular lumen.The lumen diameter was calculated after subtracting the tubular diameter from the seminiferous epithelium height.
The total length of the seminiferous tubules (STL) per testis, was estimated from previous knowledge of the volume occupied by them within the parenchyma, as well as from the average tubular diameter: STV/πr 2 (STV = seminiferous tubule volume; πr 2 = area of the tubule cross-section; r = diameter/2) (Attal 1963).The length of seminiferous tubules (per gram of testis) was calculated by dividing STL by the testes weight.

Germ and Sertoli cells count
In order to evaluate the populations of germ cell types located in the seminiferous epithelium in stage I of the seminiferous epithelium cycle (Amann & Schanbacher 1983, Swierstra 1968), twenty tubule cross-sections were used for each animal to count the following cell types: type A spermatogonia (SGA); spermatocytes at preleptotene/leptotene (PL/L) and pachytene (P), round spermatids (RS), and Sertoli cells (SC).The results were corrected for the nuclear/nucleolar diameter and the thickness of the histological section (Amann & Almquist 1962).The diameters of 30 nuclei/nucleoli of the mentioned cell types were measured for each animal.Furthermore, to evaluate the efficiency of the spermatogenic process and the support capacity of the Sertoli cells, the following ratios were calculated: efficiency of the mitotic process ((PL/L)/SGA); spermatogenesis yield (RS/SGA), meiotic index (RS/P), Sertoli cell index (RS/S), total support capacity of the Sertoli cell (SGA+PL/L+P+RS/S).
The number of Sertoli cells was estimated from the corrected number of Sertoli nucleoli and the total length of seminiferous tubules per testis (Courot et al. 1970).From this calculation, the number of Sertoli cells per gram of testis was estimated.
The daily sperm production (DSP) was estimated using the seminiferous tubule volume, the number of round spermatids, the seminiferous tubule cross-sectional area in stage 1, the duration of the cycle of the seminiferous epithelium, and the thickness of the histological section according to Amann (1970).Bras Cienc (2022) 94(2) e20211123 4 | 15 Leydig cell morphometry and stereology The nuclear diameter of the Leydig cell was measured in 30 cells per animal (400X magnification).Cells with spherical nuclei were considered, from which the nuclear volume was calculated (VN = 4/3 πR 3 , R = nuclear radius).Leydig's cytoplasmic volume was calculated using the formula VC =% cytoplasm x NuV /% nucleus.The volume of each Leydig cell was calculated from the sum of the NuV and the VC.
The Leydig cell volume per testis calculated using the formula LC vol = Leydig cells (%) in the parenchyma x parenchyma weight / 100.The number of Leydig cells per testis was calculated using the formula LC n = LC vol / volume of each Leydig cell.The total number of Leydig cells per gram of testis was calculated using the formula LC n/g = LC vol/g / volume of each Leydig cell (LC n/g = number of Leydig cells per gram of testis).The Leydigsomatic index (LSI, %) was calculated by the formula LSI = LC vol / BW x 100 (BW = body weight).

Oxidative stress
Briefly, the testicular tissue was homogenized in potassium phosphate buffer (pH 7.4, 0.2 M) with EDTA 1M, and centrifuged (13,800g, 4°C, 10 min).The supernatant was used for the calculation of superoxide dismutase (SOD), catalase (CAT), and glutathione-S-transferase (GST) levels.Data were normalized concerning the total protein levels in the supernatant.
The enzymatic activity was determined by duplicate, using a spectrophotometer (UV-Mini 1240, Shimadzu) or ELISA reader (Thermo Scientific, Waltham, MA, USA).CAT activity was assessed by measuring the decomposition rate of H 2 O 2 (Aebi 1984).SOD activity was determined according to Siddiqui et al. (2005).GST activity was analyzed through the formation of 1-chloro-2,4-dinitrobenzene conjugate (CDNB) (Habig et al. 1974).The total protein concentration was measured using bovine serum albumin as a standard curve (Lowry et al. 1951) and used to homogenize stress data.

