Introduction
Advances in oncological diagnosis and treatments have provided a considerable increase in survival of cancer patients (Jemal et al., Reference Jemal, Siegel, Xu and Ward2012), however, chemotherapy and/or radiotherapy treatments, including alkylating agents may compromise their future fertility (Anderson & Wallace, Reference Anderson and Wallace2013). Furthermore, as the number of young cancer survivors augments, the demand for fertility preservation before cancer therapy increases progressively (Donnez et al., Reference Donnez, Martinez-Madrid, Jadoul, Van Langendonckt, Demylle and Dolmans2006b). Cryopreservation and transplantation are, therefore, the main and most viable options to preserve and thus regenerate the fertility of women who will undergo cancer treatment (Donnez et al., Reference Donnez, Dolmans, Pellicer, Diaz-garcia, Serrano, Schmidt, Ernst, Luyckx and Andersen2013).
Especially for children or young patients, the available option to preserve fertility is the cryopreservation of ovarian tissue (slices/fragments) that allows the storage of a large number of primordial and primary follicles (Meirow et al., Reference Meirow, Baum, Yaron, Levron, Hardan, Schiff, Nagler, Yehuda, Raanani and Hourvitz2007) and can ensure the restoration of the ovarian endocrine function (Donnez et al., Reference Donnez, Dolmans, Pellicer, Diaz-garcia, Serrano, Schmidt, Ernst, Luyckx and Andersen2013). The cryopreserved ovarian tissue needs, therefore, to be grafted to the patient after a period of storage at low temperatures.
More than a half century ago, a study conducted in mice described the first birth after a whole ovary orthotopic transplantation with previous cryopreservation (Parrot, Reference Parrot1960). In humans, the first birth obtained using ovarian tissue cryopreservation and grafting was reported by Donnez et al. (Reference Donnez, Dolmans, Demylle, Jadoul, Pirard, Squifflet, Martinez-Madrid and Van Langendonckt2004) and was a landmark in human reproductive medicine. Since then, the birth of 70 healthy babies (Silber, Reference Silber2016) has been reported after transplantation of cryopreserved ovarian tissue. Despite these encouraging results, the techniques of cryopreservation and ovarian transplantation to restore reproductive function in women are still considered experimental. New studies are therefore being performed around the world to investigate the best way to restore fertility, either in humans (Burmeister et al., Reference Burmeister, Kovacs and Osianlis2013), non-human primates (Amorim et al., Reference Amorim, Jacobs, Devireddy, Langendonckt, Vanacker, Jaeger, Luyckx, Donnez and Dolmans2013), in domestic animals (Fassbender et al., Reference Fassbender, Hildebrandt, Paris, Colenbrander and Jewgenow2007) or in laboratory animals (César et al., Reference César, Petroianu, Vasconcelos, Cardoso, Mota, Barbosa, Soares and De Oliveir2015).
The natural plasticity of the ovary facilitates grafting to different sites in which they can be revascularized and rapidly restore its normal physiology. Furthermore, ovarian tissue can be transplanted orthotopically to the pelvis (Demeestere et al., Reference Demeestere, Simon, Moffa, Delbaere and Englert2010) or heterotopically in subcutaneous areas, kidney capsule or fat pad (Youm et al., Reference Youm, Lee, Lee, Jee, Suh and Kim2015) as well as in other sites (the rectus muscle: Kim et al., Reference Kim, Lee, Chung, Lee, Lee and Hill2009; subperitoneal tissue: Stern et al., Reference Stern, Toledo, Hale, Gook and Edgar2011). Regardless of the site, the ovarian graft can undergo ischemia and potentially follicular atresia that represent a challenge to the success of this technique. It is necessary, therefore, to make sure that there is good cell communication between graft and host tissue when the ovary is grafted.
The practice of cryostorage followed by transplantation has been restoring fertility and hormone production (in humans: Silber, Reference Silber2016; and others species: Amorim et al., Reference Amorim, Jacobs, Devireddy, Langendonckt, Vanacker, Jaeger, Luyckx, Donnez and Dolmans2013; Campbell et al., Reference Campbell, Hernandez-Medrano, Onions, Pincott-Allen, Aljaser, Fisher, McNeilly, Webb and Picton2014). This option is elective for patients who have high risk of premature ovarian failure (POF) in which oncologists and physicians work together to achieve the patient's global well-being (Revelli et al., Reference Revelli, Marchino, Dolfin, Molinari, Delle Piane, Salvagno and Benedetto2013).
Thus, this review will provide an insight into the different factors that affect ovarian functionality after transplantation and some relevant advance to date.
