Introduction
Previous studies (Ogura et al., Reference Ogura, Matsuda and Yanagimachi1994; Kimura & Yanagimachi, Reference Kimura and Yanagimachi1995a) have shown that mouse oocytes fused or injected with round spermatids were able to develop into normal offspring. This technique is referred to as round spermatid injection (ROSI). To date, viable offspring have been obtained after ROSI in several species such as mouse (Ogura et al., Reference Ogura, Matsuda and Yanagimachi1994), rabbit (Sofikitis et al., Reference Sofikitis, Miyagawa, Agapitos, Pasyianos, Toda, Hellstrom and Kawamura1994), rat (Hirabayashi et al., Reference Hirabayashi, Kato, Aoto, Ueda and Hochi2002) and humans (Tesarik et al., Reference Tesarik, Mendoza and Testart1995). This technique has been demonstrated to be a powerful tool for examining the developmental potential of spermatids in mutants that lack the ability to become fertilized (Romero et al., Reference Romero, Meikar, Papaioannou, Conne, Grey, Weier, Pralong, De Massy, Kaessmann, Vassalli, Kotaja and Nef2011; Sharan et al., Reference Sharan, Pyle, Coppola, Babus, Swaminathan, Benedict, Swing, Martin, Tessarollo, Evans, Flaws and Handel2004; Meng et al., Reference Meng, Akutsu, Schoene, Reifsteck, Fox, Olson, Sariola, Yanagimachi and Baetscher2002). Furthermore, it can also be used to rescue mutant lines of mice that have become infertile with age. The current authors believe that under circumstances in which other assisted reproduction techniques (ART), such as in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI), are unsuccessful, ROSI may be the last resort.
Despite the potential of ROSI as a useful ART, there are some limitations. For example, spermatogenic cell populations are frequently contaminated by somatic cells that are present in germ cell suspensions recovered after the dissection and enzymatic digestion of seminiferous tubules. To counteract this problem, various procedures have been developed for isolating purified populations of germ cells from the testis in humans and rodents, including: unit-gravity sedimentation (Wykes & Krawetz, Reference Wykes and Krawetz2003; La Salle et al., Reference La Salle, Sun and Handel2009); centrifugal elutriation (Barchi et al., Reference Barchi, Geremia, Magliozzi and Bianchi2009); separation using immuno-panning (Pelengaris & Moore, Reference Pelengaris and Moore1995); removal of contaminating somatic cells through adhesion to plastic dishes coated with lectin (Morena et al., Reference Morena, Boitani, Pesce, De Felici and Stefanini1996); and fluorescence-activated cell sorting (FACS; Lassalle et al., Reference Lassalle, Ziyyat, Testart, Finaz and Lefevre1999; Chang et al., Reference Chang, Lee-Chang, Panneerdoss, MacLean and Rao2011; Getun et al., Reference Getun, Torres and Bois2011). The use of FACS in particular is able to provide large numbers of high-purity and viable round spermatids, which are essential for the success of ROSI.
The purpose of the current study was to isolate relatively pure populations of viable round spermatids from mouse testicular cell populations by phase contrast microscopy (PCM) or FACS, and use the selected round spermatids for ROSI to ‘rescue’ (i.e. produce live-born offspring using spermatids from mutant male mice) two genetically altered mouse lines for which live-born offspring could not be derived using natural mating, IVF or ICSI. It was found that besides PCM, FACS technology could also be successfully utilized to derive live-born mice with high effectiveness and reliability, therefore ROSI should be considered a powerful addition to the armamentarium of ART applicable in the mouse, especially when the fertility of males is affected by age or genetic mutation.
Materials and methods
Reagents and media
Pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG) were purchased from the Tianjin Animal Hormone Factory (Tianjin City, China) and the Second Hormone Factory of Ningbo City (Ningbo City, China), respectively. All other chemicals were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA) unless otherwise stated. The potassium simplex optimization medium with amino acids (KSOMaa) used for culturing oocytes and embryos before and after ROSI was purchased from Specialty Medium (Phillipsburg, NJ, USA). The medium used for ROSI was HEPES-buffered CZB medium (Kimura & Yanagimachi, Reference Kimura and Yanagimachi1995a) and 10% polyvinyl alcohol (PVA; 360 kDa) in HEPES-buffered CZB medium.
