| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Articles |
a Laboratory for Reproductive Medicine, Marine Biological Laboratory, Woods Hole, Massachusetts 02543
b BioCurrents Research Center, Marine Biological Laboratory, Woods Hole, Massachusetts 02543
c Department of Biological Sciences, University of Missouri-Rolla, Rolla, Missouri 65409
d Women and Infants Hospital, Brown University, Providence, Rhode Island 02905
ABSTRACT
The self-referencing electrode technique was employed to noninvasively measure gradients of dissolved oxygen in the medium immediately surrounding developing mouse embryos and, thereby, characterized changes in oxygen consumption and utilization during development. A gradient of depleted oxygen surrounded each embryo and could be detected >50 µm from the embryo. Blastocysts depleted the surrounding medium of 0.6 ± 0.1 µM of oxygen, whereas early cleavage stage embryos depleted the medium of only 0.3 ± 0.1 µM of oxygen, suggesting a twofold increase in oxygen consumption at the blastocyst stage. Mitochondrial oxidative phosphorylation (OXPHOS) accounted for 60–70% of the oxygen consumed by blastocysts, while it accounted for only 30% of the total oxygen consumed by cleavage-stage embryos. The amount of oxygen consumed by non-OXPHOS mechanisms remained relatively constant throughout preimplantation development. By contrast, the amount of oxygen consumed by OXPHOS in blastocysts is greater than that consumed by OXPHOS in cleavage-stage embryos. The amount of oxygen consumed by one-cell embryos was modulated by the absence of pyruvate from the culture medium. Treatment of one-cell embryos and blastocysts with diamide, an agent known to induce cell death in embryos, resulted in a decline in oxygen consumption, such that the medium surrounding dying embryos was not as depleted of oxygen as that surrounding untreated control embryos. Together these results validate the self-referencing electrode technique for analyzing oxygen consumption and utilization by preimplantation embryos and demonstrate that changes in oxygen consumption accompany important physiological events, such as development, response to medium metabolites, or cell death.
The mammalian oocyte contains approximately 100 000 mitochondria [1, 2], each of which harbors DNA (mtDNA) that encodes essential components of oxidative phosphorylation (OXPHOS) machinery [3]. Mitochondrial DNA can be damaged by oxidative stress [4] and a wide variety of mtDNA rearrangements appear in oocytes from infertile women [5]. The heteroplasmy of oocyte mtDNA challenges currently available genetic technology and therefore limits the ability of molecular genetic approaches to predict mitochondrial function. Measurements of oxygen consumption from oocytes and embryos should provide a more direct assay of mitochondrial function and dysfunction.
More than 60 yr have past since the first oxygen measurements were taken from mammalian oocytes or embryos [6–8]. Using a Cartesian diver technique, initial studies demonstrated that groups of blastocysts consumed markedly more oxygen than did groups of cleavage-stage embryos and that changing the concentration of metabolic substrates in culture medium altered the amount of oxygen consumed by embryos [6–11]. Subsequent studies employed a microspectrophotometric technique to analyze spectral shifts in extracellular oxyhemoglobin resulting from oxygen consumption by embryos [12–14]. Stationary solid-state oxygen electrodes also have been employed to measure oxygen consumption from groups of embryos and from individual embryos. Using these electrodes, Overstrom et al. [15] documented, as had Magnusson et al. [12] using spectrophotometric methods, that embryos consuming more oxygen developed to blastocysts and exhibited higher survival rates than those that consumed less oxygen, although these results have not been subsequently confirmed. Also using oxygen electrodes, Benos and Balaban [16] determined that between 50 and 70% of the oxygen consumed by blastocysts was utilized by OXPHOS to generate ATP for the plasma membrane Na/K-ATPase, and Manes and Lai [17] demonstrated that a portion of the total oxygen consumed by embryos was not utilized by mitochondrial OXPHOS, but by other oxygenases. More recently, Houghton et al. [18] developed an ultrafluorescence technique to measure oxygen consumption and expanded upon the relationship between medium metabolites and oxygen consumption by embryos.
