5.2.1 “Synthetic” Oviduct Fluid

In 1972, Tervit and colleagues reported the successful culture of cattle and sheep ova in vitro (Tervit et al., 1972). In this paper, the authors described in detail the composition of a then-novel culture medium, which they named SOF – Synthetic Oviduct Fluid, an appropriate name since it was synthetic (i.e., created in the laboratory) and able to replicate the role of the fluid within the oviduct. Importantly, the composition of SOF was based on data on the composition of the fluid from the sheep oviduct (Restall & Wales, 1966). Using this medium, combined with an oxygen tension of 5%, Tervit and colleagues were able to generate ovine and cattle blastocysts past the so-called “development block” (which we now know to equate to zygotic genome activation), and, following transfer into suitable recipients, generate pregnancies. Indeed, almost 50 years later, SOF remains the definitive medium for the culture of cattle embryos.

The importance of the development of SOF lies not only in findings but the approach: to produce a medium whose composition was based on the fluids of the genital tract. Such an approach was adopted by Pat Quinn and colleagues (Quinn et al., 1985), who developed an embryo culture medium (human tubal fluid, HTF) based on earlier published reports of the composition of fallopian tube fluid (Lippes et al., 1972; Lopata et al., 1976). In what was probably a world first, Quinn and colleagues undertook a randomized comparison of culture media using human embryos and found that the medium based on the reported composition of the fluid from the oviduct better supported embryo development. Interestingly, Quinn et al. were able to elegantly demonstrate that as the concentration of potassium ions within culture medium fell, so did the blastocyst rate, illustrating the importance of optimized ionic composition of medium for successful embryo culture, and the apparent need for a high potassium content – a phenomenon still not adequately explained (Quinn et al., 1985). By modern standards, this was a small study, comprising of 417 human oocytes; however, it does represent the first dedicated, specific human embryo culture medium. HTF remains the basis of many contemporary embryo culture media.

The concept of seeking inspiration from the in vivo environment to inform strategies for the production of human embryo culture medium underpins the second approach to be considered: sequential medium.

5.2.3 Sequential Embryo Culture

The optimal time to transfer an embryo into the recipient has been intensely debated for many years. In general, there are two schools of thought: transfer the cleavage stage embryo on day 2–3 post insemination or transfer on day 5, or later, when the embryo has formed a blastocyst. Early clinical IVF relied largely on transferring embryos at the cleavage stages, due in part to perceived shortcomings in embryo culture medium (Gardner and Lane, 1998). However, from the early 2000s, there has been a shift toward transferring blastocysts, in large part facilitated by improvements in embryo culture. In vivo, the final stages of gamete maturation, fertilization, and early cleavage occur within the fallopian tube. The embryo resides in this environment for approximately 5 days, after which it transits into the uterus, in readiness for implantation. This transit coincides with the formation of the blastocyst. Since embryos are typically transferred into the uterus, it makes a degree of sense to transfer at the stage of development when they would typically enter the uterus – the blastocyst.

During in vivo development, the movement of the embryo from one anatomical structure to a second may logically be interpreted as a change in environment: one environment for fertilization and cleavage, and a different environment to facilitate blastocyst development. Indeed, numerous attempts at characterizing the differences in the fluids of the oviduct and uterus have been made. In the late 1960s and through the 1970s, there were a series of attempts to measure the fluid composition of the genital tracts from a range of animals, including sheep (Iritani et al., 1969), rabbit (Iritani et al., 1971), pig (Iritani et al., 1974), and mouse (Borland et al., 1977) to name but a few key reports. Although there were differences between the species, a unifying observation was that the composition of fluid collected from the oviducts differed markedly from that of the uterus. Differences in pH, total protein content, nonprotein nitrogen, energy substrates, and ions were reported, further supporting the notion that the environment that supports the cleavage embryo differed from that where the blastocyst would typically reside. Furthermore, the metabolic activity of blastocysts is notably different to that of the cleavage stage embryo; thus a logical conclusion was to develop culture medium customized for stage-specific needs of early embryos: sequential media.

