Cryopreservation of human prophase I oocytes collected from unstimulated follicles

FERTILITY AND STERILITY Copyright c 1994 The American Fertility Society Vol. 61, No.6, June 1994 Printed on acid-free paper in U. S. A. Cryopreservation of human prophase I oocytes collected from unstimulated follicles Thomas L. Toth, M.D.t:\: Susan E. Lanzendorf, Ph.D.t Bruce A. Sandow, Ph.D.t Lucinda L. Veeck, M.L.T.t Waleed A. Hassen, M.D.t Keith Hansen, M.D. II Gary D. Hodgen, Ph.D.t The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Norfolk, Virginia; Massachusetts General Hospital, Boston, Massachusetts; and Portsmouth Naval Hospital, Portsmouth, Virginia Objective: To evaluate the cryopreservation of immature human oocytes obtained from unstimulated ovarian tissue. Design: Immature prophase I oocytes were obtained from unstimulated follicles and were either cryopreserved or cultured as controls. Cryopreservation was performed in a programmable freezing machine using one of two protocols. Method I (n = 133) used a one-step addition of cryoprotectant followed by a slow freeze and thaw protocol. With method II (n = 95), the cryoprotectant was added in a stepwise manner with cryopreservation performed in the presence of 0.2 M sucrose followed by rapid freezing and thawing. Setting: Basic research center at a medical school. Patients: Patients undergoing oophorectomy for nonovarian pathology. Main Outcome Measures: Rates of survival and maturation to metaphase II were compared between control oocytes and oocytes cryopreserved with methods I and II. Results: With method I, a survival rate of 15.6% was obtained with 58.3% of surviving oocytes reaching metaphase II after culture compared with 50.0% of nonfrozen control oocytes. Method II produced a survival rate of 43.3% with 27.3% maturing to metaphase II. Maturation of control oocytes for method II was 46.4%. Although the survival rate with method II was significantly higher than with method I, the rate of in vitro maturation to metaphase II showed no difference. Conclusions: These results demonstrate that human prophase I oocytes obtained from unstimulated antral follicles are capable of meiotic maturation after cryopreservation. Fertil Steril1994;61:1077-82 Key Words: In vitro maturation, oocyte, cryopreservation, transmission electron microscopy Investigations on human oocyte cryopreservation have focused on the potential benefits for the Received November 1, 1993; revised and accepted February 16,1994. * Presented in part at the 48th Annual Meeting of The American Fertility Society, New Orleans, Louisiana, November 2 to 5, 1992. t Department of Obstetrics and Gynecology, The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School. :j: Vincent Memorial Gynecology Service, Massachusetts General Hospital. Reprint requests: Susan Lanzendorf, Ph.D., The Jones Insti- Vol. 61, No.6, June 1994 treatment of infertility. Although reports of successful pregnancies after cryopreservation of unfertilized oocytes do exist (1, 2), human oocyte cryopreservation still is considered experimental. In addition, a variety of studies on animal models have demonstrated that exposure of mature metaphase tute for Reproductive Medicine, Department of Obstetrics and Gynecology, Eastern Virginia Medical School, 601 Colley Avenue, Norfolk, Virginia 23507 (FAX: 804-446-8998). II Department of Obstetrics and Gynecology, Portsmouth Naval Hospital. Toth et al. Cryopreservation of human oocytes 1077

II oocytes to low temperatures or cryopreservation materials may result in damage to the meiotic spindle. Because the microtubular spindle of the metaphase II oocyte to which the chromosomes are attached is sensitive to temperature changes, chromatid nondisjunction may occur during cooling and result in aneuploidy after fertilization of thawed oocytes (3-5). In addition, it has been suggested that cryoprotectants may induce a precocious release of cortical granules, resulting in the premature hardening of the zona pellucida (6, 7). Carroll and coinvestigators (8) also report an increased frequency of digyny in fertilized frozen-thawed mouse oocytes. For these reasons, freezing of the immature oocyte may be an alternative approach to