1 IVF Unit, Women's Hospital, Hamad Medical Corporation, P.O. Box 3050, Doha, Qatar and 2 JA Global Pty Ltd, P.O. Box 151, Curtin, ACT 2605, Australia
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: co-culture/embryos/live birth/micro-droplet/protein-free medium
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The aim of the present investigation was to formulate a culture medium completely devoid of protein or protein derivatives of biological origin. This medium should allow fertilization of human oocytes by spermatozoa prepared in the same medium. Finally the protein-free medium should also support the generation of viable day 2 cleaved human embryos from the fertilized oocytes.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Human sperm preparation
Male patients produced semen by masturbation. Spermatozoa were prepared by the standard swim-up technique and occasionally with Percoll density gradient centrifugation. In the Percoll density gradient procedure, the recovered pellet was washed twice and resuspended in culture medium [Medi-CultTM (Medi-Cult, Copenhagen, Denmark), IVF-50TM (Scandinavian IVF Science, Gothenburg, Sweden), ART-6, ART-7 or ART-7b (all formulated and prepared in our laboratory see below)]. Sperm preparations were made using test or control media as appropriate. Specifically, spermatozoa for intracytoplasmic sperm injection (ICSI) or in-vitro fertilization (IVF) were prepared in the protein-free ART-7 or ART-7b media for the experiments on the effect of protein deficiency on embryo development. The harvested spermatozoa were incubated under 5% CO2 in air at 37°C until use.
Ovarian stimulation and oocyte retrieval
Ovarian stimulation was induced by s.c. injections of gonadotrophin-releasing hormone agonist (buserelin; Suprefact; Hoechst, Frankfurt, Germany) starting in the mid-luteal phase until menstruation or down-regulation was achieved. Down-regulation was considered to have been achieved when the endometrial thickness was 4 mm, and when blood oestradiol, progesterone and luteinizing hormone (LH) concentrations had reached baseline values. Injections of follicle stimulating hormone (FSH; Metrodin; Serono, Rome, Italy) were administered for 3 days to initiate recruitment of follicles, subsequent to which human menopausal gonadotrophin (Pergonal 500; Serono, Rome, Italy) was administered according to the response of the patient to stimulate follicular development. Ovulation was induced by an injection of 10 000 IU human chorionic gonadotrophin (HCG; Pregnyl; Organon, Oss, Holland) after achieving a follicular size of
16mm. Oocyte retrieval (OR) was performed 36 h later by ultrasound-guided transvaginal aspiration.
Insemination and culture techniques
In-vitro fertilization
Oocytes were individually inseminated with 100 000/ml motile spermatozoa in micro-droplets of equilibrated culture medium under equilibrated and embryo-tested mineral oil (Sigma Chemicals, St Louis, MO, USA). Prior to the introduction of the oocyte, these microdroplets containing the spermatozoa were 200 µl in size. The inseminated oocytes were cultured in an atmosphere of 5% CO2 in air at 37°C. The following morning the oocytes were denuded with denuding pipettes (International Medical Products BV, Zutphen, The Netherlands).
Intracytoplasmic sperm injection
In cases of male factor subfertility, the oocytes were prepared for ICSI. The oocytes were exposed to hyaluronidase solution for 30 s (80 IU/ml; Medi-Cult a/s, Lerso Parkalle 42, 2100 Copenhagen, Denmark) and then transferred into equilibrated Medi-CultTM or the protein-free media. The oocytes were incubated for 57 min in the culture medium under 5% CO2 in air at 37°C and then denuded with a denuding pipette as above. The denuded oocytes were washed and finally incubated in culture medium for a further 3060 min.
Commercially available ICSI pipettes were used for micromanipulation (Laboratoire CCD, Paris, France or Humagen, Charlottesville, VA, USA). A 5 µl ultra micro-droplet of polyvinylpyrrolidone (PVP; Medi-Cult a/s) was thinly spread at the centre of a Petri dish (Falcon Plastics, Becton Dickinson, Rutherford, NJ, USA). The PVP spread was surrounded by up to five 10 µl micro-droplets of HEPES-buffered IVF medium (Gamete-100; Scandinavian IVF Sciences AB, Gothenberg, Sweden). The micro-droplets were overlaid with equilibrated and embryo tested mineral oil (Sigma). 1 to 2 µl of the sperm preparation was introduced at the centre of the PVP spread. ICSI of mature oocytes was performed by the standard procedure.
Cumulus co-culture
The cumulus cells from mature denuded oocytes were washed by centrifugation three times in the protein-free ART-7 medium. After the final wash the cumulus cells were resuspended in 0.5 ml medium. Equilibrated 20 µl micro-droplets of the ART-7 medium made in 4-well dishes (Nunc, Copenhagen, Denmark) were seeded with the cumulus cells. The dishes were incubated under of 5% CO2 in air at 37°C for ~1 h. The cumulus cells firmly attached to the base of the dish within 1 h. The attached cumulus cells were washed up to 12 times with equilibrated ART-7 medium with a finely drawn Pasteur pipette with minimal agitation. During each wash, unattached cumulus cells and red blood cells were siphoned off. When the cell monolayer was completely free of blood cells and other debris, the dishes were equilibrated for at least 1 h. Immediately after ICSI the oocytes were washed in ART-7 medium three times and then co-cultured with the cumulus cells in an atmosphere of 5% CO2 in air at 37°C.