Perhaps the best known sequential embryo culture system was developed during the 1990s, that is the so-called G-series of culture products (Gardner, 1994; Barnes et al., 1995; Gardner & Lane, 1997). Sequential media were designed to support cleavage development (G1), and compaction and blastocyst development (G2). These media differed in terms of the provision of carbohydrates and amino acids, and were again based on measurements made on the fluids from the oviduct and uterus (Gardner et al., 1996). Trialed initially in mouse embryos (Gardner, 1994), the use of a sequential approach was found to support the development of high quality blastocysts, higher implantation, and better embryo development (Gardner & Lane, 1996).

The medium intended to support the cleavage stages of development is characterized by low glucose and the nonessential amino acids (see Chapter 4 for more details). The rationale behind this was that glucose is utilized in very low amounts through cleavage, with pyruvate and lactate being the preferred energy sources at this stage. Based largely on the work of Chatot et al. (1989), who demonstrated that omission of glucose from embryo culture medium was able to support mouse embryo development through the “2-cell block,” a viewpoint flourished that glucose was actually inhibitory for cleavage stage embryos. Indeed, Quinn et al. (1985) produced a modified HTF lacking glucose and phosphate and reported improved embryo development. This modified medium contained EDTA and glutamine. Thus, EDTA was added to further suppress glycolysis and its presence during the cleavage stages of development was shown to improve development of mouse and cattle embryos (Gardner & Lane, 1997).

The medium designed to support blastocyst development (G2) was characterized by the presence of higher glucose and the nonessential and essential amino acids. Furthermore, EDTA was omitted. Considering the accepted picture of embryo metabolism, such a medium would facilitate a rise in glucose depletion and the absence of EDTA will allow glycolysis to occur – characteristic of blastocyst formation.

Although the idea of changing the culture environment to mimic the transition from the fallopian tube to the uterus is reasonable, it is notable that both essential and nonessential amino acids are present in the oviduct fluid of all commonly studied mammalian species (i.e., mice, rabbit, cattle, pig, sheep, and horse) (Aguilar & Reyley, 2005) as well as the human fallopian tube (Tay et al., 1997). Furthermore, embryos consume and release amino acids, whether they are essential or nonessential, throughout the preimplantation period (Brison et al., 2004). Therefore, it seems logical to add all the amino acids and let the embryo decide, considering that the interactions of amino acids with other nutrients are so far unknown (Ménézo et al., 2013).

Importantly, while sequential culture medium makes a degree of logic, there is no conclusive evidence to support the superiority of one culture modality over the other (Biggers et al., 2002; Macklon et al., 2002, Paternot et al., 2010, Summers et al., 2013, Dieamant et al., 2017). So far, there are no randomized controlled trial comparing all the current culture media together, and it is not possible to make conclusions about which one is the best (Chronopoulou & Harper, 2015; Youssef et al., 2015).

5.3.1 Protein Supplements

In the early days of embryo culture, the medium in which embryos were grown contained undefined biological supplements, such as egg white (Hammond, 1949), or was even composed entirely of plasma (rabbit embryos: Lewis and Gregory, 1929). Indeed, the medium used by Steptoe and Edwards for the first successful IVF contained patient serum (Steptoe and Edwards, 1978). However, a key observation was reported in 1984 by Menezo and colleagues (Menezo et al., 1984) who compared medium supplemented with human cord serum or 1% human serum albumin (HSA) and observed no difference in ongoing pregnancy rates. This led to the conclusion that serum was not necessary for human embryo development. This was an especially prescient observation, given the later reports that linked the inclusion of serum in animal embryo culture medium to imprinting defects and in utero overgrowth – the so-called large offspring syndrome (Young et al., 1998; Sinclair et al., 2000). Nonetheless, there remains the requirement for the inclusion of a macromolecule source, satisfied largely through the inclusion of protein supplements.

