Environment of the preimplantation human embryo in vivo: metabolite analysis of oviduct and uterine fluids and metabolism of cumulus cells*

Transcription

1 FERTILITY AND STERILITY Copyright e 1996 American Society for Reproductive Medicine Printed on acid free paper in U. S. A Environment of the preimplantation human embryo in vivo: metabolite analysis of oviduct and uterine fluids and metabolism of cumulus cells* David K. Gardner, D.Phil.t:j: Michelle Lane, B.Sc.t Han Calderon, M.D. II John Leeton, M.D. Monash University, Melbourne, Victoria, Australia Objective: To determine the levels of metabolites surrounding the human oocyte and embryo in vivo. Design: Oviduct and uterine fluids were collected throughout the menstrual cycle. Cumulus cells were collected at oocyte retrieval and their production of metabolites was assessed. Samples were analyzed for pyruvate, lactate, and glucose by microfluorimetry. Patients: Luminal fluids were collected from naturally cycling patients at the time of routine clinical investigation. Patient consent and hospital ethics approval were obtained for this study. Results: Pyruvate in the oviduct did not vary with the day of cycle, the mean value was Lactate and glucose concentrations varied with the day of cycle; lactate increasing from 4.87 in the follicular phase to at the time of ovulation, whereas glucose decreased from 3.11 in the follicular phase to 0.50 midcycle and subsequently increased to 2.32 in the luteal phase. The concentrations of pyruvate, lactate, and glucose in uterine fluid remained constant throughout the cycle (0.10, 5.87, and 3.15, respectively). All metabolite concentrations in uterine fluid were significantly different from those in the oviduct midcycle. Cumulus cells readily consumed glucose in vitro, with lactate being the major metabolite produced. Conclusion: These data indicate that lactate and glucose concentrations in the oviduct change with day of cycle and that the human embryo is exposed to different metabolite concentrations as it passes along the tract. Furthermore, cumulus cells readily consume glucose, producing lactate. Therefore, the early human embryo is exposed to low glucose and high lactate levels in vivo. Fertil Steril 1996;65: Key Words: Culture media, glucose, lactate, pyruvate, viability Viability of embryos conceived through IVF is compromised by suboptimal culture conditions, resulting in reduced pregnancy rates (1). The composition of culture media used in IVF clinics varies Received March 16, 1995; revised and accepted July 14, * Supported by IVF America Inc., Greenwich, Connecticut and Monash IVF Pty. Ltd., Melbourne, Victoria, Australia. t Institute of Reproduction and Development, Laboratories of Human and Animal Reproductive Biology. * Reprint requests: David K Gardner, D.Phil., Institute ofreproduction and Development, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia (FAX: ). Department of Obstetrics and Gynecology. 1\ Present address: Beth Lehem, Haglilit, Israel. considerably and has little physiological basis (2). Unfortunately, there are little data concerning the levels of metabolites within the lumen of the naturally cycling human female reproductive tract, primarily because of the lack of techniques capable of analyzing the small amounts of fluid present (3). Previous analyses of human oviduct fluid have used chronic collection techniques (4). Therefore, rather than having physiological values, the concentrations of the embryo metabolites pyruvate, lactate, and glucose in human embryo culture media have been based on those required to sustain the development of mouse embryos in culture (5-7). With the development of quantitative microfluorescence, it is now possible to analyze submicroliter volumes of fluid Gardner et at Environment of the human embryo in vivo 349

2 for metabolites (8, 9). We therefore have used this technology to analyze the metabolite composition of neat samples of human oviduct and uterine fluids throughout the menstrual cycle. A further consideration regarding the supply of metabolites to the oocyte and zygote in vivo is the role of the cumulus mass. The mouse oocyte has a requirement for pyruvate to support maturation and the first cleavage division (6). In the absence of pyruvate, cumulus cells can support maturation by the conversion of glucose to pyruvate (la, 11). Consequently, we have investigated the metabolism ofhuman cumulus cells to determine their role in human oocyte and zygote nutrition. Table 1 Composition of Incubation Medium for Cumulus Cells Compound NaCI () KCI () MgSO. 7H 20 () KH 2PO. () CaCI2 H 20 () NaHCOa () Glucose* Penicillin (gil) Streptomycin (gil) Phenol red (gil) Bovine serum albumin (mg/ml) * Glucose was added at either 1 or MATERIALS AND METHODS Collection of Oviduct and Uterine Fluids Oviduct fluids were collected from the ampullary region of the tube during laparoscopy using a Jansen-Anderson (Cook Australia, Brisbane, Queensland, Australia) catheter. All patients were aged between 25 and 40 years and were being investigated for infertility. All patients had a previously documented regular menstrual cycle of 27 to 28 days and