Neurofibromin haploinsufficiency results in altered spermatogenesis in a mouse model of neurofibromatosis type 1

The fertility of men with neurofibromatosis 1 (NF1) is reduced. Despite this observation, gonadal function has not been examined in patients with NF1. In order to assess the role of reduced neurofibromin in the testes, we examined testicular morphology and function in an Nf1+/- mouse model. We found that although Nf1+/- male mice are able to reproduce, they have significantly fewer pups per litter than Nf1+/+ control males. Reduced fertility in Nf1+/- male mice is associated with disorganization of the seminiferous epithelium, with exfoliation of germ cells and immature spermatids into the tubule lumen. Morphometric analysis shows that these alterations are associated with decreased Leydig cell numbers and increased spermatid cell numbers. We hypothesized that hyper-activation of Ras in Nf1+/- males affects ectoplasmic specialization, a Sertoli-spermatid adherens junction involved in spermiation. Consistent with this idea, we found increased expression of phosphorylated ERK, a downstream effector of Ras that has been shown to alter ectoplasmic specialization, in Nf1+/- males in comparison to control Nf1+/+ littermates. These data demonstrate that neurofibromin haploinsufficiency impairs spermatogenesis and fertility in a mouse model of NF1.


Introduction
Neurofibromatosis 1 (NF1) is a multisystem autosomal dominant condition that affects 1 in 3000 individuals. It is caused by germ-line mutations in the neurofibromin (NF1) gene, which is located on chromosome 17. There is significant variation in the clinical phenotype among affected individuals, but the most common manifestations are café-au-lait macules, iris Lisch nodules, multiple neurofibromas, axillary and inguinal freckling, skeletal abnormalities and learning disabilities [1]. Individuals with NF1 have fewer children, on average, than individuals in the general population, and this reduction in fertility is greater in males than females [2][3][4][5][6]. This reduction in reproductive fitness has been attributed to social rather than biological PLOS

Experimental animals
The NF1 mouse model used in this study has been described elsewhere [29]. Dsk9 is a missense mutation in the GTPase-activating protein related domain of neurofibromin. Nf1 Dsk9/+ (subsequently referred to as "Nf1 +/-") mice were maintained on a mixed (C3HeB/FeJ x C57bl/6) background. All surgical and animal care procedures were approved by the University of British Columbia Committee on Ethics of Animal Experiments (IACUC Approval # A11-0117), and were conducted in accordance to the University's Guidelines for Animal Experiments (https://animalcare.ubc.ca/animal-care-committee/policies-and-guidelines). For the breeding studies, male or female Nf1 +/mice were bred with Nf1 +/+ control littermates. For breeding studies, we used three haremed cages in which female Nf1 +/mice were crossed to male Nf1 +/+ control littermates and three haremed cages in which male Nf1 +/mice were crossed to female Nf1 +/+ control littermate. We also examined breeding records from additional six cages using the same crosses that were not haremed.

Tissue isolation procedure
Nf1 +/and Nf1 +/+ control littermate mice were weighed prior to surgical procedures. Animals were sacrificed using isoflurane followed by cervical dislocation at 6 to 7 months of age. The testes and epididymis were isolated from the lower abdominal cavity and placed in ice-cold HEPES-PSS buffer (PH 7.4) containing 10mM HEPES, 6mM glucose, 1.8mM CaCl, 130mM NaCl, 4mM KCl, 4mM NaHCO 3 , 1.2mM MgSO 4 , 1.2mM KH 2 PO 4 , and 0.03mM EDTA. The testes were cleaned of fat, weighed, and either fixed in 10% formalin or flash-frozen in liquid nitrogen and stored at -80˚C for histological staining and Western blotting, respectively. Within 30 minutes of isolation, the cauda epididymis, where mature sperm are stored, was cleaned of fat and dissected from the corpus epididymis and the vas deferens. The cauda epididymis was minced and placed into 500 μL of HEPES-PSS buffer for 30 minutes in order to release spermatozoa into the buffer. Spermatozoa were centrifuged at 15000 rpm for 1 minute and re-suspended in 500 μL of HEPES-PSS buffer. Sperm were counted in a hemocytometer using 10 μL of the sperm suspension. A minimum of 100 sperms per chamber were counted to reduce statistical dispersion.

Histology
Following fixation in 10% formalin, testis samples were dehydrated and embedded in paraffin. Five-micrometer transverse sections were stained using standard hematoxylin and eosin or Masson trichrome staining protocols (Sigma Aldrich, ON, Canada). Sections were viewed and imaged using an Olympus BX61 upright light microscope and Q imaging Retiga Exi camera with InVivo 3.2.0 software (Media Cybernetics). Computerized quantitative morphometric analysis was performed using Image-Pro Plus software (Media Cybernetics).

