Dynamic changes in ovarian follicles measured by ultrasonography during gonadotropin stimulation in rhesus monkeys

Available online at www.sciencedirect.com

Theriogenology 72 (2009) 560–565 www.theriojournal.com

Dynamic changes in ovarian follicles measured by ultrasonography during gonadotropin stimulation in rhesus monkeys S. Yang a,b, X. He b, Y. Niu b, X. Wang b, B. Lu b, T.B. Hildebrandt c, F. Goeritz c, K. Jewgenow c, Q. Zhou d, W. Ji b,* a Biotechnology Research Center, Faculty of Life Science and Technology, Kunming University of Science and Technology, Kunming 650093, P.R China b Kunming Primate Research Center and Kunming Institute of Zoology, Chinese Academy of Sciences, Yunnan Key Laboratory for Animal Reproduction, Kunming, Yunnan 650223, P.R China c Leibniz Institute for Zoo and Wildlife Research, PF 601103, 10252 Berlin, Germany d State Key Laboratory of Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing 100080, P.R China

Received 31 December 2008; received in revised form 19 March 2009; accepted 8 April 2009

Abstract The objective was to study dynamic changes of ovaries in rhesus macaques stimulated by gonadotropins to identify an indicator for predicting ovarian response to stimulation. Twenty-one cycling monkeys were given 36 IU/d recombinant human folliclestimulating hormone (rhFSH) for 8 d. Animals (n = 17) with 5 follicles (3 mm) in their ovaries on Day 9 of ovarian stimulation were deemed good responders, whereas those with a lesser response were poor responders (n = 4). For these two groups, the mean (SD) numbers of oocytes retrieved were 44.3  21.4 and 11.0  4.6, respectively. In retrospect, the mean diameters of the ovaries and of the largest follicles, the total number of detectable follicles (diameter >0.5 mm), and serum estradiol concentrations gradually increased during the stimulation period in the good responders but did not increase in the poor responders. Comparing good and poor responders, the number of ovarian follicles >0.5 mm already exhibited a difference (12.9  6.5 vs. 2.9  1.3, respectively, P < 0.05) on Day 1 of stimulation. However, for other end points, differences were not significant until at least Day 5. Moreover, good responders yielded a fivefold higher blastocyst development rate than that of poor responders (P < 0.01). In conclusion, the number of ovarian follicles detected with ultrasonography could be useful to predict the response to FSH stimulation in non-human primates. # 2009 Elsevier Inc. All rights reserved. Keywords: Ovarian follicle; Ovarian stimulation; Primate; Rhesus monkey; Ultrasonography

1. Introduction Substantial progress in the application of assisted reproductive technologies (ARTs) to non-human primates over the past two decades has resulted in the

* Corresponding author. Tel.: +86 871 5139413; fax: +86 871 5139413. E-mail address: [email protected] (W. Ji). 0093-691X/$ – see front matter # 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.theriogenology.2009.04.012

routine production of in vitro–derived embryos and the use of embryo transfer to establish pregnancy and produce offspring [1,2]. These advances were followed by the introduction of somatic cell nucleus transfer research and blastocyst-derived embryonic stem cell lines [3–5]. However, several factors restrict the use of rhesus monkeys as models of reproductive biology of primates, including humans. These factors include the high costs of ovarian stimulation, limited animal resources, and the availability of rhesus monkeys for

