| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Experimental Radiation Oncology,4 The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030
Department of Agricultural and Environmental Sciences and George Washington Carver Experiment Station,5 Tuskegee University, Tuskegee, Alabama 36088
Department of Morphology,6 Institute of Biological Sciences, Federal University of Minas Gerais, Belo Horizonte, Brazil 31270-901
Department of Obstetrics and Gynecology,7 Baylor College of Medicine, Houston, Texas 77030
ABSTRACT
Male mice that are homozygous for the juvenile spermatogonial depletion (jsd) mutation in the Utp14b gene undergo several waves of spermatogenesis. However, spermatogonial differentiation ceases and in adults, spermatogonia are the only germ cells that remain. To understand further the blockage in spermatogonial differentiation in Utp14bjsd mutant mice, we correlated the rate and severity of spermatogonial depletion and the restoration of spermatogenesis following the suppression of testosterone or elevation of testicular temperature with the genetic background. Testes from Utp14bjsd mutant mice on B6, C3H, and mixed C3H-B6–129 (HB129) genetic backgrounds all showed steady decreases in the numbers of normal spermatogonia between 8 wk and 20 wk of age. The percentages of tubules with differentiating germ cells were higher and the spermatogonia were more advanced in C3H- background than in B6- or HB129-background Utp14bjsd mice. Genetic crosses showed that the source of the Y chromosome was a major factor in determining the severity of spermatogonial depletion in Utp14bjsd mutant mice. When Utp14bjsd mutants were subjected to total androgen ablation or unilateral cryptorchidization, spermatogenic development recovered markedly in the C3H and HB129 background but showed less recovery in the B6-background mice. The differences noted between the strains in terms of the severity of spermatogonial depletion were not dependent upon testosterone level or scrotal temperature but correlated with the magnitudes of the effects of elevated temperature on normal and Utp14bjsd mutant spermatogenic cells. Thus, the abilities of germ cells in certain strains to survive elevated temperatures may be related to their abilities to maintain some degree of differentiation potential after the Utp14bjsd gene is mutated.
high-resolution light microscopy,, jsd mutation,, spermatogenesis,, spermatogonia,, testis,, testosterone,, Y chromosome
Male mice that are homozygous for the juvenile spermatogonial depletion (jsd) mutation initiate normal spermatogenesis, although this ceases when the mice reach adulthood [1]. In adults, type A spermatogonia are the only remaining germ cells. The mutation that causes the Utp14bjsd phenotype is a frameshift that generates a stop codon in Utp14b, the autosomal retroposon of the X-linked mouse homolog of the yeast UTP14 gene [2, 3], which is necessary for 18S ribosomal RNA processing. The mechanism of this age-dependent depletion is not yet known, although reciprocal spermatogonial transplantations have shown that the phenotype is attributable to a defect in the germ cells themselves rather than to defects in the somatic cells of the testis [4, 5].
The suppression of testosterone [6–8] or increases in testicular temperature [9], treatments that disrupt the meiotic and postmeiotic stages of spermatogenesis, surprisingly restore spermatogonial differentiation in adult Utp14bjsd mice by mechanisms that remain unknown.
The time course of spermatogonial depletion was first described in Utp14bjsd mice on the C57BL/6J (B6) genetic background, in which nearly all signs of spermatogenesis disappear by 10 wk of age [1, 10]. However, Utp14bjsd mutant mice on a mixed B6-C3H (BH) genetic background retain differentiating germ cells in more than 10% of the tubules at 12 wk of age [7, 9].
This variation in the severity of the defect with different genetic backgrounds suggests that if we are to identify the mechanisms underlying the block in spermatogonial differentiation, we must first characterize fully the stages at which the block in spermatogonial differentiation occurs on the genetic backgrounds being used. An earlier study of the differentiation of Utp14bjsd spermatogonia in B6 mice used tubule whole mounts and showed that about 3% of the mitotic clones expanded to more than 16 cells [11]. However, it is unclear whether such large clones are still A-aligned (Aal) spermatogonia [12] or whether they are more advanced in terms of differentiation [13]. To distinguish the stages of spermatogonial differentiation based on morphology, rather than clone size, we employed recently developed criteria to distinguish different spermatogonial types in semithin tissue cross-sections using high-resolution light microscopy [14]. This approach enables the morphological recognition of the undifferentiated type A (including As, Apr, and Aal), each of the differentiated type A (A1, A2, A3, and A4), intermediate, and type B spermatogonia. This allows evaluation of spermatogonial development in Utp14bjsd mice without reference to the stages of the cycle of the seminiferous epithelium, and permits the detection of morphological defects in these spermatogonia.
We found nuclear defects in the spermatogonia of Utp14bjsd mice. In addition, we discovered that both the block in spermatogonial differentiation and its reversal by hormone suppression or temperature elevation are dependent upon genetic background, and that the major genetic factor is the source of the Y chromosome. Neither testosterone levels nor scrotal temperatures were directly responsible for the interstrain differences, although the resistance of normal spermatogenesis to elevated temperature in the different strains was related to the ability to restore spermatogenesis with testosterone suppression and temperature elevation in Utp14bjsd mutants.
