Matrix-free 3D culture supports human follicular development from the unilaminar to the antral stage in vitro yielding morphologically normal metaphase II oocytes

STUDY QUESTION Can group culture with stage-specific anti-Müllerian hormone (AMH) modulation support human follicular development and oocyte maturation in vitro? SUMMARY ANSWER In the presence of FSH, AMH supplementation at the secondary-to-early antral stage followed by AMH depletion promotes the coordinated growth and function of human follicles during group culture, thereby yielding mature oocytes. WHAT IS KNOWN ALREADY Stage-specific AMH modulation promotes in-vitro development of nonhuman primate follicles. The group culture method supports nonhuman primate follicle growth from the primary to antral stage, producing developmentally competent oocytes. STUDY DESIGN, SIZE, DURATION Ovarian tissue samples were collected from 19 patients of reproductive age (22–47 years old having menstrual cycles) who underwent oophorectomy or hysterectomy for clinical purposes. Tissue pieces were cultured in a matrix-free system for 3 weeks followed by isolation of follicles for the subsequent 6-week individual or group culture. PARTICIPANTS/MATERIALS, SETTING, METHODS Pieces of ovarian cortical tissue were cultured to support primordial follicle activation and early-stage follicle growth. Secondary follicles isolated from cultured tissue were then randomly assigned to two groups for individual culture: control and AMH modulation, i.e., recombinant human AMH protein supplementation during the secondary-to-early antral stage followed by the addition of neutralizing anti-human AMH antibody. Secondary follicles were also cultured in groups with the same AMH modulation. Follicle survival, growth, steroid hormone and paracrine factor production, steroidogenic protein expression, as well as oocyte maturation and morphology were assessed. MAIN RESULTS AND THE ROLE OF CHANCE Follicles grew to the secondary stage during 3 weeks of ovarian tissue culture. In-vitro-developed follicles expressed AMH and levels of secreted AMH increased (P < 0.05) in the culture media over time. Secondary follicles isolated from cultured ovarian tissue survived and grew to the antral stage during 6 weeks of individual follicle culture. In-vitro-developed antral follicles produced granulosa and theca cell-derived steroid hormones and paracrine factors, which were detectable in the culture media. Germinal vesicle oocytes obtained from cultured follicles exhibited a perinucleolar chromatin rim configuration. AMH modulation did not alter follicle survival or oocyte maturation relative to those of the control follicles. However, follicle diameters, as well as steroid hormone and paracrine factor production, increased (P < 0.05) in the AMH-modulation group compared with the control group. Secondary follicles isolated from cultured ovarian tissue formed aggregates and grew to the antral stage during 6 weeks of group culture. In-vitro-developed antral follicles expressed steroidogenic enzymes and secreted steroid hormones were detectable in the culture media. Oocytes obtained from cultured follicle aggregates with AMH-modulation progressed to the metaphase II stage after IVM, containing a normal-sized first polar body and meiotic spindle. Oocytes exhibited a typical ultrastructure. LIMITATIONS, REASONS FOR CAUTION Follicles were obtained from fresh ovarian tissue of adult patients. Oocyte maturation rates were relatively low and oocytes were assessed by morphological evaluation. Owing to the lack of a control group, the beneficial effects of AMH modulation remained undetermined for the group culture in this study. WIDER IMPLICATIONS OF THE FINDINGS Stage-specific AMH modulation supports human follicular development in the matrix-free group culture, which is consistent with previously reported AMH actions on growing follicles in nonhuman primates. Oocytes generated by in-vitro-developed follicles achieve meiotic maturation with a typical morphology and ultrastructure, which supports in-vitro follicle maturation as a potential approach for fertility preservation in women. STUDY FUNDING/COMPETING INTEREST(S) NICHD R01HD082208 and NIH Office of the Director P51OD011092. The authors have no competing interest to declare. TRIAL REGISTRATION NUMBER N/A.

anti-Müllerian hormone / oocyte maturation / matrix free / fertility preservation / group culture

For women at risk of or experiencing primary ovarian insufficiency, fertility preservation is recommended to retain their reproductive ability (Chae-Kim and Gavrilova-Jordan, 2018). To date, available options of female fertility preservation include embryo, oocyte, and ovarian tissue cryopreservation (Practice Committee of the American Society for Reproductive Medicine, 2019). Cryopreservation of ovarian cortical tissue theoretically represents an efficient way of preserving thousands of follicles at one time. This method has been proposed principally to preserve fertility for prepubertal girls when ovarian stimulation and IVF are not options, and for female patients who cannot delay cancer treatment for ovarian stimulation and oocyte retrieval (Donnez et al., 2010; Wallace et al., 2014). Because ovarian cortical tissue contains only primordial and early-growing (intermediary, primary, and secondary) follicles, mature oocytes cannot be obtained unless those follicles resume development and grow to the antral stage. Currently, autologous transplantation is performed clinically to restore follicular development in vivo (Practice Committee of the American Society for Reproductive Medicine, 2019), and pregnancy stemming from ovarian tissue cryopreserved in early childhood (≤6 years old) remains to be achieved. However, there is a legitimate concern regarding the potential for reintroducing cancer cells present in ovarian tissue back to patients, especially those having systemic hematologic malignancies (Dolmans et al., 2010). Therefore, it is critical to establish culture systems for efficient follicular development in vitro to obtain mature oocytes. Primordial follicles in fresh or cryopreserved ovarian cortex could activate spontaneously and grow to the secondary stage during tissue culture for 15 days (Wright et al., 1999; Novella-Maestre et al., 2015). Although successful culture of early-growing follicles from cryopreserved human ovarian tissue has not been reported,. secondary follicles isolated from fresh tissue grew to the antral stage during 6 weeks of individual follicle culture and yielded metaphase II (MII) oocytes via additional IVM (Xiao et al., 2015). Therefore, a multistep