Ovarian tissue transplantation
The main goal of transplantation of ovarian tissue is the restoration of ovarian endocrine function and fertility especially in young and adult women undergoing cancer treatment (Donnez et al., Reference Donnez, Dolmans, Pellicer, Diaz-garcia, Serrano, Schmidt, Ernst, Luyckx and Andersen2013). The ovary is abundant in primordial follicles in the quiescence stage or at rest that constitute the ovarian reserve. Only a few of these will be activated and develop to an advanced follicular stage (Kim, Reference Kim2012). Primordial follicles can be cryopreserved and stored at any stage of the female reproductive life, without the need for hormonal treatment.
The ovary is a well suited place for transplantation due to its naturally abundant angiogenic factors that favor the neovascularization process. According to the site of transplantation the procedure is classified as: orthotopic or heterotopic implantation. Orthotopic implantation is defined as tissue transplanted to its place of origin or into the pelvic cavity. In the case of heterotopic implantation, it is defined as tissue transplanted in a different site or in an extra-ovarian region (Sonmezer & Oktay, Reference Sonmezer and Oktay2010), such as the abdominal wall (Rodriguez-Wallberg & Oktay, Reference Rodriguez-Wallberg and Oktay2012), forearm (Oktay et al., Reference Oktay, Economos, Kan, Rucinski, Veeck and Rosenwaks2001), kidney capsule (Youm et al., Reference Youm, Lee, Lee, Jee, Suh and Kim2015) or breast (Kim et al., Reference Kim, Hwang and Lee2004) following the cryopreservation process.
According to the graft recipient, transplantation can be classified as xenotransplantation (performed in different species), allotransplantation (performed in the same species) or autotransplantation (performed in the same individual).
Site of ovarian tissue implantation
Orthotopic site
In this type of transplantation, the tissue is reimplanted in its original physiological surroundings and the development of transplanted tissue is very effective. The main advantage of orthotopic transplantation of ovarian tissue is that natural conception could occur without the intervention of assisted reproductive techniques. According to Donnez & Dolmans (Reference Donnez and Dolmans2015), the pelvic cavity (orthotopic site) would provide the optimal environment for follicular development compared with heterotopic sites, as temperature, pressure, paracrine factors and blood supply are similar to those observed in a physiological situation. Conversely, the disadvantage would be the limited number of fragments able to be transplanted due to ovary size. In addition, orthotopic transplantation is an invasive procedure that may cause severe pelvic adhesions (Demeestere et al., Reference Demeestere, Simon, Emiliani, Delbaere and Englert2009).
In animal production, the first encouraging results were obtained by Gosden et al. (Reference Gosden, Baird, Wade and Webb1994) in sheep. These authors reported the resumption of cyclic activity, pregnancy, and producing a live birth after orthotopic autotransplantation of samples of ovarian tissue cryopreserved by slow freezing. Subsequently, Salle et al. (Reference Salle, Demirci, Franck, Rudigoz, Guerin and Lornage2002, Reference Salle, Demirci, Franck, Berthollet and Lornage2003) and Bordes et al. (Reference Bordes, Lornage, Demirci, Franck, Courbiere, Guerin and Salle2005) also reported live birth using orthotopic transplantation of a frozen–thawed and vitrified ovary in ewes. Imhof et al. (Reference Imhof, Bergmeister, Lipovac, Rudas, Hofstetter and Huber2006) also reported a live birth after orthotopic transplantation of frozen–thawed ovarian tissue in ovine. Moreover, Santos et al. (Reference Santos, Knijn, Vos, Oei, Loon, Colenbrander, Gadella, van den Hurk and Roelen2009) demonstrated complete follicular development and the recovery of endocrine function after cryopreservation and orthotopic autotransplantation of small ovarian fragments in bilaterally ovariectomized goats without administration of hormones.
It should be emphasized that the first live birth in humans obtained using the combination of ovarian tissue cryopreservation and orthotopic transplantation was reported by Donnez et al. (Reference Donnez, Dolmans, Demylle, Jadoul, Pirard, Squifflet, Martinez-Madrid and Van Langendonckt2004), and was a landmark in human reproductive medicine. Since then, various authors have reported live birth after ovary transplantation (Andersen et al., Reference Andersen, Rosendahl, Byskov, Loft, Ottosen, Dueholm, Kirsten, Andersen and Ernst2008; Silber et al., Reference Silber, De Rosa, Pineda, Lenahan, Grenia, Gorman and Gosden2008; Donnez et al., Reference Donnez, Squifflet, Pirard, Demylle, Delbaere, Armenio, Englert, Cheron, Jadoul and Dolmans2011, Silber, Reference Silber2016).
