Loss of ASPM Disrupts Female Folliculogenesis in Mice

Background: The abnormal spindle-like, microcephaly-associated (ASPM) gene is a causative gene of autosomal recessive primary microcephaly (MCPH) 5 in humans, which is characterized by a reduction in brain volume. It was previously reported that truncated ASPM proteins in transgenic mice caused major defects in the germline, a severe reduction in ovary weight and the number of follicles accompanied by reduced fertility. However; it remains unknown whether a loss of ASPM induces abnormal ovarian function, resulting in female infertility. Methods: In order to assess the ovary function, we examined vaginal smear cytology from the age of 7 weeks to 100 weeks in CAG-mediated Cre-loxP conditional Aspm -/- knockout mice and control female mice. In addition, we evaluated the ovarian size, brosis ratio and the number of follicles (primordial, primary, secondary, antral and atretic follicles) in mice from 15 weeks to 100 weeks old by image analyses. Mann-Whitney U-test was used for statistical analysis. Results: The size of the ovary was signicantly reduced in Aspm knockout mice at 15-20 weeks, 40-50 weeks and 70-80 weeks old. compared with the control mice. Furthermore, at all stages, we found a severe decrease in the number of developing follicles at 10-15 weeks, 40-50 weeks and 70-80 weeks old, accompanied by disrupted cyclic changes of vaginal cytology. Conclusion: The results showing that folliculogenesis was signicantly decreased and associated with abnormal vaginal cytology in Aspm knockout female ovaries suggested that ASPM might play an important role in the folliculogenesis and estrous cyclicity of the postnatal ovary.


Introduction
The abnormal spindle-like, microcephaly-associated (ASPM) gene is a causative gene of autosomal recessive primary microcephaly (MCPH) 5 in humans, which is characterized by a reduction of brain volume, leading to intellectual disability and varied psychomotor disorders [1]. The pathogenetic mechanisms of microcephaly have been intensively investigated in animal models, including gyrencephalic models. The results of these studies have suggested that ASPM regulates cortical surface expansion by controlling the a nity of ventricular radial glial cells, an undifferentiated cell type found in the ventricular surface, thus increasing the ratio of the undifferentiated cells to more differentiated progenitors in the cortical units [2]. ASPM functionally acts in the symmetric division of progenitor cells by maintaining the stemness, which is essential for regulating appropriate organ size and structure.
In our previous study, we reported that a loss of ASPM induced testicular hypoplasia, composed of a decreased number of seminiferous tubules accompanied by a decrease in the number of spermatogonia, spermatocytes, spermatid and sperm, and vacuolation of Sertoli cells [3]. It has been reported that truncated ASPM proteins in transgenic mice cause major defects in the germline, not only testicular hypoplasia but also a severe reduction in ovary weight and the number of follicles, accompanied by reduced fertility [4]. However; the precise function of the ovary, including cyclic changes in the number and stages of follicles, as well as temporal changes in ovarian size and histology according to aging have not yet been examined in the absence of ASPM.
In the present study, we investigated the function and morphology of the ovary in Aspm knockout (KO) female mice, and found a severe decrease in the number of developing follicles at all stages in young adult mice, accompanied by disrupted cyclic changes in the vaginal cytology. These results suggested that ASPM might play an important role in germ cell differentiation and maturation during the development of the murine ovary.