Animal husbandry and mice
All experimental procedures reported here were conducted at the Department of Laboratory Animal Sciences, Shanghai Jiao Tong University School of Medicine (China). The research activity performed was approved and monitored by the Animal Care and Use Committees of Shanghai Jiao Tong University School of Medicine, and all mice were cared for following the guidelines of Use of Animals of Department of Laboratory Animal Sciences, Shanghai Jiao Tong University School of Medicine. All mice in this project were housed singly or in groups in ventilated cages within a dedicated barrier facility under specific pathogen-free (SPF) conditions.
The following strains of mice were used as spermatid donors: transgenic EGFP and Pten-null mutant (Pten−/−) mice. The EGFP male mouse had been bred repeatedly, but was unable to produce offspring due to its extreme age, resulting in a poor physiological state. This sub-infertile mouse was used to collect round spermatids as a control. The Pten−/− male was infertile due to the dysfunction of spermiogenesis, despite showing signs of libido and the detection of vaginal plug in females after overnight mating. Seven- to 10-week-old female BDF1 mice were used as oocyte donors. Eight- to 12-week-old female ICR mice were made pseudopregnant for use as foster mothers by mating with verified-sterile vasectomized male mice the day before surgical embryo transfer.
Preparation of round spermatids for PCM selection
Pten−/− and the control EGFP male mice were euthanized, and the caudae epididymides were removed using a pair of fine scissors. To collect spermatogenic cells for ROSI, the seminiferous tubules of the testes were minced as described previously (Kimura & Yanagimachi, Reference Kimura and Yanagimachi1995a), except that the cells were suspended in HEPES-buffered CZB medium.
Isolation of haploid cells from the Pten−/− male mouse by FACS
The testes from the Pten−/− male mouse were cut into pieces after removing the tunica albuginea, and the testicular fragments were immersed in phosphate-buffered saline (PBS) and rocked to flush out Sertoli cells near the lumen. Cells were then collected and stained with Hoechst 33342 stain (5 μg/ml). After 90 min incubation, cells were resuspended in ice-cold PBS with 10% fetal bovine serum (FBS) and 2 μg/ml propidium iodide to identify living versus dead cells. All solutions from the incubation step onward contained verapamil (50 μM/ml), to decrease the speed of dye extrusion. Finally, FACS was carried out using the FACS Aria II with an ultraviolet (UV) laser (BD Biosciences, San Jose, CA, USA).
Oocyte preparation
Female mice were super-ovulated by intraperitoneal injection with 5–7 IU PMSG, followed by 5–7 IU of hCG 48 h later. At 13–14 h after hCG injection, oocytes were collected from the oviduct into HCZB medium, and cumulus cells were removed by treatment with bovine testis hyaluronidase (300 U/ml; Sigma Chemical Co.) for 3–5 min. Oocytes were then washed three times in KSOMaa and were kept at 37°C in a 5% CO2 atmosphere until ROSI, which was performed within 3 h of oocyte collection.
ROSI and oocyte activation
A 1-μl aliquot of the spermatogenic cell suspension obtained as described above was mixed with approximately 10 μl of HEPES-CZB medium that contained 12% (w/v) polyvinylpyrrolidone (PVP; 360 kDa) in a micromanipulation chamber. Round spermatid injection was carried as described previously by Kimura & Yanagimachi, (Reference Kimura and Yanagimachi1995a), except that the current experiments were performed at room temperature. Briefly, a single spermatid was drawn repeatedly in and out of an injection pipette, until its plasma membrane was completely broken and the nucleus became almost completely separated from the cytoplasm. The nucleus was injected into an oocyte using an injection pipette (internal diameter 5–8 μm) with the aid of a piezo-driven micromanipulation system (Primetech, Ibaraki, Japan). The injected oocytes were left for 10 min on the stage of the microscope at room temperature (24–26°C), after which time they were transferred to KSOMaa medium at 37°C in 5% CO2 in air. Finally, the injected oocytes were activated in Ca2+-free CZB medium that contained 5 mM SrCl2 for 20 min (Bos-Mikich et al., Reference Bos-Mikich, Swann and Whittingham1995).
Embryo transfer
Pseudopregnant ICR mice were used as recipients of ROSI embryos. Specifically, blastocysts were transferred to a day 2.5 pseudopregnant mouse that had been previously mated with a vasectomized male. Six to 10 embryos were transferred into each oviduct. At day 18.5 of gestation, the offspring were born naturally or delivered by Caesarean section.