Sperm parameters
The caput/corpus and cauda regions were sliced and homogenized as previously described for the testis.Homogenization resistant spermatids (stage 19 of spermatogenesis) in testis and spermatozoa in the caput/corpus and cauda regions of the epididymis were counted according to Dias et al. (2019).To calculate the daily sperm production (DSP), the number of spermatids at stage 19 was divided by 4.84 (the number of days that spermatids are present in the seminiferous tubules of mice).Epididymal transit time was determined by dividing the number of sperm in each region by the daily sperm production (DSP) (Dias 2018).

Statistics
The mean results were tested for normality using the Shapiro Wilk test.Parametric data were analyzed by ANOVA and the Student-Newman-Keuls post hoc test.Pearson's correlation was used to assess the relationship between two variables.Differences were significant when P <0.05.
The principal component analysis (PCA) was performed to verify clustering, to eliminate redundancies, and to define the most important variables during the separation of experimental groups.To do so, data were transformed (ranging) for standardization due to different scale magnitudes.The level of importance of each variable was determined by eigenvector values (McGarigal et al. 2000), with substantial correlation values determined for each attribute concerning the principal components (PC) 1 and 2. The level of importance of each PC was determined by the Broken-stick method, where eigenvalues exceeding the expected values were kept for interpretation.Analyzes were performed using the Fitopac 2.1.2.85 software (Shepherd 2010).In addition, the relative importance of the evaluated characteristics was calculated (Singh 1981).All analyzes were performed using the Genes software (Cruz 2008).

RESULTS
Biometry, morphometry and stereology Foxn1 +/-and Foxn1 -/-animals showed lower body, testicular, and parenchyma weight (P<0.05).In addition, the PSI and GSI were significantly lower in Foxn1 +/-and Foxn1 -/-animals (Figure 1).Foxn 1 +/+ animals show normal tissue microstructure, with well-structured epithelia and normal interstitium.In Foxn 1 +/-animals, it is possible to see reduced vacuoles in the seminiferous epithelium and Leydig cell nucleus.The same can be seen in Foxn1 -/-animals, although with greater alterations in the epithelial structure, larger vacuoles, and in higher proportions, causing epithelial disorganization in some seminiferous tubules (Figure 1).
The weight of the gonad showed a strong positive correlation with the GSI (r = 0.96) and the weight of the parenchyma (r = 0.92) in addition to a strong negative correlation with  the weight of the albuginea (r = -0.91)and PSI (r = -0.90).The proportion of seminiferous tubules and interstitium was not different between groups.However, the proportion of seminiferous epithelium was lower in Foxn1 -/-animals, while the proportion of tunica propria was also lower in Foxn1 +/-animals (Figure 2).The proportion of epithelium showed a strong positive correlation with the seminiferous tubule's proportion (r = 0.85).Tubulesomatic and epitheliumsomatic indexes were lower in Foxn1 +/-and Foxn1 - /-animals (Table I).Both indexes showed a positive correlation with the seminiferous tubule's proportion (r = 0.52).
The diameter of the seminiferous tubule and the height of the epithelium were lower in Foxn1 +/-and Foxn1 -/-animals, the latter having the lowest values, which was reflected in the decrease in the area of such compartments.The diameter of seminiferous tubule and the height of seminiferous epithelium show a strong positive correlation between them (r = 0.88).On the other hand, the lumen diameter and the tubule-epithelium relationship did not differ between groups (Table II).The length of seminiferous tubule per testicle and per gram of testicle was lower in Foxn1 +/-and Foxn1 -/-animals (Table II).

Germ and Sertoli cells count
The corrected number of spermatogonia, primary spermatocytes in pre-leptotene and pachytene, and round spermatids per tubule cross-section were lower in Foxn1 +/-and Foxn1 - /-.In addition, the corrected germ cells number per tubule cross-section was lower in these same groups, with Foxn1 -/-animals showing the lowest mean (Table III).Sertoli cell number was not different between groups, as was the mitotic index.However, the meiotic index, the Sertoli cell index, and the Sertoli cell support capacity were lower in Foxn1 +/-and Foxn1 -/-, with the lowest means observed in Foxn1 -/-.The number of Sertoli cells per testis and per gram of testis, as well as the spermatogenic yield, and the sperm production per testis and per gram of testis were lower in Foxn1 +/-and Foxn1 -/- compared to Foxn1 +/+ individuals (Table III) .