Animals
In this study, we used CAG-mediated Cre-loxP conditional Aspm -/-(KO) and Aspm +/+ (control: CNT) female mice, prepared as previously described [3]. Brie y, the Aspm KO mice retained exon 2-3 of the Aspm gene, encompassed by a pair of loxP signals, so that cre-recombinase activity could switch the allele from the wild-type to null zygotes. Aspm +/heterozygote was obtained by crossing the mice with CAG-Cre transgenic mice. Aspm -/-(KO) offspring were also obtained as frequently as expected from the Mendelian inheritance and maintained under a 12-h light-dark cycle (on at 7 a.m.), at a controlled temperature (24-25°C) with free access to food and water. All of the animal studies were approved by the Institutional Review Board for Biomedical Research using Laboratory Animals at Kyoto Prefectural University of Medicine, and the animals were handled in accordance with the institutional guidelines and regulations.
Vaginal squamous epithelium was obtained by washing the vaginal cavity with 20 µL of PBS using pipette tips and xed with 100% ethanol and 2% polyethylene glycol for 30 minutes. Post-xed smeared cells were stained with Papanicolaou stain.
The ovaries were dissected, xed overnight with 4% paraformaldehyde and embedded in para n. Serial coronal sections were prepared at 4 µm-thickness. Following antigen retrieval with EDTA pH 8.0, the tissue was blocked with 10% goat serum, 1% BSA, and 0.01% Triton X-100 (Sigma-Aldrich, Japan) in 0.1M PBS and incubated in primary antibody diluted in 0.1% Triton X-100 and PBS at 4°C overnight (Cleaved Caspase-3(Asp175), rabbit polyclonal, 1:200, Cell Signaling TECHNOLOGY, Danvers, MA, USA). The sections were rinsed with PBS and then incubated in the secondary antibody.

Image analyses
Morphometry Four µm-thick serial para n-embedded sections were prepared and 10 sections at intervals of 40 µm were used for assessing the total number of follicles in the ovary. The entire area of the sections stained with hematoxylin-eosin or Masson-trichrome stain was scanned, captured on computer, and image acquisition was conducted (× 40 magni cation mode) with a NanoZoomer C9600 (Hamamatsu Photonics, Tokyo, Japan). One cytologist (MM) counted the number of each type of follicle, according to the criteria as follows: Primordial follicles contain immature oocytes surrounded by at, squamous granulosa cells. In the primary follicle, the granulosa cells change from a at to a cuboidal structure and the zona pellucida forms around the oocyte, separating it from the surrounding granulosa cells. In the secondary follicles, stroma-like theca cells encircle the follicle's outermost layer and differentiate to the theca externa and theca interna. The late-term secondary follicle is composed of a fully grown oocyte surrounded by a zona pellucida, granulosa cells layer, a basal lamina, a theca interna, a capillary net, and a theca externa. The antral follicles showed the formation of a uid-lled cavity, which has been called the antrum. In the atretic follicles, the granulosa cells undergo apoptosis, showing more than two pyknotic nuclei, an irregular arrangement and pulling away from the basement membrane.
Using the same digital HE images, the size and area of the ovary, speci cally the longest diameter and the major area was measured. Quanti cation of the brosis was assessed using Masson-trichrome stained sections. For the image analysis, × 10 magni ed images were exported using NDP.view 2 software (Hamamatsu Photonics, Tokyo, Japan). Fiji software was also employed (ImageJ 2.0.0-rc-61/1.5n, http:// ji.sc) for the computer-based image analysis. First of all, background correction was carried out to reduce any color/luminance variations in the image. The tissue area (region of interest, ROI) from each stained section was extracted based on luminance information using an auto-thresholding technique. An automatically calculated threshold was used to overcome non-uniform staining conditions [5].
Filamentous blue-stained areas of each specimen were extracted based on color information by autothresholding techniques. Finally, the amount of brosis in each tissue specimen was quanti ed by the following formula: brosis (%) = actual brosis area × 100 / ROI area.

Statistical analysis
All results are expressed as either mean ± SD or mean ± SEM. The Mann-Whitney U-test was used to analyze any differences between the two groups. Values of P < 0.05 were considered to be statistically signi cant.

Vaginal smear analyses
The vaginal smears were classi ed into four categories: Proestrus, Estrus, Metaestrus, or Diestrus (Fig.1). The smears showed regular cyclic changes of E, M, D and proestrus P stage with 4-5 days intervals at the age of 30-40 weeks in control females (Fig.1). On the other hand, the vaginal smears exhibited a disrupted cycle and remarkable elongation of the estrous or diestrous stages at 30-40 weeks old in KO females (Fig.1). The frequency of the abnormal estrous cycles was shown in Fig