the cryopreservation of the female gamete. At this stage, meiosis is arrested in prophase I at the dictyate stage with the chromosomes located within the membrane-bound nucleus or germinal vesicle. The study presented here was designed to evaluate cryopreservation of immature prophase I human oocytes obtained from unstimulated ovaries. Two cryopreservation protocols were used and transmission electron microscopy (TEM) was performed to evaluate post-thaw morphology. MATERIALS AND METHODS Ovarian tissue was received from the Pathology Department at Eastern Virginia Medical School, Norfolk Community Hospital, and Portsmouth Naval Hospital after approval by the Institutional Review Board. Ovarian tissue was obtained from 36 patients undergoing gynecological procedures not involving ovarian pathology. Patients ranged in age from 26 to 45 years (mean, 34.5 ± 5.5 years). No significant difference in age was observed between patients whose oocytes were cryopreserved using method I compared with method II. The average number of oocytes per patient was 6 (range, 1 to 20). Tissue, kept at room temperature (RT), was received in sterile saline within 4 hours of the onset of the surgical procedure. Oocytes were isolated by cutting the tissue into 5-mm square sections, which were then teased apart with blunt needles. Germinal vesicle stage oocytes surrounded by at least three layers of coronal cells were isolated using a dissecting microscope (6 to 40X magnification), rinsed, and held at 37 C in phosphatebuffered saline (PBS) supplemented with 7.5% heat-inactivated human fetal cord serum until all oocytes were collected (-1 hour). At that time, oocytes were distributed into two groups and were cultured either as controls or cryopreserved. 00- cytes from a single patient were cryopreserved using one of the two methods evaluated. Both control and frozen-thawed oocytes were cultured (37 C, 5% CO 2 ) in Ham's F-10 medium (GIBCO Laboratories, Grand Island, NY) supplemented with 7.5% human fetal cord serum. Oocytes in culture were scored daily for the presence or absence of a germinal vesicle and first polar body and classified as either prophase I, metaphase I, or metaphase II stage. Cryopreservation The two methods of oocyte cryopreservation used in this study differed in cooling rates, seeding temperature, thawing methods, and presence or absence of sucrose in freezing medium. Both protocols used 1.5 M 1,2 propanediol as cryoprotectant and a programmable controlled-rate freezer (TS Scientific, Perkasie, PA). Method I Using this method (9), oocytes were loaded directly into cryovials containing 0.3 ml of 1.5 M 1,2 propanediol in PBS. After equilibration in the freezing medium for 30 minutes at RT, the vials were loaded into the controlled-rate freezer and cooled from RT to -6 C at a rate of 1.0 C/min. Oocytes were held at -6 C for 10 minutes during which time they were internally seeded with a 9- inch pasteur pipette dipped in freezing medium and cooled in liquid nitrogen. Oocytes were cooled to -80 C at a rate ofo.5 C/min and then plunged and stored in liquid nitrogen for 1 to 7 days. Oocytes were thawed by placing the cryovials in the controlled-rate freezer precooled to -100 C. Oocytes were warmed to RT at a rate of 8 C/min, removed from the vials, and washed free of cryoprotectant in decreasing concentrations of 1,2 propanediol (1.0, 0.5, and 0.0 M; 5 minutes each) in PBS. Oocytes surviving the procedure were placed in culture. Surviving oocytes had distinct, intact plasma and nuclear membranes. Degenerate oocytes had dark, granular cytoplasm and/or ruptured oolemma and were discarded. Method II With this protocol (10), oocytes were cryopreserved in PBS supplemented with 20% heat-inactivated fetal bovine serum. After the stepwise addition of cryoprotectant (0.5, 1.0, and 1.5 M 1,2 propanediol; 5, 5, and 10 minutes, respectively), 1078 Toth et al. Cryopreservation of human oocytes Fertility and Sterility