culture strategy has been suggested to support unilaminar (primordial, intermediary and primary) follicle development to the antral stage to generate mature oocytes. Currently, fresh ovarian cortical tissue culture followed by individual culture of isolated secondary follicles has produced oocytes capable of resuming meiosis via IVM (McLaughlin et al., 2018). However, the oocytes extruded unusually large polar bodies indicating possible abnormal spindle position and/or inter-chromosomal spacing (Coticchio et al., 2013). Optimization of culture conditions is needed to further enhance coordinated follicular development in vitro with improved oocyte morphology. Intraovarian regulatory factors have been suggested to facilitate folliculogenesis in vitro, of which anti-Müllerian hormone (AMH) exhibits stage-dependent actions. AMH appears to be stimulatory during the early stage of follicular development, but is inhibitory after antrum formation, as suggested by studies in humans and nonhuman primates (Schmidt et al., 2005; Xu et al., 2016a). Promoting primary and secondary follicle growth in vitro by AMH supplementation was demonstrated in human ovarian tissue culture and rhesus macaque individual follicle culture (Schmidt et al., 2005; Xu et al., 2016a), although enhancing antral follicle maturation in vitro by blocking endogenous AMH action was only reported in rhesus macaque individual follicle culture (Xu et al., 2016a). A recent study established a matrix-free 3D culture system to support in-vitro development of nonhuman primate follicles with AMH modulation. AMH addition at the secondary-to-early antral stage followed by AMH depletion promoted the growth of individually cultured rhesus macaque follicles with increased follicle diameters, paracrine factor production (e.g. vascular endothelial growth factor, VEGF), and oocyte diameters (Xu et al., 2018a). Compared with an encapsulation approach, the matrix-free system allows for efficient diffusion of macromolecules during culture (Xu et al., 2018b). Therefore, a protocol with similar tissue handling, media composition, and oxygen tension was applied to human follicle culture to test the hypothesis that stage-specific AMH modulation enhances follicular development from the unilaminar to antral stage in vitro to generate mature oocytes via a multistep culture. Experiments were performed to assess follicular endocrine and paracrine function, steroidogenic enzyme expression, as well as oocyte maturation and morphology.

Ethical approval for the collection and use of human tissue Ovarian cortical tissue samples and clinical information were obtained from the Oregon Ovarian Cancer Registry and Tissue Repository at the Oregon Health & Science University (OHSU). Patients recruited were women of reproductive age not experiencing menopause (22–47 years old; n = 19) who underwent oophorectomy or hysterectomy due to benign or malignant gynecologic conditions (Table I) (Supplementary Fig. S1). Discarded ovarian tissue (∼10–15% of one ovary) was collected intraoperatively, with written informed consent, and de-identified. The experimental protocol was approved by the Institutional Review Board (#921) at OHSU. Table I. | Experiment | Patient | Age years | BMI kg/m2 | Racial/ethnic category | Menstrual cycle | Diagnosis | Follicles (n) | |---|---|---|---|---|---|---|---| | Individual follicle culture | 1 2 3 4† 5 6*,† | 47 38 36 47 36 34 | 48 21 22 NA NA NA | White/non-Hispanic White/non-Hispanic White/non-Hispanic White/Hispanic White/Hispanic White/non-Hispanic | NA NA NA NA NA NA | Benign ovarian cyst Ovarian fibroma Lymphangioleiomyomatosis Breast cancer Breast cancer Breast cancer | - 3 10 3 16 8 | | 7 8 9 10*,† | 43 41 38 31 | 41 23 57 28 | White/non-Hispanic White/non-Hispanic White/non-Hispanic White/Hispanic | NA Luteal phase Amenorrhea/Mirena IUD Amenorrhea | Pelvic and renal mass Adnexal mass Malignant neoplasm of uterus Cervical cancer | 10 10 16 - | | | Group follicle culture | 11* 12 13 14 15 | 45 40 37 30 22 | 48 15 26 NA 50 | White/non-Hispanic White/non-Hispanic White/non-Hispanic White/non-Hispanic NA | Menometrorrhagia Luteal phase Amenorrhea/Mirena IUD NA Amenorrhea/Mirena IUD | Ovarian cystadenoma Uterine leiomyosarcoma Adnexal mass Endometriosis Vaginal mass | 3 8 3 10 3 | | 16 17 18 | 45 39 37 | 35 25 22 | White/non-Hispanic White/non-Hispanic White/non-Hispanic | Luteal phase Luteal phase NA | Endometrial adenocarcinoma Endometrial cancer Adnexal mass | - - - | | | 19 | 23 | 23 | White/non-Hispanic | Amenorrhea | Pelvic mass | – | Patients received chemotherapy prior to the surgery. †Patients received radiation therapy prior to the surgery; -Patients excluded from the study because of absence of follicles in the collected ovarian tissue sample. IUD, intrauterine device; NA, not available. Ovarian tissue samples were kept in HEPES-buffered holding media (Cooper Surgical, Trumbull, CT, USA) at 37°C and transferred to the laboratory within 1 h. Tissue samples from six of the 19 patients did not contain any follicles, and therefore were excluded (Table I) (Supplementary Fig. S1). Ovarian tissue culture Tissue culture Ovarian cortical tissue was cultured to initiate primordial follicle activation and to support follicle growth to the secondary stage. The cortical tissue layer was obtained using a Stadie-Riggs tissue slicer (Thomas Scientific, Swedesboro, NJ, USA), and was cut mechanically into ∼1.4 × 10−2 mm3 pieces (diameter = ∼300 μm) using needles, with each piece containing >5 follicles as determined under a dissecting scope. Tissue pieces (n = 1–7/patient) were distributed individually into wells of the ultra-low attachment microplates (Corning, New York, NY, USA) for matrix-free culture at 37°C and 5% CO2 for 3 weeks. Each well contains 200 µl alpha minimum essential medium (Thermo Fisher Scientific, Waltham, MA, USA) with 15 mIU/ml recombinant human FSH (Merck & Co., Kenilworth, NJ, USA), 6% (v/v) human serum protein supplement (Cooper Surgical), 5 µg/ml insulin, 5 µg/ml transferrin, 5 ng/ml sodium selenite, 0.5 mg/ml bovine fetuin, and 10 µg/ml gentamicin (Sigma-Aldrich, St Louis, MO, USA). Media (100 µl) was collected and replaced every other day throughout 3 weeks of culture. It is known that primordial follicles do not produce AMH, and AMH expression begins when primordial follicles are activated to grow and become primary follicles (Weenen et al., 2004). AMH secreted by a primary or secondary follicle is detectable in the media during individual follicle culture (Xu et al., 2013). In order to rule out the presence of primary and secondary follicles in cultured ovarian