Testicular morphometric analysis
Testicular volume was measured using the formula for a prolate (elongated) ellipsoid ( 4 3 pab 2 ) [30], where "a" is the major axis (polar diameter) and "b" is the minor axis (equatorial radius). The width and equatorial length of the testes were measured on testicular histological images using Image J1.45s, version 1.44p (NIH, Bethesda, MD). We found that testicular volume was significantly correlated with mouse size (R2 = 0.5, p = 0.02). Therefore, we divided testicular volume by the body weight of each mouse to yield volume per unit weight to normalize for mouse size.
Seminiferous tubule size was measured at a final magnification of 200X using Image-Pro Plus software (Media Cybernetics, Inc., Rockville, MD). To ensure comprehensive coverage of the testes, five rectangles of equal dimensions were randomly placed on the image. Within each of five grids placed on the histology image, 400 equidistant points were counted by manually tagging the cell type on which the point fell. A total of 2000 points were counted in each testis. If the point fell on white space within the seminiferous tubule, it was characterized as 'intratubular space', and if the point fell on white space outside the tubules, it was characterized as 'interstitial space'. The lumen of the tubules was included as part of the spermatid count. Distinction was not made between a round or elongating spermatid. Germ cell count included both spermatogonia and spermatocytes. Sertoli cells, germ cells, and round and elongated spermatids in the seminiferous tubules were expressed as a percentage of the total cell population of the tubule. The cross section of the tubule in which cells were enumerated also included intracellular space. The density and distribution of Sertoli cells, germ cells, spermatids and Leydig cells within the testis were measured on a complete section at a final magnification of 200X using Image-Pro Plus software. Within each of the five rectangles, five tubules were randomly selected and the diameter of each axis (i.e. the lengths and widths) were measured. The five rectangles covered over 70% of the section and each rectangle had between 15-25 tubules within it so five tubules covered at least 20% of the tubules present. Tubules with a length-to-width ratio less than or equal to 1.5 were considered to be round or nearly round and were included in the analysis of seminiferous tubule size. At least 25 tubules were measured per testis (50 tubules per mouse) for both Nf1 +/+ control and Nf1 +/mice.

Western blotting
Flash-frozen isolated testes from Nf1 +/+ and Nf1 +/mice were homogenized in a pre-chilled stainless-steel mortar and pestle. The resulting tissue powder was mixed in 50 μL of ice-cold lysis buffer containing 50mM pyrophosphate, 50mM NaF, 50mM NaCl, 5mM EDTA, 5mM EGTA, 100μM Na 3 VO 4 , 10mM HEPES (pH 7.4), 0.1% Triton X-100, 10 μg/mL leupeptin, and 1mM phenylmethylsulfonyl fluoride. Extracted protein (40 μg) was fractionated by gel electrophoresis in 9% sodium dodecyl sulfate-polyacrylamide gels, transferred to nitrocellulose membranes, and then blocked for 1 hour with PBS containing 5% skim milk and 0.2% Tween-20. Following overnight incubation (at 4˚C) with specific primary antibody, the membrane was incubated with secondary antibody for 1 hour at room temperature. Immunoblots were visualized with an enhanced chemiluminescence detection system following the protocol provided by the manufacturer (Pierce Biotechnology, Rockford, IL, USA).

Statistical analysis
Data are reported as means ± SEM from at least five mice. Differences between Nf1 +/+ and Nf1 +/groups were analyzed by 2-tailed Student's t tests and were graphed using GraphPad 5 Prism software (San Diego, Calif., USA). Statistical significance was defined as p � 0.05.

Effects of Nf1 mutation on mouse litter size and sperm count
To assess the fertility of Nf1 +/male mice, we determined the number of pups produced when Nf1 +/males were crossed to Nf1 +/+ control females, and Nf1 +/+ control males were crossed to Nf1 +/females. As shown in Table 1, the average number of pups per litter from Nf1 +/males was significantly reduced compared to that of Nf1 +/+ control male littermates (p = 0.007); however, the amount of time between litters was not significantly different (p = 0.75). We saw no evidence of post-natal death or other indications that Nf1 +/males were more aggressive with pups than Nf1 +/+ control males. Epididymal sperm counts in Nf1 +/mice were significantly higher than those of Nf1 +/+ control mice (p = 0.02) ( Table 1).

Histological changes in seminiferous tubules
To determine whether these changes in fertility were associated with morphological changes in the testes, we performed histological examination of the testis in Nf1 +/males compared to Nf1 +/+ control males. We found varying amounts of mild to moderate degeneration of seminiferous tubules in all Nf1 +/males. Mild alterations include detachment of the germinal epithelium from the basal lamina, vacuolation, and residual bodies (Fig 1A). Moderate alterations include disorganization of the seminiferous epithelium, with exfoliation of germ cells and immature spermatids into the lumen (Fig 1B), suggesting that spermiation may be affected.