oocytes were added to cryovials containing 0.5 ml of the last solution supplemented with 0.2 M sucrose. Oocytes were allowed to equilibrate for 15 minutes at RT before being loaded into the controlled-rate freezer and cooled from RT to -7 C at a rate of 2 C/min. Oocytes were held at -7 C for 10 minutes, during which time they were seeded internally. Oocytes were cooled to -30 C at a rate of 0.3 C/min, plunged, and stored in liquid nitrogen for 1 to 7 days. Oocytes were thawed by gentle agitation in a 37 C water bath followed by dilution of cryoprotectant at RT in the presence of 0.2 M sucrose (1.0, 0.5, 0.0 M; 5 minutes each). As with method I, surviving oocytes were placed in culture. Electron Microscopy Analysis was performed on eight oocytes. 00- cytes were fixed by addition of2.5% glutaraldehyde in 0.1 M cacodylate buffer for ::2:30 minutes. Fixed oocytes were placed on a slide with very little excess fluid and melted agar (1.5%) was placed over the oocytes and allowed to harden (11). A clean razor blade was used to cut the agar into a rectangle of approximately 0.5 X 1.0 mm, in which the oocyte was located. The agar blocks were fixed in glutaraldehyde, and then postfixed for 1 hour in 1 % osmium tetroxide in 0.1 M cacodylate buffer, dehydrated with alcohol, and embedded in Poly/Bed 812 (Polysciences, Inc., Warrington, PA). Thick sections were cut on an LKB Ultratome III Ultramicrotome (LKB Instruments, Inc., Rockville, MD) and stained with toluidine blue. Thin sections were cut with a diamond knife, stained with lead citrate and uranyl acetate, and viewed in a Philips 301 electron microscope (Philips Electonic Instruments, Inc., Mahwah, NJ). Statistics An unpaired Student's t-test was used to determine significant differences between patient's ages. In all other cases, comparisons between cryopreserved and control groups were made using X 2 twoby-two contingency tables. Confidence intervals were calculated by binomial distribution. In all cases, a P value ~ 0.05 was selected as the minimal criterion for statistically significant differences. RESULTS A total of 133 oocytes were used in the evaluation of method 1. Seventy-seven oocytes were frozenthawed, with 12 oocytes surviving (15.6%). After Vol. 61, No.6, June 1994 culture, eight oocytes (66.7%) underwent germinal vesicle breakdown (GVBD), with seven oocytes (58.3%) reaching the metaphase II stage after culture (Table 1). Fifty-six oocytes were used for control purposes to evaluate normal in vitro maturation, with 44 (78.6%) undergoing GVBD and 28 (50.0%) reaching the metaphase II stage. Evaluation of method II used a total of 95 oocytes. Sixtyseven oocytes were frozen and, after thaw, 29 00- cytes (43.3%) survived. Twenty-two surviving oocytes were cultured, with 21 (95.5%) undergoing GVBD and 6 (27.3%) maturing to the metaphase II stage (Table 1). Twenty-eight oocytes were cultured as controls, with 22 (78.6%) undergoing GVBD and 13 (46.4%) maturing to metaphase II. The rates of G VBD and maturation to metaphase II showed no significant difference between each method and its corresponding control (Table 1). The survival rate after freeze-thaw with method II (43.3%) was found to be significantly higher (P < 0.05) than with method I (15.6%). In addition, the rate of GVBD in method II (95.5%) was significantly higher (P < 0.05) than that of method I (66.7%). However, the rate of in vitro maturation of prophase I oocytes to metaphase II after cryopreservation between the two methods was not significantly different. As demonstrated in Figure 1, the projected number of metaphase II oocytes resulting from the cryopreservation of 100 oocytes by each method is similar (9.1 for method I versus 11.8 for method II). For method I, this number can be computed by dividing the number of cryopreserved 00- cytes reaching metaphase II by the total number of oocytes cryopreserved then multiplying by 100 (cryopreservation metaphase II index). However, with method II, seven surviving thawed oocytes underwent microscopic evaluation and cannot be included in the data. The cryopreservation metaphase II index then was computed using the rates of survival, GVBD, and maturation to metaphase II achieved by the data. Eight oocytes were evaluated after processing for electron microscopy and included two prophase I oocytes fixed after freeze-thawing, two control prophase I oocytes fixed at collection from ovarian tissue, one metaphase II oocyte fixed after cryopreservation at the prophase I stage and maturation in vitro, and three metaphase II control oocytes matured in vitro. All frozen-thawed oocytes were cryopreserved using method II. Upon evaluation, both control and thawed prophase I oocytes contained mitochondria, numerous Golgi complexes, and aggregates of smooth endo- Toth et al. Cryopreservation of human oocytes 1079

