In vivo oxygen concentration

Since the 1950s, intense research has been performed to identify the oxygen tension in the female reproductive tract. Historically, embryo culture has been performed at atmospheric oxygen. Several studies have investigated the luminal fluids of mammals oviduct, and have reported an oxygen concentration between 2% and 8% (Mastroianni and Jones, Reference Mastroianni and Jones1965; Fischer and Bavister, Reference Fischer and Bavister1993; Kovačič, Reference Kovačič2012). There is good evidence to suggest that, in the female reproductive tract, the developing embryo is exposed to concentrations below 5% (Fischer and Bavister, Reference Fischer and Bavister1993; Ng et al., Reference Ng, Mingels, Morgan, Macklon and Cheong2018). There is increasing evidence in the published literature on mammalian species, including humans, to suggest the adverse effects of atmospheric oxygen on embryo development by changes to the transcriptome (Gardner and Lane, Reference Gardner and Lane2005) and alterations to the proteome (Katz-Jaffe et al., Reference Katz-Jaffe, Linck, Schoolcraft and Gardner2005) and gene expression (Rinaudo et al., Reference Rinaudo, Giritharan, Talbi, Dobson and Schultz2006) that compromise both carbohydrate and amino acid metabolism (Wale and Gardner, Reference Wale and Gardner2012), impacting the epigenome (Morgan et al., Reference Morgan, Santos, Green, Dean and Reik2005; Ventura-Juncá et al., Reference Ventura-Juncá, Irarrázaval, Rolle, Gutiérrez, Moreno and Santos2009; Li et al., Reference Li, Goossens, Van Poucke, Forier, Braeckmans, Van Soom and Peelman2014a; Marcho et al., Reference Marcho, Cui and Mager2015) and inducing premature X-chromosome inactivation (Lengner et al., Reference Lengner, Gimelbrant, Erwin, Cheng, Guenther, Welstead, Alagappan, Frampton, Xu, Muffat, Santagata, Powers, Barrett, Young, Lee, Jaenisch and Mitalipova2010). It has been suggested that culture in reduced oxygen tension is crucial to keep physiological embryo development and increase reproductive competence. In the cytoplasm, the oxidative stress resulting from accumulation of ROS is likely to be a mechanism by which high oxygen concentration weakens the embryo, reducing its implantation potential and its capacity to generate a pregnancy (Kwon et al., Reference Kwon, Yang, Hwang, Yoo, Kim, Lee, Ryu and Oh1999). Meuter and colleagues (Reference Meuter, Rogmann, Winterhoff, Tchkonia, Kirkland and Morbeck2014) showed that cultured mouse blastocysts might express markers of senescence, generated by oxidative stress when cultured in suboptimal culture conditions.