Ultra micro-droplet culture
Ultra micro-droplets of culture medium were made in 4-well dishes previously filled with equilibrated mineral oil. A volume of 1.52 µl was used for culture of three to seven human embryos or 5 µl for culture of up to 20 mouse embryos. The dishes were incubated under 5% CO2 in air at 37°C. The medium was changed daily using a finely drawn Pasteur pipette and equilibrated culture medium.
Assessment of fertilization
Fertilization was determined 1820 h after ICSI or IVF. The oocytes were considered fertilized when two distinct pronuclei were visible. Fertilized oocytes were cultured in micro-droplets or ultra micro-droplets of equilibrated medium in an atmosphere of 5% CO2 in air at 37°C. Cleavage and embryo quality were assessed 24 h later.
Assessment of the quality of day 2 cleaved embryos
Two parameters were employed to determine the quality of day 2 embryos: the average blastomere score and the average embryo grade. These were calculated as follows:
|
|
Randomization of human oocytes
All oocytes collected were pooled and subjected to hyaluronidase treatment. Following this, eggs were apportioned roughly in the ratio of 3:7 or 4:6 with the larger portion for the control group. This was performed to safeguard the interest of the patients (experiments 10 and 11). However, once the safety of the protein-free medium was firmly established, oocytes were apportioned equally (experiment 12). The oocytes were apportioned at random without bias. Separate personnel were involved in denuding and allocation of denuded oocytes for separate treatments to prevent bias.
Following the evaluation of the protein-free medium with mouse embryos and with 1PN and 3PN human embryos, it appeared that the protein-free medium was safe for human application. The patients were informed that their embryos were to be cultured in different culture media. Most patients, especially those with repeated failures, were keen to try new media and techniques to help establish pregnancy.
IRB approval was not sought as the composition of the protein-free medium did not contain substances or compounds that have never been used to culture human embryos in the past except for one component. The single component of the protein-free medium that has never been used in human embryo culture media in the past has, however, been consumed by humans for decades as a food additive, in pharmaceutical preparations and in eye drops without any detrimental effect.
Media formulation
The formulation of protein-free medium involved the substitution of albumin with other components that could perform the same functions in culture. Optimal concentrations of various media components were determined, including amino acids (alanine, aspartate, glutamate, glutamine, glycine, serine, taurine), antioxidants and chelators (catalase, citric acid, desferrioxamine, EDTA, glutathione, magnolol, pentoxifylline, probucol, superoxide dismutase, uric acid), energy sources (fructose, glutamine, sodium pyruvate), osmolytes (mannitol, myoinositol), vitamins (ascorbic acid, cyanocobalamin, folic acid, tocopherol) and elemental iron. These were tested in a 1-cell Swiss outbred (SO) and or F1 [(CBAxC57BL/6J)xSO] mouse embryo assay in Earle's balanced salt solution (Sigma Chemical Co.). Optimal concentration was defined as that which supported the highest number of hatched blastocysts. Three media (ART-1, ART-2 and ART-3) were formulated and investigated in the present study.
Media
Media ART-1, -2 and -3 contained bovine serum albumin (BSA) (Sigma; 5 mg per ml), while the ART-6 medium contained HSA (Irvine Scientific, Santa Ana, CA, USA; 5 mg per ml). The ART-7 series of media contained no added protein or protein derivatives. ART-1, -2, -3 and -6 media contained Earle's salts in addition to other components. The ART-7 series contained additional amino acids, antioxidants, osmolytes and vitamins and did not resemble the ART-1, -2, -3 or -6 media. The mineral salt composition of the ART-7 series of media was not similar to Earle's salts. The control media containing protein (HSA) were the commercial IVF-50TM and Hatch-50TM media (Scandinavian IVF Science AB, Gothenburg, Sweden), and Medi-CultTM medium (Medi-Cult a/s, Copenhagen, Denmark). The exact composition of the various media employed is not disclosed for commercial reasons.
Specific experiments
Experiment 1
The objective of this experiment was to determine the ability of the three ART media to support the development of SO mouse 1-cell embryos. The control comprised IVF-50TM and Hatch-50TM media. The IVF-50TM medium was replaced with a mixture of 50% IVF-50TM and 50% Hatch-50TM medium on day 3 of culture, i.e. at about 72 h post-copulation. On day 4 and beyond, the IVF-50TM medium was completely replaced with Hatch-50TM medium. In contrast, the embryos were cultured continuously in the three test ART media. The end points were embryo cavitation on day 4, and blastulation and hatching/hatched rates on days 5 and 6.
Experiment 2
Experiment 1 was repeated with the blocking strain of (SO) embryos to confirm the previous finding that the ART-1 medium could overcome developmental blocks during culture from the 1-cell stage. The control medium used was IVF-50TM/Hatch-50TM.
Experiment 3
The ART medium identified as the optimal formulation (ART-1) in experiment 1 was compared with IVF-50TM and Medi-CultTM media in an assay of 1-cell mouse embryos resulting from a cross between (CBAxC57BL/6J) females and SO males. In this experiment the Hatch-50TM and M3TM media which are the day 3 sequential media for IVF-50TM and Medi-CultTM media respectively were not used. Instead the embryos were cultured continuously in the same medium. This experiment was intended to expose the differences between the various media tested when the more complex and enriched day 3 sequential media were not used.
Experiment 4
This experiment was performed to determine the ability of the ART-6 medium (identical to ART-1 medium except for the protein source BSA which was replaced with HSA) to support human oocyte fertilization and subsequent development in vitro and in utero. The control medium was Medi-CultTM. Spermatozoa for insemination were prepared in the respective control and test media.
