Granulocyte colony-stimulating factor in conjunction with vascular endothelial growth factor maintains primordial follicle numbers in transplanted mouse ovaries.

As anticipated, 2 weeks after grafting frozen-thawed ovaries into the ovarian bursas of recipient females injected with saline, there occurred a near 80% decline ( P <.05) in primordial follicle numbers when compared with primordial follicle numbers in age-matched nontransplanted ovaries. Injection of transplanted mice with VEGF alone had no effect on the degree of primordial follicle loss in the transplanted ovaries. In contrast, a combination of VEGF and G-CSF significantly ( P =.003) increased primordial and total ( P = .012) follicle numbers over those detected in transplanted ovaries of saline-injected control mice. Although primordial follicle numbers in transplanted ovaries of mice injected with VEGF and G-CSF were approximately 50% lower than those of nontransplanted ovaries, this difference was not statistically significant ( P = .058) ( Fig. 1 ). A similar, but not statistically significant, beneficial effect of injecting G-CSF with SCF on the maintenance of primordial follicle numbers in transplanted ovaries was also observed (see Fig. 1 ). With regard to primary and preantral follicle numbers, no statistically significant effects were noted across any of the treatment groups ( Fig. 2 ). Statistically significant differences in ovarian histology were noted between the four treatment groups. There were almost no follicles in VEGF-treated group, mostly preantral follicles in placebo-treated group, and many follicles at different developmental stages in the G-CSF–treated ovaries ( Fig. 3 ).

Our study has established that VEGF in conjunction with G-CSF maintains primordial follicles in transplanted mouse ovaries. This is the first study that we are aware of that has assessed the influence of growth factors on the number of follicles after ovarian transplantation. To date, little is known about factors that orchestrate the maintenance and, possibly the renewal, of oocytes in postnatal ovaries. Numerous cytokines and growth factors including VEGF, transforming growth factor- β (TGF- β ), SCF, and growth differentiation factor-9 have been implicated in oocyte maturation, follicular development, ovulation, and corpus luteum formation ( 30 , 35 – 39 , 46 ). A crucial angiogenic factor, VEGF plays an important role in cell proliferation and sex steroid–dependent angiogenesis in the ovary during the estrous cycle and pregnancy ( 30 , 31 ). Matrix metalloproteinase-9 mediated stem cell factor (SCF; Kit ligand) processing is essential for cell mobilization induced by chemokines/cytokines, VEGF, placental growth factor (PlGF), and stromal cell derived factor-1 ( 47 ). Stem cell factor was originally characterized because of its ability to influence stem cell growth and differentiation. Previous studies on mice with mutations at the Steel gene locus and c-kit (SCF receptor) locus have demonstrated deficient gametogenesis ( 38 , 39 ). In situ hybridization and immunocytochemistry for c-kit with normal ovarian sections revealed high levels of expression in developing oocytes. Analysis of SCF expression revealed high levels of SCF in granulosa cells ( 37 ). Genetic evidence and localization studies have suggested an important interaction between granulosa cells and oocytes via SCF. Stem cell factor was also found to induce the primordial to primary follicle transition ( 37 – 39 ). In combination with cytokines, SCF results in a synergistic enhancement of the proliferation, differentiation, and survival of various hematopoietic lineages. Its synergy with G-CSF is of particular importance in normal hematopoiesis ( 46 ). Granulocyte colony-stimulating factor is an effective mobilizer of peripheral blood progenitor cells and is used to mobilize and provide a source of stem cells for autologous and allogeneic bone marrow transplantation ( 40 ). In vivo, the combination of SCF and G-CSF has been shown to increase the mobilization of peripheral blood progenitor cells over that seen with G-CSF alone ( 47 ). Several possible ovarian growth factors have been proposed in the past. Yang et al. ( 48 ) demonstrated that the messenger RNA (mRNA) and protein expression of VEGF isoforms associated with angiogenesis increased in rat ovarian grafts after transplantation. The VEGF-188 mRNA level increased already 2 days after transplantation, while VEGF-120 and VEGF-164 did not rise significantly until day 30 after transplantation. It is interesting that gonadotropin treatment significantly increased VEGF-188 isoform expression. These findings suggest that different isoforms of VEGF may play different role during angiogenesis ( 48 ). In our experiment, we used VEGF-164 isoform and did not achieve follicle preservation in the group treated with VEGF only. However, when combined with G-CSF, a statistically significant effect was achieved. Similarly, Maltaris et al. ( 49 ) examined the effect of gonadotropin stimulation on the primordial follicle pool of cryopreserved human ovarian tissue after xenotransplantation in SCID mice. It is interesting that gonadotropin administration after grafting stimulated the primordial follicles to maturity but prolonged stimulation caused accelerated depletion of the ovarian follicular reserve. The findings suggest that gonadotropin administration may be beneficial during the initial neovascularization period as suggested by Yang et al. ( 48 ) but may be harmful with prolonged use ( 49 ). The results of our study reveal that loss of primordial follicles due to transplantation is prevented by VEGF and G-CSF in mice. These findings suggest that these growth factors may play a crucial role in regulating germ cell survival in the ovary. However, only the combination of VEGF and G-CSF statistically significantly impacted the number of primordial follicles. No effect was observed with VEGF alone, and a borderline effect was noticed with combination of G-CSF and SCF. We postulate that the increased number of the primordial follicles in the VEGF and G-CSF cotreated group was mostly related to a reduction in the incidence of ischemia-induced primordial follicle loss. Only the resting follicle pool was increased 2 weeks after treatment, but the number never exceeded the starting number before transplantation. No influence was observed on early growing follicles compared with those receiving the vehicle alone. The main focus of this preliminary study was to assess ovarian histology in transplanted ovaries exposed to cytokines. A long-term follow-up and assessment of function of the transplanted ovaries is still needed to ascertain whether the maintenance of primordial follicle numbers in transplanted animals receiving VEGF and G-CSF is functionally important. Nevertheless, our study suggests that treatment of mice after ovarian transplantation with VEGF and G-CSF maintains primordial follicle numbers in transplanted ovaries. Our findings suggest that growth factors may play a crucial role in preventing or minimizing ischemia-induced follicle loss and possibly may enhance oogenesis. However, the exact mechanisms underlying these findings are not clearly understood and need further evaluation. Future directions should include defining the mechanism by which growth factors maintain primordial follicle numbers and an assessment of the long-term function of the transplanted ovaries, such as their hormonal function and fertility.