S. Yang et al. / Theriogenology 72 (2009) 560–565

561

2. Materials and methods

routinely examined, and those that exhibited multiple increases in serum estradiol (E2) compared with the baseline values and at least five follicles of at least 3 mm diameter for both ovaries combined were defined as ‘‘good responders’’; otherwise, they were characterized as ‘‘poor responders’’ [10,12]. Animals were anesthetized with ketamine (10 to 12 mg/kg im; Lianyungang International Trade Co. Ltd., Lianyungang City, China), and cumulus-oocyte complexes were collected by laparoscopic follicular aspiration, 32 to 35 h after administration of hCG. Follicular contents were placed in HEPES-buffered TALP (modified Tyrode solution with albumin, lactate, and pyruvate) medium [14] containing 0.3% bovine serum albumin (BSA) at 37 8C. Oocytes were stripped of cumulus cells by pipetting after brief exposure (<1 min) to hyaluronidase (0.5 mg/ mL) at 37 8C and were classified as metaphase II (MII; with first polar body, PB1); metaphase I (MI; no germinal vesicle, no PB1); or prophase I (PI or GV, intact germinal vesicle). Numbers of MI, MII, and GV oocytes were counted for statistical analysis, and then oocytes were used for in vitro fertilization (IVF) or other studies.

2.1. Animals and chemicals

2.3. IVF and embryo culture

Adult female rhesus macaques were housed in individual cages in a controlled environment (20 to 24 8C, humidity 40% to 60%) and an 0800 to 2000 light cycle at the Kunming Primate Research Center (KPRC). Vaginal bleeding was monitored daily to detect the onset of menses. All animal procedures were approved in advance by the Institutional Animal Care and Use Committee of KPRC. Unless stated otherwise, all chemicals were obtained from Sigma Chemical Co. (St. Louis, MO, USA).

Freshly collected mature oocytes were inseminated as in our previous reports [10]. Briefly, hyperactivated spermatozoa and mature oocytes (MII) from the collection were coincubated for 12 to 16 h at 37 8C in a humidified atmosphere of 5% CO2 in air. For embryonic development, fertilized oocytes were cultured under mineral oil in 50-mL drops of hamster embryo culture medium-10 [15] containing 10% fetal bovine serum (HyClone, Logan, UT, USA) at 37 8C in humidified air containing 5% CO2. The culture medium was changed every other day. Embryo development was monitored daily using Nomarski optics (200 to 400) on a Nikon Diaphot TMD microscope (Nikon, Tokyo, Japan).

only half of the year due to their natural breeding season, even in captivity [6]. Protocols for ovarian stimulation of the rhesus monkey originally used gonadotropins extracted from animal or human blood serum, urine, and/or pituitaries [7–9]. More recently, stimulation protocols have used human recombinant gonadotropins with improved reliability and results in terms of quality and quantity of oocytes, as well as reduced stress on the animals [10–12]. However, the dynamic changes in ovarian follicles in animals given exogenous gonadotropins are not clear. Moreover, there is still a proportion of monkeys that respond poorly to ovarian stimulation [10,12], probably due to failed synchronous growth of multiple follicles or reduced numbers of follicles recruited into the growth pool. It has not been possible to predict which animals will be poor responders prior to or soon after the start of ovarian stimulation. Therefore, the objective of this study was to evaluate the dynamic changes of ovarian follicles in stimulated rhesus monkeys and to find a predictor of good versus poor responses to ovarian stimulation.

2.2. Ovarian stimulation and oocyte recovery During the physiologic breeding season (November to March), 21 adult females (5 to 10 yr old) were chosen for this study. Treatment with recombinant human follicle-stimulating hormone (rhFSH; Gonal F; Laboratories Serono SA, Aubonne, Switzerland) was initiated 1 to 3 d after the onset of menses. All animals received 18 IU rhFSH given intramuscularly twice daily (total of 36 IU/day) for 8 d; Day 1 was the first day of FSH treatment, as previously described [10,13]. On Day 9, all monkeys were given 1000 IU of human chorionic gonadotrophin (hCG) intramuscularly (Laboratories Serono SA, Aubonne, Switzerland) to support oocyte nuclear maturation [10]. On the same day, animals were

2.4. Ultrasonography Ultrasonographic examinations were conducted every other day during the stimulation period. Animals were anesthetized with ketamine (10 to 12 mg/kg) given intramuscularly, and the lower abdominal and inguinal regions were shaved. Ovaries and their follicles were imaged transcutaneously using a Diasus ultrasound system (Dynamic Imaging Ltd., Livingston, Scotland, UK), equipped with a 10- to 22-MHz linear-array transducer. To help locate the ovaries, the middle finger of the operator, wearing a sterile glove, was inserted into