Experiments 57
The objective of these experiments was to determine whether the protein-free ART-7 medium could support the development of 1-cell, 2-cell and 4-cell embryos of the blocking strain of SO mice. These experiments were performed by one of us (J.A.) at the Laboratory of Dr P.J.McCullagh, Developmental Physiology Group, John Curtin School of Medical Research (JCSMR), The Australian National University, Canberra, Australia. The Animal Breeding Establishment of the JCSMR maintains a genetically pure and uncontaminated colony of the SO mice. The founders of this strain had been obtained in 1985 from the Animal Research Centre, Perth, Western Australia, which had originally purchased them from Charles River, New York. In JCSMR they have been bred as a closed colony using the systematic five unit breeder rotation (Poiley, 1960). The ART-7 medium (without added protein) and the control, which was identical to ART-7 medium with added protein (BSA added just prior to use), were made up in Qatar and transported in dry ice to Australia by air. The embryos were cultured communally in groups not exceeding 20 in ultra micro-droplets of ~5 µl.
Experiment 8
Human day 1 unipronucleate (1PN) and tripronucleate (3PN) embryos generated in Medi-CultTM or IVF-50TM media were cultured in protein-free medium. The objective of this investigation was to determine the ability of the protein-free ART-7 medium to support the development of human embryos and the developmental stages that can be attained by human embryos in this medium.
Experiment 9
The survival of motile human sperm preparations prepared by the swim-up method was determined over a 48 h period when incubated under an atmosphere of 5% CO2 in air at 37°C. Semen samples were equally apportioned. One portion was processed with the test ART-7 medium while the other was processed with the control IVF-50TM medium.
Experiment 10
Human sibling oocytes were cultured in micro-droplets (20 µl) of the ART-7 protein-free medium immediately after ICSI. The development of the fertilized embryos until day 2 was compared with that of sibling ICSI embryos cultured under similar conditions in the control IVF-50TM medium.
Experiment 11
This experiment was similar to experiment 10 except that co-culture with homologous cumulus cells was incorporated in an attempt to improve further the quality of the sibling embryos generated in the protein-free ART-7 medium after ICSI when cultured in micro-droplets. The sibling embryos in the control medium (IVF-50TM) were cultured under similar conditions but not co-cultured with cumulus cells.
Experiment 12
This experiment was similar to experiment 10 but ART-7b medium (modified ART-7 medium) was used instead of ART-7 medium. The modification required the formulation and assessment of another 16 media formulations. The composition of ART-7b medium did not resemble those of ART-1, -2, -3 and -6. This modification was made to improve the quality of the medium and to overcome dependency on cumulus cells to generate better quality embryos. Furthermore in this experiment instead of micro-droplets, ultra micro-droplets of 1.52 µl were used to culture the embryos. This was done to determine whether the quality of the sibling embryos cultured in the protein-free milieu can be improved by the influence of paracrine and autocrine factors released by the embryos. The low volume was employed to prevent dilution of these factors (Paria and Dey, 1990). Three to seven embryos were cultured communally in the ultra micro-droplets. Control sibling embryos were initially cultured in micro-droplets of Medi-CultTM medium due to difficulty in stabilizing the ultra micro-droplet at the bottom of the dish under oil. Later, by using a finely drawn glass pipette, ultra-microdroplets of the control medium containing protein were introduced and stabilized at the bottom of the well and these were employed in all subsequent cases. This problem was not encountered with the protein-free media.
Clinical trial
A clinical trial was approved by the IVF Unit Management of the hospital for the transfer of embryos generated in the protein-free ART-7 and ART-7b media. Initially only six transfers of three embryos per transfer were performed with embryos generated in the ART-7 medium with cumulus co-culture. Of these six transfers, two recipients became pregnant and proceeded to term uneventfully. Following the birth of a normal singleton and a normal set of twins, the ART-7 series of media was felt to be safe for human application. The clinical trial was continued from October 1998 with embryos generated in ART-7b medium cultured in ultra micro-droplets under oil. The only selection criterion was that a patient had 12 follicles. Patients fulfilling this criterion were selected at random for the study. In general, a larger proportion of oocytes was allocated to the control group for every patient to ensure that there was no compromise of her interests in the evaluation of the protein-free medium. This trial is currently ongoing.
Statistical analysis
Statistical comparisons were performed with the two-sample t-test or two-by-two tables. P 0.05 was considered statistically significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
|
|
|
|
Experiment 12
The fertilization rate of sibling oocytes after ICSI was similar in the control Medi-CultTM and ART-7b media (Table XI). The quality of ICSI sibling embryos generated in the ART-7b medium, cultured in ultra micro-droplets was significantly superior in terms of blastomere score on day 2 compared to those generated in the control medium. The average grades of embryos generated in the ultra micro-droplet ART-7b medium and the control medium were similar. Developmental arrest at the zygote stage was significantly higher in Medi-CultTM medium (19.5%) compared to ART-7b medium (5.9%).
|
|
The remaining two patients (nos 19 and 20) each received three and two embryos generated in the ultra micro-droplet cultures of the ART-7b medium as well as one embryo each generated in the control Medi-CultTM medium. Both the patients are pregnant with a set of twins and triplets respectively. Therefore at least one implantation in the twin pregnancy and two implantations in the triplet pregnancy were attributed to embryos generated in the ultra micro-droplet cultures of ART-7b medium.