Two techniques were successfully used to reimplant fresh or frozen–thawed ovarian tissue in an orthotopic site, within a specially created window on the peritoneum (Donnez et al., Reference Donnez, Dolmans, Demylle, Jadoul, Pirard, Squifflet, Martinez-Madrid and Van Langendonckt2004) or on the remaining ovary (Donnez et al., Reference Donnez, Dolmans, Demylle, Jadoul, Pirard, Squifflet, Martinez-Madrid and Van Langendonckt2006a). The large tissue strips (8–10 × 5 mm) of ovarian fragments could be sutured into the remaining ovary after the removal of the native cortex. However, the transplantation of small pieces (2 × 2 mm), in the medulla space was difficult because it could not be sutured (Donnez et al., Reference Donnez, Squifflet, Van Eyck, Demylle, Van Langendonckt and Dolmans2008). A normal reproductive lifespan was demonstrated after orthotopic grafting of vitrified ovary in the mouse (Liu et al., Reference Liu, Xie, Zhang, Xu, Bujard and Jun2008). Furthermore, Silber et al. (Reference Silber, De Rosa, Pineda, Lenahan, Grenia, Gorman and Gosden2008) reported reinitiation of ovulatory menstrual cycles and normal serum FSH levels after 77–142 days after transplantation of fresh and cryopreserved ovarian tissue between a series of monozygotic (MZ) twin pairs. Moreover, two live births were reported following fresh ovary transplantation between two identical twin sisters and another one live birth after autotransplantation of cryopreserved ovarian tissue (Silber et al., Reference Silber, De Rosa, Pineda, Lenahan, Grenia, Gorman and Gosden2008). Burmeister et al. (Reference Burmeister, Kovacs and Osianlis2013) also reported human pregnancy after ovarian tissue cryopreservation and subsequent orthotopic autotransplantation.
Heterotopic site
The potential advantages of heterotopic implantation include: avoidance of invasive procedures; easy accessibility of the graft; increased capacity for cortical slices; and feasibility for grafting even if severe pelvic adhesions preclude orthotopic transplantation (Kim, Reference Kim2012). In addition, the use of general anaesthesia is not required and the removal of the transplanted fragments is not difficult (Filatov et al., Reference Filatov, Khramova, Kiseleva, Malinova, Komarova and Semenova2016). However, unlike orthotopic transplantation, natural conception cannot be expected after transplantation of ovarian tissue to heterotopic sites and, therefore, in vitro fertilization (IVF) is required for conception.
As mentioned above, ovarian grafts can be transplanted in several different sites, including the bursa cavity, the kidney capsule, and subcutaneous sites. Transplantation in the kidney capsule has been often used due its excellent blood supply that would enhance graft survival (Youm et al., Reference Youm, Lee, Lee, Jee, Suh and Kim2015). Graft recovery and oocyte yield are significantly higher from the bursal cavity and kidney capsule compared with subcutaneous site in murine species (Youm et al., Reference Youm, Lee, Lee, Jee, Suh and Kim2015). Regardless of the site of transplantation, the ovarian graft will undergo ischemia and potential follicular atresia during the period after transplantation before tissue revascularization; this process remains a challenge to the success of this technique (Wang et al., Reference Wang, Chang, Sun, Dang, Ma, Hei, Shen, Zhao, Cai, Pei, Zhang and Jiang2012). However, by virtue of their low metabolic rate, primordial follicles are relatively resistant to the effects of oxygen deprivation (Schmidt et al., Reference Schmidt, Ernst, Byskov, Nyboe Andersen and Yding Andersen2003).
Heterotopic ovarian transplantation experience in humans and non-human primates has been very limited (Amorim et al., Reference Amorim, Jacobs, Devireddy, Langendonckt, Vanacker, Jaeger, Luyckx, Donnez and Dolmans2013). Several heterotopic sites have been tested in humans, including the uterus broad ligament (Gosden et al., Reference Gosden, Robert and Morris2010), the rectus muscle (Kim et al., Reference Kim, Lee, Chung, Lee, Lee and Hill2009), the forearm (Oktay et al., Reference Oktay, Economos, Kan, Rucinski, Veeck and Rosenwaks2001), breast tissue (Kim et al., Reference Kim, Hwang and Lee2004), the subcutaneous tissue of the abdomen (Youm et al., Reference Youm, Lee, Lee, Jee, Suh and Kim2015) and subperitoneal tissue (Stern et al., Reference Stern, Toledo, Hale, Gook and Edgar2011). In studies with primates, other heterotopic sites (omentum, retroperitoneal iliac fossa, uterine serosa, mesosalpinx, and the pelvic wall) were also used (Suzuki et al., Reference Suzuki, Hashimoto, Igarashi, Yamanaka, Yamochi, Takenoshita, Hosoi, Morimoto and Ishizuka2012). In addition to intraperitoneal (omentum) or subperitoneal sites, the subcutaneous tissue of the abdomen could also provide an adequate environment for follicle growth in both humans and primates (Kim, Reference Kim2014). However, heterotopic sites may not provide an optimal environment for follicular development due to differences in temperature, paracrine factors and blood supply compared with the intraperitoneal environment (Donnez et al., Reference Donnez, Jadoul, Squifflet, Van Langendonckt, Donnez, Van Eyck, Marinescu and Dolmans2010a). Primate models have been used to find suitable locations to overcome some of these challenges (Igarashi et al., Reference Igarashi, Suzuki, Hashimoto, Takae, Takenoshita, Hosoi, Morimoto and Ishizuka2010). High rates of follicular survival after heterotopic xenotransplantation of human cryopreserved prepubertal ovarian tissue in mice have been recently demonstrated with the retention of a large pool of dormant primordial follicles in the graft (Luyckx et al., Reference Luyckx, Pharm, Soares, Scalercio, Jadoul, Amorim, Soares, Donnez and Dolmans2013). Furthermore, Stern et al. (Reference Stern, Gook, Hale, Agresta, Oldham, Rozen and Jobling2013) reported the first clinical pregnancy following heterotopic grafting of cryopreserved ovarian tissue in a woman after a bilateral oophorectomy.
