Coculture of human zygotes on fetal bovine uterine fibroblasts: embryonic morphology and implantation*

Transcription

1 FERTILITY AND STERILITY Copyright c 1989 The American Fertility Society Printed on acid-free paper in U.S.A. Coculture of human zygotes on fetal bovine uterine fibroblasts: embryonic morphology and implantation* Klaus E. Wiemer, Ph.D.t Jacques Cohen, Ph.D.:!: 11 Sharon R. Wiker, B.S.:!: Henry E. Malter, B.S.:!: Graham Wright, B.S.:!: Robert A. Godke, Ph.D. t Louisiana State University, Baton Rouge, Reproductive Biology Associates, and Emory University Medical Sclwol, Atlanta, Georgia Zygotes from in vitro fertilization patients (n = 116) were randomly allocated to culture in either conventional plastic petri dishes or coculture on a monolayer of fetal bovine uterine fibroblasts. Embryos (n = 288) remained 26 to 32 hours in these culture systems. Video tape recording for later morphological analysis (11 parameters) was performed on 117 conventionally cultured and 104 cocultured embryos, shortly before replacement, by an independent observer, unaware of the culture conditions for each embryo. A significantly greater number of cocultured embryos (52%) had "good" morphology (zero or only one abnormal characteristic) as compared with conventionally cultured embryos (30%). The most outstanding morphological characteristic of cocultured embryos was the expanded appearance of their blastomeres. The incidence of implantation per embryo increased from 13% to 19% when the coculture rather than conventional culture system was used, and the incidence of ongoing pregnancy per patient after coculture doubled to 35%. Fertil Steril52:503, 1989 Mature human oocytes have a limited chance of implantation after in vitro fertilization (IVF).1-3 Embryonic wastage can be caused by a variety of factors, such as decreased receptivity of the endometrium, trauma during embryo replacement, and genetic disorders. The embryo culture system may be suboptimal as well, but it has been difficult to provide evidence for this condition. Certain batches of tissue culture disposables or culture medium ingredients may have a noticeable effect Received February 6, 1989; revised and accepted April 20, * Prize paper, presented at the Forty-Fourth Annual Meeting of The American Fertility Society, October 10 to 13, 1988, Atlanta, Georgia. t Department of Animal Science, Louisiana State University. :j: Reproductive Biology Associates. Department of Gynecology and Obstetrics, Emory University Medical School. II Jleprint requests and present address: Jacques Cohen, Ph.D., The Center for Reproductive Medicine and Infertility, Cornell University Medical College, P.O. Box I-HT, 306; 505 East 70th Street, New York, N.Y on embryonic development. 4 Generally, 25% of cleaved human embryos develop into expanded blastocysts in vitro, and only a minority of those will hatch, providing additional evidence that culture conditions may be insufficient. 5 A recent preliminary study6 demonstrated that human monospermic zygotes, cultured for 1 day on a monolayer offetal bovine uterine fibroblasts, had a higher rate of cleavage and implantation. Cy~ toplasmic fragmentation was also diminished when helper cells were used. Moreover, three-pronucleate embryos cavitated more frequently when the coculture system was compared with conventional culture. These results suggest that a substantial number of embryos could be rescued after the use of helper cells. The cross-species suitability of fetal bovine uterine fibroblasts has been shown to be successful with growth of equine embryos in a similar coculture system. 7 The insufficiency of the human IVF culture conditions has also been confirmed by an increased incidence of implantation when human zygotes were immediately replaced into the fallopian tube. 8 The objective of the current study was to deterwiemer et ai. Human embryos and helper cells 503

2 mine differences in embryonic morphology and rate of implantation when human zygotes were either cultured using the conventional system or incubated with a monolayer of fetal bovine uterine fibroblasts. Video tape recordin~ was performed on embryos shortly before replacement. Videotapes were retrospectively analyzed by an observer unaware of the origin of the embryos and the outcome of the procedures. MATERIALS AND METHODS Patients and IVF Methodology Couples were treated with IVF for a variety of infertility disorders. The study was performed during November 1987 and November Patient selection and allocation to the different culture methods were performed after at least one oocyte was fertilized. Zygotes from 116 patients, assessed 21 to 25 hours after oocyte collection, were randomly allocated to one of two culture systems: (1) Earle's