The Utp14bjsd mutant mice on the C57BL6 (B6) background were obtained from the Jackson Laboratory (Bar Harbor, ME).
The Utp14bjsd mutant mice on the C3H background were developed by backcrossing Utp14bjsd mutant mice on a C3H-B6 mixed hybrid background (designated BH) [7] to C3H/KamLaw for ten generations.
The Utp14bjsd mutant mice on a hybrid background (designated HB129) were developed by crossing male Fshb-null mice on a 129-B6 mixed background [15], which were originally derived from 129 XY ES cells (AB2.1 line), with C3H females, with the offspring with the C3H haplotype at the Utp14bjsd locus being selected [16]. These mice were intercrossed and then crossed with Utp14bjsd mutant females on the BH hybrid background [7]. Offspring were intercrossed and eventually, a stock that carried the Utp14bjsd mutation, but not the Fshb-null mutation, was selected and maintained by random intercrossing.
Historical data from Utp14bjsd mutant mice on a BH mixed hybrid background were also used. This line was derived from mice (received from the Baylor College of Medicine) that were used to map the Utp14bjsd locus [17]. The line was produced by crossing a Utp14bjsd/Utp14bjsd female on the B6 background with a C3H/HeJ male, to yield F1 mice that were intercrossed in all the subsequent generations.
The mice were provided with food and water ad libitum. In initial studies, both breeding and nonbreeding mice were fed LabDiet Purina 5001 chow. In the later studies (genetic crosses), the breeding mice were fed LabDiet Purina 5015 chow and the LabDiet Purina 5001 diet was only used for the experimental mice after weaning. All the experimental procedures were approved by the M. D. Anderson Cancer Center Institutional Animal Care and Use Committee.
The Utp14bjsd mice were genotyped by PCR of tail DNA using the primers and HphI digestion of the PCR product, as previously described [2].
Reciprocal crosses were performed as follows: homozygous Utp14bjsd/Utp14bjsd females on the B6 background were crossed with +/Utp14bjsd males on the C3H background, to generate (BxH)F1-Utp14bjsd/Utp14bjsd mutant mice. Similarly, Utp14bjsd/Utp14bjsd females on the C3H background were crossed with +/Utp14bjsd males on the B6 background, to generate (HxB)F1-Utp14bjsd/Utp14bjsd mutant mice. In addition, the N2 backcross generation was produced by mating (BxH)F1-+/Utp14bjsd males with Utp14bjsd/Utp14bjsd females on the B6 background. Mice were killed at 12 wk of age for the assessment of tubule differentiation indices and hormone measurements, as described below.
Testosterone production and action were suppressed by treating 8- or 9-wk-old Utp14bjsd mice with the GnRH antagonist acyline (obtained from the Contraceptive Development Branch of NICHD, North Bethesda, MD) and, except as noted, the androgen receptor antagonist flutamide for 4 wk [16]. Acyline was prepared at a concentration of 2 mg/ml in water and administered as two s.c. injections at 10 mg/kg each at separate sites, followed 2 wk later by a single injection at a dose of 10 mg/kg. Flutamide was administered via s.c. implantation of two 2-cm Silastic capsules (Dow Corning, Midland, MI). Unless otherwise stated, mice were killed 4 wk after the start of treatment. Additional B6 mice were treated and killed at shorter intervals.
Utp14bjsd mutant mice and +/Utp14bjsd littermates were unilaterally cryptorchidized at 12 wk of age. The adipose tissue of the right caput epididymis was sutured to the lower part of the inner peritoneal wall at a position that placed the testis at the level of the urinary bladder, as described earlier [9]; note that the gubernacular cord was not cut. The left testis remained in the scrotum and was used as the control. Mice were killed 8 wk later for analysis.
Blood was collected by cutting the axillary vein of anesthetized mice and drawing the blood up in a syringe. Serum was separated and stored at –20°C until testosterone was measured. Both testes were weighed. The left testis was fixed in Bouin solution and embedded in methacrylate (JB4; Polysciences Inc., Warrington, PA) for histological evaluation, unless otherwise noted. Testis homogenates were obtained by manual homogenization of the right testis and centrifugation [7]; the supernatant was stored at –20°C until intratesticular testosterone (ITT) was measured.
For the evaluations of spermatogonial morphology and differentiation, testes of adult mice on each genetic background were perfusion-fixed for high-resolution light microscopy [14]. Normal mice (+/+ and +/Utp14bjsd) were killed at 12 wk of age and Utp14bjsd/Utp14bjsd mice were killed at 8, 12, and 20 wk of age. Heparin was injected into each animal at least 15 min prior to perfusion. Saline was initially perfused for 10–15 min to clear the blood, followed by 5% glutaraldehyde in 0.05 M sodium cacodylate buffer (pH 7.4) (Electron Microscopy Sciences, Hatfield, PA) for approximately 30 min. The testes were then removed, weighed, and sliced into 0.5-mm slabs perpendicular to the long axis of the testis. The slabs were then cut into 2-mm portions, fixed overnight in 5% glutaraldehyde, and washed three times for 15 min each in cacodylate buffer. The slabs were then post-fixed in an osmium-ferrocyanide mixture, dehydrated, and embedded in Araldite. From the Araldite-embedded blocks, 1-µm sections were cut with a Leica RM 2165 microtome and stained with toluidine blue-borate for high-resolution light microscopy.