Therefore, the choice of the transplantation sites constitutes an essential factor involved in future graft viability and in the subsequent oocyte competence (Demeestere et al., Reference Demeestere, Simon, Emiliani, Delbaere and Englert2009).
Graft recipient of ovarian tissue implantation
Xenotransplantation
Complex biological processes often require in vivo analysis, and several important research studies have been made using mice as a model for the study of various biological systems. Humanized mice, or mouse–human chimeras were developed to overcome these constraints and are now an important research tool for in vivo study of human cells and tissues (Shultz et al., Reference Shultz, Ishikawa and Greiner2007). Immunodeficient mice have been used to evaluate follicle development and survival after cryopreservation of ovarian tissue. Some studies (Table 1) have shown that xenografting of human ovarian tissue into mice has been an effective model to study ovarian function and follicle development in vivo (Van Eyck et al., Reference Van Eyck, Jordan, Gallez, Heilier, Van Langendonckt and Donnez2009). Since 1960, athymic mice (nude) have been a standard for establishing in vivo models of human malignancies. Due to the lack of a thymus, nude mice cannot generate mature T lymphocytes and therefore are unable to mount most types of immune responses (Fransolet et al., Reference Fransolet, Henry, Labied, Masereel, Blacher, Noël, Foidart, Nisolle and Munaut2015). This absence of functioning T cells prevents nude mice from rejecting not only allografts, but even xenografts (Fransolet et al., Reference Fransolet, Henry, Labied, Masereel, Blacher, Noël, Foidart, Nisolle and Munaut2015). Another attractive model is the use of severe combined immunodeficient (SCID) mice (Fransolet et al., Reference Fransolet, Henry, Labied, Masereel, Blacher, Noël, Foidart, Nisolle and Munaut2015). SCID mice were the first and are the most commonly used model for ovarian xenografts (Aubard, Reference Aubard2003) and were characterized by their scid mutation, which leads to a defect in the recombination of antigen receptor genes, impairing their capacity to generate functional B and T lymphocytes (Custer et al., Reference Custer, Bosma and Bosma1985). Thus, SCID mice can maintain tissues from foreign species without demonstrating a graft-versus-host response (Bosma et al., Reference Bosma, Custer and Bosma1983) and, consequently, they can serve as a good model for transplantation studies. A few years after SCID mouse generation, the scid mutation was transferred onto a non-obese diabetic (NOD) background. This transfer leads to NOD–SCID mice, which have reduced natural killer cell (NK cell) activity in addition to the deficiency in functional B and T cells (Prochazka et al., Reference Prochazka, Gaskins, Shultz and Leiter1992). Moreover, their ability to activate some components of the complement system is impaired, and these mice are markedly deficient in macrophages (Shultz et al., Reference Shultz, Ishikawa and Greiner2007).
Table 1 Summary of some relevant results obtained after xenotransplantation of ovarian tissue in different species
Pc: piece.