Statistics
The rates of surviving oocytes, embryonic development and pups born were analysed using a chi-squared test with SPSS software (SPSS Inc., Chicago, IL, USA). Significant differences were accepted at P-values <0.05.
Results
Enrichment of spermatogenic cells by FACS
Spermatogenic cell suspensions were collected from male mice. These cell suspensions contained spermatozoa, elongating and round spermatids and primary pachytene spermatocytes (Fig. 1A–E ).
Accurate selection of round spermatids using phase contrast microscopy (PCM) and fluorescence-activated cell sorting (FACS). Representative PCM images of mouse spermatogenic cells are shown, as selected by PCM only (A–E), or FACS (F). Mature spermatozoon (A); maturing spermatozoon (B); elongating spermatids (C); round spermatid and dividing round spermatids (D); primary pachytene spermatocyte (E); round spermatids selected by FACS (F). Scale bar = 10 μm. *: acrosome; arrowhead: dividing round spermatids; arrow: primary pachytene spermatocyte.
In the current study, attempts were made to collect round spermatids from the adult Pten−/− mouse testis by the FACS technique. The results for mouse spermatogenic cells showed the distribution histograms of Hoechst 33342-stained testicular germ cells based on DNA flow cytometry (Fig. 2). Testicular cells were fractioned into five populations (P1–P5 in Fig. 2A,C ) including predicted round spermatid population (P5). The FACS experiments revealed that about 15% of the cells were round spermatids in the total cell population of interest (Fig. 2C ).
Fluorescence-activated cell sorting (FACS)-based selection of round spermatids isolated from a Pten−/− mouse. Representative FACS data for the selection of haploid round spermatids. (A) P1, P2, P3 and P4 gates around the haploid cell population of interest in the forward scatter (FSC) versus side scatter (SSC) plots. P1 represents total haploid cell population of interest; P2 and P3 represent P1 population without adhesive cells; P4 represents P3 population without dead cells. (B) FACS histogram plots of pure round spermatids. (C) Distribution proportion of cell populations. P5 representing pure haploid cells was about 15% among all cells.
Processing by FACS was used to facilitate the collection of round spermatids. Approximately 99% of cells were identified morphologically as round spermatids in the P5 fraction (Fig. 1F ), demonstrating that round spermatids can be enriched efficiently from testicular cells by FACS based on the side scatter (SSC) and forward scatter (FSC) phenotypes.
ROSI and embryo transfer
The numbers of oocytes injected, survival rate (number of oocytes surviving injection), fertilization rate (number of embryos surviving injection that developed to zygote stages), embryonic developmental rates, total number of pregnant foster mothers and birth rate (number of pups born) are presented in Table 1. For all strains of mice studied, approximately 96.88–98.67% of injected oocytes survived the ROSI procedure. Of these cells, 98.18–100% of surviving embryos were artificially activated, 98.39–100% of which developed to 2-cell embryos, and 50.82–52.05% developed into blastocysts.
Embryonic development and reproductive success of ROSI in two mouse strains using PCM- and FACS-based selection methods
a Values with different superscripts within columns are significantly different at P < 0.05 (chi-square test).
FACS: fluorescence-activated cell sorter; PCM: phase contrast microscope.
For conventional PCM-selected ROSI-derived embryos, two out of two recipients became pregnant using round spermatids from EGFP males, and one of two for Pten−/− mice. These pregnancies yielded 17 and 11 pups from EGFP and Pten−/− male mice respectively. The birth rate was 44.74% and 35.48% for EGFP and Pten−/− mice respectively. For FACS-selected ROSI-derived embryos from Pten−/− mouse, both recipient females became pregnant and gave birth to 12 pups (42.86% birth rate). There was no significant difference in birth rate between all strains studied, or the selection method used.
Discussion
Advances in reproductive techniques have allowed infertile males to gain fertility by micro-injection of spermatozoa and pre-spermatozoal cells into oocytes (Palermo et al., Reference Palermo, Joris, Devroey and Van Steirteghem1992; Van Steirteghem et al., Reference Van Steirteghem, Nagy, Joris, Liu, Staessen, Smitz, Wisanto and Devroey1993; Ogura et al., Reference Ogura, Ogonuki, Takano and Inoue2001; Yanagimachi, Reference Yanagimachi2001). The nuclei of round spermatids are able to participate in normal embryonic development, which therefore suggests that genomic and epigenetic elements needed for embryonic development are already established in round spermatids. The present study succeeded in rescuing two strains of sub-infertile or infertile mice with the help of conventional ROSI and validated the feasibility of this approach.