562

S. Yang et al. / Theriogenology 72 (2009) 560–565

Table 1 Mean (SD) number of oocytes recovered in gonadotropin-stimulated rhesus monkeys designated as good and poor responders according to the number of ovarian follicles 3 mm on Day 9 of the stimulation cycle. Responder group Good Poor

Proportion of animals (%) 17 of 21 (81) 4 of 21 (19)

Number of follicles 3 mm on Day 9 of stimulation cycle 15.3  9.1 2.3  1.1 b

a

Mean number of oocytes retrieved per animal Total* 44.3  21.4 11.0  4.6 b

MII a

20.1  9.5 1.8  1.0b

MI a

GV a

7.7  4.3 0.8  1.0b

19.6  13.3a 8.5  4.2b

a,b

Within a column, means without a common superscript differed (P < 0.01). Total includes MII and MI and GVoocytes; MII represents oocytes that were mature at collection, whereas MI and GV refer to oocytes that were immature at collection. *

the animal’s rectum after flushing out feces with warm water. The maximal resolution provided by the ultrasound system was 100 mm. To reduce the time required for the examinations, the entire ultrasonographic examination was videotaped with a digital recorder (GV-D900E PAL; Sony Inc., Tokyo, Japan) on miniDV cassettes (Sony Inc.) and analyzed retrospectively. Representative sonograms were created by computerizing the ultrasound sequence with Adobe Premier Pro 1.5 (Adobe Systems Inc., Amsterdam, The Netherlands), and the regions of interest were measured on the sonograms in JPEG format with the AnalySIS 3.1 software program (Olympus Inc., Munster, Germany). 2.5. Hormone assays To quantify serum concentrations of estradiol (E2), blood samples were drawn from anesthetized animals by saphenous venipuncture while ultrasonography was being performed. Serum concentrations of E2 were determined by radioimmunoassay [10,16]. The intraassay and interassay coefficients of variation were all <10%. 2.6. Statistical analysis Results are presented as the mean  SD, unless stated otherwise. Prior to analysis with SPSS 11.5 software (SPSS Inc., Chicago, IL, USA), development rates of IVF oocytes were square root transformed, and concentrations of serum E2 were transformed by logtransformation. For follicle numbers, diameters of ovaries and follicles, and serum estradiol concentrations during ovarian stimulation, comparisons between good and poor responders were evaluated by the generalized linear model (GLM) repeated-measures ANOVA. Moreover, an unpaired Student’s t-test was used to compare, between the two groups, the numbers of oocytes retrieved and the developmental rate of IVF oocytes at specific time points during stimulation

[13,17]. For all analyses, P < 0.05 was considered different. 3. Results 3.1. Ovarian responses and oocytes recovered and subsequent embryonic development Based on the responses to ovarian stimulation on Day 9, 21 monkeys were retrospectively divided into two groups: good responders (n = 17, 81%) and poor responders (n = 4, 19%; Table 1). Good responders yielded at least sevenfold more follicles 3 mm in both ovaries compared with that of poor responders (15.3  9.1 vs, 2.3  1.1, P < 0.01). Similarly, good responders had more than fourfold higher mean numbers of oocytes (44.3  21.4 vs 11.0  4.6, P < 0.01), 11-fold higher mean numbers of mature (MII) oocytes (20.1  9.5 vs. 1.8  1.0, P < 0.01), and a greater mean number of MI (7.7  4.3 and 0.8  1.0, P < 0.01) and GV (l9.6  13.3 and 8.5  4.2, P < 0.01) oocytes. In addition, 31 MII oocytes randomly selected from eight good responders (3 to 5 oocytes/animal) and all MII oocytes retrieved from poor responders were used to assess their development after IVF. The remainders of the oocytes from good responders were used for other studies (data not shown). It was noteworthy that the developmental potential of oocytes in the good responders was also higher than that in the poor responders (Table 2). Fertilization and blastocyst development rates were threefold and fivefold higher in oocytes from good versus poor responders (P < 0.01). 3.2. Dynamic changes in ovaries and follicles in stimulated rhesus monkeys Representative ultrasound images of ovaries in one good responder and one poor responder are shown in