It is upon this historical backdrop that we apply a novel technology for assessing oxygen consumption from individual oocytes and embryos. The self-referencing electrode technique [19, 20] allows noninvasive measurements of physiological parameters from individual cells by detecting differences between the composition of the medium immediately surrounding cells relative to that of the bulk medium. The exquisite sensitivity of this methodology arises by moving an electrode between two positions, 10 µm apart. One pole of the electrode oscillation measures the physiological signal near the cell while the other pole of the oscillation measures the physiological signal at a position remote from the cell. By obtaining the difference between the measurements at these two sites, the contribution of electronic drift is reduced and the resultant signal obtained is proportional to the difference in the concentration of the measured parameter at these two positions. This technique has been used to measure oxygen consumption from a variety of cell types [21] and recently was employed to characterize the baseline oxygen consumed by developing embryos [22].
Here, we present the application of this technique to investigate the modulation of embryonic oxygen consumption by a number of factors, including; mitochondrial toxins, a metabolic substrate in the culture medium, and a cell death-inducing agent known to evoke changes in mitochondria [23]. To our knowledge, significant levels of mtDNA rearrangements have not been reported in wild-type or genetically engineered mouse oocytes. Therefore, we have employed pharmacological agents to perturb OXPHOS in mouse oocytes and, by this approach, have determined the contribution of mitochondrial OXPHOS-dependent and OXPHOS-independent processes to the total oxygen consumed by individual embryos.
One benefit of the self-referencing electrode technique is that oxygen measurements can be noninvasively obtained from individual embryos and therefore this technology can interface with other single-cell techniques, such as polymerase chain reaction (PCR) of mtDNA deletions [5], metabolite-uptake/lactate-secretion assays [24, 25], mitochondria redistribution dynamics [26], preimplantation genetic diagnosis [27, 28], in vivo protein dynamic studies using green-fluorescent protein (GFP) [29], and fluorescence resonance energy transfer (FRET) [30]. The self-referencing electrode technique also will permit exploration of subtle changes in oxygen consumption that otherwise would be unattainable from groups of asynchronous embryos and may eventually provide a diagnostic tool for assessing the level of mitochondria dysfunction and developmental potential of individual embryos.
MATERIALS AND METHODS
The self-referencing system used to monitor oxygen gradients around embryos was identical to that previously described [19–22]. Briefly, this system was composed of a Zeiss Axiovert 100 TV inverted microscope (Carl Zeiss, Inc., Thornwood, NY) with a modified stage plate to which computer-controlled micromanipulators were affixed (BioCurrents Research Center, MBL, Woods Hole, MA). The microscope rested on a Kinetic Systems Vibraplane vibration-resistant table (Boston, MA) enclosed in a stainless steel insulated chamber. The temperature of the chamber and its contents, including the microscope and manipulators, was maintained at 37°C. Physiological measurements were conducted in 4 ml of HEPES-buffered KSOM (HKSOM) containing reduced NaHCO3 (4 mM) and elevated HEPES (14 mM) absent of an oil overlay and at atmospheric gas pressures. Recordings were obtained from embryos in plastic petri dishes with a coverglass bottom (MatTek Corp., Ashland, MA) that facilitated embryo positional stabilization. Typically, a set of four to six embryos was placed in a dish, and at least three independent recordings were obtained over several minutes from each embryo in a semirandomized order. Approximately 60 min was required to complete the measurements from a set of embryos. We have shown that mouse embryos exposed to these conditions developed to blastocysts and live pups at a frequency similar to control embryos not physiologically examined (unpublished data), and oxygen gradients around embryos were stable throughout the measurement period.
Oxygen-sensitive electrodes (tip diameter of 2–4 µm) were purchased from Diamond General Corp. (Ann Arbor, MI) and used as previously described [21, 22]. A silver/silver chloride reference electrode completed the circuit in solution by way of a 3 mol L-1 KCl/3% agar bridge. All electrodes were calibrated in saturated oxygen and saturated nitrogen solutions. During recording, the electrode was oscillated in a square wave parallel to the electrode axis over a distance of 10 µm with a frequency of 0.3 Hz. Initially, the near position of this oscillation was 1.5–2.5 µm from the zona pellucida or plasma membrane in cases where the zona pellucida was removed by brief pronase (0.5% in HKSOM) digestion. Mathematical modeling of the gradients around embryos suggested that more reliable signals could be obtained at a distance of 6 µm away from embryos. Accordingly, all subsequent recordings were obtained at this standard distance. Data acquisition and manipulation were performed as described previously [19–22]. The hardware and software controlling electrode movements, signal amplification, and data acquisition were designed and constructed by the BioCurrents Research Center at the Marine Biological Laboratory, Woods Hole, MA (www.mbl.edu/BioCurrents [case sensitive]). The lower limit of detection, with a signal-to-noise ratio of three, was approximately a 0.01 µM difference in oxygen concentration between the two excursion points, and the oxygen signals we document here were greater than 10 times this lower detection limit. All data are presented as mean ± standard deviation unless otherwise specified.