In vitro-produced embryos at the pre- and post-compaction stages differed notably in several crucial features. Cleavage stage embryos have a restricted mechanism to control metabolic homeostasis, and therefore are highly susceptible to conditions that may generate ROS (Umaoka et al., Reference Umaoka, Noda, Narimoto and Mori1992; Johnson and Nasr-Esfahani, Reference Johnson and Nasr-Esfahani1994; Takahashi et al., Reference Takahashi, Keicho, Takahashi, Ogawa, Schultz and Okano2000; Guérin et al., Reference Guérin, El Mouatassim and Ménézo2001; Favetta et al., Reference Favetta, St John, King and Betts2007). Post-compaction embryos have more complex systems and their regulation competence could be significantly increased (Lane and Gardner, Reference Lane and Gardner2007). Several studies conducted on animal models have demonstrated that reduced oxygen concentration is associated with better pregnancy outcome compared with atmospheric oxygen, especially when embryos are transferred at the blastocyst stage (Umaoka et al., Reference Umaoka, Noda, Narimoto and Mori1992; Booth et al., Reference Booth, Holm and Callesen2005). The fact that oxygen levels differ throughout the reproductive tract is not a new discovery. This phenomenon has been well described in multiple species. In the human, it has been suggested that during the journey from oviduct to the uterus, the preimplantation embryo proceeds through decreasing oxygen concentrations and reaches the lowest concentration when it starts the compaction process (Thompson et al., Reference Thompson, Simpson, Pugh, Donnelly and Tervit1990; Kovačič, Reference Kovačič2012). This notion comes from histological data from salpingectomy specimens, demonstrating that the preimplantation embryo crosses the uterotubal junction late on day 3 (Croxatto, Reference Croxatto2002), and evidence that the oxygen tension in the uterus is lower than 5–7% (Fischer and Bavister, Reference Fischer and Bavister1993). At the morula stage there is a shift in the metabolic pathway from oxidative phosphorylation for energy production in the pre-compaction stage to increased dependence on ATP production during glycolysis in the post-compaction stage (Leese, Reference Leese1995; Thompson et al., Reference Thompson, Partridge, Houghton, Cox and Leese1996). In rabbits and hamsters, uterine oxygen is similar to that in the oviducts and declines to 3–5% at the point of implantation. Oxygen concentration in the oviduct of rhesus monkey is 5–8.7%, while in the uterine level it is 1.5–2% (Fischer and Bavister, Reference Fischer and Bavister1993). Indeed, the uterus is a hypoxic environment and remains such for the entire first trimester of pregnancy (Rodesch et al., Reference Rodesch, Simon, Donner and Jauniaux1992). Hypoxia is important not only to support vasculogenesis, but also for promoting rapid cell proliferation, similar to tumour growth (Krisher and Prather, Reference Krisher and Prather2012). Limited studies are currently available that illustrate the progression of the human embryo through the female reproductive tract and analyze oxygen concentration in the uterine cavity in non-pregnant women (Yedwab et al., Reference Yedwab, Paz, Homonnai, David and Kraicer1976; Ottosen et al., Reference Ottosen, Hindkaer, Husth, Petersen, Kirk and Ingerslev2006). The study by Yedwab and co-workers (Reference Yedwab, Paz, Homonnai, David and Kraicer1976), reported an average oxygen concentration below 2% in the human uterus. In another study the oxygen tension was measured at the endometrial surface in 21 patients at the time of insemination, with a mean value around 2% (Ottosen et al., Reference Ottosen, Hindkaer, Husth, Petersen, Kirk and Ingerslev2006). To our knowledge, no studies have been published reporting the exact oxygen tension in the human oviduct.

Oxygen tension and cell physiology

In the in vivo environment, the embryo is surrounded by oviductal or uterine fluid, and from these it acquires the substrates and oxygen required for development. The main activity of the early embryo is the metabolic production of energy and gene activation. At an early stage, embryos use aerobic respiration to obtain the oxygen required for the oxidation of substrates such as pyruvate, lactate and amino acids. The production of energy, as ATP molecules, starts once the embryo begins to grow. At the blastocyst stage, it switches from oxidative phosphorylation to aerobic glycolysis to support protein synthesis and ion transport systems (Martin, Reference Martin2000; Thompson et al., Reference Thompson, Sherman, Allen, McGowan and Tervit1998). The embryo obtains its oxygen via passive diffusion with an efficiency linked to the oxygen concentration in the gas phase, the solubility of oxygen in the medium, presumptive and yet full appreciated boundary layers or concentration differences at the embryo/cellular surface, influenced by rates of diffusion, utilization and fluid movement (Smith et al., Reference Smith, Takayama and Swain2012) around the embryo, and the rate of diffusion through the cytoplasm. Baltz and Biggers (Reference Baltz and Biggers1991) analyzed the oxygen transport in embryo culture in a system in which mouse embryos were contained in microdroplets of culture medium surrounded by a thin layer of mineral oil to prevent evaporation. The source of oxygen for the embryos lay beyond two liquid phases: medium and oil. Transport needs to be sufficiently rapid to replace the oxygen consumed by the embryos, otherwise the medium could become depleted of oxygen. This could generate problems, particularly when large numbers of embryos are present in a single drop, or if the layer of oil is thick, so that oxygen must cross a large barrier before reaching the embryos. They concluded that diffusion through a layer of oil appears not to be rate limiting, even if a large number of embryos are cultured in the same drop, suggesting no risk of anoxia (Baltz and Biggers, Reference Baltz and Biggers1991). In the mouse embryo, more than 70% of the oxygen is metabolized via oxidative phosphorylation at the blastocyst stage and less than 30% when the embryo is at the 2- to 4-cell stage (Trimarchi et al., Reference Trimarchi, Liu, Porterfield, Smith and Keefe2000). It has been established that trophectoderm cells use significantly more oxygen, produce more ATP and contain a greater number of mitochondria compared with the inner cell mass (ICM) (Houghton, Reference Houghton2006).

