Key words fertility preservation - ovarian tissue cryopreservation - prepubertal ovarian tissue
- xenotransplantation - metaphase II oocyte
Schlüsselwörter Fertilitätserhalt - Kryokonservierung von Eierstockgewebe - präpubertäres Ovargewebe
- Xenotransplantation - Metaphase-II-Oozyte
Introduction
Advances in diagnosis and treatment of oncological diseases have increased the survival
prognosis in recent years, particularly in children and juvenile cancer patients.
Unfortunately, treatment strategies are frequently gonadotoxic and may lead to a depletion
of the ovarian reserve accompanied by premature ovarian failure. Consequently, various
strategies have been developed for fertility preservation; the most advantageous option
usually depends on the patientʼs particular situation, including age as well as potential
arising risk in case of postponing anti-cancer treatment. The only possibility in
preserving fertility in prepubertal females is ovarian tissue cryoconservation, while
there are multiple additional options for adult females. Thawing and re-transplantation
of frozen tissue can be performed to restore fertility after anti-cancer treatment
[1 ], [2 ], [3 ], [4 ], [5 ]. This preservation method profits from two main advantages; ovarian cortex usually
incorporates a huge amount of follicles [6 ], [7 ] and provides a high developmental potential, especially in prepubertal females [8 ], [9 ]. Two transplantation studies have been performed, reporting that frozen-thawed prepubertal
ovarian cortex could induce puberty [10 ], [11 ]; potentially, even restore fertility. Demeestere et al. [12 ] reported a case of ovarian tissue auto-transplantation with fertility restoration,
resulting in a live birth; the tissue had been collected at the age of 13 years and
11 months, before the first menstrual bleeding. These results and worldwide more than
85 reported live births after re-transplantation of autologous frozen-thawed ovarian
tissue to date [13 ] emphasize the potential of this technique; according to the criteria of the European
Society of Human Reproduction and Embryology (ESHRE) special interests groups “Ethics
and Law” as well as “Safety and Quality in Assisted Reproductive Technology” and the
American Society for Reproductive Medicine (ASRM), this method should be graded as
innovative [14 ], [15 ], on the way to routine clinical practice [16 ].
We could demonstrate that a maturing oocyte can be obtained from prepubertal human
ovarian tissue; achieved via xenotransplantation in oophorectomized severe combined
immunodeficiency (SCID) mice without any exogenous hormone stimulation [17 ]. Our following aim was to investigate if spontaneous antral follicle formation as
well as oocyte maturation would occur again from the same xenograft, still without
administering any hormones. Herein, we report repetitive collection of oocytes from
the same xenotransplant; to our knowledge, it has not been published before.
Material and Methods
Patient
A detailed description has been given by Lotz et al. [17 ]. In brief, ovarian tissue from a 6-year-old patient affected by nephroblastoma was
removed in the course of laparotomy for nephroblastoma excision and cryopreserved
for fertility preservation in 2009; informed consent as well as approval of the local
ethical committee (University Womenʼs Hospital, Bonn, Germany) had been given prior
to the intervention. About half of the ovarian cortex was removed from both ovaries
during the surgery for nephroblastoma removal. Histological examination of the ovarian
cortex revealed a sufficient amount of primordial follicles and excluded the presence
of malignant cells. Less than 5% of the frozen tissue was used for this experiment.
Ovarian tissue freezing and thawing
Cryopreservation and thawing procedures for ovarian cortex tissue are identical to
our first report [17 ] and were previously described by Isaschenko et al. in 2008 [18 ]. In summary, biopsies from ovarian tissue were cut into small pieces of approximately
1 × 2 × 1 mm and equilibrated in a freezing solution including 1.5 mol/L dimethylsulfoxyde
(DMSO) as cryoprotectant in Leibovitz medium. The tissue fragments were frozen in
1.8 mL Nunc cryovials via slow freezing protocol in a closed freezing system (Icecube
14S; Synlab) with auto-seeding. Cryopreserved tissue pieces were shipped to the Erlangen
University Hospital (Erlangen, Germany) for thawing as well as transplantation experiments.
Rapid warming was performed in a 37 °C water bath, subsequently exposing the tissue
fragments to stepwise decreasing sucrose concentrations (release from cryoprotectants).