S. Yang et al. / Theriogenology 72 (2009) 560–565

563

Table 2 Development potential of mature oocytes recovered in gonadotropin-stimulated rhesus monkeys designated as good and poor responders according to the number of ovarian follicles 3 mm on Day 9 of the stimulation cycle. Responder group

Number of animals useda

Number of oocytes usedb

Development (mean  SD) Fertilized oocytes, n (%) c

Good Poor

8 4

31 7

a

28 (90.0  14.6 ) 3 (29.0  34.4b)

Blastocyst, n (%) d 17 (64.0  16.6a) 1 (13.0  25.0b)

a,b

Within a column, means without a common superscript differed (P < 0.01). Eight monkeys were randomly selected from 17 good responders. b Three to five MII oocytes were used for assessing their developmental potential in each monkey of eight good responders, and all MII oocytes retrieved were used for the same protocol in poor responders. c % Fertilized = (ova exhibiting two pronuclei/MII oocytes)  100. d % Blastocyst = (blastocysts/fertilized oocytes)  100. a

Fig. 1 to illustrate dynamic changes in ovaries and follicles. Mean diameters of ovaries and of the largest follicle and the number of detectable follicles (diameter >0.5 mm) were significantly greater in good responders. In that regard, those end points gradually increased and reached maximal values at the completion of ovarian stimulation in good responders, and the serum concentration of E2 reached its maximal value on Day 9. In contrast, all of these end points were virtually unchanged over time in the poor responders (Fig. 2).

Most interestingly, the mean number of follicles (diameter >0.5 mm) detected ultrasonographically was already higher on Day 1 in animals that subsequently exhibited good responses to FSH stimulation compared with that of the poor responders (12.9  6.5 vs. 2.9  1.3, P < 0.05). In contrast, there were no significant differences between good and poor responders for ovarian diameter, serum E2 concentrations, and diameters of the largest follicles until Days 5, 7, and 9 of ovarian stimulation, respectively.

Fig. 1. Ultrasonographic images of left and right ovaries of gonadotropin-stimulated rhesus monkeys. (A) Ovaries of a good responder from which 57 oocytes were retrieved (16 at MII, 7 at MI, and 34 at GV) and (B) ovaries of a poor responder that yielded 19 oocytes (2 at MII, 1 at MI, and 16 at GV) are shown after stimulation. Day 1 is the first day of FSH treatment. Human chorionic gonadotropin was given on Day 9, and oocyte collection was performed on Day 11, immediately after the ultrasonographic examination. The yellow, red, and gray lines in the images were used for the measurement of the size of ovaries. The scale bars represent 10 mm.

564

S. Yang et al. / Theriogenology 72 (2009) 560–565

Fig. 2. Changes in ovaries and serum E2 concentrations in rhesus monkeys during ovarian stimulation. Shown are mean diameters of (A) each pair of ovaries and of (B) the largest follicle, (C) plasma E2 concentrations, and (D) the numbers of detectable follicles (diameters >0.5 mm) in good responders (n = 17) and poor responders (n = 4) to ovarian stimulation. All animals received hCG on Day 9 to induce oocyte nuclear maturation, and oocyte aspiration was performed on Day 11. *Differences between good versus poor responders on these days were different (P < 0.05).