In order to develop a novel protocol for the establishment of human ovarian stroma within a mouse model subcutaneously, normal human ovarian tissues were subcutaneously implanted into SCID mice and then the implants were identified by immunohistochemistry. The results demonstrated that human ovarian tissue successfully survives in a SCID mouse host and retains the properties of the original normal ovarian tissues (Fu et al., Reference Fu, Wang, Sun, Xu, Zhou and Cheng2014). Previously, Luyckx et al. (Reference Luyckx, Pharm, Soares, Scalercio, Jadoul, Amorim, Soares, Donnez and Dolmans2013) reported that frozen–thawed preantral follicles from prepubertal patients can successfully survive and develop after long-term ovary xenografting. The study conducted by Henry et al. (Reference Henry, Labied, Fransolet, Kirschvink, Blacher, Noël, Foidart, Nisolle and Munaut2015) after xenotransplantation of sheep ovary to SCID and NOD–SCID showed that they were both suitable for studying graft recovery, however, based on histologic analysis, the overall tissue morphology was better preserved in SCID mice. A significant alteration of the gene responsible for ovarian metabolism and function was reported after xenotransplantation of rat ovary into the kidney capsule of immune-deficient mice (Agca et al., Reference Agca, Lucy and Agca2009). Ishijima et al. (Reference Ishijima, Abe and Suzuki2009) also showed high follicular loss after xenotransplantation of canine cryopreserved ovary in immune-deficient mice. Moreover, it was demonstrated that encapsulation of ovarian tissue with VEGF165 in a collagen matrix during xenografting in SCID mouse produces a more rapid onset of functional vessel formation and earlier revascularization of the transplant (Henry et al., Reference Henry, Labied, Fransolet, Kirschvink, Blacher, Noël, Foidart, Nisolle and Munaut2015). Lotz et al. (Reference Lotz, Liebenthron, Nichols-Burns, Montag, Hoffmann, Beckmann, van der Ven, Töpfer and Dittrich2014) reported antral follicle formation after 122 days post xenotransplantation in a single human ovarian fragment of 6 years old without exogenous hormone stimulation. Moreover, it was shown that isolated human follicles were able to survive after encapsulation in fibrin clots and short-term xenotransplantation (Paulini et al., Reference Paulini, Vilela, Chiti, Donnez, Jadoul, Dolmans and Amorim2016).
Autotransplantation/allotransplantation
As mentioned before, the main goal of ovarian tissue transplantation is the restoration of fertility and endocrine function of women who undergo cancer treatments (chemo-/radiotherapy). In this context, the efforts made so far by research teams from at least 10 countries have proved that transplantation of cryopreserved ovarian tissue is a promising option to preserve fertility in female patients with cancer. These procedures allow immediate initiation of cancer treatment, as it does not require prior ovarian stimulation nor sperm donation compared with other technologies such as embryo or oocyte cryopreservation.
It is very difficult to know the exact number of attempts made to transplant ovarian tissue (fresh and cryopreserved), especially when it is not directly involved with the clinic or even when the main focus of research is the biotechnology of animal production. In fact, according to Donnez & Dolmans (Reference Donnez and Dolmans2015), the number of reimplantations performed worldwide is not known. However, excellent results have been seen in the USA, Brussels, Paris, Spain, Denmark, and Israel. In addition, successes are reported by research teams in Japan, Italy, Germany, and Australia (Stoop et al., Reference Stoop, Cobo and Silber2014). In addition to the excellent restoration rate of ovarian function (95%) and a reasonable gestation rate (23%) obtained when this strategy is applied, it has been observed that the duration of ovarian function has been maintained for a period from 4 to 5 years (Donnez & Dolmans, Reference Donnez and Dolmans2013, Reference Donnez and Dolmans2014).
Autotransplantation of frozen and thawed ovarian tissue is only possible if absence of cancer cells in the graft is confirmed and there is a legitimate concern for the reseeding of malignant cells when carrying out ovarian transplantation (Rodriguez-Wallberg & Oktay, Reference Rodriguez-Wallberg and Oktay2012). Disadvantages of ovarian tissue cryopreservation and autotransplantation include the limited life span of the ovarian grafts due to the potential post-transplantation window of ischemia responsible for follicular atresia. In addition, for patients with autoimmune or genetic disorders, gonadal autotransplantation is ineffective in preserving fertility. In such cases, allotransplantation might be the solution (Yi-Hsin et al., Reference Yi-Hsin, Yu-Chi, Chii-Ruey, Wei-Jen, Jah-Yao and Chi-Huang2011). Silber and colleagues reported a series of monozygotic twins (Table 2) who underwent ovarian isotransplantation to rescue the sterile sister (Silber et al., Reference Silber, Lenahan, Levine, Pineda, Gorman, Friez, Crawford and Gosden2005, Reference Silber, De Rosa, Pineda, Lenahan, Grenia, Gorman and Gosden2008). Furthermore, orthotopic ovarian allotransplantation has been performed in patients diagnosed with Turner's syndrome. This technique was able to restore regular menstruation and ovulation, it raised hormonal concentrations and led to the development of secondary sexual characters (Mhatre & Mhatre, Reference Mhatre and Mhatre2006). However, Scott et al. (Reference Scott, Hendrickson, Lash and Shelby1987) have shown that immunosuppressive therapy, such as cyclosporine A plus steroid, is needed for ovarian allografts to survive.
Table 2 Summary of some relevant results obtained after auto/allotransplantion of ovarian tissue in different species
Pc: piece.