Sperm transit
The number of spermatids per testis was lower in the Foxn1 +/-and Foxn1 -/-groups, however, the relative number per testis was lower only in Foxn1 -/-(P<0.05,Table IV).The daily sperm production and the relative sperm count in all epididymal segments were lower in Foxn1 +/-and Foxn1 -/-animals (P<0.05),although alterations in the transit time were only noticed in the cauda epididymis of the Foxn1 -/-group (Table IV).The number of sperm in the cauda epididymis per organ (r = 0.93) or per gram of organ (r = 0.92) showed a strong positive correlation with the transit time in each segment.The number of spermatids per testis (r = 1.00) and per gram of testis (r = 0.91) show a positive correlation with sperm production.

Oxidative stress
Glutathione levels were not different between groups; however, SOD and CAT levels were significantly lower in Foxn1 +/-and Foxn1 -/- animals, with a strong correlation between them (r = 0.96, Figure 3)

The relative importance of the variables
The relative importance of the variables pointed out that all variables were equally important for the separation of the experimental groups (Figure 4b).Removing one of the variables would alter the group's distribution.

DISCUSSION
Although Foxn1 -/-mice are widely used in reproduction studies, there are few reports with details regarding its reproductive biology.Thus, in the present study, we evaluated the testicular parenchyma of wild-type, Foxn1 +/-and Foxn1 -/-mice.
The reduction in body weight observed in Foxn1 +/-and Foxn1 -/-animals is common when dealing with different species, however, significant reductions in testicular weight and consequent parenchyma weights, and in the parenchymasomatic index, may indicate changes in spermatogenesis, thus in sperm production (França & Russell 1998).GSI and PSI were reduced in Foxn1 +/-and Foxn1 -/-groups, indicating that important changes within the organ are taking place.
The proportion between the spermatogenic and steroidogenic compartments is quite variable in mammals (França & Russell 1998), being one of the main factors responsible for variation in sperm production among species.The tubular compartment constitutes the major part of the testicular parenchyma, ranging from 70 to 90% of the whole organ area and all values observed in the present study remained within this range (Clermont & Trott 1969).The seminiferous tubules show a direct relationship with the tubular length, as well as with germ and Sertoli cells' populations, culminating in sperm production (França & Russell 1998).
The seminiferous epithelium proportion was lower in Foxn1 -/-animals than in the Foxn1 +/+ or Foxn1 +/-, consequently reducing the epitheliumsomatic index in Foxn1 -/-animals.Although in Foxn1 +/-animals there was no significant reduction in the proportion of seminiferous epithelium, this proportion was lower, reflecting the reduction in the epitheliumsomatic index in these animals concerning Foxn1 +/+ .Such alterations would inevitably compromise sperm production since the seminiferous epithelium shows a direct correlation with the number of germ cells and sperm produced.In addition, significant reductions in the proportions of the tunica propria both in Foxn1 +/-and Foxn1 -/-animals were noticed.Myoid cells, which are important components of the tunica propria have a functional correlation with the Leydig cells, acting in the fine regulation of spermatogenesis (Zhou et al. 2019), thus such reduction of the tunica propria might influence the spermatogenic process.
The average tubular diameter does not change significantly after the establishment of sexual maturity and remains constant  et al. 1986), which was observed in Foxn1 +/-and Foxn1 -/-animals.Interestingly, Foxn1 - /-animals showed the lowest tubule diameter, also noticed for other seminiferous epithelium parameters, such as the length of the tubules, as well as in their respective areas.Due to such reductions, the spermatid's populations were significantly lower than in the other groups, reflecting in lower sperm production and the sperm number present in all epididymis' segments.
The reduction in the spermatogonia, spermatocytes and spermatids counts led to a significant reduction in the number of germ cells in both Foxn1 +/-and Foxn1 -/-animals.Such reduction is in accordance with the reduction in height of the seminiferous epithelium observed in individuals of the Foxn1 +/-and Foxn1 -/-strains.Furthermore, although the Sertoli cell count was not altered, there were lower Sertoli cell support capacity and lower Sertoli cell indexes, which culminated in a reduced sperm reserve, daily sperm production and spermatogenesis yield between Foxn1 +/-and Foxn1 -/-strains, as well as between both and Foxn1 +/+ .Despite the lower sperm count within the epididymis, the spermatic transit, which is the transport of sperm through the epididymal duct, was only altered in the cauda in Foxn1 -/-animals.Such transport is dependent on the differential hydrostatic pressure gradient between the proximal and distal portions of the epididymal duct, the contractile activity of the duct wall, controlled by the autonomic nervous system, and the action of androgens (Robaire & Viger 1995, Cosentino & Cockett 1986, Klinefelter 2002).The delay in transit time through the epididymis does not alter the fertile capacity of gametes, but when it is accelerated, fertility is compromised since the time available for the processes required for the acquisition of fertile capacity is reduced (Kempinas et al. 1998).Changes in sperm transit time also change the amount of sperm available for ejaculation (Klinefelter 2002).
The individual volume and number of Leydig cells, as well as the leydigsomatic index, were reduced in Foxn1 -/-and Foxn1 +/- animals, which might be associated with low testosterone production.The transcription factor Foxn1 is translocated to the cell nucleus after phosphorylation, where it regulates the expression of several genes (Mecklenburg et al. 2005).Foxn1 can play a role in maintaining the physiology of the testes, however, it is poorly expressed in the nucleus of Leydig cells in Foxn1 -/-animals (Oliveira et al. 2020).Thus, the cytoplasmic distribution of such transcription factor suggests its participation in the regulation of the function of the steroidogenic pathway in Leydig cells (Oliveira et al. 2020).
Both Foxn1 -/-and Foxn1 +/-animals showed reduced SOD and CAT activity.SOD is an antioxidant enzyme considered to be the first line of defense against the deleterious effects of reactive oxygen species (ROS), catalyzing the reaction of conversion of the superoxide radical to H 2 O 2 (Barbosa et al. 2010) which is easily degraded by CAT and GST (Aitken & Roman 2008).CAT converts H 2 O 2 to water and oxygen (Barreiros et al. 2006).Glutathione levels did not differ between groups.GST is involved in the reactions of phase II of antioxidant activity, reducing the production of lipid peroxidation through the reduction of hydroperoxides (Hayes et al. 2005).Thus, it is responsible for cellular detoxification through glutathione conjugated with xenobiotics and aldehyde products produced in lipid peroxidation, making them more soluble in water (Habig et al. 1974).Its activity, as well as that of CAT, can be directly inhibited by high concentrations of NO (Kostic et al. 2000, Wong et al. 2001).Thus, we can suggest that Foxn1 -/-and Foxn1 +/- animals have an inherent deficiency in natural antioxidant defenses, which can cause changes to reproductive functionality (the quality and quantity of sperm), showing increased structural changes in the seminiferous epithelium.