Morphology of the ovaries
The ovaries were slightly smaller in size, with a reduced number of primordial, primary and secondary follicles at 5 weeks of age in the KO females. The ovaries of the KO females were much smaller in size, compared with those of the control females at both 15-20 weeks and 40-50 weeks old. The ovaries of the KO females exhibited a brous stromal mass composed of a very few follicles from 15 weeks to 70-80 weeks old. Although the control ovaries showed multiple cysts, atretic follicles and brous tissue at 70-80 weeks of age, a few primordial, primary and secondary follicles were still preserved. At 100 weeks of age, ovaries from both the control and the KO mice showed massive brosis. However, a few atretic follicles were found in the control mice (Fig.3).
In the control and KO groups, respectively, the longest diameters of the ovary were 1.82 ± 0.57 (mm) and 1.51 ± 0.25 (mm) at 5 weeks, 2.24 ± 0.21 (mm) and 1.38 ±0.18 (mm) at 15-20 weeks, 1.97 ± 0.24 (mm) and 1.92 ± 0.57 (mm) at 40-50 weeks, 1.69 ± 0.21 (mm) and 1.92 ± 0.94 (mm) at 70-80 weeks and 1.55 ± 0.95 (mm) and 2.20 ± 0.55 (mm) at 100 weeks of age. The longest diameters were signi cantly less in the KO group, compared with the control group at 15-20 weeks and 40-50 weeks of age. The diameters of the ovaries showed a trend for a decrease in the KO group. However, the difference did not approach the signi cance level at 5 weeks old (Fig. 4B).
The brosis ratio was analyzed at 15-20, 40-50 and 70-80 weeks old. The ratio was signi cantly increased in the KO group, compared with the control mice (Two-way ANOVA). A slight tendency was shown for an increase in the brosis ratio in the KO mice, and a decrease in the control mice during aging, but those differences did not reach the level of signi cance (Fig. 4C).
At each follicle stage, the average number of the primordial, primary, secondary, antral and atretic follicles per ovary decreased with aging in the control mice ( Table 1). The number of primordial, primary and secondary follicles showed a tendency for a reduction in the KO females, however the difference did not reach the signi cance level at 5 weeks old (Fig. 5A). The number of follicles at each stage, speci cally, primordial, primary, secondary, antral and atretic follicles, was signi cantly reduced in KO mice at 15-20 (Fig. 5B), 40-50 (Fig. 5C) and 70-80 weeks old (Fig. 5D), compared with the control mice. At 100 weeks of age, the number of atretic follicles only showed a signi cant decrease in the KO mice, compared with the control mice (Fig. 5E).
The cleaved caspase-3-immunoreactivity of the ovaries showed no signi cant difference between the Aspm KO and the control mice at 5 weeks old (Supplement Fig. 1), suggesting that apoptosis of the granulosa cells was not involved in the histological ovarian changes observed along with the loss of Aspm.