4. Discussion To our knowledge, this is the first report on sequential, ultrasonographic assessment of ovarian follicular growth throughout the entire gonadotropin stimulation cycle in rhesus monkeys. Our most important finding was that high-resolution ultrasonographic images of the number of follicles detected in both ovaries at the time of menses was predictive of whether the female would be a good or poor responder to ovarian stimulation. We inferred that the size of the growing pool of follicles at menses determined the number of follicles recruited and the number of oocytes obtained after ovarian stimulation. Non-human primates are frequently promoted as ideal models for human ART. However, unlike in humans [18–21], there are few reports in non-human primates concerning responses to controlled ovarian stimulation that could predict or lead to successful IVF cycles. In primates, the growth of primary follicles begins after the FSH surge. Because inhibition of sibling follicular growth by the dominant follicles develops later in the menstrual cycle, the administration of exogenous gonadotropins during menses supports the growth and development of many additional follicles by upregulating their FSH receptors [22]. Typically, however, in non-human primate ART, there are always

some females that respond poorly or not at all to a standard gonadotropin stimulation regimen [10,12], such as the ones used in this study, which are usually identifiable only after 7 d or more of gonadotropin administration [10,12,23]. This represents a considerable waste of staff time and expensive rhFSH. As shown in the current study, it was not possible to discriminate between good and poor responders until at least Day 5 and as late as Day 9 of the stimulation cycle using ovarian size, serum E2 concentrations, or diameters of the largest follicles as response indicators. In contrast, ultrasonographic monitoring of follicular diameter is noninvasive and, as shown in the current study, was a very simple and reliable method for monitoring ovarian stimulation in ART cycles in rhesus monkeys. High-resolution ultrasonography of rhesus monkeys demonstrated that a cohort of follicles is recruited into the growth pool at the onset of menses. Recruitment and growth of these additional follicles, even at the start of menses when they were only approximately 0.5 mm in diameter, could be detected using just a single ultrasonographic examination performed on stimulation Day 1. We found that the numbers of these small follicles at the beginning of menses was significantly associated with conventional measures of good versus poor ovarian responses to FSH stimulation. Not only were numbers and quality of

S. Yang et al. / Theriogenology 72 (2009) 560–565

oocytes collected much higher in good versus poor responders, but also IVF and embryo development to blastocysts were also much better using oocytes from good responders, though a partial number of total oocytes in the ovarian stimulation cycles were used for assessing the developmental potential in the current study. Therefore, we inferred that ultrasonographic monitoring of the numbers of these small follicles could be used as a noninvasive predictor of good versus poor responses to ovarian stimulation. In conclusion, we have shown that the number of ovarian follicles at menses can be a prognostic factor for retrieving large numbers of high-quality oocytes from rhesus monkeys. In contrast, discrimination between good versus poor responders on the basis of conventional indicators could not be done until the ovarian stimulation regimen was well under way or almost completed. Therefore, high-resolution ultrasonography was a valuable tool for helping to improve the efficiency of ART in non-human primates and also could improve the ability to control ovarian stimulation. Acknowledgments This work was supported by research grants from the Chinese Major State Research Development Program 2007BAI33B05, the Chinese National Basic Research Program 2007CB947701 and 2006CB701505, the Chinese Academy of Sciences KSCX2-YWR-47 and KSCX2-YWR-89, and the Chinese National Science Foundation 30871232. We thank Dr. Barry Bavister for editing the manuscript. References [1] Wolf DP. Assisted reproductive technologies in rhesus macaques. Reprod Biol Endocrinol 2004;2:37. [2] Bavister BD. The role of animal studies in supporting human assisted reproductive technology. Reprod Fertil Dev 2004;16: 719–28. [3] Niu Y, Yang S, Yu Y, Ding C, Yang J, Wang S, et al. Impairments in embryonic genome activation in rhesus monkey somatic cell nuclear transfer embryos. Cloning Stem Cells 2008;10:25–36. [4] Zhou Q, Yang SH, Ding CH, He XC, Xie YH, Hildebrandt TB, et al. A comparative approach to somatic cell nuclear transfer in the rhesus monkey. Hum Reprod 2006;21:2564–71. [5] Byrne JA, Pedersen DA, Clepper LL, Nelson M, Sanger WG, Gokhale S, et al. Producing primate embryonic stem cells by somatic cell nuclear transfer. Nature 2007;450:497–502. [6] Bavister BD. ARTs in action in nonhuman primates: symposium summary—advances and remaining issues. Reprod Biol Endocrinol 2004;2:43.