balanced salt solution (EBSS) supplemented with 15% maternal or donor serum (conventional culture); or (2) serum-supplemented EBSS on a monolayer of fetal bovine uterine fibroblasts (coculture). Embryos remained 26 to 32 hours in one of these two culture systems. Before replacement, embryos were videotaped for subsequent morphological analysis. Embryos were washed two or three times in droplets of EBSS shortly before replacement. The replacement medium was composed of 25% EBSS and 75% serum. A maximum-of three embryos was replaced at a time. Clinical pregnancy was defined when fetal heart activity could be visualized with ultrasonography. Follicular stimulation, ovulation-monitoring procedures, oocyte collection, and embryo cryopreservation techniques have been extensively described for this laboratory Preparation of the Fetal Uterine Fibroblast Monolayer The methodology for preparing fetal bovine uterine fibroblasts and maintenance of these cells is provided elsewhere. 6 Uterine explants were removed from the endometrium of healthy bovine fetuses during the last trimester of gestation. The "Explant Phase" of the cell preparation procedure was maintained with Ham's F-10 culture medium (Gibco, Grand Island, NY, or Flow Laboratories, McLean, VA) supplemented with 10% heat-inactivated fetal calf serum (Sigma, St. Louis, MO), penicillin, streptomycin, and amphotericin-b (HF-10). Monolayers were established 14 to 20 days after initial incubation of the explants. Trypsinization was performed with phosphate-buffered saline solution containing 0.05% trypsin (Sigma). Flasks (250 ml Falcon; Becton Dickinson and Co., Lincoln Park, NJ) were sequentially seeded with onethird of the original cell concentration. Cells were grown at 37 C under 5% CO 2 in air 4 to 7 days between each subpassage. Cells were placed in dimethylsulfoxide and frozen in liquid nitrogen after the seventh subpassage. All fibroblasts used in this study were previously stored in liquid nitrogen for at least 30 days. Fibroblasts were thawed and screened for bovine viruses. 6 Coculture of Zygotes on Fibroblasts Monolayers were prepared by plating 1 X 10 5 viable fibroblasts into each well of a four-well tissue culture plate (Nunc, Copenhagen, Denmark), 48 to 72 hours before the onset of coculture. Monolayered wells were checked before coculture for debris and percentage of cell growth covering the bottom of the well. Only wells with 50% to 100% of the surface area overgrown were used for embryo culture. Wells were washed once with 4 ml of proteinfree HF-10 and then twice with protein-free EBSS. Antibiotics and antimycotics were not used in either medium. EBSS had a ph of 7.4, and the osmolarity ranged from 284 to 287 mosm. Zygotes allocated to the coculture treatment were removed from the insemination dish, washed twice in 15% serum-supplemented EBSS, and individually moved into wells containing the 0.75 ml of EBSS and the monolayer. Each well was covered with equilibrated warm mineral oil (Squibb, Princeton, NJ). Zygotes allocated to the conventional culture system were washed twice in EBSS and pipetted individually in monolayer-free well-plates (Nunc, Copenhagen, Denmark). All plates were placed in dessicators at 37 C under 5% CO 2, 5% O 2, and 5% N 2 Video Tape Recording of Embryos.Embryos (n = 294) were videotaped for 30 to 90 seconds on VHS tape using a black-and-white camera (CCTV; Panasonic, Tokyo, Japan) mounted on an IMT-2 microscope (Olympus, Tokyo, Japan).9 Illumination and magnification was kept constant 504 Wiemer et al. Human embryos and helper cells Fertility and Sterility

3 ported to be of importance and is based on all absolute criteria (negative scores), a percentage fragmentation exceeding 20 and a percentage zona pellucida thickness variation lower than 25Y Not all parameters could be determined in each embryo due to visual obstruction of either the blastomeres or the zonae pellucidae. Figure 1 Two four-cell embryos from a patient in which embryos were either cocultured on a monolayer of fetal bovine uterine fibroblasts or cultured without helper cells. Embryos with the lowest number of abnormal characteristics from each culture group are photographed here together. The microscope was focused on three blastomeres in each embryo to compare blastomere size. The embryo on the left (conventionally cultured) has a few fragments. The blastomeres from the embryo on the right (cocultured) are slightly larger and fragments are absent. during this study. Embryos were recorded at several focal points. The tapes were analyzed on a 32-cm black-andwhite monitor using the "still" and "slow" modes of a commercial videorecorder. The overall magnification was X1,400. The observer was unaware of the outcome of the IVF