Factors affecting graft function
Follicle loss is one of the major limitations after ovary transplantation. This phenomenon occurs in several steps of the procedure as cited below.
Ovarian tissue size
The thickness of ovarian cortical strips prior to freezing is critical to enable the perfusion of cryoprotectants and the ensuring of the graft survival. The surface-to-volume ratio of the graft has to be high in order to ensure good penetration and removal of cryoprotectants agents. It reduces the probability of damage caused by ice crystals during freezing and thawing, and by the ischemia and hypoxia that occurs during the initial steps of graft acceptance (Gavish et al., Reference Gavish, Ben-Haim and Arav2008). Currently, ovarian cortical tissue is cryopreserved in strips of 1–2 mm thickness either by slow freezing or vitrification protocols (Kagawa et al., Reference Kagawa, Silber and Kuwayama2009). It has been suggested (Gavish et al., Reference Gavish, Ben-Haim and Arav2008; Revel et al., Reference Revel, Laufer, Ben Meir, Lebovich and Mitrani2011) that reducing graft dimensions might enhance diffusion of oxygen, growth factors and nutrients into the ischemic graft and improve follicle's survival. However, a study conducted by Gavish et al. (Reference Gavish, Peer, Hadassa, Yoram and Meirow2014) showed an extensive neovascularization 1-week post xenotransplantation of bovine ovarian tissue in mice. No beneficial effect was found when reducing the graft thickness beyond 1 mm, including extensive primordial follicle loss and increased follicle ‘burn out’ without improving neovascularization. Moreover, the transplanted fragments of ovarian cortex contain only a fraction of an individual's ovarian reserve and as such can only provide the recipient with a relatively brief fertile window before the supply of oocytes contained within their graft is depleted (Andersen et al., Reference Andersen, Rosendahl, Byskov, Loft, Ottosen, Dueholm, Kirsten, Andersen and Ernst2008; Ernst et al., Reference Ernst, Bergholdt, Jorgensen and Andersen2010; Silber, Reference Silber2012; Donnez et al., Reference Donnez, Dolmans, Pellicer, Diaz-garcia, Serrano, Schmidt, Ernst, Luyckx and Andersen2013). This limitation means that cryopreservation and autografting of pieces of cortex appear to be less effective as a means to restore the fertility of older patients in whom follicle density is already low at the time of tissue preservation and in which it is associated with endocrine disturbance (Campbell et al., Reference Campbell, Telfer, Webb and Baird2004). In this context, cryopreservation of the whole ovary (complete with vascular pedicle) for later autotransplantation provides an attractive alternative strategy for fertility preservation as it involves restoration of all of the primordial follicles within the ovary. Furthermore, as transplantation requires vascular anastomosis rather than cortical revascularization, this intervention should result in no marked reduction of ovarian reserve due to ischemia, preventing endocrine imbalance and reducing the age constraint which limit the efficacy of this fertility preservation technology (Campbell et al., Reference Campbell, Hernandez-Medrano, Onions, Pincott-Allen, Aljaser, Fisher, McNeilly, Webb and Picton2014). However, autotransplantation of frozen–thawed whole ovary has also some disadvantages including:
-
(1) It has a higher risk of cryoinjury during freezing due to inadequate diffusion of cryoprotectants throughout the entire ovary and non-homogenous cooling rate between the core and the periphery of the ovary as well.
-
(2) Cryoinjury of ovarian vasculature
-
(3) Surgical difficulty of vascular anastomosis due to the small size of ovarian artery (~0.5 mm in diameter), short ovarian vascular pedicle (~5 cm in length), discrepancy between the diameters of graft and recipient vessels, and possible failure of microvascular anastomosis.
-
(4) Higher risk of post-operative vascular complications including anastomotic bleeding, pseudoaneurysm, stenosis, or microvascular thrombosis.
-
(5) Its vascular complications can severely affect the survival of the entire ovary leaving no other attempt for transplantation (Bedaiwy & Falcone Reference Bedaiwy and Falcone2010; Kim, Reference Kim2010; Zhang et al., Reference Zhang, Sheng, Cao, Wang and Chen2011).