CONCLUSIONS
Foxn1 +/-and Foxn1 -/-strains resulting from the mating between Foxn1 -/-male and heterozygous Balb/c female, are quite similar concerning reproductive parameters.They show compromised spermatogenesis, with reduced spermatids' population within the also reduced seminiferous epithelium, reflecting in low daily sperm production, especially on the Foxn1 -/- animals.Such behavior might be related to the lower Leydig cell volume and populations within the interstitium, therefore affecting the fertility rate of the strains.
FERNANDA C.R. DIAS et al.Foxn1 AND THE TESTICULAR PARENCHYMA OF MICE An Acad Bras Cienc FERNANDA C.R. DIAS et al.Foxn1 AND THE TESTICULAR PARENCHYMA OF MICE An Acad

Table I .
Proportions and indexes of the seminiferous tubules components.
b ESI= epitheliumsomatic index, TSI= tubulesomatic index.Different superscripts ( a,b ) in the same row are P<0.05.Values are mean ± S.D. ANOVA with SNK post hoc test.

Table III .
Germ and Sertoli cells count in seminiferous tubules cross-sections (Stage I of the seminiferous epithelium cycle).

Table IV .
Spermatids and sperm count in testis and epididymis.

Table V .
Leydig cell morphometry and stereology.LC = Leydig cell, LSI: leydigsomatic index.Different superscripts ( a,b ) in the same row are P<0.05.Values are mean ± S.D. ANOVA with SNK post hoc test.