The embryos transferred in one of the singleton pregnancies were generated in frozenthawed ART-7 medium suggesting that the protein-free medium can be stored frozen. This pregnancy proceeded to term with the birth of a normal boy.
In addition, one patient (no. 8) who received two embryos generated in the ART-7 medium with cumulus co-culture and two embryos in the ultra micro-droplet ART-7b culture systems is pregnant with a set of twins. The overall clinical pregnancy rate was 50% and the implantation rate for embryos generated in the ART-7 (with cumulus co-culture) and ART-7b media is 31.6% (18/57).
Finally, among another group of six patients (nos 2126) who received embryos generated in both the ART-7b ultra micro-droplet cultures and in control Medi-CultTM medium, two are pregnant with singletons and one with a set of twins. These pregnancies cannot be attributed with certainty to any particular medium. The clinical trial is ongoing.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Blastocyst hatching and outgrowths are considered the most accurate assessment of embryo viability (Drakakis et al., 1996).
There are a number of reports of development of mouse blastocysts from zygote and cleaved stages in protein-free media. The earliest report was probably that Cholewa and Whitten (1970) who used polyvinylpyrrolidone (PVP), which is a high molecular weight colloid, as a lubricant and for increasing the viscosity of the medium. Subsequent to this work a number of other workers have successfully cultured mouse and other mammalian embryos in protein-free media (Dandekar and Glass, 1990; Spindle, 1995
; Li et al., 1996
; Schramm and Bavister, 1996
). Recently Biggers et al. (1997) investigated the effect of replacing BSA with PVA and/or amino acids on mouse zygote development. They observed that PVA could not be substituted completely for BSA in the mouse embryo culture medium. Blastocyst development was only slightly less than with BSA, but the rate of partial hatching was significantly less. Substitution of BSA with PVA lowered the overall response but did not lead to major perturbations.
PVP and PVA are not sources of fixed nitrogen and they do not perform the various biological roles of protein. In addition to its many biological roles, BSA also conferred the useful physical attributes of lubrication and viscosity in the culture medium. The incorporation of PVP and PVA merely serve to duplicate the physical attributes of BSA. Increased viscosity and lubrication in the culture medium promotes ease of handling and manipulation of the embryo and prevents it adhering to the walls of the culture dish. The teratological properties of PVP and PVA have not been fully examined, raising questions about their use in human therapeutic assisted reproduction (Gardner and Lane, 1998).
Some workers have shown that replacement of serum protein with a single antioxidant and chelator such as EDTA does not impair fertilization and cleavage in the mouse and human (Mehta and Kiessling, 1990; Serta et al., 1997
). However, Serta et al. (1997) suggested that the embryo transfer catheter be rinsed extensively with the protein-free medium which implies the tendency of embryos to stick to the inner wall of embryo transfer catheter. Serum protein also has a role in maintaining pH in the culture medium (Moessner and Dodson, 1993
). Beside its role as a nutrient in biological systems, protein has a number of other roles such as a chain-breaking antioxidant and a chelator of metal ions (Barber, 1961
; Vidlakova et al., 1972
; Wayner et al., 1987
).
The physiological functions of albumin and plasma proteins in general are well documented. The role of albumin in preventing membrane peroxidation indicates a direct role in membrane stability. It is involved in capillary membrane permeability and in osmoregulation. Albumin provides 80% of the total colloid osmotic pressure in plasma. Albumin is involved in the transport of carbon dioxide and acts as a pH buffer. Albumin accounts for the greatest (95%) portion of the non-bicarbonate buffer value of plasma. Proteins also serve as a source of energy. Deaminated alanine is pyruvate, which can be either converted to acetyl-CoA or glucose and glycogen. Albumin may help to solubilize lipids and transports hormones, vitamins and metals, serving as a reservoir for the release and use of these components.
Any attempt at substituting serum albumin in culture medium should therefore take into consideration its role in vivo and as well as its physical attributes, which are useful for embryo handling and manipulation in vitro. In the present study, we have examined individual components such as amino acids, antioxidants and chelators, osmolytes, vitamins and alternative energy sources that could substitute in part the various roles of protein in vivo and in vitro. A single component may not fulfil all the functions of serum protein. However, many components with varying properties collectively could perform most of the functions of serum protein in the culture medium. This approach to the formulation of a chemically defined embryo culture medium has resulted in protein-free media (ART-7 and ART-7b) capable of supporting fertilization of human oocytes with spermatozoa prepared in the same medium. The resultant zygotes developed in the protein-free medium to form viable early cleavage-stage human embryos. The transfer of these embryos led to normal pregnancies and live births.
The present study has demonstrated that the ART-1, -2 and -3 media can support the continuous development of the mouse embryos from day 1 up to the last day of culture. Whereas the control IVF-50TM medium had to be replaced with Hatch-50TM medium on day 3 to increase the number of blastocysts generated from ~61% in embryos of F1 mice to ~8486%, blastocysts generated in the ART media attached to the base of the dish and continued to expand until the last day of observation (day 7/8). Hatched and hatching blastocysts generated in the control IVF-50TM medium or IVF-50TM and Hatch-50TM sequential media system collapsed, and detached from the base of the dish on day 5/6. The reason for the appearance of the black spots on collapsed blastocysts is not clear. Based on these findings it appears that the quality of embryos and blastocysts generated in the ART media was healthier than those generated in the control IVF-50TM/Hatch-50TM media. Developmental blocks in SO mouse embryos did not occur in ART-1, -2 and -3, and control media, although the SO embryos were known to block at time of activation of the embryonic genome (Du and Wales, 1993).