Effects of Nf1 mutation on spermatid maturation
Further suggesting an important role for neurofibromin during spermiation, we also observed abnormal spermatids with enlarged heads and atypical residual bodies in Nf1 +/mice, which appear to be associated with the release of immature spermatids into the lumen (Fig 2).
Progressive differentiation and migration of spermatocytes is required for the production of mature spermatozoa. To further examine this process, we quantified the number of Leydig cells, Sertoli cells, spermatocytes, and spermatids in the seminiferous epithelium of Nf1 +/and Nf1 +/+ control mice ( Table 2).
Leydig cells secrete testosterone and cytokines required for spermatogenesis. The number of Leydig cells present in Nf1 +/mice was significantly lower than in Nf1 +/+ control littermates (p = 0.02) ( Table 2). This reduction did not result in a significant increase in the interstitial space in Nf1 +/mice and was not associated with any difference in the number of Sertoli cells. The total number of germ cells also did not differ between Nf1 +/and Nf1 +/+ control mice, although the number of immature (round or elongated) spermatids was significantly increased in Nf1 +/mice compared to Nf1 +/+ control (p = 0.03) ( Table 2). Testicular volume was also significantly increased among Nf1 +/mice (p = 0.05) ( Table 2). This difference does not appear to be a result of alterations in the size of the seminiferous tubules, as we did not observe a significant difference in the seminiferous tubule area between Nf1 +/and Nf1 +/+ control mice ( Table 2).

Effects of the Nf1 mutation on signaling pathways within seminiferous epithelium
As loss of neurofibromin may cause hyperactivation of the RAS/MAPK/ERK pathway [16,19,28], which is involved in regulation of seminiferous epithelium [20][21][22][23]31]. We examined the effects of neurofibromin deficiency on expression and activation of the signaling components of the phosphoinositide 3-kinase (PI3K) and ERK pathway in the testes, and showed that Nf1 mutation resulted in increased phosphorylation and activation of ERK, while having no effect on total ERK protein expression (Fig 3). In addition, we observed a significant increase in total PI3K expression in Nf1 +/mice when compared to Nf1 +/+ control mice (Fig 3).