tissue, media samples were pooled by week for each patient, and AMH concentrations were measured using an AMH ELISA kit (AL-105; AnshLabs, Webster, TX, USA) (Xu et al., 2016b). Follicle growth To evaluate follicle growth during culture, tissue pieces were photographed weekly using an Olympus CK-40 inverted microscope and an Olympus DP11 digital camera (Olympus Imaging America, Center Valley, PA, USA). At the end of culture, selected pieces were fixed in 4% paraformaldehyde-PBS for 3 h at room temperature, embedded in HistoGel (Thermo Fisher Scientific) and paraffin. Five micrometer sections were cut and stained with hematoxylin and eosin (Xu et al., 2018a). Growing follicles were further assessed for AMH protein production using immunohistochemistry (Xu et al., 2016c). Briefly, tissue sections were incubated with mouse anti-human AMH antibody (1:75; MAB1737; R&D Systems, Minneapolis, MN, USA) at 4°C overnight. Antibody preabsorbed with blocking peptide (1737-MS; R&D Systems) was used as the negative control. Sections were then incubated with the secondary antibody, processed using a VECTASTAIN Elite ABC Kit (Vector Laboratories), and counterstained using hematoxylin. Histological images were captured using an Olympus BX40 microscope and an Olympus DP72 digital camera (Olympus Imaging America). Individual follicle culture Follicle culture To prevent a potential negative impact from atretic follicles in cultured ovarian tissue, growing follicles were isolated for subsequent follicle culture. Individual follicle culture was performed to study the direct effects of AMH modulation, precluding follicle–follicle interactions. At the end of ovarian tissue culture, secondary follicles (diameter = 125.6–223.2 μm) were isolated mechanically using needles without enzyme treatment to maintain basement membrane integrity, distributed individually into wells of the ultra-low attachment microplates, and cultured in the media described above for 6 weeks. Based on studies in nonhuman primate follicle culture (Xu et al., 2018a), follicles from each of the eight patients were randomly assigned to two treatment groups with 38 follicles/group (1–8 follicles/patient/group; Table I) (Supplementary Fig. S1): a control group, and an AMH-modulation group: 100 ng/ml recombinant human AMH protein (1737-MS; R&D Systems; dose based on a previous study in rhesus macaques; Xu et al., 2018a) supplementation during weeks 0–3 (secondary-to-early antral stage) followed by 100 ng/ml neutralizing anti-human AMH antibody (MAB1737; R&D Systems; dose based on a previous study in rhesus macaques; Xu et al., 2018a) addition during weeks 4–6 (antral stage). Media (100 µl) was collected and replaced every other day throughout 6 weeks of culture. Follicle survival and growth Follicles were photographed weekly to evaluate follicle survival and antrum formation. Follicles were considered to undergo atresia if the oocyte was degenerated or extruded from the follicle (Supplementary Fig. S2), and were excluded from the subsequent culture. Follicle diameters were measured using Image J software (National Institutes of Health, Bethesda, MD, USA) (Xu et al., 2016a). At the end of culture, selected follicles were fixed, embedded, and sectioned for hematoxylin and eosin staining, as described above. It is known that AMH expression and secretion increases during early-stage follicle growth, and peaks after follicles form an antrum and become small antral follicles (Weenen et al., 2004; Xu et al., 2013). To further assess follicle growth, media samples from the control culture were pooled by week for each patient, and AMH concentrations were measured. Follicular steroidogenic and paracrine function To determine the function of antral follicles developed in vitro, media samples from culture week 6 were pooled for each patient and treatment for the analyses of known ovarian steroid hormones and paracrine factors by the Endocrine Technologies Core at the Oregon National Primate Research Center, OHSU (Xu et al., 2018a). Granulosa cell-derived estradiol (E2) was assayed using an Immulite 2000 platform (Siemens Healthcare Diagnostics, Deerfield, IL, USA). Theca cell-derived androstenedione (A4) was assayed by ELISA using an AA E-1000 kit (Rocky Mountain Diagnostics, Colorado Springs, CO, USA). Activin A and VEGF were assayed using an Activin A ELISA Kit (AnshLabs) and a Human VEGF Quantikine ELISA Kit (R&D Systems), respectively. Insulin-like growth factor (IGF) 1 and 2 were measured using a Human IGF-I/IGF-1 Quantikine ELISA Kit (R&D Systems) and a RayBio Human IGF-2 ELISA Kit (RayBiotech, Peachtree Corners, GA, USA), respectively, based on the manufacturer’s instructions. Assays were validated for follicle culture media, and measurement from blank media without culture was used for background subtraction. The intra-assay co-efficient of variation (CV) and inter-assay CV were 4% and 8%, respectively, for IGF1, and 10% and 12% for IGF2. Oocyte development At the end of follicle culture, antral follicles were treated with 100 ng/ml recombinant hCG (EMD Serono, Rockland, MA, USA) and 10 ng/ml epidermal growth factor (BD Biosciences, San Jose, CA, USA) in the media for 34 h to test oocyte maturation capacity in intact follicles (Xu et al., 2018a). Cumulus–oocyte complexes (COCs) were dissected out using needles in Tyrode’s albumin lactate pyruvate–HEPES–bovine serum albumin (0.3% v/v) media. Oocytes were denuded mechanically by gentle pipetting to assess diameter and meiotic status using the same inverted microscope described above (Xu et al., 2018a). To assess oocyte morphology, selected germinal vesicle (GV) oocytes were fixed in 4% paraformaldehyde–PBS for 30 min at room temperature for immunofluorescence staining (Xu et al., 2011). Briefly, oocytes were incubated with anti-α-tubulin antibody (1:100; clone DM1A; Sigma-Aldrich) overnight at 4°C followed by incubation with Alexa 633 goat anti-mouse IgG (1:500; Invitrogen, Carlsbad, CA, USA). F-actin was probed with Alexa Fluor 488 Phalloidin (1:50; A12379; Invitrogen) and chromatin was labeled with 5 μM DAPI (Invitrogen). Images were captured using a Leica SP5 confocal-equipped DM6000 CFS microscope (Leica Microsystems, Bannockburn, IL, USA). Group follicle culture Follicle culture To maintain interfollicular interactions, group culture was performed based on results from studies in nonhuman primates (Xu et al., 2018a). At the end of ovarian tissue culture, secondary follicles (diameter = 125.0–198.4 μm) were isolated from tissue of five patients, distributed into wells of the ultra-low attachment microplates with 3–5 follicles per well (3–10 