Digital images of embryos were captured periodically throughout the experiment using a Cohu analog video camera (Cambridge Research Instruments, Cambridge, MA) and a personal computer running Metamorph Software (Universal Imaging Corp., West Chester, PA). The morphometric features of embryos were analyzed from these images using Metamorph Software.
Stock solutions of mitochondrial toxins, carbonyl cyanide P-trifluoro-methoxyphenyl-hydrazone (FCCP, 1 mM) and antimycin A (1 mM), were prepared in ethanol. On the day of use, these stocks were further diluted to a concentration of 100 µM in ethanol. Stock solutions of sodium cyanide (1 M) were prepared in water, daily. While recording oxygen gradients from embryos, stock solutions containing mitochondrial toxins were added to the physiological medium to yield the final desired concentrations: 1 µM FCCP, 1 µM antimycin A, 1 mM sodium cyanide, 0.01% solvent. Cyanide at 1 mM is more than adequate to inhibit mitochondrial cytochrome oxidase [31]. To induce disruption in mitochondria associated with cell death, one-cell embryos and blastocysts were treated with 50 µM diamide for 3 h, according to the procedures previously reported [23]. To investigate the influence of the metabolic substrate, pyruvate, on oxygen consumption, one-cell embryos, harvested 22–23 h after hCG injection (see below), were transferred into pyruvate-free KSOM 6–7 h after retrieval (28–29 h after hCG injection) and cultured overnight. Several hours after culture in pyruvate-free medium these embryos cleaved in a manner similar to control sibling embryos that were incubated in standard KSOM containing 0.2 mM pyruvate, and although the blastomeres were smaller (see Results) they were otherwise morphologically indistinguishable. Oxygen consumption measurements from the resulting two-cell embryos cultured overnight in pyruvate-free medium were conducted in HEPES-buffered pyruvate-free KSOM and in a similar manner as measurements from control embryos conducted in HEPES-buffered KSOM containing 0.2 mM pyruvate.
Female B6C3F1 mice (6 wk old) were purchased from Charles River (Boston, MA) and subjected to a 14L:10D cycle for at least 1 wk before use. Animals were cared for according to procedures approved by the Marine Biological Laboratory and Women and Infants Hospital Animal Care Committees. Male B6C3F1 mice of proven fertility were used for mating. Female mice were superovulated by intraperitoneal injection of 7.5 IU PMSG (Calbiochem, La Jolla, CA) followed 46–48 h later by injection of 7.5 IU hCG, and mated individually. Females exhibiting mating plugs were selected the next morning and killed by cervical dislocation at 22–23 h after hCG injection. Zygotes (~30/animal) enclosed in cumulus masses were released from the ampullae into HKSOM supplemented with 0.03% hyaluronidase. Cumulus cells were gently removed from one-cell embryos by pipetting. Cumulus-free one-cell embryos were washed in HEPES-buffered KSOM three times and then in pre-equilibrated modified KSOM three times. Modified KSOM used for in vitro culture was supplemented with nonessential amino acids (1 ml of 100x stock/100 ml medium) and 2.5 mM HEPES. Embryos were pooled, randomly distributed, and cultured in 50-µl droplets (7–16 embryos/drop) under mineral oil at 37°C in a humidified atmosphere of 7% CO2 in air. These embryo harvest and in vitro culture procedures sustained development of control one-cell embryos to blastocysts (>90%) and subsequent development to term. For most experiments, the zona pellucida was removed mechanically following mild treatment with pronase. No difference in physiological signals was observed if the embryos were examined immediately following ZP removal or after a 0.5–2.5-h incubation/recovery period. Zona-free embryos cleaved in a manner similar to zona-intact sibling embryos. All reagents were purchased from Sigma Chemical Co. (St. Louis, MO), unless stated otherwise.