Reactive oxygen species

Preimplantation development is a time of dynamic alteration, involving modifications of the genome, proteome, metabolome and epigenome (Marcho et al., Reference Marcho, Cui and Mager2015), and therefore embryos have high sensitivity to their external environment. The rationale against the use of atmospheric oxygen tension is that it can be potentially toxic for cells through formation of ROS (Catt and Henman, Reference Catt and Henman2000), and contribute to defective embryo development, with higher rates of fragmentation (Bedaiwy et al., Reference Bedaiwy, Falcone, Mohamed, Aleem, Sharma, Worley, Thornton and Agarwal2004). Gametes and embryos are cultured in a medium that can itself be the source of ROS (Martín-Romero et al., Reference Martín-Romero, Miguel-Lasobras, Domínguez-Arroyo, González-Carrera and Alvarez2008) and this may negatively influence pregnancy outcomes (Thompson et al., Reference Thompson, Simpson, Pugh, Donnelly and Tervit1990; Johnson and Nasr-Esfahani, Reference Johnson and Nasr-Esfahani1994; Blondin et al., Reference Blondin, Coenen and Sirard1997; Kwon et al., Reference Kwon, Yang, Hwang, Yoo, Kim, Lee, Ryu and Oh1999; Takahashi et al., Reference Takahashi, Keicho, Takahashi, Ogawa, Schultz and Okano2000; Meuter et al., Reference Meuter, Rogmann, Winterhoff, Tchkonia, Kirkland and Morbeck2014). Oxidative stress is considered to be a principal cause of the 2-cell developmental block (Johnson and Nasr-Esfahani, Reference Johnson and Nasr-Esfahani1994; Favetta et al., Reference Favetta, St John, King and Betts2007). Although embryos are equipped with several endogenous means of countering oxidative stress, many antioxidant genes are not expressed until the later stages of preimplantation development (Harvey et al., Reference Harvey, Arcellana-Panlilio, Zhang, Schultz and Watson1995). Production and accumulation of ROS from exposure to atmospheric oxygen concentrations during culture may therefore overwhelm embryo defence mechanisms. Physiologically, ROS may originate from embryo metabolism and/or the surroundings and are produced during oxygen reduction in the mitochondria (Guérin et al., Reference Guérin, El Mouatassim and Ménézo2001). Therefore, when cells are exposed to ROS, DNA damage frequently occurs, producing distinctive patterns of chemical modification. Oxidative stress opens the mitochondrial permeability transition pores. As a result, a large loss of ions and metabolites from the mitochondrial matrix occurs, causing mitochondrial alterations, ATP depletion, and finally triggers the apoptotic mechanism (Takahashi et al., Reference Takahashi, Keicho, Takahashi, Ogawa, Schultz and Okano2000; Guérin et al., Reference Guérin, El Mouatassim and Ménézo2001). Electron transport chain respiration has been suggested to be a significant pathway of mitochondrial superoxide production (Barja, Reference Barja1999). Importantly, superoxide and hydroxyl anions, being polar molecules, do not pass the inner mitochondrial membrane, and their efflux is regulated by ion channels (Lustgarten et al., Reference Lustgarten, Bhattacharya, Muller, Jang, Shimizu, Shirasawa, Richardson and Van Remmen2012). Furthermore, superoxide is rapidly converted to hydrogen peroxide by superoxide dismutases antioxidant enzymes (SOD) (Takahashi and Asada, Reference Takahashi and Asada1983; Missirlis et al., Reference Missirlis, Hu, Kirby, Hilliker, Rouault and Phillips2003). Together these barriers make it unlikely that matrix-derived superoxide could escape and exit from the mitochondria. It has been reported that elevated mitochondrial superoxide production might negatively affect the genome, causing DNA damage and tumourigenesis (Ishii, Reference Ishii2007). Nakada and colleagues (Reference Nakada, Kubota, Sasagawa, Yagisawa, Watanabe and Ishigooka1995) described an inverse association between decreased SOD activity and the presence of malignant neuroendocrine tumours located in the medulla. This inverse association suggests that deficits in SOD, and potentially, elevated superoxide, may be related to the pathobiology of the tumour (Nakada et al., Reference Nakada, Kubota, Sasagawa, Yagisawa, Watanabe and Ishigooka1995).