follicles/patient; Table I), and cultured in media with the same AMH modulation as described above for 6 weeks (Supplementary Fig. S1). Media (100 µl) was collected and replaced every other day throughout 6 weeks of culture. Follicle growth During the matrix-free group culture, secondary follicles attach to each other and form aggregates, as described in nonhuman primate follicle culture (Xu et al., 2018a). To evaluate follicular antrum formation during culture, follicle aggregates were photographed weekly. At the end of culture, selected follicle aggregates were fixed, embedded, and sectioned for hematoxylin and eosin staining, as described above for ovarian tissue fixation. Follicular development in vitro was further assessed by steroid hormone production. Media samples from culture week 6 were pooled for each patient for analyses of E2 and A4. Follicular protein expression To determine follicular cell differentiation in vitro, selected follicle aggregates were assessed for steroidogenic enzyme expression using immunohistochemistry (Xu et al., 2013). Briefly, 5 µm follicle aggregate sections were incubated with goat anti-human CYP19A1 antibody (1:100; sc-14245 targeting aromatase in granulosa cells; Santa Cruz Biotechnology, Santa Cruz, CA, USA) and mouse anti-human CYP17A1 Antibody (1:200; sc-374244 targeting 17α-hydroxylase/17,20-lyase in theca cells; Santa Cruz Biotechnology) at 4°C overnight. Sections were then incubated with secondary antibodies, processed and imaged, as described above for ovarian tissue immunohistochemistry. Oocyte maturation At the end of group culture, COCs were dissected out from antral follicles in follicle aggregates for individual IVM to prevent potential impact from adjacent follicles (McLaughlin et al., 2018). COCs were cultured individually in 100 μl SAGE In-Vitro Maturation Media (CooperSurgical) supplemented with 75 mIU/ml FSH and 75 mIU/ml LH (EMD Serono) in Tri-Gas (90% N2, 5% CO2, and 5% O2) and 100% humidity at 37°C for 48 h. Oocytes were denuded mechanically by gentle pipetting to assess meiotic status (Xu et al., 2018a). The meiotic spindle of MII oocytes was visualized by polarized microscopy using an Oosight Imaging System (Cambridge Research & Instrumentation, Woburn, MA, USA) (Tachibana et al., 2009). To further assess oocyte ultrastructure, MII oocytes were fixed and sectioned for transmission electron microscopy using an FEI Tecnai with iCorr microscope (Thermo Fisher Scientific) by the Multiscale Microscopy Core at OHSU (Segovia et al., 2017). Statistical analyses Statistical analysis was performed using SAS software (SAS Institute, Cary, NC, USA). Data represent 13, 8 and 5 patients in ovarian tissue, individual follicle, and group follicle culture experiments, respectively. Because follicles from each patient were randomly distributed into the control and AMH-modulated groups during individual follicle culture, the Wilcoxon signed-rank test was used to evaluate differences in follicle survival and antrum formation with each patient as an experimental unit. The two-sample Student’s t-test was used to analyze diameters of follicles and oocytes, as well as media steroid hormone and paracrine factor concentrations with each follicle or oocyte as an experimental unit. One-way ANOVA was used to analyze media AMH concentrations with each tissue piece or follicle as an experimental unit. Differences were considered significant at P < 0.05 and values are presented as the mean ± SEM.

Ovarian tissue culture Follicle growth In all ovarian tissue pieces cultured (n = 40 from 13 patients), primordial/intermediary follicles developed to the primary/secondary stage in 3 weeks (Fig. 1A and B). Follicles with a centrally-located oocyte surrounded by a single layer or multiple layers of granulosa cells were visible in cultured tissue at week 3, indicating the presence of primary and secondary follicles, respectively (Fig. 1B and D). While AMH protein was not detectable (<0.023 ng/ml) in the media during the first week of culture, it became measurable with increasing (P < 0.05) concentrations at weeks 2 and 3, suggesting the presence of primordial and/or intermediary follicles in the beginning of tissue culture and the subsequent follicle growth (Fig. 1C). Figure 1D depicts the early stage of follicular development during ovarian cortical tissue culture. Histological staining revealed a follicular morphology similar to that observed in human follicles developed in vivo (Gougeon, 1996). Primary follicles contained a single layer of cuboidal-shaped granulosa cells surrounding a morphologically normal GV oocyte. Secondary follicles contained a relatively large GV oocyte surrounded by multiple layers of granulosa cells without exhibiting an antral cavity. Follicles resided in typical stroma cells of the ovarian tissue. Specific immunostaining for AMH protein concentrated in the cytoplasm of granulosa cells in growing follicles (Fig. 1E). Positive AMH staining was weak in primary follicles, and became intense when follicles progressed to the secondary stage. AMH staining was absent in the negative control tissue section (Fig. 1F). Individual follicle culture Follicle survival and growth When secondary follicles isolated from cultured ovarian tissue (n = 38 from 8 patients) were further cultured individually (Fig. 2A), ∼50% of them survived for 6 weeks under the control conditions (Table II). Surviving follicles, if they grew, started to form an antrum around culture week 3. Only ∼5% of the surviving follicles remained at the secondary stage throughout 6 weeks of culture, which were known previously as ‘no-grow follicles’ (Xu et al., 2013; Yin et al., 2016). The majority (∼95%) of surviving follicles grew to the antral stage with average diameters greater than 600 µm, which were growing follicles including ‘slow-grow’ and ‘fast-grow’ follicles (Fig. 2B) (Table II). AMH protein was detectable in the media during the first week of culture. Media AMH concentrations increased (P < 0.05) in the following weeks and plateaued at week 4, indicating a typical follicle growth pattern from the secondary to antral stage (Fig. 2C). Figure 2D depicts an antral follicle that developed during individual follicle culture. Histological staining revealed a follicular morphology similar to that observed in human follicles developed in vivo (Gougeon, 1996), i.e. a spherical shape with an antral cavity, an oocyte surrounded by cumulus cells, an intact granulosa layer, and presumptive theca cells. Table II. | Parameters | Control | AMH-modulation | P value | |---|---|---|---| | Survival (%) | 50.0 ± 21.8 | 63.2 ± 14.8 | 0.32 | | Antrum formation (%) | 94.7 ± 8.0 | 95.8 ± 2.8 | 0.22 | | Diameter (μm) | 617.7 ± 23.8 | 676.1 ± 20.8 | 0.04* | | Estradiol (pg/ml) | 2095.9 ± 776.6 | 4136.4 ± 693.0 | 0.04* | | Androstenedione (pg/ml) | 210.8 ± 105.9 | 277.5 ± 56.1 | 0.23 | | Activin A (ng/ml) | 0.04 ± 0.01 | 0.06 ± 0.01 | 0.05* | | IGF2 (ng/ml) | 0.6 ± 0.1 | 1.5 ± 0.3 | 0.02* | | VEGF (ng/ml) | 2.0 ± 0.4 | 3.4 ± 0.5 | 0.02* | Follicles from each of the eight patients were randomly assigned to two treatment groups with 38 follicles/group for follicle culture. The Wilcoxon signed-rank test was used to evaluate differences in follicle survival and antrum formation with each patient as an experimental unit. The two-sample Student’s t-test was used to analyze follicle diameters and steroid hormone/paracrine factor concentrations in the media with each follicle as an experimental unit. Values are the mean ± SEM. Anti-Müllerian hormone (AMH)-modulation, AMH supplementation during weeks 0–3 and depletion during weeks 4–6 of culture. IGF2, insulin-like growth factor 2; VEGF, vascular endothelial growth factor. Significant difference between culture groups (P < 0.05). AMH modulation during individual follicle culture (n = 38 from eight patients) did not alter follicle survival or antrum formation rates relative to those of the control follicles at week 6 (Table II). Although secondary follicle diameters were comparable between the control and AMH-modulation group in the beginning of culture (160.3 ± 7.6 vs. 156.9 ± 5.2 µm; P = 0.35), in-vitro-developed antral follicles from the AMH-modulated group had greater (P < 0.05) diameters at week 6 than those of the control group (Table II) (Supplementary Fig. S3). Follicular steroidogenic and paracrine function When cultured under the control conditions, in-vitro-developed antral follicles produced ovarian steroid hormones E2 and A4, as well as paracrine factors activin A, IGF2 and VEGF, which were measurable in the culture media at week 6 (Table II). However, IGF1 protein was not detectable (<0.06 ng/ml) in any media samples from the control follicle culture. Media concentrations of E2, activin A, IGF2, and VEGF increased (P < 0.05) in the AMH-modulated group compared with those of the control group at culture week 6 (Table II). There were no differences in week 6 media A4 concentrations between follicles cultured under the control and AMH-modulation conditions. IGF1 protein remained undetectable in any follicle culture media samples, despite AMH modulation. Oocyte development All oocytes harvested from antral follicles developed in the control culture were morphologically normal and remained at the GV stage without resuming meiotic division (Table III) (Supplementary Fig. S4). Immunofluorescence staining of a GV oocyte exhibited a typical plasma membrane and abundant transzonal projections extending from the surrounding cumulus cells into the zona pellucida (Fig. 2E and F). Condensed chromatin formed a single ring around the nucleolus-like bodies, i.e. perinucleolar chromatin rim configuration (Fig. 2F) (Sánchez et al., 2015). Table III. | (Number (n) of) | |||| |---|---|---|---|---| | Culture conditions | Antral follicles | Oocytes harvested | Healthy GV | Oocyte diameter (µm) | | Control | 18 | 18 | 18 | 98.3 ± 1.6 | | AMH modulation | 23 | 23 | 23 | 104.8 ± 3.1* | The two-sample Student’s t-test was used to analyze oocyte diameters with each oocyte as an experimental unit. Values are the mean ± SEM. Anti-Müllerian hormone (AMH)-modulation, AMH supplementation during weeks 0–3 and depletion during weeks 4–6 of culture. GV, germinal vesicle oocyte. Significant difference between culture groups (P < 0.05). Oocytes harvested from antral follicles developed in AMH-modulated culture were all morphologically normal GV oocytes without resuming meiotic division (Supplementary Fig. S4). However, oocytes from the AMH-modulated group had larger (P < 0.05) diameters compared with those of the control group (Table III). Group follicle culture Follicle growth Group-cultured secondary follicles in every well of the plate formed a follicle aggregate (n = 7 from five patients), with at least one follicle in each follicle aggregate growing to the antral stage at week 6 (Fig. 3A and B). In-vitro-developed antral follicles in the follicle aggregate exhibited a morphology similar to that of in-vivo-developed human follicles (Gougeon, 1996), including a morphologically normal GV oocyte surrounded by cumulus cells, an antral cavity, multiple granulosa layers, presumptive theca cells, and an intact basement membrane (Fig. 3D). Follicle aggregates secreted an appreciable amount of E2 (28773.8 ± 5372.8 pg/ml) and A4 (4372.2 ± 939.3 pg/ml) into the culture media at week 6, suggesting the presence of well-developed antral follicles. Follicular protein expression While cytoplasmic CYP19A1 staining was detected in granulosa cells, specific cytoplasmic immunostaining for CYP17A1 appeared to concentrate in limited areas of theca cells in antral follicles developed during group culture (Fig. 3E and F) (Supplementary Fig. S5). CYP19A1 and CYP17A1 staining was absent in the negative control sections (Fig. 3E and F inserts). Oocyte maturation At the end of group culture, 14 COCs were harvested from in-vitro-developed antral follicles (Supplementary Table SI). All oocytes were morphologically normal after IVM, with 78.6% of them remaining at the GV stage without resuming meiotic division (Supplementary Table SI) (Supplementary Fig. S4). Three (21.4%) oocytes matured to the MII stage by extruding the first polar body (Figs 3C and 4A and C), with diameters greater than 110 µm (119.4 ± 3.3 μm) (Supplementary Table SI). The first polar body and the meiotic spindle were observed with normal sizes and positions in the MII oocytes, as indicated by polarized microscopy (Fig. 4B) and electron microscopy (Fig. 4D) (Supplementary Fig. S6) (Tachibana et al., 2009; Segovia et al., 2017). Electron microscopy imaging showed chromosomes in the first polar body and the meiotic spindle, as well as a typical plasma membrane with abundant microvilli projecting into the perivitelline space and spherical cortical granules located immediately below the plasma membrane (Fig. 4D insert) (Supplementary Fig. S6). There were numerous organelles dispersed in the cytoplasm, including mitochondria (Fig. 4E). Mitochondria were rounded or oval with few transversal cristae (Fig. 4E insert), and were often associated with the smooth endoplasmic reticulum (SER) and small vesicles to form mitochondria-SER aggregates and mitochondria-vesicle complexes, respectively (Fig. 4E) (Segovia et al., 2017).