The tubule differentiation index (TDI) was determined for each testis by microscopic analysis, as described previously [7]. The TDI represents the percentage of tubules that were clearly differentiating, which meant, unless otherwise stated, that they contained three or more germ cells at the type B spermatogonial stage or beyond. In some instances, it was also useful to calculate the percentage of tubules that had reached or passed the primary spermatocyte stage, which we refer to as TDI-Spermatocytes. However, in 8-wk-old Utp14bjsd mice, some tubules had type A spermatogonia and spermatids, while the intermediate developmental stages were missing because spermatogonial differentiation had proceeded at a younger age and had ceased by this time-point. To ensure that the TDI reflected the status of spermatogonial differentiation immediately preceding the 8-wk time-point, these tubules were not scored as differentiating tubules unless they contained some B spermatogonia or primary spermatocytes.
Historical data on the TDI from mice on the BH hybrid background [7] were also included in some of the analyses. However these data were obtained for paraffin-embedded tissues in which the identification of intermediate and B spermatogonia is was not clear. Therefore, the TDI values for these tissues could not be compared quantitatively with those obtained for plastic-embedded tissues. Nevertheless, spermatocytes could be readily identified in both preparations and the TDI-Spermatocyte values were comparable.
For the morphological evaluation of spermatogonial subtypes, these cells were photographed with a Q-Color 3 video camera using an Olympus BX-60 photomicroscope. All digital images received the same adjustment of resolution and contrast using the Photoshop software (Adobe Systems Inc., Mountain View, CA). Images were grouped according to the morphological characteristics described previously [14], which distinguish spermatogonia based on their nuclear features, i.e., the shape of the nucleus, presence of a vacuole, amount and disposition of the heterochromatin, granularity of the euchromatin, and the morphology and degree of nucleolar compaction. The numbers of undifferentiated type A spermatogonia (As, Apr, and Aal), differentiated type A spermatogonia (A1, A2, A3, and A4), intermediate spermatogonia, type B spermatogonia, abnormal spermatogonial nuclei, and the numbers of Sertoli cell nucleoli were counted in 20–30 cross-sectioned seminiferous tubules per animal. The spermatogonial counts are expressed as the number of cells per 100 Sertoli cell nucleoli.
The serum testosterone and ITT concentrations were measured using coated-tube RIA kits (DSL 4000; Diagnostics Systems Laboratories Inc., Webster, TX) [18]. For serum testosterone, the standards were prepared in GnRH-suppressed rat serum that had been stripped with dextran-coated charcoal (Sigma Chemical Co., St. Louis, MO). For ITT, the standards were prepared in PBS that contained 0.1% gelatin. The ITT is expressed as amount per gram of testis rather than amount per testis to reflect the concentrations of testosterone to which the testicular cells were exposed, which in turn determined the responses. The intraassay and interassay coefficients of variation were 10% and 16%, respectively.
Temperature measurements were performed on mice that were anesthetized with a 15–30-sec exposure to isoflurane and placed on an insulated pad under continuous exposure to anesthesia. Thermocouple microprobes and a BAT-8 digital thermometer were used (PhysiTemp, Clifton, NJ). In normal (+/Utp14bjsd) mice, a small incision was made in the scrotal skin and a 29G needle probe was inserted into the testis through the cremaster sac. In the Utp14bjsd mutants, due to the small size of the testis, the needle was inserted through the skin into the scrotum adjacent to the testis. A separate 26G needle was inserted into the peritoneal cavity. The temperature was recorded as quickly as possible, since body temperature decreases during anesthesia. Intraperitoneal temperatures were based on measurements carried out on 2 or 5 separate days, and intratesticular or intrascrotal temperature measurements were performed on 2 days (each side on separate days).
The data for cell counts and TDI are presented as the arithmetic mean ± SEM. The data on testosterone measurements are the averages ± SEM calculated from the log-transformed data. The significance of differences between genotypes, ages of mice, and treatments were initially evaluated by ANOVA (when multiple groups were simultaneously being compared) and, if ANOVA indicated a value of P < 0.05, we then used a Student's t-test, with P < 0.05 to indicate significance between groups. A computer-assisted statistics program (SPSS ver. 11.5; SPSS Inc., Chicago, IL) was used. The statistical analysis of the testosterone levels was performed on log-transformed data. Due to variability between mice of the cryptorchidization procedure, the data were not distributed normally and the statistical analysis was performed with a nonparametric Mann-Whitney test, with P < 0.05 to indicate significance between groups.