Table 1 In Vitro Maturation of Human Prophase I Oocytes After Cryopreservation: Comparison of Methods I and II Method I 95% confidence Cryopreserved interval Control No. of oocytes 77 56 Post-thaw survival* 12 (15.6) 6to 23 GVBD* 8 (66.7)t 33 to 92 44 (78.6) Maturation to metaphase II* 7 (58.3) 25 to 83 28 (50.0) Method II 95% 95% 95% confidence confidence confidence interval Cryopreserved interval Control interval 67 28 22 (43.3) 30 to 55 66 to 86 21 (95.5):1: 82 to 100 22 (78.6) 61 to 93 36 to 62 6 (27.3)11 5 to 41 13 (46.4) 25 to 64 * Values in parentheses are percentages. t Germinal vesicle breeakdown in method I not significantly :I: Germinal vesicle breakdown in method II not significantly Maturation to metaphase II in method I not significantly II Maturation to metaphase II in method II not significantly plasmic reticulum evenly distributed throughout the ooplasm. Corona cells extending through the zona pellucida to the oolemma were present in both groups. Nuclei were not present in any of the sections evaluated. Control oocytes contained numerous small vesicles throughout the cytoplasm and appeared to result from swelling of the vesicular smooth endoplasmic reticulum. Frozen-thawed prophase I oocytes contained fewer small vesicles but did contain larger vacuoles located with the cortical region. Few cortical granules were present within the cortex of both prophase I control and frozen-thawed oocytes. No evidence of cortical granule extrusion was seen within the perivitelline space of these oocytes. A lipofuscin-like body (12) was observed in one prophase I control oocyte. One control metaphase II oocyte demonstrated normal organelle and cortical granule distribution with a minimal number of vesicles. The other control oocyte contained numerous expanded vesicles, clumping of organelles, and evidence of cortical.a 70.. "" 0 80 0 0,; Z 100 90 '\ 80 \\ 50 40 30 20 10 Frozenl TMwecI '-.\ '-.\ '-. \ '-. \ \. \---""'...,...... "',..., ~... ~ av80 Method I... Method H --- MI Figure 1 Projected outcome of 100 human prophase I oocytes after cryopreservation: method I versus method II. granule extrusion into the perivitelline space. Sections through chromatin were not obtained. Evaluation of the metaphase II oocyte matured in vitro after cryopreservation at the prophase I stage also demonstrated numerous expanded vesicles with few cortical granules within the cortex. All other organelles were normal in appearance and distribution. This oocyte also contained a single large vacuole and residual bodies were noted within the ooplasm. The metaphase plate was located in the region of the first polar body near the oocyte surface. Chromosomes were visible associated with microtubules cut lengthwise. Overall, no major ultrastructural differences were noted between frozen-thawed and control oocytes, but this may be due to the small sample sizes. DISCUSSION Cryopreservation of the immature oocyte followed by maturation in vitro has been performed successfully in the mouse and rat models (4, 13). Successful fertilization of frozen-thawed oocytes has occurred in the mouse (14). Mandelbaum and co-workers (15) also report the successful cryopreservation of immature human oocytes, although it is unclear as to the technique of freezing and whether the oocytes were collected from stimulated or unstimulated follicles. In the study, 10 of 27 oocytes (37%) survived the procedure, with 2 (20%) maturing to the metaphase II stage after culture in vitro. It is also unclear in the report whether the two maturing oocytes were inseminated or fertilized. The report presented here confirms that immature human oocytes are capable of surviving cryopreservation and maturing to metaphase II after 1080 Toth et al. Cryopreservation of human oocytes Fertility and Sterility