Applying matrix-free culture techniques to the multistep in-vitro follicle maturation, the present study achieved human follicular development from the unilaminar to the antral stage. Primordial/intermediary follicles developed to the primary/secondary stage during 3 weeks of ovarian tissue culture. In the presence of FSH and insulin, stage-specific AMH modulation, i.e. AMH supplementation at the secondary-to-early antral stage followed by AMH depletion, improved follicle growth during subsequent follicle culture, with increased antral follicle diameters and coordinated steroid/paracrine factor production. Group follicle culture followed by IVM yielded MII oocytes exhibiting a typical morphology and ultrastructure. Cultured in serum-free media, primordial and intermediary follicles in the ovarian cortical tissue developed spontaneously in vitro. Data are consistent with previous findings in human ovarian tissue culture with or without serum supplementation (Wright et al., 1999). Because AMH expression by granulosa cells is not initiated until after primordial follicles are recruited for active growth (Weenen et al., 2004) and AMH secreted by individually cultured primary/secondary follicles is detectable in the media (Xu et al., 2013). AMH is an excellent marker for the presence of growing follicles beyond the intermediary stage in culture. The fact that AMH was undetectable in the culture media indicated that primordial follicles, if activated, and intermediary follicles had yet to grow actively during the first week of tissue culture. Follicle growth proceeded in the subsequent 2 weeks of culture with increased granulosa cell AMH expression and secretion into the media. One of the primordial follicle activation regulatory mechanisms in the ovary is the Hippo signaling pathway (Kawamura et al., 2013). Previous studies demonstrated that the fragmentation process disrupted Hippo signaling leading to increased primordial follicle activation in human ovarian cortex cultured in vitro and grown in vivo (Kawamura et al., 2013; Grosbois and Demeestere, 2018). In addition, augmented primordial follicle activation in vitro could be due to the loss of inhibitory signals, potentially from the non-cortical regions of the ovary, based on the evidence that a greater proportion of primordial follicles activated during primate ovarian cortical tissue culture than that of rodent whole ovary organ culture (Fortune et al., 1999). Secondary follicles isolated from cultured ovarian tissue developed to the antral stage during individual culture and demonstrated steroid hormone/paracrine factor production. Follicle growth in vitro was improved by the combination of AMH supplementation during the secondary-to-early antral stage and blocking endogenous AMH actions during the antral stage in culture. Data are consistent with previous studies in nonhuman primate follicles isolated from fresh ovarian tissue, wherein AMH promoted secondary follicle growth but inhibited antral follicle maturation in vitro (Xu et al., 2016a). The stimulatory effect of AMH on the early stage of follicular development was also demonstrated in human ovarian tissue culture (Schmidt et al., 2005). E2 and paracrine factors produced by in-vitro-developed antral follicles were elevated following AMH modulation, suggesting increased granulosa and theca cell numbers and/or steroidogenic and paracrine function. E2, activin A, and VEGF are the well-known factors critical for antral follicle maturation, and their increased production by cultured antral follicles following AMH modulation has been reported in nonhuman primate studies (Xu et al., 2018a). IGF1 production was not evident in in-vitro-developed antral follicles, which is consistent with previously observed absence of IGF1 protein expression in the human ovary (el-Roeiy et al., 1993). In contrast to IGF1, antral follicles secreted IGF2 into the culture media, which mimics IGF2 protein expression in antral follicles developed in vivo (el-Roeiy et al., 1993). Despite AMH modulation, GV oocytes in cultured antral follicles did not respond to chorionic gonadotrophin treatment to resume meiosis. Nevertheless, the abundant transzonal projections supported paracrine communication between cumulus cells and the oocyte during in-vitro follicular development (Albertini et al., 2001). The condensed chromatin configuration of the GV nucleus suggested the great potential for these oocytes to acquire meiotic competence with additional growth and cytoplasmic maturation (Sánchez et al., 2015). Interfollicular interactions were maintained when secondary follicles isolated from cultured ovarian tissue grew in groups. Follicles formed aggregates, reassembling ovarian tissue-like morphology with well-developed theca cell layers surrounding antral follicles, which enhanced steroidogenesis. The follicle aggregate structure provided support for, and built connections between, developing follicles, and the paracrine regulation appeared to be achieved to benefit follicular development and oocyte growth with AMH modulation. Following additional IVM, oocytes derived from in-vitro-developed antral follicles resumed meiosis. Results are consistent with previous studies in nonhuman primates, wherein primary follicles developed to the antral stage in group culture with AMH modulation and produced MII oocytes supporting preimplantation embryonic development to the morula stage after IVF (Xu et al., 2018a). Owing to the lack of a control group, the beneficial effects of AMH modulation remained undetermined for the group culture in this study. Although oocyte developmental competence could not be assessed in this study, the typical spindle position in the MII oocytes supported extrusion of a normal-sized first polar body (Coticchio et al., 