Morphological Characterization of Spermatogonia in Utp14bjsd Mice
In the adult Utp14bjsd mutant mice, the seminiferous epithelium was populated with Sertoli cells, undifferentiated type A (As, Apr, and Aal) spermatogonia, some differentiated spermatogonia, and, on the C3H background, some early primary spermatocytes. The absence of later-stage germ cells in the Utp14bjsd mutants made it easier to find the spermatogonial types. It was even possible to distinguish two types of undifferentiated type A spermatogonia, i.e., those with dark cytoplasm (Fig. 1A) and those with light cytoplasm (Fig. 1B). Approximately 8% of the spermatogonia were in mitosis and about 5% were apoptotic; it was not possible to associate these cells with any particular stage of differentiation. In about 35% of the spermatogonia of the Utp14bjsd mutant mice of all genetic backgrounds and ages, the nuclear features (Fig. 1, A–H), which are the main factors in discriminating different spermatogonial types, were very similar to those observed in normal mice [14] in the present study (data not shown). However, in the Utp14bjsd mutant mice, about 65% of the type A spermatogonia had abnormal morphologies. Most of these type A spermatogonia had nuclear vacuoles that were not membrane-bounded but nevertheless represented areas with lower densities of nucleic acids and proteins [19]. In the normal mice, vacuoles were found only in the undifferentiated type A spermatogonia (Fig. 1B), while these vacuoles in the Utp14bjsd spermatogonia were frequently larger (Fig. 1I) and more numerous (Fig. 1J) than those in normal undifferentiated type A cells. Furthermore, many of the nuclei with vacuoles had reticulated nucleoli that are typical of A2 to A4 spermatogonia (Fig. 1, I and J), in contrast to the compact nucleoli of undifferentiated type A spermatogonia (Fig. 1B). Some of the cells had large nuclei with vacuoles and heterochromatin, and they appeared to be degenerating (Fig. 1K). In addition, a few nuclei had abnormal, highly dispersed, reticulated nucleoli (Fig. 1L).
|
Differential Decline in Spermatogonial Differentiation in Different Genetic Backgrounds
The TDI was determined for male mice that were homozygous for the Utp14bjsd mutation on the B6, C3H, and HB129 genetic backgrounds. The first measurement was taken at 8 wk of age, at which time the Utp14bjsd mutant mice on the B6 and HB129 backgrounds already had low TDI values of 3% to 12%. These values declined further to less than 3% at 12–15 wk of age (Fig. 2, A and B). In contrast, for the Utp14bjsd mice on the C3H genetic background, the TDI was 41% at 8 wk of age, but this declined steadily to 5% at 20 wk of age. At 12 wk of age, the TDI values for Utp14bjsd mice on the B6 and HB129 backgrounds were similar to each other, although both were significantly lower than the TDI for Utp14bjsd mice on the C3H background. The percentages of tubules that contained germ cells that had differentiated to the spermatocyte stage or further (TDI-Spermatocytes) were compared in the three strains and in the BH hybrids. The historical TDI-Spermatocyte values for mice on the BH hybrid background at 12 wk of age and older [7] were similar to those for mice on the C3H background and higher than those for mice on the B6 or the HB129 backgrounds (Fig. 2C).
|
Analysis of Stage of Differentiation as a Function of Age for Different Strains
The different spermatogonial types were counted in the wild-type (+/+) and heterozygous (+/Utp14bjsd) mice of the three different strains at 12 wk of age. The results for the wild-type mice are shown in Figure 3, A–C; the results for the heterozygous mice (data not shown) were not significantly different. The numbers of spermatogonia at the type A1 stage were approximately 50% lower than the numbers of undifferentiated spermatogonia (Fig. 3, A–C), and these numbers increased steadily through the A2 to intermediate stages and dramatically at the B stage. Essentially no abnormal spermatogonia were observed.
|
The most striking observation for the Utp14bjsd mutants was the high frequency of abnormal spermatogonia (Fig. 3, A–C), which was correlated with the age- and strain-dependence of the severity of the differentiation defect. In Utp14bjsd mice on the B6 background, 72%, 76%, and 78% of the spermatogonia were abnormal at 8, 12, and 20 wk, respectively. The mutants on the HB129 background were similar to those on the B6 background with 66%, 73%, and 76% of the spermatogonia being abnormal at 8, 12, and 20 wk, respectively. In contrast, the C3H background showed only 49%, 63%, and 64% abnormal spermatogonia at 8, 12, and 20 wk, respectively.
There was clear evidence of defective spermatogonial differentiation in the Utp14bjsd mice. In mice on the B6 or HB129 backgrounds, aged 12 wk or more, the numbers of A1 spermatogonia were reduced to less than 20% of the undifferentiated spermatogonia (Fig. 3, E and F). In contrast to the progressive increase in numbers observed from the A1 to B spermatogonia in normal mice, the numbers of spermatogonia were constant or declined in the mutants, except for some progression to B spermatogonia and early spermatocytes (data not shown) in the Utp14bjsd mice on the C3H background at 8 wk and 12 wk of age (Fig. 3, D and E). Although some B spermatogonia remained in Utp14bjsd mice on the C3H background at 20 wk of age, A4 spermatogonia were the most mature cells observed in mice on the B6 background (Fig. 3F). These data further demonstrate the age- and background strain-dependence of the decline in spermatogonial differentiation in Utp14bjsd mice.