Animals
Female SCID mice (6 weeks old, Harlan-Winkelmann, Borchen, Germany) were housed for
a study comparing prepubertal ovarian tissue from several females in vivo versus in
vitro; the herein reported findings were part of this larger study. One mouse out
of 7 which had been xenografted with tissue from the 6-year-old nephroblastoma patient
showed the herein reported findings; 2 mice died after transplantation and 4 were
sacrificed after a much shorter grafting time for the main study (before discovery
of the first lump). Animals lived in groups of five mice per cage with water and food
(Altromin 1314, Altromin, Lage, Germany) ad libitum; a daily rhythm of 12 h light
and 12 hours dark. All tests and procedures were carried out under laminar flow conditions
in a positive pressure room. Approval for the study was obtained from the local ethical
committee on animal experimentation (Erlangen University Hospital, Erlangen, Germany)
and animals were maintained in accordance with Animal Care and Use Committee regulations.
A detailed description of the housing conditions is available in [17 ].
Transplantation procedure, xenograft treatment and oocyte retrieval
The whole approach was already described after the first oocyte retrieval [17 ]. Surgery for xenotransplantation was performed under isoflurane (Isoflo® ad us. vet., Abott AG, Baar, Germany) anesthesia without regard to the estrous cycle
stage. Study animals were oophorectomized following xenotransplantation by placing
one frozen-thawed piece of human ovarian tissue in an intramuscular pocket of the
neck. Exogenous hormones were not administered during the whole experiment; neither
in the first 122 days between transplantation and puncture of the first oocyte, nor
during the following period of 37 days until the second oocyte retrieval. All mice
were checked daily for health and behavior; follicle growth was monitored through
palpation of the neck. Each time a lump could be distinctly identified at the graft
site, the animal was anesthetized (Isoflo® ad us. vet., Abott AG, Baar, Germany) for subsequent follicle aspiration by using
a modified sterile needle (20 G Supra, Karl Lettenbauer GmbH, Erlangen, Germany);
needles had been specially beveled as well as sharpened before, to minimize injury
during intervention. Collected fluid was then inspected for the presence of cumulus
oocyte complexes (COC); all complexes were harvested and placed into warmed, pre-equilibrated
culture medium without exogenous gonadotropins (Universal IVF Medium, Origio, Malov,
Denmark). After a rest period of one hour in a CO2 incubator (5% CO2 in air), COC were enzymatically denudated (Hyaluronidase, Origio, Malov, Germany)
to examine meiotic stage and then placed back in the incubator for further in vitro
culture. The animal was returned to its cage after first follicle puncture for continued
monitoring of another follicle growth.
Graft recovery and histological assessment
The mouse was sacrificed by cervical dislocation immediately after the second oocyte
retrieval. The xenograft was recovered and fixed in 4% paraformaldehyde (Sigma-Aldrich
Chemistry Ltd., Munich, Germany). After routine paraffin embedding, the entire sample
was cut in serial sections and stained with hematoxylin and eosin for histological
examination. Classification of follicles based on their morphological appearance was
as follows:
primordial follicles; oocyte surrounded by a single layer of flattened granulosa cells,
primary stage; oocyte surrounded by one layer of cuboidal granulosa cells,
preantral follicles; the granulosa cells are cuboidal with a minimum of two cell layers
around the oocyte,
antral stage; an antrum is visible within multilayered cuboidal granulosa cells.
Results
Retrieval of a mature oocyte
Another lump started to grow at the graft site during ongoing observation following
our previously reported retrieval of a metaphase I oocyte from a 7-mm lump (oocyte
diameter 150 µm; Zona pellucida width 14.6 µm); this first oocyte reached the mature
metaphase II stage within 22 h of additional in vitro culture ([Fig. 1 a ] to [c ]). This time, the lump again exhibited a diameter of approximately 7 mm, 37 days
after the first puncture; the follicle fluid was aspirated and one COC, containing
one metaphase II oocyte (oocyte diameter and Zona pellucida width similar to the first
oocyte), was harvested ([Fig. 2 a ] to [c ]).
Fig. 1 First oocyte retrieval from an unstimulated prepubertal human ovarian tissue xenograft
in a SCID mouse. This data has been published previously: One frozen-thawed fragment
of prepubertal human ovarian tissue was transplanted in the neck of an oophorectomized
SCID mouse. A lump of approximately 7 mm in diameter was visible (a ; lump indicated by an arrow) after 122 days grafting time. This lump was punctured
and a cumulus oocyte complex (COC) could be identified under a stereo-microscope in
the aspirated fluid (b ; magnification at 35 ×, arrow pointing to the COC). The COC contained one maturing
metaphase I oocyte (c ; magnification at 320 ×), which matured to metaphase II within 22 hours of additional
in vitro culture (oocyte diameter 150 µm; Zona pellucida width 14.6 µm).