thawing. This study used oocytes collected from ovaries not stimulated with exogenous gonadotropins and compared two different cryopreservation techniques. Method I, a slow freeze-slow thaw protocol, resulted in a low post-thaw survival rate (15.6%) but a high rate of maturation in surviving oocytes (58.3%). In contrast, method II, a more rapid freeze-thaw protocol, demonstrated a higher rate of cryosurvival (43.3%) but a lower rate of maturation (27.3%). However, it should be noted that the large confidence intervals seen in the data from method I (Table 1) may suggest a different outcome had the sample size been larger. Evaluation of the data also demonstrates that both protocols result in similar numbers of mature oocytes (Fig. 1), suggesting a cryopreservation event distinct from survival. It is, therefore, possible that method I may be selecting against those oocytes with less developmental potential. In the present report, TEM was performed to evaluate oocyte ultrastructure after cryopreservation. Both control (n = 4) and cryopreserved (n = 3) oocytes showed numerous swollen, enlarged vesicles of smooth endoplasmic reticulum, a phenomenon also reported in human oocytes cooled to O C (16). Thawed oocytes also contained large vacuoles, which may result from the coalescence of these vesicles and may also indicate atresia (12). As reported in previous studies (12, 17), few cortical granules were observed in prophase I stage oocytes, including those fixed after cryopreservation. In addition, the metaphase II oocyte resulting from a cryopreserved prophase I oocyte and one metaphase II control oocyte also had few cortical granules within the cortex. Because of the low number of oocytes evaluated, it cannot be determined if exposure of immature oocytes to cryoprotectants also will result in a precocious loss of cortical granules, as reported to occur in cryopreserved metaphase II oocytes (6, 7). The meiotic spindle was observed in one metaphase II oocyte after cryc...,reservation and in vitro maturation and appeared normal. It is yet to be determined whether cryopreservation of immature oocytes will result in damage to the meiotic spindle, as in mature oocytes. In the mouse, cryopreserved prophase I oocytes did not demonstrate any major abnormalities in the meiotic spindle at second metaphase after thawing and maturation in vitro (18). Studies to evaluate spindle morphology by immunocytochemical staining in human oocytes are being conducted. In addition, evaluation of the fertilizability of these human oocytes has been initiated in a clinical setting (19). In these studies, method II is used with a slight modification of the cryopreservation medium to PBS supplemented with 15% human fetal cord serum. In the second study (19), the storage time in liquid nitrogen was shortened to 15 minutes to allow for insemination of the thawed oocytes at the same time as control oocytes. Prophase I oocytes collected from hyperstimulated follicles of IVF patients demonstrated a 58.5% survival rate and no significant differences were noted in rates of maturation, fertilization, and cleavage compared with control nonfrozen oocytes. The decreased survival and maturation seen in the present study may be due to the collection of oocytes outside the normal cohort recruited by the ovaries during the menstrual cycle. Oocytes may have been collected in stages of growth and atresia that made them either more susceptible to cryodamage or with less potential for in vitro maturation. The collection of prophase I oocytes from stimulated ovaries may provide access to large cohorts of grown 00- cytes recently recruited by the ovary. Therefore, the purposeful collection of prophase I oocytes after ovarian hyperstimulation may provide a larger, clinically useful pool of oocytes for cryopreservation. The results of this study demonstrate preliminarily that human prophase I oocytes obtained from unstimulated antral follicles are capable of meiotic maturation after cryopreservation and storage at -196 C. Although the fertilizability and developmental capacity of these oocytes was not determined, it is hoped that these results and ongoing experiments may demonstrate that cryopreservation of immature human oocytes to be a clinically