procedure and the type of culture system used. A total of 10 previously defined morphological parameters was used. 9 Although the criteria for assessing embryos were subjective in nature, the reliability of the procedure was maximized by using absolute scoring ("positive" or "negative") rather than measured parameters. The absolute parameters were as follows: identical blastomere size, presence of vacuoles, visibility of cell organelles, darkness of cytoplasm, extruded membranes, patched membrane, contracted blastomeres, and cell-cell adherence. In addition, two measured parameters were expressed as percentages per embryo: (1) extracellular fragmentation was scored as a value percentage of the perivitelline space; and (2) percent thickness of zona pellucida variation within each embryo was determined by a minimum of two measurements of the zona pellucida. The percentage was calculated by taking the difference between the average and the greatest thickness of the zona pellucida, dividing the result by the average thickness, and multiplying by looy It was noted that fibroblast-exposed embryos had overtly expanded blastomeres. Blastomere expansion was, therefore, added as an absolute parameter during this study. Positive (full) expansion was defined when all blastomeres virtually touched the zona pellucida, leaving little perivitelline space (Fig. 1). The total number of abnormal morphological characteristics was previously re- Statistics Data were analyzed with the x 2 test in all instances unless the expected frequency was less than five, in which case the Fisher's Exact Test was used. RESULTS Less than 5% of zygotes did not cleave, irrespective of the culture system. All 116 patients had at least one embryo for replacement. The incidence of clinical pregnancy was higher after replacement of cocultured embryos (40%) than after replacement of conventionally cultured embryos (24%). However, this difference was not significant. Furthermore, the incidence of pregnancy after coculture seemed enhanced irrespective ofthe number of embryos replaced (Table 1). However, subgroups were rather small and more data has to be obtained to substantiate these results. Most importantly, however, the incidence of pregnancies ongoing beyond week 14 of gestation doubled (from 17% to 35%; P = 0.05) when embryos were cocultured on fibroblasts. Four of 14 (29%) patients whose embryos were cultured conventionally miscarried, whereas only 3 of 23 (13 %) with cocultured embryos miscarried. So far, 12 women have delivered healthy infants from cocultured embryos. Eight patients had a multiple pregnancy (7 twins and 1 set of triplets), most of those occurring after coculture (Table 1). The incidence of implantation per embryo increased from 13% to 19% after coculture. Significant differences in embryo morphology between conventionally and co cultured embryos were identified in 4 of 11 parameters (Table 2). Three of these parameters were previously found to be important for prediction of implantation of both fresh and thawed cleaved human embryos. These parameters were cell-cell adherence, cytoplasmic fragmentation, and zona thickness variation. These morphological characteristics were improved in cocultured embryos when compared with those from conventional culture. Some of the embryos could not be assessed due to the presence Wiemer et al. Human embryos and helper cells 505

4 Table 1 Frequency of Clinical Implantation, Miscarriage, and Multiple Pregnancy after Coculture of Human Zygotes on Fetal Bovine Uterine Fibroblasts Proportion of clinically pregnant patients Type of zygote culture 1 ~o.ofembryosreplaced 2 3 Incidence of Proportion of Incidence of implantation All ongoing multiple per embryo replacements pregnancies pregnancy replaced Conventional Coculture 00. 0/12 1/5 (%) (0) (20) 00. (%) 00. (%) 5/14 (36) 9/32 (28) 5/12 (42) 17/41 (42) 00. (%) 00. (%) 00. (%) 00. (%) 14/58 (24) 10/58 (17)" 2/14 (14) 17/136 (13) 23/58 (40) 20/58 (35)" 6/23 (26) 29/152 (19) "p= of fragments or corona cells, hence the total number of observations varied from parameter to parameter (Table 2). The most overt morphological improvement after coculture was the expansion of blastomeres. Only 18% of all blastomeres appeared swollen when the embryos were conventionally cultured, compared with 51% of cocultured embryos having expanded blastomeres. Furthermore, the number of atypical findings per embryo was greater in the conventionally cultured embryos. Fifty-two percent of cocultured embryos had a single morphological abnormality or none, compared with 30% of conventionally cultured embryos (Table 3). DISCUSSION The morphology of human embryos markedly improved when they were cocultured for 1 day on a Table 2 Morphological