In experiment 4, the BSA in ART-1 medium was removed and replaced with HSA. This medium was referred to as ART-6 medium. The fertilization rate, zygote cleavage arrest rate and embryo quality were similar in both control (Medi-CultTM) and ART-6 media. In a separate retrospective study of oocyte retrievals performed during the same period (19961997), we observed similar pregnancy rates in Medi-CultTM (22.1%, n = 925, P = 0.87) and IVF-50TM media (20.7%, n = 305, P = 1.00) compared to ART-6 medium (20.0%, n = 50) (J.Ali, unpublished observations). These findings suggest that the quality of embryos generated in the ART-6 medium was similar to that in Medi-CultTM and IVF-50TM media.
In experiments 57, the SO 1-cell mouse embryos generally blocked at the 2- and 4-cell stages when cultured in the ART-7 medium with and without supplemented protein. When in-vivo generated 2- and 4-cell embryos were cultured in the same media, most progressed to the blastocyst stage and subsequently herniated and hatched. The proportion of blastocysts formed when cultured from the day 2 cleaved stages was similar in the protein-free and protein-supplemented ART-7 media. The proportion of hatching or hatched blastocysts that developed in the protein-deficient ART-7 medium was less than in the protein-supplemented medium, but this difference was not statistically significant.
In contrast to observations on the SO mouse zygotes, human zygotes did not show developmental blocks during early cleavage in the protein-deficient ART-7 medium, but a large proportion of 1PN and 3PN embryos blocked at 1216-cells stages on day 4 of culture. Only a small proportion of these embryos compacted (29.2%) and a smaller proportion cavitated (12.5%), developing into poor quality blastocysts. It appears that ~50% of polyploid zygotes cleave at a rate similar to that of diploid embryos. Some of the remaining polyploid zygotes may cleave subsequently but a third of all polyploid zygotes fail to cleave at all (Boyers et al., 1987). In our experience, however, most polyploid zygotes cleaved and developed at a rate similar to that of diploid embryos. Balakier (1993) reported that 93% of tripronucleate embryos cleaved, 25% reached the 8-cell stage, 19% progressed to the 13-cell stage and 6% became blastocysts. In the present study, all the unipronucleate and triploid zygotes cleaved and a high proportion attained the 8-cell stage (95.8%), 87.5% progressed to the 1216-cell stages and 12.5% formed poor blastocysts. In general these findings suggest that ART-7 medium could support development of human embryos to the 1216-cell stages, which is similar to that which can be attained in most simple culture media. In this experiment, the medium was not changed and embryos were cultured continuously. The developmental arrest at the compacted morula stage observed in the present study in the 1PN and 3PN embryos could therefore be due to accumulation of toxic levels of ammonium. The inclusion of amino acids in embryo culture medium could be detrimental to embryo development due to the production of embryotoxic amounts of ammonium. To overcome this problem, the medium was changed daily in subsequent experiments (from experiment 10 onwards) as previously suggested (Lane and Gardner, 1997
). It is also plausible that another medium more suited to the human morula stage could replace ART-7 on day 4 of culture to obtain more blastocysts. The need for a sequential culture system to achieve optimal culture of mammalian embryos in vitro is well recognized (Gardner, 1994
; Barnes et al., 1995
). Future research needs to be directed toward formulation of a day 4 protein-free medium to replace ART-7 medium on day 4 of culture. However, human embryos are generally transferred at the day 2 or 3 cleaved stages. In the present study, ART-7 medium supported the development of most zygotes to the cleaved day 2 (100%) or day 3 (95.8%) or day 4 (87.5%) morula stages and therefore could serve a practical purpose in human IVF treatment procedures.
Human spermatozoa processed in the protein-free medium survived after 24 and 48 h in an atmosphere of 5% CO2 in air at 37°C. The proportion that survived in the ART-7 medium was significantly greater at 24 h and similar at 48 h compared to the spermatozoa processed in the control IVF-50TM medium (79% versus 50.1%, P = 0.0113; 41.7% versus 26.2%, P = 0.222 respectively; n = 22). A deficiency of protein in the medium did not impair fertilization of human oocytes by ICSI or conventional IVF (Ali, 1997; Ali et al., 1998
).
In experiment 10, the embryos generated in ART-7 medium were slightly slower in development (average blastomere score: 2.7) compared to those generated in the control IVF-50TM medium (average blastomere score: 3.2) but this difference was not statistically significant. The quality of the embryos in ART-7 medium (average grade: 2.7) significantly improved with homologous cumulus cell co-culture (average grade: 3.3) (Table X). Moreover, the cleavage rate of embryos generated in the ART-7 cumulus co-culture system was significantly faster (average blastomere score: 3.5) than in the control medium (average blastomere score: 2.8). It has been demonstrated that cumulus cells secrete proteins which could have been utilized by the embryos for development activities (Tesarik et al., 1988
). Cumulus cells may also have conferred positive or negative conditioning to the protein-free culture milieu (Saito et al., 1994
).