feasible technique. This technique, combined with IVF, may be beneficial to women desiring future fertility who are anticipating loss of gonadal function from extirpative therapy, radiation, or chemotherapy. Continued research is needed to accomplish these aims. Such would represent a possible alternative to human embryo freezing, thus avoiding some of the moral and ethical problems encountered by IVF programs today. Acknowledgments. The authors thank Charles Coddington, M.D., Howard W. Jones, Jr., M.D., Suheil Muasher, M.D., Howard Seltman, M.D., and William E. Gibbons, M.D., for their contributions to this study. We also thank Martha Wilson, B.A., for graphics preparation and Ms. Dara Willett-Leary for her editorial contribution and manuscript preparation. REFERENCES 1. Chen C. Pregnancy after human oocyte cryopreservation. Lancet 1986;2:884-6. VoL 61, No.6, June 1994 Toth et al. Cryopreservation of human oocytes 1081

2. van Uem JFHM, Siebzehnrubl ER, Schuh B, Koch R, Trotnow S, Lang N. Birth after cryopreservation of unfertilized oocytes. Lancet 1987;1:752-3. 3. Van der Elst J, Van den Abbeel E, Jacobs R, Wisse E, Van Steirteghem A. Effect of 1,2-propanediol and dimethylsulphoxide on the meiotic spindle of the mouse oocyte. Hum Reprod 1988;8:960-7. 4. Sathananthan AH, Ng SC, Trounson AO, Bongso A, Ratnam SS, Ho J, et al. The effect of ultrarapid freezing on meiotic and mitotic spindles of mouse oocytes and embryos. Gamete Res 1988;21:385-401. 5. Pickering SJ, Braude PR, Johnson MH, Cant A, Currie J. Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertil Steril 1990;54:102-8. 6. Trounson A, Kirby C. Problems in the cryopreservation of unfertilized eggs by slow cooling in dimethyl sulfoxide. Fertil Steril 1989;52:778-86. 7. Vincent C, Pickering SJ, Johnson MH. The zona hardening effect of dimethylsulfoxide on the mouse zona pellucida requires the presence of an oocyte and is associated with a reduction in the number of cortical granules present. J Reprod Fertil 1990;89:253-9. 8. Carroll J, Warnes GM, Matthews CD. Increase in digyny explains polyploidy after in vitro fertilization of frozenthawed mouse oocytes. J Reprod Fertil 1989;85:489-94. 9. Veeck LL. Atlas of the human oocyte and early conceptus. Vol. 2. Baltimore: Williams & Wilkins, 1991. 10. Kuzan FB, Quinn P. Cryopreservation of mammalian embryos. In: Wolf DP, Bavister BD, Gerrity M, Kopf GS, editors. In vitro fertilization and embryo transfer: a manual of basic techniques. New York: Plenum Press, 1988:301-47. 11. Sandow BA, Namy CA. Light and electron microscopic analysis of human oocytes. In: Jones HW Jr, Jones GS, Hodgen GD, Rosenwaks Z, editors. In vitro fertilization-norfolk. Baltimore: Williams & Wilkins, 1986:94-105. 12. Sathananthan AH, Trounson AO, Wood C. Atlas of fine structure of human sperm penetration, eggs and embryos cultured in vitro. New York: Praeger, 1986. 13. Pellicer A, Lightman A, Parmer TG, Behrman HR, De Cherney AH. Morphologic and functional studies of immature rat oocyte-cumulus complexes after cryopreservation. Fertil SteriI1988;50:805-10. 14. Schroeder AC, Champlin AK, Mobraaten LE, Eppig JJ. Developmental capacity of mouse oocytes cryopreserved before and after maturation in vitro. J Reprod Fertil 1990;89:43-50. 15. Mandelbaum J, Junca AM, Plachot M, Alnot MO, Salat Baroux J, Alvarez S, et al. Cryopreservation of human embryos and oocytes. Hum Reprod 1988;3:117-9. 16. Sathananthan AH, Trounson A, Freemann L, Brady T. The effects of cooling human oocytes. Hum Reprod 1988;3: 968-77. 17. Sathananthan AH, Trounson AO. Ultrastructure observations on cortical granules in human follicular oocytes cultured in vitro. Gamete Res 1982;5:191-8. 18. Van der Elst J, Nerinckx S, Van Steirteghem AC. In vitro maturation of mouse germinal vesicle-stage oocytes following cooling exposure to cryoprotectants, and ultrarapid freezing: limited effect on the morphology of the second meiotic spindle. Hum Reprod 1992;7:1440-6. 19. Toth TL, Baka SG, Veeck LL, Jones HW Jr, Muasher S, Lanzendorf SE. Fertilization and in vitro development of cryopreserved human prophase I oocytes. Fertil Steril 1994;61:891-4. 1082 Toth et al. Cryopreservation of human oocytes Fertility and Sterility