RESULTS
Oxygen Gradients Around Embryos
Using an oxygen-sensitive electrode in self-referencing mode, the concentration of dissolved oxygen in the medium surrounding individual mouse embryos could be measured reliably and was stable over the recording duration (~1 h). The dissolved oxygen concentration of the medium near one-cell embryos was reduced by 0.33 ± 0.07 µM relative that of the bulk medium (0.21 mM), such that a gradient of depleted dissolved oxygen surrounded each embryo (Fig. 1A). The concentration of oxygen in the medium was most depleted nearest the embryos and the medium was less depleted of oxygen with increasing distance from the embryo. The gradient of dissolved oxygen concentration could be measured >50 µm away from the embryo surface (Fig. 2).
|
|
The dissolved oxygen concentration nearest embryos of different developmental stages was characteristic of the particular stage and the medium surrounding later developmental stages was more depleted of oxygen than that surrounding earlier stages. On average, the medium near a two-cell embryo was more depleted of oxygen (0.43 ± 0.10 µM below that of the bulk medium, n = 22) than the medium surrounding a one-cell embryo (0.33 ± 0.07 µM below that of the bulk medium, n = 8) (Fig. 1B). However, the radius of a two-cell blastomere (26.9 ± 0.6 µm; n = 22) was less than that of a one-cell embryo (37.3 ± 0.7 µm; n = 16), and therefore, the position of the probe relative to the geometric center of the specimen was not equivalent between embryos of different developmental stages. A mathematical function was determined that allowed the oxygen gradients surrounding embryos of different sizes to be adjusted geometrically and thus compared quantitatively (Fig. 2B). The only assumption intrinsic to this function was that the oxygen consumption was relatively homogeneous around the embryo. Measurements from different locations around embryos have not revealed regional difference in oxygen gradients and thus experimentally confirmed this assumption. Perhaps surprisingly, oxygen consumption measurements around blastocysts also did not reveal regional differences (e.g., inner cell mass versus polar trophectoderm), and we are presently investigating whether the diffusional properties of oxygen underlie this signal normalization. A series of 120 measurements were obtained at various distances (>6 µm) from nine blastocysts, and a standard curve was generated with the function
where x is the distance from the electrode tip to the geometric center of the cell as determined from images of embryos taken at an equatorial focal plane (Fig. 2B). Scaling to a common radius of 40 µm occurred by Eq. 2
This set of equations has been verified in more than a dozen subsequent experiments from more than 80 embryos of a variety of developmental stages, including some devoid of their zona pellucida, and in each case the data could be fit with this function.
The mathematical function was used to normalize the oxygen concentration values obtained near embryos of different sizes and yielded the following relationships; all cleavage-stage embryos (one-, two-, and four-cell) depleted the surrounding medium of oxygen to a similar degree (change in dissolved [O2] = ~0.3 µM below that of the bulk medium) and the medium surrounding blastocysts was depleted of oxygen twofold more than that surrounding cleavage-stage embryos (change in dissolved [O2] = 0.6 ± 0.1 µM below that of the bulk medium) (Fig. 1B).
These values highlight the sensitivity of the self-referencing electrode technique. The concentration of dissolved oxygen in bulk medium was 0.2102 mM, and the medium immediately surrounding individual blastocysts was depleted of oxygen such that the oxygen concentration declined to 0.2096 mM or a change of only 1/350th. Yet, the oxygen electrode in self-referencing mode could reliably and unequivocally detect this minute change in oxygen concentration.
It was likely that the region of depleted oxygen surrounding individual embryos resulted from oxygen consumption by the embryo, and oxygen consumption by individual embryos could be calculated using Fick's law: flux = oxygen diffusion coefficient x (
[O2]/probe excursion) [19–22]. The degree of oxygen depletion we observed around embryos corresponded to an oxygen consumption by individual cleavage-stage embryos of 7.7 pmol O2 cm-2 h-1 (or 0.1 nl embryo-1 h-1) and by individual blastocysts of 15.8 pmol O2 cm-2 h-1 (or 0.3 nl embryo-1 h-1).