565

[7] Zelinski-Wooten MB, Hutchison JS, Hess DL, Wolf DP, Stouffer RL. Follicle stimulating hormone alone supports follicle growth and oocyte development in gonadotrophin-releasing hormone antagonist-treated monkeys. Hum Reprod 1995;10:1658–66. [8] Bavister BD, Boatman DE, Collins K, Dierschke DJ, Eisele SG. Birth of rhesus monkey infant after in vitro fertilization and nonsurgical embryo transfer. Proc Natl Acad Sci USA 1984;81: 2218–22. [9] Abbasi R, Kenigsberg D, Danforth D, Falk RJ, Hodgen GD. Cumulative ovulation rate in human menopausal/human chorionic gonadotropin-treated monkeys: ‘‘step-up’’ versus ‘‘stepdown’’ dose regimens. Fertil Steril 1987;47:1019–24. [10] Yang S, He X, Hildebrandt TB, Jewgenow K, Goeritz F, Tang X, et al. Effects of rhFSH dose on ovarian follicular response, oocyte recovery and embryo development in rhesus monkeys. Theriogenology 2007;67:1194–201. [11] VandeVoort CA, Tarantal AF. Recombinant human gonadotropins for macaque superovulation: repeated stimulations and post-treatment pregnancies. J Med Primatol 2001;30:304–7. [12] Schramm RD, Paprocki AM, Bavister BD. Features associated with reproductive ageing in female rhesus monkeys. Hum Reprod 2002;17:1597–603. [13] Yang S, Shen Y, Niu Y, Hildebrandt TB, Jewgenow K, Goeritz F, et al. Effects of rhFSH regimen and time interval on ovarian responses to repeated stimulation cycles in rhesus monkeys during a physiologic breeding season. Theriogenology 2008;70:108–14. [14] Bavister BD, Boatman DE, Leibfried L, Loose M, Vernon MW. Fertilization and cleavage of rhesus monkey oocytes in vitro. Biol Reprod 1983;28:983–99. [15] McKiernan SH, Bavister BD. Culture of one-cell hamster embryos with water soluble vitamins: pantothenate stimulates blastocyst production. Hum Reprod 2000;15:157–64. [16] Wolf DP, Vandevoort CA, Meyer-Haas GR, Zelinski-Wooten MB, Hess DL, Baughman WL, Stouffer RL. In vitro fertilization and embryo transfer in the rhesus monkey. Biol Reprod 1989;41: 335–46. [17] Yang S, He X, Hildebrandt TB, Zhou Q, Ji W. Superovulatory response to a low dose single-daily treatment of rhFSH dissolved in polyvinylpyrrolidone in rhesus monkeys. Am J Primatol 2007;69:1278–84. [18] Aubriot FX, Chapron C. FSH and IVF management: the best and worst case scenarios. Gynecol Obstet Fertil 2008;36:636–40. [19] Ficicioglu C, Kutlu T, Baglam E, Bakacak Z. Early follicular antimullerian hormone as an indicator of ovarian reserve. Fertil Steril 2006;85:592–6. [20] Ito M, Muraki M, Takahashi Y, Imai M, Tsukui T, Yamakawa N, et al. Glutathione S-transferase theta 1 expressed in granulosa cells as a biomarker for oocyte quality in age-related infertility. Fertil Steril 2008;90:1026–35. [21] van Montfoort AP, Geraedts JP, Dumoulin JC, Stassen AP, Evers JL, Ayoubi TA. Differential gene expression in cumulus cells as a prognostic indicator of embryo viability: a microarray analysis. Mol Hum Reprod 2008;14:157–68. [22] La Marca A, Carducci Artenisio A, Stabile G, Rivasi F, Volpe A. Evidence for cycle-dependent expression of follicle-stimulating hormone receptor in human endometrium. Gynecol Endocrinol 2005;21:303–6. [23] VandeVoort CA, Tarantal AF. The macaque model for in vitro fertilization: superovulation techniques and ultrasound-guided follicular aspiration. J Med Primatol 1991;20:110–6.