Genetic Mapping of the Loci Responsible for the Interstrain Differences
At the time the genetic crosses were done, additional Utp14bjsd mice on the parental backgrounds were analyzed, so that comparisons could be made with contemporaneous samples. Surprisingly, although the same investigator performed all of the analyses, the TDI values at 12 wk of age for the contemporaneous controls (Fig. 4, filled symbols) for both the B6 and C3H strains were much lower than those obtained in earlier experiments (Fig. 2, and Fig. 4, open symbols). This may be attributable to a change in the diet used for the breeding mice, from a general purpose diet (LabDiet Purina 5001, 4.5% fat) to a high fat diet (LabDiet Purina 5015, 11% fat). Regardless of the change from the earlier values, the TDI values for the B6 mice were all still below those for the C3H mice obtained during the same time period.
|
A reciprocal cross with B6 and C3H mice was performed to determine the contribution of the sex chromosomes, imprinting, and the mitochondrial genome to the interstrain differences in the severity of the Utp14bjsd phenotype. Histological analysis of F1 mice at 12 wk of age revealed that all the (BxH)F1-Utp14bjsd/Utp14bjsd mice had higher TDI values than all the (HxB)F1-Utp14bjsd/Utp14bjsd mice (Fig. 4). Since the severities of the F1 phenotypes varied in the same direction as the paternal rather than the maternal phenotypes (B6 and HxB-F1 more severe than C3H and BxH-F1, respectively), the major contributor to the phenotype must have been either the paternal Y chromosome or male-imprinted paternally derived autosomal loci.
However, the (BxH)F1 mice had a significantly higher TDI than the concurrently analyzed C3H mice (P < 0.001); similarly, the (HxB)F1 mice had a significantly higher TDI than the concurrently analyzed B6 mice (P = 0.002) (Fig. 5A). These results demonstrate that recessive autosomal genes also contribute to the severity of the phenotype, since the F1 mice had a less severe phenotype than the corresponding inbred lines with the same Y chromosome.
|
To determine whether the paternally mediated mild phenotype of the (BxH)F1 mice was caused by the C3H Y chromosome or male-imprinted C3H autosomal alleles, we backcrossed (BxH)F1-+/Utp14bjsd males to Utp14bjsd/Utp14bjsd females on the B6 background. If the Y chromosome was the cause of the paternally transmitted effect, we would expect all N2 generation mice to have a phenotype similar to the (BxH)F1 mice. However, if the paternally transmitted effect was due to male imprinting of a C3H autosomal allele, half of the N2 offspring should have a phenotype similar to that of the (BxH)F1 mice and half should have a phenotype similar to that of the (HxB)F1 mice. The resultant N2 male mice had a mean TDI value similar to that of the (BxH)F1 mice (Fig. 5A). The individual TDI values for the N2 mice ranged from 15% to 76% (Fig. 4), values that were higher than the TDI for all the (HxB)F1 mice, which suggests that the Y chromosome, and not imprinting, is the major source of the variation due to strain background. The significantly (P < 0.05) greater variance in the N2 TDI values compared to those for the (BxH)F1 mice confirms that there are contributions from autosomal loci, which are identical in the individual F1 mice but differ in the individual N2 mice.
The conclusion that the Y chromosome carries dominant factors that affect the severity of juvenile spermatogonial depletion phenotype is supported by the less severe phenotype of mice on the BH than on the HB129 genetic background (Fig. 2C). The BH hybrid mice carry the C3H Y chromosome, which is known to confer a less severe phenotype, whereas the HB129 mice carry the 129 Y chromosome, which resembles the B6 Y chromosome and confers a more severe phenotype.
Since spermatogenic arrest in jsd mutant mice can be reversed by suppression of ITT and serum testosterone levels, we measured the basal testosterone levels in the Utp14bjsd mice on different genetic backgrounds, to determine whether they correlated with the strain differences with respect to the degree of spermatogonial arrest. Contrary to this hypothesis, the ITT and serum testosterone levels were higher in C3H mice, which had the highest TDI, than in B6 or HB129 mice, which had low TDI values (Fig. 6, D and E). Furthermore, the two different types of F1 mice from the genetic crosses had similar testosterone levels (Fig. 5, B and D) but very different TDI values (Fig 5A). The N2 mice, which had very high TDI values, had low testosterone levels similar to those of the B6 mice, which had low TDI values. Figure 5C illustrates the complete lack of correlation between the TDI and ITT levels in mice on these different genetic backgrounds.
|
Stimulation of Spermatogenic Recovery by Hormone Treatment
Previous studies have shown that spermatogenic recovery in Utp14bjsd mutant mice on the B6 [6, 8], BH-hybrid [7], and HB129-hybrid [16] genetic backgrounds can be induced by suppressing testosterone levels with different GnRH antagonists. In the present study, we compared the extents of spermatogenic progression following the suppression of testosterone levels and activities following treatment for 4 wk with acyline plus flutamide of 8-wk-old Utp14bjsd mutant males on the different genetic backgrounds.
Testosterone suppression dramatically stimulated spermatogonial differentiation in the Utp14bjsd mice on the C3H and HB129 backgrounds, with 80% of the tubules showing differentiated cells (Fig. 6, A and B). However, differentiation was not significantly stimulated by the 4-wk hormone suppression treatment in Utp14bjsd mice on the B6 background, although transient stimulation was noted at 2 wk (Fig. 6C). The transient nature of this stimulation on the B6 background was perhaps due to the increased severity of the juvenile spermatogonial depletion phenotype with age, which overwhelmed the beneficial effects of testosterone suppression [6, 7]. The degree of stimulation did not correlate with the spontaneous level of differentiation at 12 wk of age, in that untreated Utp14bjsd mice on both the B6 and HB129 backgrounds showed little differentiation at this age, and stimulation by androgen suppression was strong in the HB129 but not in the B6 background.