Oxygen gradient measurements from 22 individual two-cell embryos exhibited a normal distribution with a median (±95% confidence intervals) of 0.31 ± 0.03 µM below that of the bulk medium and ranged from 0.27 to 0.41 µM below that of the bulk medium (Fig. 3). Subsequent to obtaining the oxygen measurements, 16 of these two-cell embryos were cultured and >90% developed to blastocysts. There appeared no obvious correlation between the magnitude of the oxygen gradient around an embryo and its developmental competence; however, almost all mouse embryos in this study exhibited robust development, thereby limiting the power of the analysis to determine a relationship between oxygen consumption and developmental potential. Present experiments employing other model systems continue to investigate this relationship (see Diamide Oxidation Modulates Oxygen Consumption below).
|
Oxygen Gradients Result, in Part, from Metabolic OXPHOS
The oxygen gradients surrounding embryos arose from a combination of mitochondrial OXPHOS and oxygen consumption by other oxygenases not associated with OXPHOS. To determine the proportion of the oxygen gradients resulting from OXPHOS, we measured oxygen gradients around individual embryos exposed to the following mitochondria uncouplers and toxins: FCCP that increases oxygen consumption by mitochondria through uncoupling the proton gradient from oxidative phosphorylation; sodium cyanide and antimycin A that reduce oxygen consumption by mitochondria through blocking electron flow at different sites within the oxidative phosphorylation cascade [31]. Embryos treated with FCCP (1 µM) consumed more oxygen as indicated by a greater depletion of oxygen from the surrounding medium (Fig. 4A,B). The medium surrounding individual one-cell embryos treated with FCCP was more depleted of oxygen (89.2 ± 31.2% more) compared to the baseline oxygen concentration prior to FCCP treatment (0% in Fig. 4B), indicating that the mitochondria of FCCP-treated embryos consumed more oxygen than did those of control cleavage-stage embryos (Fig. 4). Similarly, the medium surrounding individual blastocysts treated with FCCP was more depleted of oxygen than that surrounding control untreated blastocysts, but only 49.5 ± 25.9% more depleted than the baseline oxygen concentration prior to FCCP treatment (0% in Fig. 4B), indicating a relatively more modest increase of OXPHOS in blastocysts following FCCP treatment.
|
Both cyanide (1 mM) and antimycin A (1 µM) decreased oxygen consumption by embryos as observed by a return of the medium oxygen concentration near embryos toward concentrations more similar to that of the bulk medium (Fig. 4A,B). The medium surrounding individual two-cell embryos treated with cyanide or antimycin A was less depleted of oxygen (~23% less) than that surrounding control untreated embryos, indicating a decline in oxygen consumption by treated embryos (Fig. 4B). Expressed another way, ~77% of the oxygen consumed by cleavage-stage embryos was cyanide/antimycin A resistant. The medium surrounding individual blastocysts treated with cyanide or antimycin A was substantially less depleted of oxygen (60–70% less) than that surrounding control blastocysts, indicating that antimycin A and cyanide evoked a pronounced decrease in oxygen consumption by blastocysts (Fig. 4B). By contrast to cleavage-stage embryos, only ~30% of the oxygen consumed by blastocysts was cyanide/antimycin A resistant. These numbers could be used to calculate the contribution of OXPHOS-independent oxygen consumption to the total oxygen depleted from the surrounding medium and yielded an interesting result; the medium surrounding both cleavage-stage embryos and blastocysts was similarly depleted of 0.2 µM dissolved oxygen by OXPHOS-independent mechanisms. Therefore, OXPHOS-independent oxygen consumption remained relatively constant throughout embryo development, while mitochondrial OXPHOS-dependent oxygen consumption substantially increased between cleavage-stage embryos and blastocysts (Fig. 4C).