Protein supplements, such as human serum albumin (HSA) and serum substitute supplement (SSS), are extracted from human blood; thus the chemical composition varies between batches and manufacturers (Morbeck et al., 2014b). These extracts are not manufactured specifically to be used for embryo culture but rather for in vivo clinical use. It is therefore a challenge for manufacturers to source batches of HSA or SSS. A chemical analysis of different culture media revealed the presence of a number of bioactive substances that are not declared on the label or in the product insert (Dyrlund et al., 2014). The origin of these bioactive substances is most likely the protein supplement that was added to the culture media. One study has found that culturing embryos in media containing different protein supplements resulted in a difference in the birthweight of the offspring suggesting that the source of protein supplement is not trivial (Zhu et al., 2014). Even more startlingly, most preparations contain stabilizers such as octanoic acid that is shown to inhibit growth of embryos (Fredrickson et al., 2015). Thus, the inclusion of protein supplements presents somewhat of a paradox: to date their inclusion remains a necessity for viable clinical embryo production, yet the inclusion of chemically undefined products offers a route for embryos to be exposed to unexpected compounds. Fortunately, significant research in the animal embryo field persists on identifying chemically defined alternatives for embryo culture with some promising results (e.g., Gómez et al., 2020).

Beyond the presence of protein supplements, some culture media contain added hormones and growth factors in addition to the various protein supplements.

5.3.2 Growth Factors and Hormones

The fluid within the genital tract contains a myriad of bioactive proteins and their receptors (Figure 5.1). It is not unreasonable to assume that the concentration of these growth factors in vivo is carefully controlled and may change during the menstrual cycle. Additionally, there is likely a difference between the proximal and distal end of the fallopian tube as well as in the uterus. In vivo, these growth factors may exert their putative effect on gametes, zygotes, and embryos acting both directly and indirectly by modulating the functions of the various cells in the genital tract. However, there is an almost complete absence of data enabling a systematic approach to identify which growth factor(s) to add to embryo culture media and at which concentration. Therefore, most embryo culture media do not contain added growth factors despite the prediction that, in vivo, they play an important role in regulating growth and differentiation of early embryos.

Figure 5.1 Cytokines and cytokine receptors detected in human genital tract and/or in human embryos.

Adapted from data published by Wydooghe et al., 2017.

Cytokines are molecules primarily regulating the function of cells involved in the immune response. Several cytokines have been shown to be present in semen and the female genital tract (Figure 5.1, Wydooghe, 2017). It has been demonstrated that leukemia inhibitory factor (LIF) is present in the fallopian tube, and both embryos and tubal cells express LIF receptors. It is thought that LIF is involved in the implantation process. Addition of LIF to embryo culture media has been shown to enhance the quality of human embryos (Dunglison et al., 1996). Glycoprotein 130 (Gp130) is involved in mediating the effect of a range of cytokines and the addition of Gp130 to embryo culture media might have a positive effect embryo development in vitro (Hambiliki et al., 2013).

Animal studies have shown that the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF) in culture media may exert a positive effect on embryo development, leading to significant interest in this molecule as a putative supplement to embryo culture. However, in a large prospective randomized study, the addition of GM-CSF to the embryo culture media revealed no overall positive effect on embryo development, implantation rates, or delivery rates. However, a post hoc subgroup analysis indicates a possible positive effect in the subgroup of patients with previous failed cycles (Ziebe et al., 2013). Indeed, in some of the early data on the mechanism of GM-CSF, it was concluded that one function was to suppress apoptosis in embryos (Sjöblom et al., 1999). However, whether suppression of apoptosis, which represents a mechanism by which an embryo may eliminate damaged cells, is of benefit is unclear.

A more recent prospective randomized study compared the outcome after culturing human embryos either in a standard medium or the same medium with addition of a mixture of cytokines (LIF, GM-CSF, heparin-binding epidermal growth factor (HB-EGF)). The addition of cytokines to the medium improved embryo quality (Fawzy et al., 2019), but as a mixture of cytokines were used in this study, it is not known whether the effect was due to an individual cytokine or the specific mixture used.