Measurements of ITT and serum testosterone showed that GnRH antagonist treatment was effective at reducing the ITT levels in Utp14bjsd-B6 mice (Fig. 6D), while androgen ablation was weakly effective at stimulating spermatogonial differentiation in this strain. Serum testosterone was suppressed to similar levels in the mice on all three genetic backgrounds (Fig. 6E), although they differed with respect to the degree of induced recovery of spermatogenesis (Fig. 6, A and B).
Further evaluation of the effects of testosterone suppression indicated that in addition to inhibiting spermatogonial differentiation from type A to type B, testosterone also inhibited the differentiation of type B spermatogonia to spermatocytes. When Utp14bjsd-C3H mice were treated with GnRH antagonist alone from 8–12 wk of age, thereby reducing the serum testosterone and ITT levels to 0.3 ng/ml and 8 ng/g of testis, respectively, there was a significant increase in the number of differentiating spermatogonia but not in the number of spermatocytes (Fig. 7). However, further suppression of the residual androgen activity with flutamide increased the number of spermatocytes from 8 to 35 per 100 Sertoli cells but did not produce a significant increase in the number of spermatogonia. Thus, total androgen ablation is required to reverse in a significant manner the inhibition of differentiation of type B spermatogonia to spermatocytes in Utp14bjsd mice using a 4-wk treatment regimen.
|
Since the reduced testicular temperature of scrotal testes is also a major factor in the blockade of spermatogonial differentiation in Utp14bjsd mutant mice [9], we tested whether the variations in the severity of the spermatogenic defect in mice on different genetic backgrounds were related to differences in the testicular temperatures. The testicular temperatures of the 8- to 10-wk-old +/Utp14bjsd mice on the B6, C3H, and HB129 backgrounds at 40 sec (standard deviation = 10 sec) after anesthesia were 34.5 ± 0.3°C, 34.3 ± 0.2°C, and 34.3 ± 0.4°C, respectively. The corresponding peritoneal temperatures of the B6, C3H, and HB129 mice were similar at 37.3 ± 0.2°C, 37.5 ± 0.2°C, and 37.7 ± 0.2°C, respectively.
Stimulation of Spermatogenic Recovery by Temperature Elevation
Cryptorchidism negatively affects the later stages of normal spermatogenesis in normal mice. The testis weights of the heterozygous (+/Utp14bjsd) mice on the C3H, HB129, and BH backgrounds were markedly reduced to a similar extent (to 45% of that of the control) by cryptorchidization; testis weight was reduced to 36% of that of the control in mice on the B6 background, although this value was not significantly different from the degrees of reduction in the other strains (Fig. 8B). However, the extent of tubule differentiation in the cryptorchidized testes of the heterozygotes differed between the various genetic backgrounds. The cryptorchidized heterozygotes on the B6 background showed a significant number of tubules in which differentiated cells were absent, whereas the TDI was essentially 100% in the other strains (Fig. 8A). In addition, there was a reduced ability to produce spermatids in mice on the HB129 background, as spermatids were observed in only 2% of the tubules scored, compared to at least 35% in the other genetic backgrounds analyzed.
|
There were also significant interstrain differences in the enhancement of spermatogonial differentiation in Utp14bjsd mutant mice by cryptorchidization (Fig. 8A). In mice on the B6 background, only 38% of the tubules from the cryptorchid testes showed differentiation. In contrast, in Utp14bjsd mice on the BH-hybrid [9], C3H, and HB129 backgrounds, unilateral cryptorchidization stimulated differentiation in 100% of the tubules of the Utp14bjsd mutants. However, there were quantitative differences among the later three strains; in the Utp14bjsd mutant mice on the C3H and BH-hybrid backgrounds, the numbers of spermatocytes were higher and the appearance of spermatids was more frequent than in the Utp14bjsd mutant mice on the HB129 background. The increase in testis weights noted following cryptorchidization of the Utp14bjsd mice on different genetic backgrounds (Fig. 8B) reflected this extent of recovery of spermatogenesis. Cryptorchidization increased testis weights about 1.5-fold in the Utp14bjsd mice on the C3H and BH-hybrid backgrounds, whereas there were no significant changes in the testis weights of the mice on the B6 and HB129 backgrounds.
Effects of Hormone Suppression on Scrotal Temperature
Since both the suppression of testosterone and increased testicular temperature stimulated spermatogenic recovery in Utp14bjsd mice, we investigated whether testosterone suppression acts indirectly by increasing the testicular temperature. Thus, 8-wk-old Utp14bjsd mice (HB129 background) were treated with GnRH antagonist and flutamide, to produce androgen ablation. Scrotal temperatures measured during the fourth week of androgen ablation on average 22 sec (standard deviation = 11 sec) after anesthesia indicated that hormone suppression induced a highly significant (P < 0.001) increase in scrotal temperature, from 33.7 ± 0.2°C in sham-treated mice to 35.2 ± 0.2°C in the androgen-ablated mice. We have not yet assessed whether interstrain differences exist for the elevation of temperature with hormone suppression.