In general, the survival of cumulus cells that remained attached in the ART-7 medium beyond 24 h (experiment 11) appear to be of limited duration. The presence of protein in medium appears to be necessary for cells to adhere to the base of the dish. Thus, absence of protein from the ART-7 medium may have caused the cells to detach. It is unlikely that all the detached cells were dead. However, ~1015% are expected to die overnight. Since very few cells remained after washing, their contribution to the quality of embryos ~24 h after oocyte retrieval appears unlikely. This suggests that the conditioning of the medium by the cells occurs prior to and a few hours after fertilization. The influence of cumulus cells in culture was short (calculated at ~22 h) but the long-term effects and impact on the quality of embryos generated was significant. Cumulus cell-conditioned medium does not appear to be necessary for cleavage, and it is therefore not necessary to continue with co-culture beyond the first 24 h of culture to generate quality day 2 human embryos. Indeed it has been shown that intact cumulus cells are beneficial up to at least 7 h after IVF, but not at 20 h (Zhang et al., 1995).
Cumulus co-culture could be useful in the generation of better quality embryos. Co-culture is, however, unlikely to remain a permanent feature in embryo culture procedures because it is not chemically defined. It introduces numerous confounding variables that make investigation of the nutrient requirements and metabolism of embryos very difficult. The risk of contamination of the co-culture system with microorganisms is high. Co-culture is demanding on staff time. In experiment 12, ultra micro-droplets of the ART-7b protein-free medium were used in an attempt to eliminate the dependency on cumulus cells for the generation of better quality and viable human embryos in a protein-free culture system. The ART-7b medium was a modification of the ART-7 medium. As observed in experiment 4, a significantly higher proportion of zygotes underwent developmental arrest in the control medium as compared to ART-7b medium. The quality of embryos generated in the ART-7b culture medium and ultra micro-droplet culture system was better than those generated in the control medium containing protein. This difference was statistically significant with regard to average blastomere score but not to embryo grade. The development of embryos in the ART-7b ultra micro-droplet culture system was significantly faster (average blastomere score: 3.9) than that in the control medium (average blastomere score: 3.1) (P = 0.0401). Furthermore communal culture of embryos in ultra micro-droplets of ART-7b medium without cumulus co-culture appears to produce good quality and viable human embryos. Clinical pregnancies (58%; 7/12 transfers) were achieved with this culture system. The implantation rate of 41.7% (15/36) obtained in the clinical trial with embryos generated in the ART-7b medium without cumulus co-culture was similar to that reported for day 3 transfers (37%) for embryos generated in medium containing protein (Gardner et al., 1998).
Improved quality of embryos generated in micro-droplets has been previously shown to be due to autocrine and paracrine effects of certain growth factors released by the embryo into the medium (Paria and Dey, 1990; Paria et al., 1991
; Canesco et al., 1992
). Indeed the addition of human insulin-like growth factor-1 (hIGF-1) to culture medium significantly enhanced human preimplantation development, in particular, blastocyst formation (Lighten et al., 1998
). Recently, communal culture of human embryos was reported to result in higher implantation and pregnancy rates in human assisted reproduction (Almagor et al., 1996
). The effect of growth factors in preimplantation development has been extensively reviewed by Kane et al. (1997). Evidence from previous studies on the effect of growth factors indicates a lack of clear growth pattern on embryo development in vitro. Autocrine growth factors may be essential for embryo development under stressful culture conditions. Growth factors that may show positive effects on embryo preimplantation development include colony stimulating factor-1 (CSF-1), fibroblast growth factor-4 (FGF)-4, IGF-I/IGF-II, transforming growth factor-
(TGF)-
/epidermal growth factor (EGF) and possibly leukaemia inhibitory factor (LIF).
Some investigators have attempted to generate viable human embryos in a protein-free culture system of which the most recent reports were from Serta et al. (1997) and Parinaud et al. (1998a). However, Serta and co-workers (1997) prepared their spermatozoa for insemination by swim-down through a column of BSA but subsequent culture was performed in a protein-free medium. They achieved a pregnancy rate of 31% (n = 45) some of which have proceeded to term with the birth of normal children. Parinaud et al. (1998a) obtained fertilization with spermatozoa prepared in a protein-free medium when insemination was performed in the same medium. However, the resultant zygotes were cultured in BM1 medium. It is not clear whether their BM1 medium contained protein. A previous publication from the same group suggests that BM1 medium contained 1% HSA (Parinaud et al., 1998b).
In the present study, we have demonstrated that fertilization in human oocytes could occur in a protein-free culture environment with spermatozoa prepared in a protein-free medium. Protein is therefore not essential in the medium used for preparation of spermatozoa for insemination, nor for fertilization to occur during ICSI nor conventional IVF. Based on these findings, it is reasonable to assume that capacitation of spermatozoa may occur and the acrosome reaction may not be affected in the absence of protein in the medium.
Ten clinical pregnancies from a total of 20 transfers were achieved from embryos generated in the ART-7 series of media, with and without cumulus co-culture. The birth of normal babies conceived from embryos generated in the protein-free culture system indicates that viable day 2 human embryos can be generated in a medium devoid of added protein. The authors are not aware of any such previous report of births in which the laboratory procedures of assisted reproduction from oocyte retrieval, sperm preparation, fertilization and culture of resultant embryos were performed with a medium devoid of added protein.