Ischemia
Ischemia is defined as the insufficient perfusion and reduction of the arterial or venous blood flow, depletion of cellular energy storages, and the accumulation of toxic metabolites which can lead to cell death (Ingec et al., Reference Ingec, Isaoglu, Yilmaz, Calik, Polat, Alp, Kurt, Gundogdu and Suleyman2011). Ischemic times that extend for several days most likely induce profound and irreversible ischemia–reperfusion injuries. They are responsible for follicle loss, as the graft needs 4–5 days to be reoxygenated (Van Eyck et al., Reference Van Eyck, Jordan, Gallez, Heilier, Van Langendonckt and Donnez2009). A study with sheep ovarian tissue reported that 65% of the follicles are lost after grafting of fresh tissue while subjecting the tissue to cryopreservation and thawing would add another 7% of follicle loss (Baird et al., Reference Baird, Webb, Campbell, Harkness and Gosden1999). Moreover, fresh or frozen human ovarian grafts transplanted into nude mice showed similar reduction in follicular density, demonstrating that ischemia is the main factor behind follicular depletion (Demeestere et al., Reference Demeestere, Simon, Emiliani, Delbaere and Englert2009). Thus, duration of warm ischemia before neovascularization is crucial for follicular survival and may be responsible for the major loss of follicles after grafting rather, than after freezing.
The mechanisms behind ischemic injury involves energy depletion and reperfusion oxidative stress (Demeestere et al., Reference Demeestere, Simon, Emiliani, Delbaere and Englert2009), which can cause damage to lipids, DNA, enzymes and structural proteins, therefore leading to cell death (Kim et al., Reference Kim, Qian and Lemasters2003). Gene expression of several inflammatory factors is initiated by hypoxia-sensitive response elements and nuclear factor-kappa beta (NF-κB), resulting in transmigration of neutrophils and macrophages into the tissue (Vollmar et al., Reference Vollmar, Glasz, Menger and Messmer1995) that causes tissue destruction and fibrosis. However, primordial follicles still survive for relatively long periods of ischemia after avascular transplantation due to their very low metabolic demands. Moreover, they are distributed just under the ovarian surface and they may be the first to benefit from the ingrowth of new vessels. In a study with cryopreserved bovine ovarian cortex it was demonstrated that stromal cells are more vulnerable to ischemia compared with primordial follicles (Kim et al., Reference Kim, Hwang and Lee2004). Due to the negative effects of ischemic time in the follicular pool, angiogenesis is the key mechanism for follicle survival.
Addition of exogenous substances to improve angiogenesis and ovarian graft
Many animal experiments have focused on improving ovarian tissue survival. Transplantation of frozen–thawed ovarian tissue in animal models led to antral follicle development and live birth (Schubert et al., Reference Schubert, Canis, Darcha, Artonne, Smitz and Grizard2008). However, despite the successful results obtained with cryopreservation and autotransplantation of ovarian tissue, it is important to highlight the large follicle loss resulting from the latter procedure (Gavish et al., Reference Gavish, Peer, Hadassa, Yoram and Meirow2014). In this context, many studies have been conducted with different substances such as antioxidants agents or angiogenic factors. Local antioxidant injection of vitamin E before grafting could improve follicular survival rate (Martinez-Madrid et al., Reference Martinez-Madrid, Donnez, Van Eyck, Veiga-Lopez, Dolmans and Van Langendonckt2009). Ischemia and oxidative stress could be reduced by using drugs to stimulate revascularization of the graft, such as vascular endothelial growth factor or sphingosine-1-phosphate (S1P), as described by Donnez & Dolmans, (Reference Donnez and Dolmans2013). A study carried out by Zelinski et al. (Reference Zelinski, Murphy, Lawson, Jurisicova, Pau, Toscano, Jacob, Fanton, Casper, Dertinger and Tilly2011) showed that S1P and its analogues protected the primate ovary against radiation-induced damage. In addition, Scalercio et al. (Reference Scalercio, Amorim, Brito, Percario, Oskam, Domingues and Santos2015) showed that trolox, an analogue of vitamin E improved follicular quality and avoided apoptosis in stromal cells when ovarian tissue is pre-treated with this substance. Moreover, a study showed that autografted mice ovary followed by daily administration of l-carnitine until day 7 inhibited follicle apoptosis, relieved oxidative damage, and increased follicular survival and function in ovarian graft (Zhang et al., Reference Zhang, Wang, Yao, Zhang, Zhang, Chen, Fu and Zhang2015).
To reduce the hypoxic period after tissue transplantation and improve follicular preservation, angiogenesis can be stimulated by vascular endothelia growth factor (VEGF), which is the main signaling protein that regulates new vessel formation from pre-existing vessels (Henry et al., Reference Henry, Labied, Fransolet, Kirschvink, Blacher, Noël, Foidart, Nisolle and Munaut2015).