Fig. 2 Second oocyte retrieval from the same unstimulated xenograft. Another lump grew at
the graft site during ongoing observation (a ; lump indicated by an arrow); we aspirated the follicle fluid 37 days after the first
oocyte retrieval and examined the aspirate under a stereo-microscope. One oocyte could
be harvested, COC already dropped out (b ; 35 × magnification). Further inspection under an inverted microscope identified
one mature metaphase II oocyte (c ; arrow indicating the first polar body, magnification at 320 ×).
Identification of follicles in all developmental stages during histological evaluation
of the graft
All developmental stages of follicles could be identified during investigation of
histological serial sections; primordial, primary, preantral as well as antral follicles
with intact morphology ([Fig. 3 a ] to [f ]) and very occasionally with multinucleated oocytes ([Fig. 4 a ]). In addition, some large atretic follicles were present in the ovarian tissue part
as well as multinucleated oocytes ([Fig. 4 b ]). Furthermore, several clusters with dormant primordial follicles could be identified
([Fig. 5 a ] and [b ]).
Fig. 3 Histological sections of the xenograft showing different developmental stages of
follicles. All developmental stages of follicles could be identified during investigation
of histological serial sections of the prepubertal ovarian tissue xenograft. Representative
sections showing primordial, primary and preantral follicles (a –d ; developing follicles in [a ] are indicated with an arrow, star in [a ] marks tissue injury caused by the follicle puncture) as well as antral follicles
(e, f ; Cumulus oophorus is indicated with an arrow in f ). Sections were stained with hematoxylin–eosin; magnification at 100 × in a and d –f ; 200 × in b, c ).
Fig. 4 Histological sections of the xenograft depicting a multinucleated oocyte as well
as an atretic follicle. Extremely rare, multinucleated oocytes were present (a ; 200 × magnification) inside the xenograft. In addition, some large atretic follicles
could be identified (b ; 100 × magnification).
Fig. 5 Histological sections illustrating a cluster of dormant follicles. Several clusters
with dormant primordial follicles could be identified. A representative example is
given in (a ; magnification at 50 ×). Parts of the same site are given in (b ) with a higher magnification (100 ×). In both cases, arrow indicates the primordial
follicles and the graftʼs muscle tissue is marked with a star.
Discussion
Ovarian tissue cryopreservation in combination with later re-transplantation for fertility
preservation in prepubertal females is currently the only realistic preservation option
for these patients [1 ], [2 ], [3 ]. Nevertheless, only few autografting cases can be found for prepubertal tissue in
the literature [10 ], [11 ]. As ovarian function and developmental potential in frozen thawed tissue can be
tested very well in vivo via xenotransplantation into immunodeficient mice [20 ] or in vitro by using the LIFE/DEAD assay [21 ], this reluctant usage of frozen thawed tissue might be caused by safety concerns.
In terms of safety, previous studies show contrasting results: Lotz et al. [22 ] as well as Greve et al. [23 ], for instance, could not detect any malignant cell contamination and/or reintroduction
of malignancy, whereas Dolmans et al. [24 ] reported leukemia induction after xenotransplantation. Hence, more studies would
be necessary to clarify if ovarian tissue re-transplantation after cancer treatment
is safe or not; for each malignancy entity separately.
Using the xenografting method to restore fertility would be a potential capability
to circumvent the risk of reintroducing cancer through reimplantation of tissue which
contains malignant cells; in short, xenografting potentially contaminated tissue with
the objective of receiving mature oocytes for in vitro fertilization and later embryo
replacement. From the methodical point of view, it would be a convenient technique,
as cancer cells are not able to pass the Zona pellucida. Nevertheless, there are ethical
and potential safety issues, which are not answered yet and have to be addressed in
further research [20 ].
A further approach for safety reasons is in vitro maturation (IVM) of oocytes; different
options are imaginable and already subject of intensive research: For instance, in
vitro culturing of whole human ovarian tissue or isolated follicles alone with the
aim to activate follicle development and receive mature oocytes for cryopreservation
or subsequent in vitro fertilization; an additional IVM period might be necessary
in case collected oocytes are not completely mature. This method has successfully
been used in some animal models, but not with human tissue so far (among others, reviewed
in [25 ], [26 ], [27 ], [28 ]). Another possibility is oocyte aspiration from removed ovarian tissue prior to
cryopreservation, combined with IVM culturing of oocytes either before freezing or
after warming; this has already been done with prepubertal tissue [29 ].
Another, very promising alternative to autografting of frozen-thawed ovarian tissue
is transplanting an artificial ovary. During this method, isolated follicles from
ovarian tissue attached to or encapsulated in an artificial matrix are provided for
transplantation instead of whole tissue; the main idea is to get rid of potential
malignant cells during this processing and thus avoid reintroducing cancer. Altogether,
there are many different basics approaches, most of them reviewed in [30 ] and a recent report by Laronda et al. including 3D printed scaffold as artificial
ovary matrix [31 ].