Discussion
In this study, we evaluated the function and morphology of murine ovaries lacking ASPM and found a signi cant decrease in the number of follicles and altered follicular maturation accompanied with a loss of estrous cyclicity in vaginal smears.
The abnormal spindle-like, microcephaly-associated (Aspm) gene is the mouse ortholog of human ASPM [3]. Aspm mRNA is widely expressed in organs during fetal development, including the liver, heart, lung, kidney, spleen, stomach, colon, bladder, skin, and skeletal muscle, in addition to the brain, testis and ovary [6]. The mutation of Aspm induces infertility in males and females. However, only a few studies have been conducted on infertility related with Aspm function [3,4]. It was reported that truncated ASPM proteins in transgenic mice caused major defects in the germline, a severe reduction in ovary weight and follicle numbers, accompanied by reduced fertility [4]. It has been suggested that the underlying mechanism of reduced fertility is related to the primary reduction of follicles caused by a massive loss of germ cells [4]. The Aspm gene encodes a centrosomal molecule, which is involved in spindle organization and orientation during cytokinesis [7]. ASPM is functionally active in the symmetric division of progenitor cells, which might be essential for ovarian development.
The cytological assessment of vaginal smears, composed of samples obtained in the proestrus, estrus, metaestrus and diestrus stages, showed that the ovarian function was regulated by secretary hormones such as estrogen and progesterone [23,24]. The cyclicity was relatively irregular at 7-11 weeks old in both control and Aspm KO mice, which might be due to sexual immaturity, individual differences, and no exposure to male animals. A regular cyclicity was established in the control group at 30-40 weeks of age when female mice were sexually matured for reproduction. On the contrary, the cyclicity was absolutely disrupted at 30-40 weeks of age in Aspm KO mice. Furthermore, an histological analysis con rmed that the size of the ovaries and the number of follicles were signi cantly decreased, and furthermore, that maturation of the follicles was altered even at 15-20 weeks old in Aspm KO female mice. It is suggested, therefore, that ASPM is an important molecule involved in the formation of an appropriate number of primordial follicles and in the cyclic regulation and maturation of the primordial follicles in murine ovaries. The initial decrease in the number of primary follicles was induced by a loss of ASPM, suggesting that ASPM is required for normal germ cell proliferation and differentiation during fetal ovarian development. In addition, the nding that an altered maturation of follicles was observed in Aspm KO females suggested that normal maturation of the primordial follicles might partly be regulated by ASPM after birth. As shown by cleaved caspase 3 immunoreactivity, the apoptotic cells in the granulosa cells of the antral follicles, showed no signi cant differences in Aspm KO, compared with the control females, suggesting that accelerating early cell death of granular cells might not be induced by a loss of ASPM. However; it is speculated that an increase in cell death of germ cells or any excessive loss of primordial follicles early in postnatal life might have contributed to the decrease in the number of primordial follicles shown at 5 weeks and 15-20 weeks of age in Aspm KO mice. This appears consistent with an observation in the cortical development of the present Aspm KO mice, showing increased apoptosis of stem/progenitor cells in the fetal brain (our unpublished data). Ovarian development in fetal and early postnatal stages should be further evaluated to show that loss of germ cells might be involved in the underlying mechanisms related to the decreased number of primordial follicles demonstrated in the present Aspm KO mice.
Although the size of the ovaries decreased along with normal aging, no signi cant difference was shown in the ovarian diameters between the control and the Aspm KO mice at ages older than 40-50 weeks, and likewise, no signi cant difference was shown in the ovarian areas at ages older than 70-80 weeks. The ovarian areas from Aspm KO mice were signi cantly smaller than those from control mice at the age of 15-20 weeks, and at the age of 40-50 weeks. Regarding brosis of the ovaries, we did not nd any signi cant differences between the control and the Aspm KO mice. Since it is argued that the ovaries of Aspm KO mice demonstrate hypoplasia as well as accelerating aging, the KO mice can be used as a model animal for primary ovarian insu ciency in order to develop therapeutic strategies. In Aspm KO mice, the hypothalamus pituitary axis might be affected in association with abnormal brain development. Unfortunately, we did not analyze the hypothalamus or the pituitary of the Aspm KO mice. However, they showed no problems in growth, metabolism, immune system and autonomic functions, which are regulated by growth hormone, thyroid hormone and steroid hormones through the hypothalamus pituitary axis. Although we analyzed gonadotropins including lutenizing hormone and follicle stimulating hormone in serum, we found no signi cant differences between the control and the Aspm KO mice.
In this study, we demonstrated altered function and histology of murine ovaries associated with a loss of Aspm, suggesting that Aspm might play an important role in ovary development and ovulation homeostasis, and thus, it could be a target molecule for the treatment of primary ovarian insu ciency.

Conclusions
A loss of ASPM protein was associated with a reduction in the ovarian size and the number of follicles, accompanied with disrupted cyclic changes in the vaginal cytology in adult female mice, suggesting that ASPM might play an important role in germ cell differentiation and maturation.

Declarations
Ethics approval: All of the animal studies were approved by the Institutional Review Board for Biomedical Research using Laboratory Animals at Kyoto Prefectural University of Medicine, and the animals were handled in accordance with the institutional guidelines and regulations.
Consent to participate: not applicable Consent for publication: not applicable Availability of data and material: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interest:
The authors hereby declare that there are no con icts of interest related to this study.
All of the authors agree to the publication. All of the authors agree to the order of authors.  Tables   Table 1 CNT KO Follicle type