In 1998, we performed 557 oocyte retrievals using a flushing medium devoid of added protein. The protein-deficient flushing medium did not appear to affect the quality of the eggs. Subsequent fertilization and embryo culture was performed in Medi-CultTM and IVF-50TM media. During this period the clinical pregnancy rate in our laboratory was 30.9% per oocyte retrieval (J.Ali, unpublished observations) which was above the world average for the year 1995 (20.4% and 21.7% for IVF and ICSI respectively) (de Mouzon and Lancaster, 1997). Recently papain, a proteolytic enzyme of plant origin, was successfully substituted for hyaluronidase which is of animal origin, for dispersion of cumulus cells in the ICSI procedure (Parinaud et al., 1998b
). Papain was substituted for hyaluronidase to eliminate the risk of transmission of diseases to IVF patients. Hyaluronidase was extracted from bovine or ovine testes. Stringent purification and sterilization measures employed for preparations of animal origin cannot exclude with absolute certainty the possibility of transmission of unknown pathogens (Truyen et al., 1995). With the formulation of protein-free media and the substitution of hyaluronidase with papain it is now possible to perform the entire laboratory assisted reproduction procedure with no known risk of transmission of disease to IVF patients. In conclusion this study showed that it is possible to generate viable human embryos in a medium devoid of added protein in a cell-free culture system.
![]() |
Note addded at proof |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Acknowledgments |
---|
![]() |
Notes |
---|
* This report was presented in part at the XVI Annual Scientific Meeting of the Fertility Society of Australia, 24 December 1997, Adelaide, Australia and the 5th Middle East Fertility Society Annual Meeting, November 1820, 1998, Amman, Jordan.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Ali, J., Whitten, W.K. and Shelton, J.N. (1993). Effect of culture systems on mouse early embryo development. Hum. Reprod., 8, 11101114.[Abstract]
Ali, J., Abdulkader, A., Shahata, M.A.M. et al. (1998) Term deliveries from human embryos conceived in a novel culture medium devoid of added protein. Middle East Fertil. Soc. J., 3, 40 (Abstract).
Almagor, M., Bejar, C., Kafka, I. et al. (1996) Pregnancy rates after communal growth of preimplantation human embryos in vitro. Fertil. Steril., 66, 394397.[ISI][Medline]
Balakier, H. (1993) Tripronuclear human zygotes: the first cell cycle and subsequent development. Hum. Reprod., 8, 18921897.[Abstract]
Barber, A. A. (1961) Inhibition of lipid peroxide formation by vertebrate blood serum. Arch. Biochem. Biophys., 92, 3843.[ISI][Medline]
Barnes, F.L., Crombie, A., Gardner, D.K. et al. (1995) Blastocyst development and pregnancy after in vitro maturation of human primary oocytes, intracytoplasmic sperm injection and assisted hatching. Hum. Reprod., 10, 32433247.[Abstract]
Biggers, J.D., Summers, M.C. and McGinnis, K.L. (1997) Polyvinyl alcohol and amino acids as substitutes for bovine serum albumin in culture media for mouse preimplantation embryos. Hum. Reprod. Update, 3, 125135.
Boyers, S.P., Diamond, M.P., Lavy, G. et al. (1987) The effect of polyploidy on embryo cleavage after in vitro fertilization in humans. Fertil. Steril., 48, 624627.[ISI][Medline]
Canesco, R.S., Sparks, A.E., Pearson, R.E. et al. (1992) Embryo density and medium volume effects on early murine embryo development. J. Assist. Reprod. Genet., 9, 454457.[ISI][Medline]
Carrillo, A.J., Lane, B., Pridham, D.D. et al. (1998) Improved clinical outcomes for in vitro fertilization with delay of embryo transfer from 48 to 72 h after oocyte retrieval: use of glucose- and phosphate-free media. Fertil. Steril., 69, 329334.[ISI][Medline]
Cholewa, J.A. and Whitten, W.K. (1970) Development of 2-cell mouse embryos in the absence of a fixed nitrogen source. J. Reprod. Fertil., 22, 553555.[Medline]
Dandekar, P.V. and Glass, R.H. (1990) Development of two-cell mouse embryos in protein-free and protein-supplemented media. J. In vitro Fertil. Embryo Transfer, 7, 107113.[ISI][Medline]
De Mouzon, J. and Lancaster, P. (1997) World collaborative report on in vitro fertilization. Preliminary data for 1995. J. Assist. Reprod. Genet. (Suppl.), 14, 250S265S.
Drakakis, P., Loutradis, D., Milingos, S. et al. (1996) The in vitro development of mouse embryos beyond the blastocyst stage into the hatching and outgrowth stage using different energy sources. J. Assist. Reprod. Genet., 13, 786792.[ISI][Medline]
Du, Z.F. and Wales, R.G. (1993) Effect of culture from the zygote stage on the metabolism of glucose and glutamine by 2-cell embryos and blastocysts recovered from outbred or F1 female mice. Reprod. Fertil. Dev., 5, 555565.[ISI][Medline]
Gardner, D.K. (1994) Mammalian embryo culture in the absence of serum or somatic cell support. Cell Biol. Int., 18, 11631179.[ISI][Medline]
Gardner, D.K. and Lane, M. (1998) Culture of viable human blastocysts in defined sequential serum-free media. Hum. Reprod. (Suppl.), 13,148160.[Medline]
Gardner, D.K., Schoolcraft, W.B., Wagley, L. et al. (1998) A prospective randomized trial of blastocyst culture and transfer in in-vitro fertilization. Hum. Reprod., 13, 34343440.[Abstract]
Kane, M.T., Morgan, P.M. and Coonan, C. (1997) Peptide growth factors and preimplantation development. Hum. Reprod. Update, 3, 137157.
Kemmann, E. (1998) CreutzfeldtJakob disease (CJD) and assisted reproductive technology (ART). Hum. Reprod., 13, 1777.