Medium Pyruvate Concentration Modulates Oxygen Consumption
Embryos cultured overnight from the late one-cell to the two-cell stage in pyruvate-free KSOM (pyruvate-free embryos), consumed less oxygen than did control embryos cultured overnight in the presence of 0.2 mM pyruvate (Fig. 5). The medium surrounding individual pyruvate-free, two-cell embryos was less depleted of oxygen than that surrounding control embryos (Fig. 5A). This difference in the degree of oxygen depletion was partially masked by the smaller size of pyruvate-free embryos (pyruvate-free embryo radii = 26.6 ± 0.9 µm; control embryo radii = 28.5 ± 0.7 µm; single factor ANOVA, P << 0.005) that allowed positioning of the electrode closer to the center of pyruvate-free embryos and thereby methodologically increased the signal (Fig. 5A). By using the mathematical function described earlier (Eqs. 1 and 2, Fig. 2B and associated text), the oxygen gradients around pyruvate-free and control embryos could be scaled such that embryos varying in size could be compared quantitatively. Upon scaling the average oxygen concentration of the medium surrounding pyruvate-free embryos was 0.12 ± 0.02 µM below that of the bulk medium, while the oxygen concentration of the medium surrounding control embryos was 0.16 ± 0.02 µM below that of the bulk medium (single factor ANOVA, P << 0.005) (Fig. 5B). Embryos cultured overnight in the absence of pyruvate consumed 26% less oxygen than those cultured in medium containing pyruvate.
|
Diamide Oxidation Modulates Oxygen Consumption
Diamide, a glutathione oxidizing agent, evokes changes in mitochondria associated with cell death in mammalian embryos [23]. Embryos treated with 50 µM diamide for 3 h consumed less oxygen than did untreated, control embryos (Fig. 6). The medium surrounding individual diamide-treated one-cell embryos was less depleted of oxygen than that surrounding control embryos (Fig. 6A). Scaling of these data using the mathematical function described earlier (Eqs. 1 and 2, Fig. 2B and associated text) allowed embryos varying in size, to be compared quantitatively. The medium surrounding diamide-treated one-cell embryos was less depleted of oxygen (0.12 ± 0.02 µM below that of the bulk medium) than that surrounding control one-cell embryos (0.18 ± 0.01 µM below that of the bulk medium) (single factor ANOVA, P << 0.005) (Fig. 6B). One-cell embryos treated with diamide and undergoing cell death consumed 32% less oxygen than did control healthy embryos.
|
Similarly, the medium surrounding individual diamide-treated blastocysts was less depleted of oxygen than that surrounding untreated, control blastocysts (Fig. 6C). Upon mathematical scaling, the oxygen concentration of the medium surrounding diamide-treated blastocysts was substantially less depleted of oxygen (0.17 ± 0.03 µM below that of the bulk medium) than that surrounding control blastocysts (0.30 ± 0.04 µM below that of the bulk medium) (single factor ANOVA, P << 0.005) (Fig. 6B). Blastocysts treated with diamide consumed 43% less oxygen than do control blastocysts.
DISCUSSION
The self-referencing electrode technique was used to measure the amount of oxygen consumed by individual preimplantation embryos by noninvasively measuring the amount of oxygen depleted from the surrounding medium. A gradient of depleted dissolved oxygen surrounded each embryo and embryos had a profound influence on the concentration of available oxygen at distances of 50 µm and more from the embryo surface. This gradient of depleted oxygen extends away from the embryo into the adjacent medium to distances greater than one-half the embryo's diameter. Neighboring embryos communally cultured in a microdrop in vitro presumably could, therefore, influence the amount of oxygen available to each other. This physiological interaction between embryos may account, in part, for the enhanced cooperative development seen by groups of embryos cultured together versus embryos cultured singly [32, 33].
The measurements of oxygen consumption from individual embryos confirmed that blastocysts consume two to four times more oxygen than do cleavage-stage embryos [8–11, 18, 34]. More importantly, the oxygen consumed by embryos at different developmental stages contributed to different physiological processes. The portion of the total oxygen consumed by OXPHOS-dependent and OXPHOS-independent processes changed during preimplantation development [9, 11, 18, 34–36]. The majority (70%) of oxygen utilized by blastocysts was cyanide-sensitive [9, 16], consistent with oxygen utilization by OXPHOS within mitochondria. Benos and Balaban [16] suggested that 50–70% of all oxygen consumed by blastocyst-stage embryos was utilized by mitochondria to generate ATP for the Na/K-ATPase. Conversely, the minority (20–30%) of oxygen consumed by cleavage-stage embryos was cyanide sensitive, suggesting that OXPHOS-independent oxygenases contributed greatly to the total oxygen consumed by cleavage-stage embryos. Manes [37] suggested that a substantial portion of the OXPHOS-independent oxygen consumption by blastocysts contributes to surface H2O2 production and, indeed, Manes and Lai [17] determined that 30% of the total oxygen consumed by blastocysts was utilized by H2O2-producing oxygenases. Because cleavage-stage embryos produce only limited H2O2 [38] (unpublished observations), the two mixed multifunctional oxidase systems maturing in cleavage-stage embryos [38, 39] might account for the OXPHOS-independent oxygen consumption at this developmental stage.