All studies were approved by the Institutional Animal Care and Use Committee at the Cleveland Clinic. Young adult (6- to 8-week-old) C57BL/6 female mice were purchased from Charles River Laboratories (Wilmington, MA). Animals were kept in light and temperature controlled conditions (12 hours of light; 12 hours of darkness; 22 ± 2°C) and were given chow pellets and water ad libitum. Animals were anesthetized by intraperitoneal injection using a xylazine-ketamine mixture: 1 mL of ketamine (100 mg/mL), 0.1 mL of xylazine (100 mg/mL), and 8.9 mL of sterile water. Skin of the abdominal wall was incised in the V pattern, and the incision carried through the subcutaneous tissue, fascia, and muscles to enter the peritoneal cavity. Ovaries were then identified on both ends of the uterus, excised, and transferred to M2 medium (Sigma-Aldrich, St. Louis, MO) containing 0.25 g/L of calcium chloride CaCl 2 .2H 2 O, 0.16 g/L of magnesium sulfate (anhydrous), 0.35 g/L of potassium chloride, 0.16 g/L of monobasic potassium phosphate, 0.35 g/L of sodium bicarbonate, 5.53 g/L of sodium chloride, 4.0 g/L albumin (bovine fraction V), 1.0 g/L of D-glucose, 0.036 g/L of pyruvic acid, 0.01 g/L of 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid, 0.06 g/L of potassium penicillin-G, and 0.05 g/L of streptomycin sulfate. The mice were killed by carbon dioxide inhalation followed by cervical dislocation. After being removed from the ovarian bursa, whole ovaries were placed in M2 medium, held at room temperature for 10 minutes, and placed into cryovials (one ovary per vial) containing 200 μ L of cryoprotectant (1.5 mol/L dimethyl sulfoxide [DMSO] in M2 medium supplemented with 10% fetal bovine serum). Cryovials were placed in a programmable freezer at 20°C and cooled at a rate of 2°C per minute to −6.5°C and held 10 minutes for seeding. The temperature was then decreased by 0.3°C per minute to −34°C and held several minutes, after which the cryovials were transferred directly into liquid nitrogen (−196°C) and stored for at least 96 hours until thawing and grafting. After at least 96 hours, the cryovials were thawed at room temperature, and ovaries were transferred into 200 μ L of fresh M2 medium and were rehydrated for 10 minutes before grafting. Ovaries were transplanted back into the ovarian bursa (orthotopic transplantation) of female C56BL/6 mice after bilateral ovariectomy, essentially as already described. In brief, one of the thawed intact ovaries was placed in the vacated right ovarian bursa (to maintain consistency) and was secured with one stitch of 7-0 Prolene suture. The body wall was then sutured in an interrupted pattern using 6-0 Vicryl, and the skin incision was reapproximated with interrupted suture of 6-0 Vicryl. Intraperitoneal buprenorphine injections were used for postoperative analgesia. The dosages of VEGF, SCF and G-CSF were established based on previously published data regarding use of these agents in vivo ( 42 – 44 ). Mice were divided into four groups. Group 1 received intraperitoneal (IP) injections of VEGF-164 (R&D Systems, Minneapolis, MN) for 5 days (8.0 μ g/kg/day). Group 2 received subcutaneous (SC) injections of SCF (R&D Systems) (100 μ g/kg/day) and SC G-CSF (R&D Systems) (50 μ g/kg/day) for 5 days. Group 3 received IP injections of VEGF (8.0 μ g/kg/day) and SC injections of G-CSF (50 μ g/kg/day) for 5 days. Group 4 received SC injections of saline for 5 days (vehicle controls). Animals were killed by carbon dioxide inhalation followed by cervical dislocation 14 days after the first injection. The number of nonatretic immature (primordial, primary and preantral) follicles per ovary was determined by serial section histomorphometry, as detailed previously elsewhere ( 45 ). All statistical tests were performed using R version 2.9.1 statistical software ( http://cran.r-project.org ). One-way analysis of variance (ANOVA) was done for each type of follicle (i.e., primordial, primary, preantral, and total), followed by multiple comparisons using Tukey’s and Dunnett’s methods for adjustment. P <.05 was considered statistically significant.