In addition, ROS can also modify most types of bio-molecules such as lipids and proteins; this alteration negatively affects membrane stability and permeability and induces damage to several sub-cellular organelles (Guérin et al., Reference Guérin, El Mouatassim and Ménézo2001; Batty et al., Reference Batty, Jensen and Freemont2009). A study published by Ma and co-workers (Reference Ma, Chen and Tzeng2017) on a mouse model investigated the mechanisms behind the beneficial effects of reduced oxygen tension in embryogenesis. The study showed that low oxygen tension might improve embryo viability by increasing expression of antioxidant enzymes and glucose transporter activities. Analysis of expression of antioxidant genes showed that they were 8- to 10-fold higher in the 3% oxygen group compared with the 20% group. This upregulation could enhance implantation potential and reduce apoptosis in mouse blastocysts (Ma et al., Reference Ma, Chen and Tzeng2017).

Embryo culture

To obtain an optimal embryo development and improve pregnancy outcomes, one approach is to keep conditions as similar as possible to the in vivo environment. It is evident from the literature that the reduction of oxygen in culture is directly proportional to the reduction of ROS. The negative effect of atmospheric oxygen and oxidative stress on embryo development in vitro has been reported (Bontekoe et al., Reference Bontekoe, Mantikou, van Wely, Seshadri, Repping and Mastenbroek2012, Catt and Henman, Reference Catt and Henman2000; Bedaiwy et al., Reference Bedaiwy, Falcone, Mohamed, Aleem, Sharma, Worley, Thornton and Agarwal2004; Marcho et al., Reference Marcho, Cui and Mager2015). However, oxygen concentration studies on human embryos reported controversial results. No significant differences were found in fertilization, cleavage, pregnancy and implantation rates during the culture of embryos up to day 2 or day 3 when using oxygen concentrations of 5% or 20% (Dumoulin et al., Reference Dumoulin, Vanvuchelen, Land, Pieters, Geraedts and Evers1995; Dumoulin et al., Reference Dumoulin, Meijers, Bras, Coonen, Geraedts and Evers1999). Conversely, significantly more surplus embryos reached the blastocyst stage when cultured at 5% oxygen compared with 20%. A study published by Waldenström and associates (Reference Waldenström, Engström, Hellberg and Nilsson2009), described a better blastocyst outcome and a significant improvement in pregnancy and birth rates when embryos were cultured under low oxygen concentrations. Kea et al. (Reference Kea, Gebhardt, Watt, Westphal, Lathi, Milki and Behr2007) showed that different oxygen concentrations did not significantly influence fertilization rates, blastocyst formation and quality, or pregnancy rates, but that there was a significant difference in the mean embryo score on day 3 in favour of the reduced oxygen concentration. Kovačič and Vlaisavljević (Reference Kovačič and Vlaisavljević2008) showed that a lower oxygen concentration increased the proportion of embryos reaching the blastocyst stage. Additionally, the same authors in a subsequent study reported that although the ongoing pregnancies and implantations were similar in the two oxygen concentration groups, the cumulative pregnancy rate (fresh and frozen–thawed embryos) was significantly in favour of the 5% oxygen group (Kovačič et al., Reference Kovačič, Sajko and Vlaisavljević2010). A study published by Ciray and colleagues (Reference Ciray, Aksoy, Yaramanci, Karayaka and Bahceci2009) on sibling oocytes demonstrated that 5% oxygen tension significantly improved the total blastocyst yield as well as the quality of day 3 and day 5 embryos. Meintjes et al. (Reference Meintjes, Chantilis, Douglas, Rodriguez, Guerami, Bookout, Barnett and Madden2009) showed an overall increase in live births when embryos were cultured in reduced oxygen concentrations. Finally, a meta-analysis on four studies confirmed a beneficial effect of culturing embryo in low oxygen concentrations for live birth rates (Bontekoe et al., Reference Bontekoe, Mantikou, van Wely, Seshadri, Repping and Mastenbroek2012). These studies analyzed certain characteristics of embryo development, and only compared the clinical outcome, implantation, pregnancy and live birth rates. However, none of these showed a negative correlation with low oxygen culture (Kovačič and Vlaisavljević, Reference Kovačič and Vlaisavljević2008; Ciray et al., Reference Ciray, Aksoy, Yaramanci, Karayaka and Bahceci2009; Kovačič et al., Reference Kovačič, Sajko and Vlaisavljević2010; de los Santos et al., Reference de los Santos, Gámiz, Albert, Galán, Viloria, Pérez, Romero and Remohï2013). Another important aspect to consider is that exposure of embryos to atmospheric oxygen may make them more vulnerable to other in vitro stressors, such as ammonium accumulation in the surrounding medium (Wale and Gardner, Reference Wale and Gardner2013). Therefore, the negative repercussion could be that when minor problems arise in the embryology laboratory, extended exposure to atmospheric oxygen will compromise embryo development. This molecular mechanism has been investigated by Wale and Gardner (Reference Wale and Gardner2013). The concept is that when stressors adhere together they bring synergistically adverse effects to embryo development, this is a crucial issue to be taken into consideration to optimize and enhance IVF outcome (Swain et al., Reference Swain, Cabrera, Xu and Smith2012, Reference Swain, Lai, Takayama and Smith2013).