2013). The appropriate microvilli distribution on the oocyte surface could contribute to effective sperm–oocyte fusion; the formation of mitochondria-SER aggregates is essential for oocyte activation upon sperm binding to induce cortical granule exocytosis; and sufficient numbers of mitochondria–vesicle complexes are required for the nuclear envelope formation of pronuclei and their migration (Swain and Pool, 2008). The typical ultrastructure in the MII oocytes obtained from multiple-step culture exhibited the potential for successful fertilization upon in-vitro insemination. In summary, the present multistep in-vitro follicle maturation suggests that 3 weeks of tissue culture facilitates sufficient early-stage follicle growth, and that matrix-free group culture with stage-specific AMH modulation supports continued follicular development and oocyte maturation. AMH protein and neutralizing antibody doses in culture may be further optimized in a dose–response study using human follicles. Utilizing AMH modulation to regulate primordial follicle activation in vitro also requires additional investigation, since previous studies in human ovarian cortical tissue culture yielded inconsistent results (Schmidt et al., 2005; Carlsson et al., 2006). This improved protocol could be applied to the in-vitro development of follicles from cryopreserved ovarian tissue with a goal of obtaining mature oocytes. Owing to the high heterogeneity of patients, studies with an increased sample size are needed to determine the relation between ovarian tissue/follicle parameters and patient characteristics, and to identify patients who may benefit from this in-vitro follicle maturation approach for fertility preservation. Future studies in epigenetic profiling are warranted to evaluate possible alterations in genomic imprinting following prolonged in-vitro manipulation. Transcriptomic analysis by single-cell RNA sequencing could also be performed to provide a deeper analysis of the maturity and competence of oocytes obtained from in-vitro-developed follicles compared to oocytes generated in vivo. The findings may increase our understanding of the efficacy and safety of in-vitro follicle maturation following ovarian tissue cryopreservation for clinical applications in female patients desiring fertility preservation because of medical conditions or treatment. Data availability The data underlying this article are available in the article. Supplementary Material Acknowledgments We are grateful for the assistance provided by members of the Assisted Reproductive Technologies Core, the Endocrine Technologies Core, and the Integrated Pathology Core at the Oregon National Primate Research Center, and the Multiscale Microscopy Core at OHSU. We thank Dr. Sonali Joshi at OHSU for managing patients’ clinical information. Authors’ roles F.X., M.S.L., Y.B., A.Y.T., T.P., K.D., M.M. and S.M.M. contributed to acquisition of data. T.P. and J.X. contributed to conception and design. F.X., T.P. and J.X. contributed to analysis and interpretation of data. F.X. and J.X. contributed to drafting the article. M.S.L., Y.B., A.Y.T., T.P., K.D., M.M. and S.M.M. contributed to revising the article critically for important intellectual content. All authors provided the final approval of the version to be published. Funding Research reported in this publication was supported by the National Institutes of Health (NIH) Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD) R01HD082208 and NIH Office of the Director P51OD011092 (ONPRC). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Conflict of interest The authors have none to declare.

- Albertini DF, Combelles CM, Benecchi E, Carabatsos MJ. Cellular basis for paracrine regulation of ovarian follicle development. Reproduction 2001;121:647–653. [DOI] [PubMed] [Google Scholar] - Carlsson IB, Scott JE, Visser JA, Ritvos O, Themmen APN, Hovatta O. Anti-Müllerian hormone inhibits initiation of growth of human primordial ovarian follicles in vitro. Hum Reprod 2006;21:2223–2227. [DOI] [PubMed] [Google Scholar] - Chae-Kim JJ, Gavrilova-Jordan L. Premature ovarian insufficiency: procreative management and preventive strategies. Biomedicines 2018;7:2. [DOI] [PMC free article] [PubMed] [Google Scholar] - Coticchio G, Guglielmo MC, Dal Canto M, Fadini R, Mignini Renzini M, De Ponti E, Brambillasca F, Albertini DF. Mechanistic foundations of the metaphase II spindle of human oocytes matured in vivo and in vitro. Hum Reprod 2013;28:3271–3282. [DOI] [PubMed] [Google Scholar] - Dolmans MM, Marinescu C, Saussoy P, Van Langendonckt A, Amorim C, Donnez J. Reimplantation of cryopreserved ovarian tissue from patients with acute lymphoblastic leukemia is potentially unsafe. Blood 2010;116:2908–2914. [DOI] [PubMed] [Google Scholar] - Donnez J, Jadoul P, Squifflet J, Van Langendonckt A, Donnez O, Van Eyck AS, Marinescu C, Dolmans MM. Ovarian tissue cryopreservation and transplantation in cancer patients. Best Pract Res Clin Obstet Gynaecol 2010;24:87–100. [DOI] [PubMed] [Google Scholar] - el-Roeiy A, Chen X, Roberts VJ, LeRoith D, Roberts CT Jr, Yen SS. Expression of insulin-like growth factor-I (IGF-I) and IGF-II and the IGF-I, IGF-II, and insulin receptor genes and localization of the gene products in the human ovary. J Clin Endocrinol Metab 1993;77:1411–1418. [DOI] [PubMed] [Google Scholar] - Fortune JE, Kito S, Byrd DD. Activation of primordial follicles in vitro. J Reprod Fertil Suppl 1999;54:439–448. [PubMed] [Google Scholar] - Gougeon A. Regulation of ovarian follicular development in primates: facts and hypotheses. Endocr Rev 1996;17:121–155. [DOI] [PubMed] [Google Scholar] - Grosbois J, Demeestere I. Dynamics of PI3K and Hippo signaling pathways during in vitro human follicle activation. Hum Reprod 2018;33:1705–1714. [DOI] [PubMed] [Google