The results presented in the present study show that the age-dependent development of the blocks in spermatogonial and spermatocyte differentiation in Utp14bjsd mice [1] occurs in all the genetic backgrounds tested. The similar numbers of undifferentiated type A spermatogonia in Utp14bjsd mice at different ages and on different genetic backgrounds in conjunction with the varying numbers of differentiating spermatogonia (Fig. 3) prove that the age and strain differences in the effect of the Utp14bjsd mutation result from quantitative differences in the yield of differentiating cells from the undifferentiated spermatogonia.
Despite the block, there was some differentiation to the intermediate and B spermatogonial stages at 12 wk of age in all the genetic backgrounds, especially in C3H mice, in which a few spermatocytes were also formed. Thus, there is not just a single wave of spermatogenesis in Utp14bjsd mice, as indicated by some authors [1, 2]. Even after the animals reach adulthood, spermatogonial differentiation proceeds to a limited extent, but with increasing age, the cells progressively complete less and less of the differentiation pathway before undergoing apoptotic death. Thus, the failure of spermatogonial differentiation in Utp14bjsd mice is not a unique effect due to the known differences between the first wave of spermatogenesis and subsequent waves [20]. Rather, it is a progressive effect that becomes more severe with progression through adulthood. Although the major stage at which spermatogonia are lost is the Aal to A1 transition [11], the blockage of differentiation at this stage was not absolute but was part of a continuous decline at later stages. We identified A4 spermatogonia in Utp14bjsd mice on the B6 background even at 20 wk of age, whereas a previous report has shown that at 3–4 mo of age, all the spermatogonia in Utp14bjsd-B6 mice are negative for c-kit [5], which indicates that none of these cells progresses to the A1 stage or beyond. This difference may be attributable to genetic divergence of the background strain or differences in housing and nutrition.
Abnormalities in the clonal size and presence of multinucleate type A spermatogonia in whole-mounted tubules from the testes of Utp14bjsd mutants have been observed previously [11]. In the present report, we describe for the first time how many of the spermatogonia in Utp14bjsd mice have specific morphological abnormalities in their nuclei. The most prominent abnormality was the presence of large or multiple nuclear vacuoles in spermatogonia that often had the other nuclear characteristics of differentiated spermatogonia. Subtle nuclear vacuoles normally appear in some undifferentiated type A spermatogonia, but never in normal differentiated spermatogonia, which have reticulated nucleoli and increased heterochromatin [14]. We were unable to determine with certainty whether these abnormal cells were undifferentiated type A spermatogonia with abnormal nucleoli or chromatin patterns or whether they were differentiating spermatogonia with vacuoles. The reduced numbers of normal undifferentiated type A spermatogonia in Utp14bjsd mice compared with wild-type mice indicated that some of these cells already had abnormalities. Nevertheless, the reduction in cell numbers caused by the Utp14bjsd mutation was lower among undifferentiated spermatogonia than among the later cell types, and stem cell function and self-renewal were retained, as demonstrated by the ability to restore spermatogenesis in 96% of the tubules of 1-year old Utp14bjsd mice by elevating the testicular temperature [9].
Abnormalities in the nucleoli or the presence of vacuoles may be directly related to the molecular basis of the Utp14bjsd mutation. Utp14b is a mouse homolog of a yeast gene that is necessary for 18S rRNA processing [21], and 18S rRNA processing is defective in Utp14bjsd mutant germ cells (Zhao and Meistrich, unpublished results). In some yeast mutants, unprocessed 18S rRNA is retained in the nucleolus [22], which can lead to the observation of abnormal nucleoli (Fig. 1, J and L). The nuclear vacuoles may represent accumulations of unprocessed 18S rRNA in a highly dispersed form in these cells or they could be a secondary consequence and indicative of an early step in the death of these cells. These morphological changes must be further characterized at the ultrastructural and molecular levels in order to test these hypotheses.
The major component of the genetic effect is the Y chromosome, with the B6 and 129 Y chromosomes conferring greater sensitivity to the blocking of spermatogonial differentiation in Utp14bjsd mice than does the Y chromosome of C3H mice. The Y chromosomes from all three strains are derived from Mus m. musculus, as opposed to the more divergent subspecies Mus m. domesticus [23]. However the B6 and 129 Y chromosomes are closely related to each other in DNA sequence, while the C3H Y chromosome is more distant in evolutionary terms [24]. Although the Y chromosome contains a limited number of genes, many of these genes are involved in testicular development and spermatogenesis [25]. The lack of recombination on the Y chromosome hinders the identification of the gene responsible for this genetic trait. However, if good candidate genes could be selected, we might selectively mutate them by gene targeting [26] and study their contributions to the juvenile spermatogonial depletion phenotype. The Sry gene does not appear to be a good candidate, since it is transcribed in the testis as a circular form that is unlikely to be translated [27].