Ultra-low oxygen concentration

Recently a new trend has been reported, investigating the effect of a very low oxygen concentration of 2% on in vitro human embryo development. The debate is based on evidence that culture conditions should mimic as accurately as possible in vivo conditions. At the moment every incubator used for human embryo culture provides a static environment, without any signs of movement. All the embryology procedures in the IVF laboratory are still carried out in culture dishes. New technological possibilities exist with advancement of microfluidic technology, which is based on the behaviour of liquids in a microenvironment and might be applied to embryo culture in the future and provide dynamics and movement to embryo development (Swain et al., Reference Swain, Lai, Takayama and Smith2013; Smith and Takayama, Reference Smith and Takayama2017). Regarding oxygen concentration, it is evident that, in vivo, the embryo in the reproductive tract is exposed to oxygen concentrations lower than 5% (Mastroianni and Jones, Reference Mastroianni and Jones1965; Fischer and Bavister, Reference Fischer and Bavister1993; Kovačič, Reference Kovačič2012). Therefore, it would be interesting to culture embryos in very low oxygen concentrations of 2% to evaluate embryo development and blastocyst formation. The published literature is not clear on this debate, perhaps because very few studies are available and conclusions are still inconsistent. A study published by Thompson and co-workers (Reference Thompson, Simpson, Pugh, Donnelly and Tervit1990), using sheep and cattle models is summarized here. Two-cell sheep embryos and 8-cell cow embryos were cultured for 5 days in medium previously equilibrated at different oxygen concentrations: sheep 0, 2, 4, 6, 8, 10, 12, 17 or 20%; and cow at 0, 4, 8, 12, 17 or 20%. At the end of 5 days of embryo culture, morphology was analyzed and the embryos were stained to evaluate the number of nuclei. Sheep embryos showed the highest number of morula (nearly 60%) at oxygen concentrations between 6% and 10%. Low quality and poor embryo development was detected at the 0%, as well as at the 20%, oxygen concentration indicating that oxygen is vital and necessary for embryo development. The same trend was observed for cattle embryos. Different results were reported by an Australian group (Feil et al., Reference Feil, Lane, Roberts, Kelley, Edwards, Thompson and Kind2006) in the mouse model. The study observed the effect of embryo culture under different oxygen concentrations and subsequent fetal and placental development at day 18 of pregnancy. Embryos were cultured from zygote to morula at 7% oxygen concentration followed by 20, 7 or 2% to the blastocyst stage. Implantation rates was not influenced by oxygen concentration, but resorption rates were increased in embryos cultured under 2% oxygen, compared with 7% oxygen. The retained placenta is a significant cause of maternal mortality. Day 18 fetal weights were reduced following culture under 2%, compared with 7% or 20% oxygen, or in vivo development. Placental weight in the mouse model was not influenced by oxygen tension. Another study performed in the mouse model reported better blastocyst development and hatching rates for embryos cultured in 3% oxygen compared with 20% oxygen. Those data were obtained from different experiments, in which 2-cell embryos (a total of 185) were cultured in 3% oxygen and 189 embryos were cultured in 20% oxygen. Results reported that embryos cultured in 3% oxygen tension had significantly higher rates of blastocyst development (92.3% vs. 79.4%) and hatching (80% vs. 70.4%) compared with those cultured in 20% (Ma et al., Reference Ma, Chen and Tzeng2017). Limited data are available on humans: a study published by Yang and colleagues (Reference Yang, Xu, Ding, Khoudja, Lin, Awonuga, Dai, Puscheck, Rappolee and Zhou2016) analyzed 155 embryos donated to research. All embryos in the study were frozen on day 3 and all were cultured at 20% oxygen concentration from fertilization until day 3. The surviving 120 embryos, after thawing, were allocated to 2, 5 or 20% oxygen concentration for further culture to the blastocyst stage. The rates of blastocyst formation and good quality blastocyst were not statistically different among the three oxygen concentration groups. However, the study presents a low number of embryos, (only 120) and the randomization showed uneven numbers among the groups (Yang et al., Reference Yang, Xu, Ding, Khoudja, Lin, Awonuga, Dai, Puscheck, Rappolee and Zhou2016). Another study on human embryos was presented by Kaser and collaborators at the American Society for Reproductive Medicine (ASRM) meeting in 2016 (Kaser et al., Reference Kaser, Bogale, Sarda, Farland and Racowsky2016). Donated embryos at the pronuclear (2PN) stage were randomized to either 5% oxygen from days 1 to 5 (102 embryos) or 5% oxygen from days 1 to 3 and moved to 2% oxygen from day 3 to day 5 (101 embryos). The percentage of good quality embryo on day 3 was similar between the two groups. Quite interestingly, the authors found in the 5–2% group a two-fold improvement in blastulation rate (40.2% versus 22.5%, P = 0.02). The same group published another study evaluating the effect of sequential oxygen tension on human blastocysts (5% oxygen from day 1 to day 3 and subsequently 2% from day 3 to day 5). This was a randomized controlled trial in which embryos were donated to research and a good number were tripronucleate (3PN) zygotes. Sibling zygotes were randomized to culture in 5% oxygen from day 1 until day 5 (n = 102) or 5% oxygen from day 1 to day 3, then 2% from day 3 to day 5 (n = 101). Although this was a small study, the results supported the hypothesis that the total proportion of blastocysts is superior when oxygen tension is reduced from 5% to 2% on day 3 for extended culture to day 5. In addition, they observed a significant increase in the proportion of 2PN and 3PN embryos that blastulated in 2% oxygen, demonstrating improved utilization of key anabolic amino acids and signs of reduced redox stress (Kaser et al., Reference Kaser, Bogale, Sarda, Farland, Williams and Racowsky2018). A study presented at the European Society of Human Reproduction and Embryology (ESHRE) meeting in 2017 by Ferrieres-Hoa and colleagues (Ferrieres-Hoa et al., Reference Ferrieres-Hoa, Roman, Mullet, Gala and Hamamah2017) analyzed the effect of ultra-low oxygen tensions of 2% on the development of human blastocysts. Blastocyst formation rate was 60.1% (143/238) in the 2% oxygen group compared with 52.1% (88/169) in the control group (5%). Blastocyst utilization rate was significantly higher in the 2% oxygen group (72% vs. 45.5%, P < 0.0001). According to these preliminary studies, it seem that the culture of human embryos in a very low oxygen tension of 2% may improve IVF outcomes. Culture in sequential oxygen concentration, cleavage stage until day 3 at 5% oxygen and extended culture to blastocyst stage at 2% is an interesting concept from both a biological and clinical outcome perspective. However, more work is needed to confirm this trend, before translating this regime to clinical use.