The correct identification of round spermatids represents a technical difficulty that limits the widespread use of ROSI. Several authors have succinctly described the morphology and size of live mouse round spermatids (Kimura & Yanagimachi, Reference Kimura and Yanagimachi1995a,Reference Kimura and Yanagimachib; Sasagawa & Yanagimachi, Reference Sasagawa and Yanagimachi1997; Sasagawa et al., Reference Sasagawa, Kuretake, Eppig and Yanagimachi1998). They are approximately 10 μm in diameter (range, 9 –11 μm) and round in shape with a smooth outline. The round nucleus contains a centrally located nucleolus. These characteristics mean that round spermatids can be easily distinguished from other cell types in testicular biopsies. The current results show that round spermatids can be identified using PCM, and that their morphology and size are significantly different from other spermatogenic cells, such as spermatozoon, elongating spermatids, and primary pachytene spermatocytes.
In order to simplify and accurately select round spermatids, other researchers have successfully taken advantage of the polarization of round spermatid mitochondria and used MitoTracker to select live round spermatids (Sutovsky et al., Reference Sutovsky, Ramalho-Santos, Moreno, Oko, Hewitson and Schatten1999; Hikichi et al., Reference Hikichi, Kishigami, Thuan, Ohta, Mizutani, Wakayama and Wakayama2005). This staining technique has no harmful effects on embryonic development, leading to the recommendation that this approach be used for research and infertility treatment. This approach does have its limitations however, it is technically difficult for researchers and technicians with limited experience, and may result in the selection of cells that are not round spermatids. Indeed, there is limited information available regarding the morphological features of live round spermatids to allow for their accurate selection for injection into oocytes. In order to overcome this limitation, Lassalle et al. (Reference Lassalle, Ziyyat, Testart, Finaz and Lefevre1999) suggested that viable round spermatids could be isolated from testis by FACS. The experience with this technique in the current paper shows that FACS sorted cells were almost all round spermatids, and that more than 99% of sorted round spermatids remained viable. Indeed, round spermatids were easily recognizable in the enriched spermatogenic cell populations analysed by FACS. Sorted cells were homogeneous in size (approximately 11 μm) and had the cytological characteristics of round spermatids.
In the present study, Hoechst 33342 was used to stain cell nuclei and it was clearly demonstrated that this refined method allows one to easily visualize, identify, and purify round spermatids. Parrilla et al. (Reference Parrilla, Vazquez, Cuello, Gil, Roca, Di Berardino and Martinez2004) demonstrated that Hoechst 33342 staining does not induce genotoxic effects in flow-sorted boar spermatozoa and in our procedure Hoechst 33342 was at a relatively low concentration. The membranes of all living cells have selective permeability. Propidium iodide (PI) is membrane-impermeant and is generally excluded from viable cells, therefore PI could only adhere passively to the live cells and the side effects should be minimal. Verapamil is used in cell biology as an inhibitor of drug efflux pump proteins such as P-glycoprotein and it can decrease the speed of dye (Hoechst 33342) extrusion before/in sorting (Goodell, Reference Goodell2005). The biological half-life of verapamil is also short (2.8–7.4 h) and thus would not influence the following work. Moreover, whether Hoechst 33342, PI or verapamil was used, they all could be removed through medium washing after sorting.
In addition, it was found that all the cells collected by FACS in the current study were almost the same size and had centrally located nucleoli. It was also shown that round spermatids from Pten−/− male mice injected into mature metaphase II (MII) stage oocytes were able to support embryonic development, with a satisfactory survival, fertilization and developmental rate, as well as the birth of healthy offspring. These data clearly show that FACS-sorted round spermatids can be effectively used for ROSI to rescue some infertile mouse strains. However, FACS selection of round spermatids for ROSI inevitably has the following two drawbacks: it must be based on the existing FACS platforms, otherwise the cost will be very high; the cell number for FACS should be in a larger quantity (>106), therefore it is not suitable for human ROSI.
In conclusion, FACS selection of round spermatids can not only provide high-purity and viable round spermatids for ART use, but also has no harmful effects on the developmental capacity of subsequently fertilized embryos. Therefore, round spermatids selected by FACS can be used for rederivation or rescue of infertile male mice, thus facilitating development biology research and ART. Lastly, ROSI should be considered a powerful addition to the armamentarium of ART applicable in the mouse.