Characteristics of Cleaved Embryos Assessed With Video Tape Recording of Human Zygotes Either Cultured Conventionally or on Fetal Bovine Uterine Fibroblasts Proportion of embryos Type of zygote culture Morphological characteristic Conventional Coculture 00. (%) 00. (%) Statistical significance All blastomeres expanded 22/121 (18) 54/105 (51) P < Less than 25% fragments in the perivitelline space 84/127 (66) 86/107 (80) P=0.02 Blastomere adherence 6/97 (63) 76/91 (84) P= Zona thickness variation >25% 53/131 (41) 59/99 (60) P=0.006 monolayer of fetal bovine uterine fibroblasts. Increased blastomere diameter was the most evident morphological feature of embryos that developed in the presence of helper cells. It is possible that these helper cells release metabolism -enhancing substances not normally found in culture medium. The osmotic pressures of cocultured media were not lower than conventional media. Production of such metabolic intermediates or stimulatory factors could partially explain the increased diameter of blastomeres. Blastomere expansion may be the cause of a higher frequency of variations in zona pellucida thickness in many of these embryos. The latter phenomenon may be the result of local pressure from the blastomeres on the zona pellucida or a release of zona digestive substances. An increased within-embryo variation in zona pellucida thickness is an important morphological criterion, correlated with implantation.13 Blastomere swelling was also observed in two-cell hamster embryos developing in vitro.14 The occurrence was found to be correlated with increased developmental arrest. Consequently, blastomere swelling in embryos from different species may reflect different mechanisms. Table 3 Comparisons Between ~umber of Abnormal Embryonic Characteristics When Zygotes Were Either Cultured Conventionally or on Fetal Bovine Uterine Fibroblasts ~o. of Proportion of embryos abnormal morphological Type of zygote culture characteristics (per embryo) Conventional Coculture 00. (%) /117 (30) 54/ /117 (40) 42/104 ;;.4 30/117 (26) 8/104 " Exact test trend analysis, P = (%) (52)" (40)" (8)" Statistical significance P= ~ot significant P= Wiemer et al. Human embryos and helper cells Fertility and Sterility

5 An enhanced rate of development and a low incidence of cytoplasmic fragmentation was found in cocultured human embryos during a preliminary study.6 Current results confirmed a decrease of fragmentation in cocultured embryos. The present study used an independent observer, unaware of the origin of the embryos, for assessing videotapes. The swollen appearance of many cocultured embryos may have masked the presence of cytoplasmic fragments, and this may have biased observers during the first study.6 The incidence of implantation per embryo after coculture increased from 13% to 19%, but the difference was not significant due to the small sample size. The implantation rate (16%) of embryos obtained during the natural menstrual cycle was superior when compared with embryos obtained after follicular stimulation.lo The only higher success rate per embryo (30%) was reported in a small series after cryopreservation of expanded blastocysts.s However, embryos that did not survive the thawing process (--50%) were not included in that study. The doubled success rate per patient and the improved embryonic morphology after the use of helper cells demonstrates that suboptimal IVF conditions may be, in part, responsible for the generally low success rates after embryo replacement. Based on experiments involving timing of thawing of cryopreserved embryos, it has been postulated that embryos grown in vitro become asynchronous with uterine receptivity.ll In a previous study, it was shown that cocultured human embryos develop faster than conventionally cultured embryos.6 This phenomenon may be related to the increased incidence of implantation after coculture, diminishing asynchrony between embryo and uterus. The findings in this study may be used to improve success rates after IVF by prolonged culturing of embryos on helper cells. Lavage of in vivofertilized embryos has demonstrated that embryos arrive in the uterus during compaction or early cavitation. Therefore, the uterine environment may not be suitable for optimal growth of pre compacted human embryos. Later replacement, 3 or 4 days after oocyte collection, should be considered in combination with the coculture system. The likelihood of increased embryonic rescue after prolonged coculture was demonstrated in a small series of polyspermic embryos.6 Previous investigations have shown that replacement of human embryos is more e.fficient between the one- to eight-cell stages. lo However, these studies were performed during the early development of