In brief, we found, as did others [35, 36], that the OXPHOS-dependent/OXPHOS-independent ratio is approximately 30%/70% in cleavage-stage embryos and switches to 70%/30% in blastocysts. These ratios are compatible with data obtained from somatic cells [40, 41]. Cyanide-resistant oxygen consumption can contribute to between 10 and 90% of the total oxygen consumed by somatic cells and, in rare cases, such as during the burst of oxygen consumption that is characteristic of activated neutrophils, greater than 98% of the oxygen consumed is utilized by OXPHOS-independent oxygenases [40].
Although the estimations of OXPHOS-dependent/OXPHOS-independent ratios are based upon pharmacological manipulations, and therefore should be interpreted with caution, they are consistent with a host of other related changes that occur during preimplantation embryo development including: changes in mitochondria morphology from condensed structures in cleavage-stage embryos to swollen cristae-containing organelles in blastocysts [42–47]; an increase in mitochondria number in blastocysts resulting from mitochondria DNA and structural replication [44–49]; changes in the biogenesis of key OXPHOS enzymes [50, 51]; a decrease in total ATP content and ATP/ADP+P ratio that would stimulate OXPHOS [35, 52, 53]; and a change in energy substrate preferences and waste product secretion [18, 24, 25, 34–36, 54, 55].
By assimilating our findings with those of previous studies, a more complete picture of the developmental changes in energy metabolism emerges. At the two-cell stage, when the mitochondria are condensed, the total ATP content of the cell, as well as the ATP/ADP ratio are high, pyruvate is preferentially used as an energy substrate, lactate secretion is low, the total oxygen consumed and carbon dioxide produced is low, and OXPHOS-independent mechanisms account for the majority of oxygen consumed. At the blastocyst stage, when the mitochondria exhibit cristae, the total ATP content and the ATP/ADP ratio are low, stimulating aerobic glycolysis and OXPHOS-dependent mechanisms that increase lactate secretion, the total oxygen consumed, and carbon dioxide produced.
Despite the low levels of oxygen consumed by OXPHOS in cleavage-stage embryos, the self-referencing electrode technique could, nevertheless, detect changes in oxygen consumption resulting from alteration in the concentration of pyruvate in the medium. Pyruvate is preferentially utilized as an energy source by cleavage-stage embryos [9, 18, 25, 35, 36] and incubation of cleavage-stage embryos in medium devoid of pyruvate resulted in a decline in oxygen consumption by 26% (P << 0.005) over that consumed by control embryos [34]. This decline in total oxygen consumed was consistent with a decrease in mitochondrial oxygen consumption of ~90% and suggests that these cleavage-stage embryos did not appreciably compensate for the absence of pyruvate by utilizing other available metabolic substrates such as glucose (0.2 mM).
Finally, Magnusson et al. [12] suggested that human blastocysts that consumed more oxygen developed to blastocysts at higher rates than did those that consumed less oxygen. Independently, Overstrom et al. [15] confirmed this result using bovine blastocysts. Consistent with these findings, we demonstrate here that healthy control one-cell embryos and blastocysts consumed more oxygen than did those undergoing cell death. The early phases of cell death are associated with changes in mitochondria [23, 56, 57], and the change in oxygen consumption was detectable only 3 h after treatment with an agent (diamide) known to induce death of embryos [23]. These results suggest that changes in oxygen consumption might be indicative of disruptions in mitochondria and a decline in developmental competence. Future studies to investigate this relationship directly should now be possible by employing the self-referencing electrode technique to monitor oxygen consumption from individual preimplantation embryos.
ACKNOWLEDGMENTS
We thank Richard Sanger and Katherine Hammer for assistance with the self-referencing technique and Gaudenz Danuser for his programming assistance and intellectual support.
FOOTNOTES
First decision: 6 December 1999.
1 A portion of this work was supported by grants NIH R21 RR 12718-02 to D.L.K. and P.J.S.S; KO81099 to D.L.K.; and NIH P41 RR01395 to P.J.S.S. 