A few manufacturers add insulin to some of their culture media intended for IVF. The presence of insulin has been demonstrated to alter the profile of DNA methylation in mice embryos (Shao et al., 2007). DNA methylation is one way of regulating expression of genes and a change in DNA-methylation pattern is therefore usually accompanied with a change m-RNA profile and the proteins synthesized. Whether such effects present in human embryos remains unknown. Thus, the effectiveness of supplementation of culture media with growth factors and cytokines remains unclear.

5.3.3 Hormones

In vivo, the genital tract contains gonadal steroids, such as estradiol and progesterone. These steroids are key regulators of genital tract function and influence the molecules that are secreted into the fallopian tube lumen and to the uterine cavity. Addition of steroid hormones to embryo culture media has not proven to be beneficial. A major challenge of steroid supplementation is the difficulty of dissolving them in culture media, a feature shared with other lipophilic substances. In vivo, they are predominantly bound to carriers, such as serum albumin and sex hormone binding globulin (SHBG).

5.3.4 Vitamins, Trace Metals and Lipids

Some embryo culture media are very complex and reflective of products intended for continuous culture of mammalian cells. Such formulations can include up to 80 different components, including fatty acids, vitamins, iron chelators, trace metals, and intermediates in metabolism. It has not been shown that these very complex culture media are superior to less complex media and it still unknown to what extent these varied components will affect development of human embryos in vitro.

5.3.5 Hyaluronic Acid

Hyaluronic acid may be involved in implantation and adding it to the medium used to transfer the embryo to the uterus has been shown to have a moderate positive effect on implantation rates (Bontekoe et al., 2015)

5.3.6 pH Regulators

In vivo, an important pH regulator is the H⁺ + HCO₃ ↔ H₂O + CO₂ equilibrium. High CO₂ levels results in a low pH and low CO₂ levels in a high pH. Most embryo culture media contain bicarbonate and pH is regulated by the CO₂ concentration in the incubator. The CO₂ concentration should be adjusted to give a pH in the physiological range (7.2–7.4). One study on mouse embryos indicates a beneficial effect of having a higher pH at the zygote stage and lower pH from the cleavage stage (Hentemann, 2011). Addition of a pH indicator, such as phenol red, may offer a visual check on the pH of the culture media. However, phenol red is not an inert molecule and may change the metabolism of cultured mammalian cells (Morgan et al., 2019). It is therefore perhaps advisable to use phenol red–free culture media. In culture media that will be exposed to atmospheric CO₂ concentration, pH must be stabilized by buffers, such as HEPES or PIPES.

5.3.7 Non Physiological Additives

Follicular aspirates often contain traces of vaginal flora and it is reasonable to assume that most semen samples are not sterile (Kastrop et al., 2007; Koedooder et al., 2019). Antibiotics are, therefore, added to culture media, even though it has been demonstrated that presence of antibiotics may reduce embryo quality (Magli et al., 1996). These antibiotics are usually not manufactured with assisted reproduction in mind, and it can be a challenge for media manufacturers to source antibiotics that do not contain trace amounts of molecules, such as lipopolysaccharides (LPS), that will have negative effect on embryo development. It is a matter of discussion whether to add penicillin in combination with streptomycin or gentamicin or to use gentamicin alone. Semen and vaginal flora contain both Gram-positive and Gram-negative bacteria and since penicillin primarily works against Gram-positive bacteria, it is necessary to add streptomycin or gentamicin to also inhibit the growth of Gram-negative bacteria. Vaginal flora may include Candida species. Embryo culture media does not contain antimycotics, and if vaginal fluid is aspirated during oocyte recovery, the cultures may be heavily contaminated with Candida cells.

Some culture media contain EDTA which will bind divalent cations. In some animal models, the presence of EDTA will be beneficial for embryo development. For human embryo culture using modern culture media, it is not established whether the presence of EDTA is necessary for optimal embryo development.

Many embryo culture media contain surface tension and viscosity modulators to enable simpler handling of small volumes of media. The effect, if any, on the development of human embryos is unknown.