Ovarian cryopreservation and transplantation procedures have so far been almost exclusively limited to avascular cortical fragments. The small cortical pieces grafted without vascular anastomosis are completely dependent on the establishment of neovascularization after grafting. Consequently, the cells in the graft undergo significant ischemic and reperfusion damage, which can induce a high rate of follicular loss (Zhang et al., Reference Zhang, Yang, Ma, Wu, Zheng, Hei, Sun, Ma, Chang, Wang, Cai, Xie, Zhao, Pei and Wang2016). Therefore, these cells remain quiescent until there is an angiogenic stimulus such as hypoxia or wounding, which then up-regulates proangiogenic factors, such as VEGF (Gerhardt & Betsholtz, Reference Gerhardt and Betsholtz2003). Angiogenesis is regulated in the reproductive tract and elsewhere by at least 20 angiogenic growth factors and inhibitors identified to date. A key player is VEGF also known as vascular permeability factor (Ferrara and Davis-Smyth, Reference Ferrara and Davis-Smyth1997). It has been shown that VEGF acts via two tyrosine kinase family receptors, namely flt-1 (VEGF receptor; VEGFR-1) and flk-1/KDR (Ferrara and Davis-Smyth, Reference Ferrara and Davis-Smyth1997), which are expressed in granulosa cells (Greenaway et al., Reference Greenaway, Connor, Hanne, Pedersen, Coomber, Lamarre and Petrik2004). The effects of two VEGF-A isoforms, VEGF111 and VEGF165, have been tested in a xenograft model using SCID mice (Commin et al., Reference Commin, Buff, Rosset, Galet, Allard, Bruyere, Joly, Guerin and Neto2012). VEGF111 is soluble and resistant to proteolysis, whereas VEGF165 is additionally anchored to the extracellular matrix. Both VEGF isoforms increase blood vessel recruitment and functional angiogenesis. The use of this isoform in a collagen matrix, which encapsulates the ovarian cortex at the time of transplantation was found to improve angiogenesis and decrease hypoxia, thereby enhancing the preservation of primary follicles (Labied et al., Reference Labied, Delforge, Munaut, Blacher, Colige, Delcombel, Henry, Fransolet, Jouan, D'Hauterive, Noël, Nisolle and Foidart2013). Moreover, it was demonstrated (Henry et al., Reference Henry, Labied, Fransolet, Kirschvink, Blacher, Noël, Foidart, Nisolle and Munaut2015) that encapsulation of ovarian tissue with VEGF165 in a collagen matrix during xenografting in mice produces a more rapid onset of functional vessel formation and earlier revascularization of the transplant. In addition, Campbell et al. (Reference Campbell, Hernandez-Medrano, Onions, Pincott-Allen, Aljaser, Fisher, McNeilly, Webb and Picton2014) used a post-operative regime of anti-thrombotic (aspirin) agents to reduce post-transplant ischemia. This substance prevents clot formation in the ovarian vasculature induced by cryo- and cyto-toxic damage to the arterial endothelial cells.
Final considerations
This review reinforces the fact that cryopreservation of ovarian tissue for further transplantation is helpful for prepubertal and women of reproductive age or patients for whom cancer treatment cannot be delayed to perform another assisted reproductive technique such as oocyte or embryo cryopreservation. Because preparation and stimulation for oocyte retrieval usually requires 2 to 3 weeks or longer, it is generally not feasible to freeze embryos from an adult female cancer patient for potential future use. Additionally, not all patients have partners with whom they can produce embryos to cryopreserve. Moreover, in contrast with oocytes and embryos, cryopreservation of ovarian tissue does not depend on the age and phase of the patient's estrous cycle, it also involves fewer ethical and social issues especially when this process is performed in humans.
Most of the experimental studies conducted with the transplant of cryopreserved ovarian tissue have the aim to re-establish fertility. However, in special situations, when it is necessary to remove the normal ovaries due to cancer or infectious processes of the pelvic organs, the absence of ovarian hormones may lead to several endocrine and functional disorders such as osteoporosis, reduction of libido, sexual dysfunction, and enhancement of lipoprotein levels. In such conditions, transplantation of cryopreserved ovarian tissue is the main solution to overcome this problem. However, despite the good results obtained by ovary cryopreservation following transplantation, it appears that the main obstacle to successfully restore fertility from frozen–thawed ovarian cortex are adhesions and the massive ischemic damage to follicles until neovascularization develops. Most follicles surviving cryopreservation will undergo ischemic loss during the time required for neovascularization.
Therefore, the optimization of ovary transplantation may require preparing the right thickness of tissue and finding the best site of transplantation. This could avoid ischemia and promote rapid revascularization. Angiogenesis is a key factor for graft survival because right after transplantation the lack of connective tissue between the graft and surrounded tissue is leading to hypoxia resulting in cell death. In this context reducing the ischemia period could enhance the success of transplantation. In this case, further animal model studies are required to develop optimized cryopreservation/transplant protocols for women.
Acknowledgements
Nathalie Jiatsa Donfack is a recipient of a grant from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Brazil). Ana Paula Ribeiro Rodrigues is recipient of a grant from CNPq Brazil (473968/2013-4). Johan Smitz is Special Visitor Researcher from CAPES. None of the authors has any conflict of interest to declare.