ovulation between days 12 to 16 had been confirmed in all cycles either before or after the laparoscopic cycle. Although strict hormonal evaluation was not made during the cycle of laparoscopy, nevertheless, the ovaries were scrutinized carefully for evidence of ovulation, which allowed a confirmatory assessment of the preovulatory, ovulatory, and periovulatory phases. Uterine fluids were collected using a suction pipelle. Patients included in the study did not exhibit signs of endometriosis. Approximately 0.5 ILL of clear fluid was obtained by each method. Any samples contaminated with blood were discarded. Using this criterion, more samples for analysis were collected from the oviduct. Immediately after collection, samples were snap frozen in liquid nitrogen and were analyzed within 24 hours. Multiple metabolite analysis of each of the submicroliter samples was achieved using microfluorimetry, capable of analyzing nanoliter and picoliter volumes of fluid (9). Collection and Preparation of Cumulus Cells Cumulus cells were sampled from oocyte-cumulus complexes at the time of oocyte retrieval from five patients undergoing IVF. The stimulation protocol for these patients was GnRH agonist with hmg (12). Cumulus cells were dissected from the oocyte-cumulus complex with a 35-gauge needle and subsequently were dispersed with 1 mglml hyaluronidase in a bicarbonate-buffered medium lacking pyruvate 350 Gardner et al. Environment of the human embryo in vivo and lactate (Table 1). Cumulus cells were centrifuged and resuspended twice in fresh medium and then incubated at 10,000 cells per 10 ILL at 37 C, in 5% CO2 in air for up to 24 hours. Samples taken at hourly intervals confirmed that glucose uptake by the cells was linear with respect to time (R 2 = 0.95). The production of pyruvate and lactate was determined at glucose concentrations of 0, 1, and 4. Glucose uptake was determined at 1 and 4. Metabolite analysis was determined by microfluorimetry (9). Microfluorimetry Metabolite concentrations in nanoliter samples of neat luminal fluids and culture media were determined using coupled reactions, in which the pyridine nucleotides NAD(P)H are either generated or consumed. Three nanoliters of fluid sample was added to 30 nl of reagent and the change in reagent fluorescence was quantified using a fluorescent microscope with photomultiplier and photometer attachments and was calibrated with standard curves run with each analysis (9). Statistical Analysis Differences between metabolite levels of luminal fluid samples were initially analyzed by Analysis of Yariance (ANOY A). Differences between phase of cycle were subsequently analyzed by protected F -test. RESULTS At the time of ovulation and while the embryo resides in the oviduct (midcycle; days 12 to 16), there was no significant difference in pyruvate concentration compared with the rest of the cycle. In contrast, the concentration of lactate and glucose varied with the day of cycle. Lactate increased from 4.87 in the follicular phase to between days 12 Fertility and Sterility

3 Table 2 Concentration of Metabolites in Oviduct and Uterine Fluids* Fluid No. of samples Pyruvate Lactate Glucose Oviduct (follicular) Oviduct (midcycle) Oviduct (luteal) Uterus (mean of all days)h ± 0.05t 0.32 ± ± ± ± 0.63t ± 1.48~ 6.19 ± ± ± ± 0.21 **tt 2.32 ± ± 0.31 * Values are means ± SEM. t Significantly different from uterine pyruvate, P < t Significantly different from oviduct midcycle lactate, P < Significantly different from oviduct midcycle glucose, P < II Significantly different from uterine pyruvate, P < ~ Significantly different from uterine lactate, P < ** Significantly different from oviduct luteal glucose, P < tt Significantly different from uterine glucose, P < tt Uterine samples were analyzed initially as follicular, midcycle, and luteal. As there was no significant difference between the groups, they have been treated as one. and 16 (P < 0.05), whereas glucose decreased from 3.11 to 0.50 midcycle (P < 0.01) and subsequently increased to 2.32 in the luteal phase (P < 0.01; Table 2). The concentrations of pyruvate, lactate, and glucose in uterine fluid were analyzed initially during follicular, midcycle, and luteal stages. As there was no difference between the stages, the data were grouped, the mean concentrations being 0.10, 5.87, and 3.15, respectively (n = 15; Table 2). All metabolites in the uterus were present at significantly different concentrations to those present in the oviduct midcycle (Table 2). Glucose uptake and the production of pyruvate and lactate by cumulus cells from five patients is shown in Figure 1. There was a linear increase in glucose uptake and pyruvate and lactate production with respect to glucose concentration. Lactate was the major metabolite produced. 