2 Correspondence: David Keefe, Laboratory for Reproductive Medicine, Lillie Building, Marine Biological Laboratory, 7 MBL Street, Woods Hole, MA 02543. FAX: 508 540 6902; dkeefe{at}wihri.org
Accepted: January 27, 2000.
Received: November 8, 1999.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  M. Hoelker, F. Rings, Q. Lund, N. Ghanem, C. Phatsara, J. Griese, K. Schellander, and D. Tesfaye Effect of the microenvironment and embryo density on developmental characteristics and gene expression profile of bovine preimplantative embryos cultured in vitro Reproduction, March 1, 2009; 137(3): 415 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. C. Spikings, J. Alderson, and J. C. St. John Regulated Mitochondrial DNA Replication During Oocyte Maturation Is Essential for Successful Porcine Embryonic Development Biol Reprod, February 1, 2007; 76(2): 327 - 335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Manser and F. D. Houghton Ca2+-linked upregulation and mitochondrial production of nitric oxide in the mouse preimplantation embryo J. Cell Sci., May 15, 2006; 119(10): 2048 - 2055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Van Blerkom, H. Cox, and P. Davis Regulatory roles for mitochondria in the peri-implantation mouse blastocyst: possible origins and developmental significance of differential {Delta}{Psi}m. Reproduction, May 1, 2006; 131(5): 961 - 976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H El Shourbagy, E. C Spikings, M. Freitas, and J. C St John Mitochondria directly influence fertilisation outcome in the pig Reproduction, February 1, 2006; 131(2): 233 - 245. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A S Lopes, L H Larsen, N Ramsing, P Lovendahl, M Raty, J Peippo, T Greve, and H Callesen Respiration rates of individual bovine in vitro-produced embryos measured with a novel, non-invasive and highly sensitive microsensor system Reproduction, November 1, 2005; 130(5): 669 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Thouas, A. O. Trounson, E. J. Wolvetang, and G. M. Jones Mitochondrial Dysfunction in Mouse Oocytes Results in Preimplantation Embryo Arrest in Vitro Biol Reprod, December 1, 2004; 71(6): 1936 - 1942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Van Blerkom Mitochondria in human oogenesis and preimplantation embryogenesis: engines of metabolism, ionic regulation and developmental competence Reproduction, September 1, 2004; 128(3): 269 - 280. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.C. Manser, H.J. Leese, and F.D. Houghton Effect of Inhibiting Nitric Oxide Production on Mouse Preimplantation Embryo Development and Metabolism Biol Reprod, August 1, 2004; 71(2): 528 - 533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Rivera, G. M Dahlgren, L. A. de Castro e Paula, R. T Kennedy, and P. J Hansen Actions of thermal stress in two-cell bovine embryos: oxygen metabolism, glutathione and ATP content, and the time-course of development Reproduction, July 1, 2004; 128(1): 33 - 42. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Dumollard, P. Marangos, G. Fitzharris, K. Swann, M. Duchen, and J. Carroll Sperm-triggered [Ca2+] oscillations and Ca2+ homeostasis in the mouse egg have an absolute requirement for mitochondrial ATP production Development, July 1, 2004; 131(13): 3057 - 3067. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.M. Acton, A. Jurisicova, I. Jurisica, and R.F. Casper Alterations in mitochondrial membrane potential during preimplantation stages of mouse and human embryo development Mol. Hum. Reprod., January 1, 2004; 10(1): 23 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. M Orsi and H. J Leese Ammonium exposure and pyruvate affect the amino acid metabolism of bovine blastocysts in vitro Reproduction, January 1, 2004; 127(1): 131 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Dumollard, K. Hammar, M. Porterfield, P. J. Smith, C. Cibert, C. Rouviere, and C. Sardet Mitochondrial respiration and Ca2+ waves are linked during fertilization and meiosis completion Development, February 15, 2003; 130(4): 683 - 692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Van Blerkom, P. Davis, V. Mathwig, and S. Alexander Domains of high-polarized and low-polarized mitochondria may occur in mouse and human oocytes and early embryos Hum. Reprod., February 1, 2002; 17(2): 393 - 406. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Land and A Collett Detection of Cl- flux in the apical microenvironment of cultured foetal distal lung epithelial cells J. Exp. Biol., January 2, 2001; 204(4): 785 - 795. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