Lane, M. and Gardner, D.K. (1997) Nonessential amino acids and glutamine decrease the time of the first three cleavage divisions and increase compaction of mouse zygotes in vitro. J. Assist. Reprod. Genet., 14, 398403.[ISI][Medline]
Li, J., Foote, R.H., Liu, Z. et al. (1996) Development of rabbit zygotes into blastocysts in defined protein-free medium and offspring born following culture and embryo transfer. Theriogenology, 47, 11031113.[ISI]
Lighten, A.D., Moore, G.E., Winston, R.M.L. et al. (1998). Routine addition of human insulin-like growth factor-I ligand could benefit clinical in-vitro fertilization culture. Hum. Reprod., 13, 31443150.[Abstract]
Loutradis, D., Kallianidis, K., Drakkis, P. et al. (1992) Successful pregnancy in human IVF using BSA as a protein source in the transfer medium. Assist. Reprod. Technol. Androl., 3, 233238.
Mehta, T.S. and Kiessling, A.A. (1990) Development potential of mouse embryos conceived in vitro and cultured in ethylenediaminetetraacetic acid with and without amino acids or serum. Biol. Reprod., 43, 600606.[Abstract]
Moessner, J. and Dodson, W.C. (1993) The role of growth factors and proteins in media for in vitro fertilization. Assist. Reprod. Rev., 3, 6367.
Paria, B.C. and Dey, S.K. (1990) Preimplantation embryo development in vitro: cooperative interactions among embryos and the role of growth factors. Proc. Natl. Acad. Sci. USA, 87, 47564760.[Abstract]
Paria, B.C, Tsukamura, H. and Dey, S.K. (1991) Epidermal growth factor-specific protein tyrosine phosphorylation in preimplantation embryo development. Biol. Reprod., 45, 711718.[Abstract]
Parinaud, J., Milhet, P., Vieitiez, G. et al. (1998a) Human sperm capacitation and in vitro fertilization in a chemically defined and protein-free medium (SMART1). Proceedings of the 16th World Congress on Fertility and Sterility and the 54th Annual Meeting of the American Society for Reproductive Medicine, October 49, 1998, San Francisco. Fertil. Steril. (Suppl.), pp. S195S196 (Abstract).
Parinaud, J., Vieitiez, G., Milhet, P. et al. (1998b) Use of plant enzyme preparation (coronase) instead of hyaluronidase for cumulus cell removal before intracytoplasmic sperm injection. Hum. Reprod., 13, 19331935.[Abstract]
Poiley, S.M. (1960) A systematic method of breeder rotation for non-inbred laboratory animal colonies. Proc. Anim. Care Panel, 10, 159166.
Quinn, P. (1994) Use of co-culture with cumulus cells in insemination medium in human in vitro fertilization (IVF). J. Assist. Reprod. Genet., 11, 270277.[ISI][Medline]
Saito, H., Hirayama, T., Koike, K. et al. (1994) Cumulus mass maintains embryo quality. Fertil. Steril., 62, 555558.[ISI][Medline]
Schramm, R.D. and Bavister, B.D. (1996) Development of in-vitro-fertilized primate embryos into blastocysts in a chemically defined, protein-free culture medium. Hum. Reprod., 11, 16901697.[Abstract]
Serta, R.S., Sakellariou, M., Kiessling, A.A. et al. (1997) Outcome of human embryos conceived and cleaved in protein-free culture conditions. Proceedings of the 53rd Annual Meeting of the American Society for Reproductive Medicine, October 1822, 1997, Cincinnati. Fertil. Steril. (Suppl.), S214S215 (Abstract).
Spindle, A. (1995) Beneficial effect of taurine on mouse zygotes developing in protein-free culture medium. Theriogenology, 44, 761772.[ISI]
Staessen, C., Janssenswillen, C., De Clerck, E. et al. (1998) Controlled comparison of commercial media for human in-vitro fertilization: Ménézo B2 medium versus Medi-Cult universal and BM1 medium. Hum. Reprod., 13, 25482554.[Abstract]
Tesarik, J., Pilka, L., Drahorad, J. et al. (1988) The role of cumulus cell-secreted proteins in the development of human sperm fertilizing ability: implications in IVF. Hum. Reprod., 3, 129132.[ISI][Medline]
Truyen, U., Parrish, C.R., Harder, T.C. et al. (1995) There is nothing permanent except change. The emergence of new viral diseases. Vet. Microbiol., 43, 103122. Cited in: Parinaud et al. (1998b).[ISI][Medline]
Van Os, H.C., Drogendijk, Aat, C. Fetter, W.P.F. et al. (1991) The influence of contamination of culture medium with hepatitis B virus on the outcome of in vitro fertilization pregnancies. Am. J. Obstet. Gynecol., 165, 152159.[ISI][Medline]
Vidlakova, M., Erazimova, J., Horky, J. et al. (1972) Relationship of serum antioxidative activity to tocopherol and serum inhibitor of lipid peroxidation. Clin. Chim. Acta, 36, 6166.[ISI][Medline]
Wayner, D.D.M., Burton, G.W., Ingold, K.U. et al. (1987) The relative contribution of vitamin E, urate, ascorbate and proteins to total peroxyl radical-trapping antioxidant activity of human blood plasma. Biochim. Biophys. Acta, 924, 408419.[ISI][Medline]
Zhang, L., Jiang, S., Wozniak, P.J. et al. (1995) Cumulus cell function during bovine oocyte maturation, fertilization, and embryo development in vitro. Mol. Reprod. Dev., 40, 338344.[ISI][Medline]
Submitted on February 17, 1999; accepted on September 10, 1999.