800r , c: o :::I e j 200! :::> Glucose Concentration () Figure 1 The uptake of glucose and production of pyruvate and lactate by human cumulus cells., glucose; +, pyruvate; 0, lactate. Each point on the graph represents the mean ± SEM of five samples. DISCUSSION This study shows that the concentration of metabolites to which the human oocyte and early embryo are exposed to in vivo differ along the female reproductive tract. In addition, lactate and glucose concentrations in the oviduct differ midcycle when the oocyte and embryo are present. The mean values for pyruvate and lactate in the oviduct are similar to those reported previously by Lopata et al. (13). However, in the previous study (13), only two samples were analyzed from patients undergoing gonadotropin treatment. Furthermore, pyruvate and lactate concentrations are significantly higher in oviduct fluid midcycle than in uterine fluid. In contrast, the glucose concentration within the oviduct is reduced significantly at the time when the oocyte and early embryo are present and is at its highest concentration in the uterus. The changing levels of metabolites along the reproductive tract mirror changes in the physiology of the human embryo, which has a preference for pyruvate as an energy source during the cleavage stages (14). Glucose uptake remains low until the blastocyst stage. Such observations also are consistent with in vitro studies that have shown that glucose in culture media is inhibitory to early human embryo development (15). The observed decline in glucose concentration in the oviduct lumen at the time of ovulation indicates that the early human embryo is not exposed to high glucose levels in vivo. The findings of this study are consistent with that reported for chronically collected human oviduct fluid (4) in which the glucose concentration fell from 3.04 to 2.40 after ovulation. Similarly, in the pig (16), glucose levels in the ampullary-isthmic junction were reduced significantly after ovulation. The fall in glucose concentration around the time of ovulation is consistent with it being a major energy source of the oviduct. An increase in glucose requirement by the oviduct could arise from m- Gardner et al. Environment of the human embryo in vivo 351

4 creases in both secretory activity and muscular and cilia movements around the time of ovulation (16). The decrease in glucose concentration in the oviduct at midcycle (2.6 ) can be accounted for by the relative increase in lactate concentration (5.6 ), assuming that 1 mol of glucose can form 2 mol of lactate. Alternatively, the fall in glucose concentration could be due to changes in carrier-mediated glucose transport by the oviduct epithelium. Edwards and Leese (17) showed in the rabbit that there was a significant decrease in the carrier-mediated transport of glucose into the oviduct in the days after mating. It is not feasible that the change in glucose concentration was due to changes in volume of oviduct fluid, as the concentrations of other metabolites increased. In contrast to the oviduct, glucose concentration in the uterus did not vary with day of cycle, in agreement with the findings of Casslen and Nilsson (3). However, Hughes et al. (18) did find an increase in uterine glucose concentration around days 14 to 16 of the cycle. Irrespective of this apparent discrepancy, we did observe that the glucose concentration that the embryo would be exposed to in the uterus was significantly greater than in the oviduct. Previous studies on the role of cumulus cells in providing nutritional support to the oocyte and embryos have focused on their production of pyruvate (11, 19). However, it is evident that lactate is the predominant metabolite produced by the human cumulus cell when incubated with glucose as the sole substrate. At 1 and 4 glucose, lactate production was 25 times greater than pyruvate. The significance of this observation, together with the greater lactate concentration in the oviduct (10.50 ) compared with the uterus (5.87 ), is that the human embryo will be exposed to a significant lactate gradient, which decreases as embryo development proceeds. Conversely, the oocyte and early embryo will be exposed to a low glucose concentration, which will increase as the embryo progresses along the tract and into the uterus. In the mouse it has been shown that viability of the cleavage stage embryo in culture is maintained by a high lactate concentration (23 ) in the medium, whereas for postcompaction stages, a reduced lactate concentration is required (20). Furthermore, it has been shown that exposure of the early mouse or hamster embryo to glucose is detrimental to subsequent development (21,22). The observed metabolic activity of cumulus cells, and therefore their ability to modify the composition of the culture medium, may explain in part their apparent beneficial effects in coculture of the human embryo (23). The concentrations of lactate (10.5 ) and glucose (0.5 ) measured in the oviduct tit midcycle 352 Gardner et ai. Environment of the human embryo in vivo are significantly different from those present in the human embryo culture medium, human tubal fluid (HTF) (7). This medium was formulated around the ionic environment of fallopian tube fluid, with particular emphasis on the high potassium concentration. The levels of metabolites in medium HTF are those present in conventional mouse embryo culture media (lactate 21.4 ; glucose 2.78 ) (2), which have little homology with those present in human oviduct fluid. It