Discussion
Despite reproductive fitness being reduced in individuals with NF1 [2][3][4]6], gonadal function has not been well characterized. Delayed puberty [9] and early menopause [11] suggest that reproductive function may be altered in people with NF1.
The present work is the first time that testicular morphology and function have been examined in Nf1 +/mice. Interestingly, we observed a 50% reduction in the average litter size from Nf1 +/males crossed to Nf1 +/+ control females when compared to Nf1 +/+ control males crossed to Nf1 +/females. This suggests fertility may be reduced in Nf1 +/males, but we cannot rule out the possibility that the effect observed is due to an increase in the fertility of Nf1 +/females when paired to Nf1 +/+ control males. However, as described below, the reduced litter size in these crosses was associated with changes in spermatogenesis and sperm quality in Nf1 +/males, supporting a role for Nf1 in male fertility.
Changes in spermatogenesis and/or sperm quality may be associated with alterations in the seminiferous epithelium, which are known to affect fertility rates [32,33]. In models of testicular toxicity, the disorganized seminiferous epithelium appears to result from alterations in Sertoli-spermatocyte and Sertoli-spermatid junctions [14,17,34,35]. Spermatogenesis depends on the maintenance and restructuring of cellular junctions between Sertoli cells, spermatocytes, and spermatids [36]. In the testis, cellular junctions regulate the permeability of amino acids, hormones, and growth factors, and create an immune barrier essential for prevention of antigen responses [33].  Coexisting tight junctions and a special adherens junction, known as the basal ectoplasmic specialization, form the blood-testis barrier [36]. Spermatocyte differentiation depends on the migration of immature spermatocytes across the blood-testis barrier and progressive migration toward the seminiferous tubule lumen. During this process, Sertoli-Sertoli, Sertoli-spermatocyte, and Sertoli-spermatid cell junctions are rapidly remodeled to enable differentiation and migration of spermatocytes and to maintain the integrity of the blood-testis barrier [21]. Alterations in Sertoli-spermatocyte junctions cause exfoliation of germ cells [14], while exfoliation of immature spermatids into the seminiferous tubule lumen suggests that Sertoli-spermatid junctions are affected as round immature spermatids would normally associate with Sertoli cells via the ectoplasmic specialization [37]. We observed exfoliation of spermatocytes and round and elongating spermatids into the lumen, suggesting that neurofibromin haploinsufficiency may affect junctional dynamics in the testis.
Further contributing to the exfoliation of immature round and elongated spermatids in Nf1 +/mice, we also found that the number of Leydig cells was also reduced in Nf1 +/mice. Leydig cells produce testosterone which reinforces the tight junctions in the blood-testis barrier [15,21,38] and regulates ES assembly and disassembly [14,16,37]. Testosterone withdrawal promotes stage specific-detachment of round spermatids from the rat seminiferous epithelium through an ERKdependent mechanism [39]. Nf1 +/mice also demonstrated enlarged sperm heads and atypical residual bodies, which further supports a role for neurofibromin in junctional dynamics. Alterations in spermatid morphology usually result from disassembly of tubulobulbar complexes, a specialized junction between Sertoli cells and elongated spermatids [40].
ERK phosphorylation is associated with the basal ectoplasmic specialization, a Sertoli-spermatocyte adherens junction that regulates spermatogenesis, and the apical ectoplasmic specialization, a Sertoli-spermatid adherens junction involved in spermiation [15,32]. Phosphorylation of ERK results in decreased levels of the adhesion protein complexes, cadherin/catenin and nectin/ afadin, resulting in loss of ectoplasmic specialization adhesion [14,25]. Increased phosphorylation of ERK because of neurofibromin haploinsufficiency is, therefore, likely to affect proliferation of immature spermatogonia and affect ES adhesion. In support of this hypothesis, we found that in Nf1 +/mice, there was an abnormal release of immature round spermatids from the seminiferous epithelium and an increase in spermatids were associated with a significant increase in ERK phosphorylation without affecting total ERK protein expression.
Though alterations in spermiation may result in lower sperm counts, we found that caudal epididymal sperm counts in Nf1 +/mice were significantly increased. It is conceivable that the hyperactivation of Ras affects survival of spermatids and contributes to a higher sperm count yet lower quality sperm. Activation of ERK is associated with reduction in sperm quality [21] and alterations in sperm quality are known to reduce the number of offspring produced [41,42]. The presence of abnormal spermatid morphology such as atypical residual bodies in the spermatids of Nf1 +/mice further supports the hypothesis that Nf1 +/mice produce lower quality sperm. Reduced reproductive fitness with an increased sperm count is not frequently observed, but has been demonstrated in progesterone receptor knockout mice (PRKO) with a similar phenotype to Nf1 +/male mice [43,44]. Similar to Nf1 +/mice, PRKO mice demonstrate delayed spermatogenesis [43] with alterations in Leydig cells [44]. While sperm quality was not directly assessed in this model, in humans reduced progesterone receptor expression is known to affect sperm quality [45].
Neurofibromin is a negative regulator of Ras, and neurofibromin haploinsufficiency leads to activation of the Ras pathway [1,19,28]. Activation of Ras is known to trigger phosphorylation of ERK, which promotes spermatogonial stem cell proliferation in mice [26,27] and is involved in differentiation of spermatogonia through regulation of adherens junctions. We speculate that the hyperactivation of ERK may also contribute to abnormal loss of ectoplasmic specialization adhesion during spermiation. Further highlighting the importance of ERK in spermatogenesis, MAP3K, the human ortholog to ERK1, is highly expressed in Leydig cells and moderately expressed throughout the seminiferous tubules in adult males (see https:// www.proteinatlas.org/ENSG00000102882-MAPK3/tissue/testis#img).
Together, the results from this study demonstrate that Nf1 +/mice have alterations in the seminiferous epithelium, and suggests that neurofibromin is involved in proliferation of the seminiferous epithelium, as well as regulation of the junctions vital to spermatogenesis. The precise mechanism of the alterations observed here is unclear, but Nf1 +/mice do appear to demonstrate defects in spermatogenesis. Thus, the observed alterations in the seminiferous epithelium of Nf1 +/mice may be related to reduction in fertility, but further investigation is needed to establish the causal relationship.
At this point, we do not know if similar alterations are present in patients with NF1, testosterone levels and spermatogenesis are rarely examined unless a symptomatic testicular abnormality is also present [46][47][48]. However, fertility is reduced [3], puberty is delayed, and osteoporosis appears to occur more frequently than expected in men with NF1 [11] suggesting defects in gonadal function or endocrine regulation.
Although there are discrepancies in spermatogenesis between human and mice, in light of the facts and findings detailed in this study, investigation of testicular function in patients with NF1 may be warranted. These studies should explore sperm counts, testosterone levels, and morphometric measurements of the testes as well as testicular pathology. With this knowledge, patients can be counselled to make decisions regarding their reproductive potential. Examination of ovarian structure and function in female Nf1 +/mice is also worth pursuing, as we are unaware of any studies that have examined it.