Our histological evaluation showed almost all developmental stages of follicles (primordial,
primary, preantral as well as antral follicles) and several primordial follicle clusters.
This is in accordance with the findings by Luyckx et al. [9 ] who investigated xenografting of prepubertal ovarian tissue for 21 weeks, partly
administering follicle stimulating hormone (FSH); they showed that a high number of
follicles survived the freezing-thawing as well as the transplantation process and
retained developmental capacity. Interestingly, we could observe a conformable follicular
development without any exogenous hormone stimulation. To date, this study by Luyckx
et al. is the only one showing such results for prepubertal human ovarian tissue.
The described repetitive retrieval of a maturing as well as a mature oocyte from the
same xenograft of prepubertal human ovarian tissue without exogenous hormone stimulation
is, to our best knowledge, the first report of that kind. Retrieving mature oocytes
from unstimulated xenografts corresponds with similar results from Lotz et al. [32 ]; nevertheless, they received these oocytes from frozen-thawed ovarian fragments
from adult women and not from prepubertal tissue. In summary, Lotz et al. compared
two stimulation protocols (hMG alone and hMG + GnRHa, both groups with additional
hCG administration) as well as the non-stimulated control in xenografted SCID mice
and received one maturing and two metaphase II oocytes. One of the metaphase II oocytes
originated in the control without exogenous stimulation and in absence of hCG supply
[32 ]. Receiving mature oocytes without any stimulation and/or hCG administration is in
conflict with parts of the discussion in Gook et al. [33 ]. Gook et al. also worked with frozen-thawed ovarian tissue from adult women, xenografted
in SCID mice. They reported 32 antral follicles with COC via histological evaluation,
five of them containing metaphase II oocytes; gonadotropin stimulation as well as
hCG administration were performed during grafting time. With this, Gook et al. described
the first metaphase II oocyte development from frozen-thawed human ovarian tissue
in 2003. In their discussion, they attribute the fail to gain mature oocytes in previous
studies to lacking exogenous hormone administration. Further xenografting studies
also reported the development of maturating or mature oocytes in frozen-thawed human
ovarian tissue; all of these studies worked with tissue from adult females and supplied
exogenous hormones [34 ], [35 ], [36 ]. Meanwhile, the present finding as well as those reported in [17 ] and [32 ] show that full oocyte maturation in cryopreserved human ovarian tissue without exogenous
hormones is possible. Nevertheless, the follicle maturation process in human ovarian
tissue takes a couple of months [37 ]; this makes xenografting studies with the aim to retrieve metaphase II oocytes from
human tissue difficult and therefore rare. Obtaining mature oocytes from ovarian tissue
grafts without hormonal stimulation might contribute to clinical benefit since prolonged
stimulation with FSH or luteinizing hormone (LH) can cause primordial follicle loss
in xenografts [38 ] or in vitro [39 ].
In the present study, two metaphase II oocytes could be obtained at a distance of
37 days without any administration of exogenous hormones. A full human cycle has a
length of 23 – 35 days; although, 5% of all women go beyond 35 days [40 ]. In comparison, the spontaneous cycle length in mice is 4 – 5 days; 2 days follicular
phase and 2 – 3 days luteal phase [41 ]. Accordingly, spontaneous follicle and oocyte maturation in this prepubertal xenograft
are more comparable to the human reproductive system. This indicates that the human
ovarian tissue might be able to synchronize the hypothalamus-hypophysis-axes of the
mouse to the physiological human cycle; however, the essence from the present report
is limited since a single observation is described. Future studies would be necessary
to examine such a hypothesis.
In the present report, follicular diameter was approximately 7 mm both times, which
is comparable to the findings from Solaimani et al. [34 ]; they reported antral follicles of 7, 10 and 12 mm in diameter for xenograft position
in the back muscle of SCID mice. Both harvested oocytes from our prepubertal ovarian
tissue xenograft seemed normal with regard to their morphological appearance and also
with regard to oocyte diameter and Zona pellucida width. Nevertheless, their fertilization
as well as their developmental capacity cannot be assessed by morphological criteria
alone.
Conclusion
Repetitive oocyte retrieval cycles can be achieved with non-stimulated xenografts
of prepubertal ovarian tissue; this highlights the potential of prepubertal ovarian
tissue. The results further indicate that the human ovarian tissue might be able to
synchronize the hypothalamus-hypophysis-axes of the mouse to the physiological human
cycle; future studies are necessary to investigate this hypothesis.
Funding
This work in part supported by Wilhelm Sander Foundation (reference no. 2012.127/1),
Munich, Germany.