Oxygen tension utilized in different IVF laboratories

Oxygen tension is one of the most important aspects discussed here, but sometimes is also one of the most neglected. In 1971, Edwards and Steptoe performed several experiments comparing embryo development at different oxygen concentrations. They obtained better results under reduced oxygen tension. This finding helped to establish 5% oxygen tension as the standard condition for human embryo culture. Subsequently several studies comparing embryo development under different oxygen concentrations were performed. Haidri and co-workers, (Reference Haidri, Miller and Gwatkin1971) described the negative effect of atmospheric oxygen concentration on the developmental competence of mouse oocytes, in particular oxygen has been shown to affect oocyte nuclear maturation and reduce embryo implantation potential. Eppig and Wigglesworth (Reference Eppig and Wigglesworth1995) showed how increased oxygen tension had a dramatic effect on the development of the mouse embryo. The study reported that 42% of oocytes grown in 5% oxygen cleaved to the 2-cell stage compared with only 3% when the oocytes were cultured in a 20% oxygen concentration. The use of low oxygen concentration and embryo culture widely diverged among clinics around the world. Although is not an easy task to carry out a global investigation of all IVF units, a world survey on the oxygen concentration used during embryo culture has been performed and published by Christianson and colleagues (Reference Christianson, Zhao, Shoham, Granot, Safran, Khafagy, Leong and Shoham2014). The survey was conducted online and collected data from 265 clinics and 71 different countries. Results showed that only 27% of IVF embryos are cultured exclusively at physiologic oxygen tension of 5%. Some units (34%), reported to use a combination with two gas system, using 5% oxygen only at specific stages during the embryo culture, while the majority of IVF clinics interviewed (39%) did not use 5% oxygen concentration at any time during the culture. Even if this analysis is only representative of a small part of the world’s IVF clinics, it is evident from the survey and from the literature that there is no agreement on the oxygen concentration to be used in embryo culture. In addition, the existence of geographical differences about the use of 5% oxygen are clear. Australia, New Zealand, Japan and some Northern European countries seem to use physiologic oxygen concentrations of 5%, while other countries use atmospheric oxygen concentration of 20%. The wide adoption of low oxygen tension in Australian IVF clinics can be attributed to several studies performed in that part of the world showing the advantage of reduced oxygen tension in different mammals and humans (Catt and Henman, Reference Catt and Henman2000). The most influential reason for not using low oxygen might be related to the additional cost. Reduced oxygen in the culture environment requires a nitrogen gas system, specialized incubators, and additional quality assurance associated with oxygen sensors (Higdon et al., Reference Higdon, Blackhurst and Boone2008). Finally, the oxygen tension in embryo culture may also have safety implications and long-term offspring health implication yet to be fully realized. During embryo culture, major epigenetic reprogramming takes place that is extremely important for the normal fate of the embryo and establishing developmental origin of offspring health and disease. Epigenetic reprogramming is very vulnerable to changes in environmental conditions such as the ones applied in IVF, including oxygen tension. The increased oxidative load derived from excess of oxygen may affect oxidation of methylcytosines leading to cytosine de-methylation and/or deamination to thymine and possibly uracil incorporation. Recently, several studies have reported the influence of oxygen tension and culture medium in epigenetic reprogramming and its effect on early embryo development (Morgan et al., Reference Morgan, Santos, Green, Dean and Reik2005; Lonergan et al., Reference Lonergan, Fair, Corcoran and Evans2006; Ventura-Juncá et al., Reference Ventura-Juncá, Irarrázaval, Rolle, Gutiérrez, Moreno and Santos2009; Marcho et al., Reference Marcho, Cui and Mager2015) especially of imprinted genes that play a key role in fetal and placental development (Mann et al., Reference Mann, Lee, Doherty, Verona, Nolen, Schultz and Bartolomei2004) and glucose metabolism and fetal growth in mice model (Donjacour et al., Reference Donjacour, Liu, Lin, Simbulan and Rinaudo2014). The consequences of these perturbations on the outcome of the offspring raises important questions not resolved that still need to be clarified. This calls for the need for more research and data on the effects of oxygen tension on epigenetic processes and on long-term consequences. Should future data suggest an important link between oxygen tension, embryo culture/manipulation, epigenetic modifications, and offspring health, then a shift to a more closely regulated oxygen tension during embryo culture will be needed.