Scholar] - Kawamura K, Cheng Y, Suzuki N, Deguchi M, Sato Y, Takae S, Ho C, Kawamura N, Tamura M, Hashimoto S et al. Hippo signaling disruption and Akt stimulation of ovarian follicles for infertility treatment. Proc Natl Acad Sci U S A 2013;110:17474–17479. [DOI] [PMC free article] [PubMed] [Google Scholar] - McLaughlin M, Albertini DF, Wallace WHB, Anderson RA, Telfer EE. Metaphase II oocytes from human unilaminar follicles grown in a multi-step culture system. Mol Hum Reprod 2018;24:135–142. [DOI] [PubMed] [Google Scholar] - Novella-Maestre E, Herraiz S, Rodríguez-Iglesias B, Díaz-García C, Pellicer A. Short-term PTEN inhibition improves in vitro activation of primordial follicles, preserves follicular viability, and restores AMH levels in cryopreserved ovarian tissue from cancer patients. PLoS One 2015;10:e0127786. [DOI] [PMC free article] [PubMed] [Google Scholar] - Practice Committee of the American Society for Reproductive Medicine. Electronic address: [email protected]. Fertility preservation in patients undergoing gonadotoxic therapy or gonadectomy: a committee opinion. Fertil Steril 2019;112:1022–1033. [DOI] [PubMed] [Google Scholar] - Sánchez F, Romero S, De Vos M, Verheyen G, Smitz J. Human cumulus-enclosed germinal vesicle oocytes from early antral follicles reveal heterogeneous cellular and molecular features associated with in vitro maturation capacity. Hum Reprod 2015;30:1396–1409. [DOI] [PubMed] [Google Scholar] - Schmidt KL, Kryger-Baggesen N, Byskov AG, Andersen CY. Anti-Müllerian hormone initiates growth of human primordial follicles in vitro. Mol Cell Endocrinol 2005;234:87–93. [DOI] [PubMed] [Google Scholar] - Segovia Y, Victory N, Peinado I, García-Valverde LM, García M, Aizpurua J, Monzó A, Gómez-Torres MJ. Ultrastructural characteristics of human oocytes vitrified before and after in vitro maturation. J Reprod Dev 2017;63:377–382. [DOI] [PMC free article] [PubMed] [Google Scholar] - Swain JE, Pool TB. ART failure: oocyte contributions to unsuccessful fertilization. Hum Reprod Update 2008;14:431–446. [DOI] [PubMed] [Google Scholar] - Tachibana M, Sparman M, Sritanaudomchai H, Ma H, Clepper L, Woodward J, Li Y, Ramsey C, Kolotushkina O, Mitalipov S. Mitochondrial gene replacement in primate offspring and embryonic stem cells. Nature 2009;461:367–372. [DOI] [PMC free article] [PubMed] [Google Scholar] - Wallace WH, Smith AG, Kelsey TW, Edgar AE, Anderson RA. Fertility preservation for girls and young women with cancer: population-based validation of criteria for ovarian tissue cryopreservation. Lancet Oncol 2014;15:1129–1136. [DOI] [PMC free article] [PubMed] [Google Scholar] - Weenen C, Laven JS, Von Bergh AR, Cranfield M, Groome NP, Visser JA, Kramer P, Fauser BC, Themmen AP. Anti-Müllerian hormone expression pattern in the human ovary: potential implications for initial and cyclic follicle recruitment. Mol Hum Reprod 2004;10:77–83. [DOI] [PubMed] [Google Scholar] - Wright CS, Hovatta O, Margara R, Trew G, Winston RM, Franks S, Hardy K. Effects of follicle-stimulating hormone and serum substitution on the in-vitro growth of human ovarian follicles. Hum Reprod 1999;14:1555–1562. [DOI] [PubMed] [Google Scholar] - Xiao S, Zhang J, Romero MM, Smith KN, Shea LD, Woodruff TK. In vitro follicle growth supports human oocyte meiotic maturation. Sci Rep 2015;5:17323. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Bishop CV, Lawson MS, Park BS, Xu F. Anti-Müllerian hormone promotes pre-antral follicle growth, but inhibits antral follicle maturation and dominant follicle selection in primates. Hum Reprod 2016a;31:1522–1530. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Hennebold JD, Seifer DB. Direct vitamin D3 actions on rhesus macaque follicles in three-dimensional culture: assessment of follicle survival, growth, steroid, and antimüllerian hormone production. Fertil Steril 2016b;106:1815–1820. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Lawson MS, Mitalipov SM, Park BS, Xu F. Stage-specific modulation of antimüllerian hormone promotes primate follicular development and oocyte maturation in the matrix-free three-dimensional culture. Fertil Steril 2018a;110:1162–1172. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Lawson MS, Yeoman RR, Molskness TA, Ting AY, Stouffer RL, Zelinski MB. Fibrin promotes development and function of macaque primary follicles during encapsulated three-dimensional culture. Hum Reprod 2013;28:2187–2200. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Lawson MS, Yeoman RR, Pau KY, Barrett SL, Zelinski MB, Stouffer RL. Secondary follicle growth and oocyte maturation during encapsulated three-dimensional culture in rhesus monkeys: effects of gonadotrophins, oxygen and fetuin. Hum Reprod 2011;26:1061–1072. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Xu F, Lawson MS, Tkachenko OY, Ting AY, Kahl CA, Park BS, Stouffer RR, Bishop CV. Anti-Müllerian hormone is a survival factor and promotes the growth of rhesus macaque preantral follicles during matrix-free culture. Biol Reprod 2018b;98:197–207. [DOI] [PMC free article] [PubMed] [Google Scholar] - Xu J, Xu F, Letaw JH, Park BS, Searles RP, Ferguson BM. Anti-Müllerian hormone is produced heterogeneously in primate preantral follicles and is a potential biomarker for follicle growth and oocyte maturation in vitro. J Assist Reprod Genet 2016c;33:1665–1675. [DOI] [PMC free article] [PubMed] [Google Scholar] - Yin H, Kristensen SG, Jiang H, Rasmussen A, Andersen CY. Survival and growth of isolated pre-antral follicles from human ovarian medulla tissue during long-term 3D culture. Hum Reprod 2016;31:1531–1539. [DOI] [PubMed] [Google Scholar] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Data Availability Statement The data underlying this article are available in the article.