Since the suppression of testosterone induces spermatogonial differentiation in Utp14bjsd mutant mice [7, 8], we tested the hypotheses that basal and GnRH antagonist-suppressed testosterone levels are also responsible for the interstrain differences in spermatogonial differentiation. Neither in the comparisons between the three genetic backgrounds (Fig. 6) nor in the genetic crosses (Fig. 5) were there any correlations between low serum testosterone or ITT levels and enhanced spermatogonial differentiation. Furthermore, there was no consistent correlation between the enhancement of spermatogonial differentiation caused by testosterone suppression and the decreased testosterone levels in the different strains. Thus, the interstrain differences in the development and reversal of the block are not related to the endogenous testosterone levels or the ability to suppress testosterone. However, since we have demonstrated that suppression of testosterone increases intratesticular temperature, it is possible that the interstrain differences in response to testosterone suppression are a result of interstrain differences in either the elevation of testicular temperature or the effects of elevated temperatures on spermatogenesis in Utp14bjsd mutants.
Since we found no significant differences in the scrotal or body temperatures of the mice on the different genetic backgrounds, these factors cannot be responsible for the differences in the decline of spermatogenesis. We suggest that variations in the reversal of the spermatogenic defect in the Utp14bjsd mice of different strains are determined by the strain-specific responses of the mutant germ cells to elevated temperatures. These responses must comprise a combination of the detrimental effects of elevated temperature on spermatogenesis and the beneficial effect on Utp14bjsd mutant germ cells. Our results indicate that spermatogonial differentiation in B6 mice is more sensitive to elevated temperature than spermatogonial differentiation in other strains. It should be noted that we did not observe a complete blockade of spermatogonial differentiation in cryptorchid B6 mice, as observed by others [11]. This is likely the result of the ‘low' cryptorchidism procedure that we used, which appears to be analogous to the type 1 cryptorchidism procedure described previously, which does not result in such a severe depletion of germ cells [28]. The greater sensitivity to elevated temperatures of spermatogenesis in B6 mice compared to other strains, including C3H, has been demonstrated previously using the more standard (type 2) cryptorchidism procedure [28], in which the testes are higher in the abdominal cavity [29]. This detrimental effect of elevated temperature accounts in part for the lower stimulation of spermatogonial differentiation in cryptorchid B6 mice. The HB129 hybrids also showed greater spermatogenesis sensitivity to elevated temperatures, as indicated by the reduced numbers of spermatocytes and spermatids in cryptorchid normal and Utp14bjsd mutant testes, as compared to the C3H mice or BH hybrids. This may explain the failure to observe an increase in testis weight in the cryptorchid Utp14bjsd-HB129 mutant mice.
Therefore, the degree of enhancement by cryptorchidism of spermatogenesis in Utp14bjsd mutant mice on different genetic backgrounds can be explained in part by the resistance of normal spermatogenesis in these strains to the abdominal temperature. In addition, the correlation between the strain-specific resistance and the age-related decline in spermatogenesis in Utp14bjsd mutants (Figs. 2, 3, and 8A), despite similar scrotal temperatures, and the ability to stimulate spermatogenic recovery by increasing the temperature (Fig. 8, A and B) indicate a strain-specific difference in the effects of temperature on Utp14bjsd mutant germ cells and that the mutant cells on the C3H and BH backgrounds function better than those on the B6 and HB129 backgrounds, even at scrotal temperatures. The failure of spermatogonial differentiation maintenance in Utp14bjsd-B6 mice treated with GnRH antagonist plus flutamide may be a consequence of the detrimental effects of elevated testicular temperature and the reduced beneficial effects of elevated temperatures on spermatogenic cells on this genetic background.
ACKNOWLEDGMENTS
We thank Jun Ju for genotyping the mice, Kuriakose Abraham, Kenneth Dunner Jr, and Adriano Moreira Ferreira for assistance with the preparation of histological samples, Walter Pagel for editorial assistance, Drs. Eva Eicher and Jan Rohozinski for discussions about the Y chromosome studies, and Drs. Richard Blye and Sheri Hild for supplying acyline.
FOOTNOTES
3Current address: Department of Medicine, Medical College of Georgia, Augusta, GA 30912. 
1Support provided by the National Institutes of Health (R01 HD-40397, Cancer Center Support Grant CA-16672, and Training Grant T32 HD-07324), Fundacao de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG; CBB-1815/05), and a Research Fellowship to H.C-G. from the Brazilian Research Council (CNPq 307172/2003-1).
Correspondence: 2Marvin L. Meistrich, University of Texas M. D. Anderson Cancer Center, Department of Experimental Radiation Oncology, 1515 Holcombe Boulevard, Box 66, Houston, TX 77030. FAX: 713 794 5369; e-mail: meistrich{at}mdanderson.org
Received: 12 January 2007.
First decision: 11 February 2007.
Accepted: 1 May 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  G. Shetty, S. H. Shao, and C. C. Y. Weng p53-Dependent Apoptosis in the Inhibition of Spermatogonial Differentiation in Juvenile Spermatogonial Depletion (Utp14bjsd) Mice Endocrinology, June 1, 2008; 149(6): 2773 - 2781. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
), and the BH and B6 backgrounds (
) for Utp14bjsd mice of the same age (or within 1 wk of each other). A significant decrease is noted (
) between the value for the given time-point and the value for the preceding time-point for the same background strain. For the mice, n = 4–8, except for HB129 at 9 wk (n = 3), BH at 9 wk (n = 2), and BH background at 25 wk (n = 3).