IVF methodology. The use of helper cells may motivate IVF programs to reconsider a later day for embryo replacement. Several suggestions can be made toward routine application of coculture systems in human IVF. Results presented here were obtained by transferring conventionally inseminated zygotes into multiwell culture plates containing fibroblast monolayers. Insemination in the presence of helper cells may improve fertilization and implantation as well. However, preliminary results in a small group of patients (n = 10) using fetal bovine uterine fibroblasts during fertilization did not increase the success rate, and only one patient became pregnant. l5 Oviductal cell lines may be more appropriate for this type of coculture. The cell types used here present our first attempts in coculturing human embryos. Other cell lines isolated from different parts of reproductive tracts from other species, including man, may be more appropriate and may further increase results. The use of fetal or neonatal cells, over those from adults, needs to be investigated. Obvious advantages are an increased cell cycle and a diminished likelihood of viral infection. The actual presence of helper cells, rather than use of conditioned medium, appears to be required for improved embryonic development. Further research into the mechanism of helper cell-embryo interaction is necessary for understanding the optimal conditions of embryonic development in vitro. The use of helper cells is now routinely applied in our IVF program and preferred over the routine application of intrafallopian zygote transfer, as this involves an extra surgical procedure. It is advised to allocate at least one incubator to cell tissue culture and to maintain two or more cell lines simultaneously, each with their own medium. Acknowledgments. We are grateful to Hilton Kort, M.D., Carlene Elsner, M.D., Joe Massey, M.D., William Graves, Ph.D., and Grace Amborski, Ph.D., for their support of this study. K. Leigh Inge, B.S., and Ms. Robin Moseley are acknowledged for their editorial and technical assistance. REFERENCES 1. Seppala M: The world collaborative report on in vitro fertilization and embryo replacement: current state of the art in In In Vitro Fertilization and Embryo Transfer, Edited by M Seppala, RG Edwards. New York, New York Academy of Science, 1985, p Cohen J, Fehilly CB, Hewitt J: New developments in invitro fertilization. In Current Problems in Obstetrics, Gy- Wiemer et ai. Human embryos and helper cells 507

6 necology and Fertility, Edited by JM Leventhal. Chicago, Year Book Medical Publishers, 1986, p Wood C, McMaster R, Rennie G, Trounson A, Leeton J: Factors influencing pregnancy rates following in vitro fertilization and embryo transfer. Fertil Steril43:245, Fishel S, Jackson P, Webster J, Faratian B: Endotoxins in culture medium for human in vitro fertilization. Fertil Steril49:108, Fehilly CB, Cohen J, Simons RF, Fishel SB, Edwards RG: Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparative study. Fertil Steril44: 638, Wiemer KE, Cohen J, Amborski GF, Munyakani L, Wiker S, Godke RA: In vitro development and implantation ofhuman embryos following culture on fetal bovine uterine fibroblast cells. Hum Reprod (Oxf). In press 7. Wiemer K, Gasey P, Devore D, Godke R: The culture of equine embryos using a new fetal uterine monolayer culture system. Theriogenology 29:327, Y ovich JL, Y ovich JM, Edirisinghe WR: The relative chance of pregnancy following tubal or uterine transfer procedures. Fertil Steril 49:858, Cohen J, Wiemer KE, Wright G: Prognostic value ofmorphologic characteristics of cryopreserved embryos: a study using videocinematography. Fertil Steril49:827, Edwards RG, Fishel SB, Cohen J, Fehilly CB, Purdy JM, Steptoe PC, Webster J: Factors influencing the success of in vitro fertilization for alleviating human infertility. J In Vitro Fert Embryo Transfer 1:3, Cohen J, DeVane GW, Elsner CW, Kort HI, Massey JB, Norbury SE: Cryopreserved zygotes and embryos and endocrinologic factors in the replacement cycle. Fertil Steril 50: 61, Malter HE, Cohen J: Partial zona dissection of the human oocyte: a nontraumatic method using micromanipulation to assist zona pellucida penetration. Fertil Steril. In press 13. Cohen J, Inge KL, Suzman M, Wiker SR, Wright G: Videocinematography of fresh and cryopreserved embryos: a retrospective analysis of embryonic morphology and implantation. Fertil Steril51:820, Bavister BD: Studies on the developmental blocks in cultured hamster embryos. In Mammalian Preimplantation Embryo, Regulation of Growth and Differentiation in Vitro, Edited by BD Bavister. New York, Plenum, 1987, p Wiker S, Cohen J: Unpublished data 508 Wiemer et al. Human embryos and helper cells Fertility and Sterility