has been documented that the concentration of metabolites in mouse embryo culture media can affect significantly their metabolism and subsequent developmental potential after transfer (20, 24). Therefore, the role of metabolite concentration in regulating human embryo development warrants further investigation. In conclusion, we have shown that the human embryo is exposed to a changing metabolite pool in vivo. Therefore, when considering extended human embryo culture, one should use more than one culture medium. Each medium should reflect the environment of the embryo in vivo and should fulfill the changing requirements of the embryo. To this end, we have formulated two culture media based around the levels of metabolites in the human female reproductive tract: medium G 1, designed to support the human embryo for the first 48 to 72 hours of growth, and medium G2, which supports development of the later stages. In a pilot study, human embryos cultured in medium Gl had significantly more cells than sibling embryos in medium HTF (25). Furthermore, we have used both media Gland G2 to support human embryo development to the blastocyst, resulting in successful pregnancy (Barnes FL, Trounson AO, Gardner DK, unpublished observations). Acknowledgments. The authors thank Antje Spitzer, B.Sc., Institute of Reproduction and Development, Monash University, Melbourne, Victoria, Australia, for technical assistance with analysis on cumulus cell metabolism, and Louise Edwards, D.Phil., Institute of Reproduction and Development, for comments on the manuscript. REFERENCES 1. Biggers JD. In vitro fertilization and embryo transfer in human beings. N Engl J Med 1981;304: Gardner DK, Lane M. Embryo culture systems. In: Trounson A, Gardner DK, editors. Handbook of in vitro fertilization. Boca Raton (FL): CRC Press, 1993: Casslen B, Nilsson B. Human uterine fluid, examined in undiluted samples for osmolarity and the concentrations of inorganic ions, albumin, glucose, and urea. Am J Obstet Gynecol 1984; 150: Lippes J, Enders RG, Pragay DA, Bartholomew WR. The collection and analysis of human fallopian tubal fluid. Contraception 1972; 5: Brinster RL. Studies on the development of mouse embryos Fertility and Sterility

5 in vitro. IV. Interaction of energy sources. J Reprod Fertil 1965; 10: Biggers JD, Whittingham DG, Donahue RP. The pattern of energy metabolism in the mouse oocyte and zygote. Proc Natl Acad Sci USA 1967;58: Quinn P, Kerin JF, Warnes GM. Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubal fluid. Fertil Steril 1985;44: Mroz EA, Lechene C. An NADH-coupled assay for femtogram of nanogram quantities of chymotrypsin. Anal Biochem 1983; 128: Gardner DK, Leese HJ. Assessment of embryo metabolism and viability. In: Trounson A, Gardner DK, editors. Handbook of in vitro fertilization. Boca Raton (FL): CRC Press, 1993: Donahue RP, Stern S. Follicular cell support of oocyte maturation: production of pyruvate in vitro. J Reprod Fertil 1968; 17: Leese HJ, Barton AM. Production of pyruvate by isolated mouse cumulus cells. J Exp ZooI1985;234: Calderon I, Healy D. Endocrinology of IVF. In: Trounson A, Gardner DK, editors. Handbook of in vitro fertilization. Boca Raton: CRC Press, 1993: Lopata A, Patullo MJ, Chang A, James B. A method for collecting motile spermatozoa from human semen. Fertil Steril 1976;27: Hardy K, Hooper MAK, Handyside AH, Rutherford AJ, Winston RML, Leese HJ. Non-invasive measurement of glucose and pyruvate uptake by individual human oocytes and preimplantation embryo. Hum Reprod 1989;4: Conaghan J, Handyside AH, Winston RML, Leese HJ. Effects of pyruvate and glucose on the development of human preimplantation embryos in vitro. J Reprod Fertil1993;99: Nichol R, Hunter RHF, Gardner DK, Leese HJ, Cooke GM. Concentrations of energy substrates in oviductal fluid and blood plasma of pigs during the peri-ovulatory period. J Reprod Fertil 1992; 96: Edwards LJ, Leese HJ. Glucose transport and metabolism in rabbit oviduct epithelial cells. J Reprod Fertil1993;99: Hughes EC, Jacobs RD, Rubulis A, Husney RM. Carbohydrate pathways of the endometrium. Effects of ovular growth. Am J Obstet GynecoI1963;85: Ahuja KK. Fertilisation: the importance of scientific research to clinical practice. Br J Fam Plann 1988; 13: Gardner DK, Sakkas D. Mouse embryo cleavage, metabolism and viability: role of medium composition. Hum Reprod 1993;8: Chatot CL, Ziomek CA, Bavister BD, Lewis JL, Torres I. An improved culture medium supports development of randombred 1-cell mouse embryos in vitro. J Reprod Fertil 1989; 86: Schini SA, Bavister BD. Two-cell block to development of cultured hamster embryos is caused by phosphate and glucose. BioI Reprod 1988;39: Saito H, Hirayama T, Koike K, Saito T, Nohara M, Hiroi M. Cumulus mass maintains embryo quality. Fertil Steril 1994;62: Gardner DK, Leese HJ. Concentrations of nutrients in mouse oviduct fluid and their effects on embryo development and metabolism in vitro. J Reprod FertiI1990;88: Gardner DK. Culture of mammalian embryos in absence of serum or somatic cells. Cell BioI Int 1994; 18: VoL 65, No.2, February 1996 Gardner et al. Environment of the human embryo in vivo 353