Mammalian preimplantation embryos experience a critical switch from an oxidative to a predominantly glycolytic metabolism. In this study, the change in nutrient metabolism between the 2-cell and blastocyst stages was followed by measuring single embryo concentrations of tricarboxylic acid (TCA) cycle and glycolytic metabolites with microfluorometric enzymatic cycling assays. When the normal values were established, further changes that occur as a result of the induction of apoptosis by exposure to high-glucose conditions were examined. From the 2-cell to the blastocyst stage, the embryos experienced an increase in TCA metabolites and a dramatic increase in fructose 1,6-bisphosphate (FBP). The high TCA metabolites may result from accumulation of substrate due to a slowing of TCA cycle metabolism as glycolysis predominates. Embryos exposed to elevated glucose conditions experienced significantly lower FBP, suggesting decreased glycolysis, significantly higher pyruvate, suggesting increased pyruvate uptake by the embryos in response to decreased glycolysis, and increased TCA metabolites, suggesting an inability to oxidize the pyruvate and a slowing of the TCA cycle. We speculate that the glycolytic changes lead to dysfunction of the outer mitochondrial membrane that results in the abnormal TCA metabolite pattern and triggers the apoptotic event.
- tricarboxylic acid cycle
- preimplantation embryo
- programmed cell death
glucose transport and metabolism are critical for mammalian blastocyst formation and further development (8, 20). At this stage, the switch occurs from oxidation of pyruvate via the tricarboxylic acid (TCA) cycle to the use of glucose as the main substrate via glycolysis (7, 14). As a result, the blastocyst exhibits extreme sensitivity to glucose deprivation. We have previously shown that any decrease in glucose transport, basal or insulin stimulated, results in enhanced apoptosis at this stage, which manifests later in pregnancy as a malformation or miscarriage (3, 4, 21, 22). This decrease in blastocyst glucose transport and resulting apoptosis occur in conditions of maternal hyperglycemia and hyperinsulinemia.
The blastocyst stage marks a new peak in cellular proliferation and growth as the first epithelial layer, the trophectoderm, is formed. These changes create new biosynthetic demands on the embryo. Maintenance of a high rate of glycolysis is thought to be important for providing a “dynamic buffer” of metabolic intermediates for the biosynthesis of macromolecules (23). For example, glucose 6-phosphate is used in the formation of the ribose 5-phosphate required for DNA and RNA synthesis. Another reason for the increased glucose demand may be that increasing amounts of glucose are converted to lactate at the blastocyst stage in humans and rodents (14). In these species in particular, the embryo resides in the uterus for a relatively short time before implantation, and the switch to an anaerobic metabolism is in response to the lack of adequate vascularization and oxygenation at the implantation sites or decidual zones. The only source of ATP for the embryo at this point would be conversion of glucose to pyruvate and lactate via glycolysis.
Recent studies have shown that cell death caused by reduced availability of glucose, as seen with growth factor withdrawal, is initiated by mitochondrial changes that result in cytochromec release (10, 27). Overexpression of GLUT1 can prevent this onset of apoptosis (27), and the regulation of outer mitochondrial membrane integrity via the voltage-dependent anion channel (VDAC) appears to depend on cellular metabolic changes associated with glycolysis (26).
In this study, we attempted to track the changes in nutrient metabolism in the blastocyst by measuring single embryo concentrations of TCA metabolites and glycolytic metabolites by use of microfluorometric enzymatic cycling assays. We postulate that perhaps some of the metabolic alterations experienced by the blastocyst exposed to hyperglycemia may be responsible for perturbations in mitochondrial physiology and thus may trigger an apoptotic cascade.
MATERIALS AND METHODS
Embryo collection and culturing.
Embryos were recovered as described previously (21) from superovulated female mice (B6 × SJL F1, Jackson Laboratories). In vivo retrieved embryos were then flushed from the oviducts 48, 72, and 96 h later, corresponding to 2-cell, morula, and blastocyst stages, respectively. In vitro cultured embryos were flushed at the 2-cell stage and cultured in KSOM mouse embryo culture media (Specialty Media, Phillipsburg, NJ) containing either 0.2 mMd-glucose, the control concentration of glucose, or 5.6 or 50 mM d-glucose as the test conditions. The embryos were cultured at 37°C in 5% CO2-5% O2-90% N2.
Embryo extraction for metabolite assays.
At each stage, embryos were washed in BSA-free medium for 1 min and then quickly frozen on a glass slide by dipping in cold isopentane equilibrated with liquid N2. After freeze-drying overnight in a vacuum at −35°C, the embryos were extracted in nanoliter volumes under oil, as previously described (21). Citrate, α-ketoglutarate, aspartate, glycerol 3-phosphate, and ATP were measured in alkaline-extracted embryos. These embryos were extracted in 1 μl of 0.1 N NaOH at room temperature for 20 min, 0.5 μl of the extract was heated to 80°C for 20 min, and a 0.2-μl mixture of 0.2 N HCl and 0.1 M Tris · HCl (pH 6.8) was added. Malate, fumarate, and glutamate were measured in acid-extracted embryos. These embryos were extracted in 1 μl of 0.1 N NaOH at room temperature for 20 min, 0.5 μl of the extract was added to 0.1 μl of 0.6 N HCl and heated to 80°C for 20 min, and finally the extract was neutralized with 0.1 μl of 0.2 M Tris base. Both treated extracts were stored at −70°C.
Metabolite microanalytic assays.
Separate assays were developed for each metabolite measured and were designed to link to reactions using NADH or NADPH (Tables1 and 2,steps 1 and 2). The NADPH/NADH by-product is then enzymatically amplified in a cycling reaction (Table 1, steps 3 and 4), and a byproduct of the amplification step is measured in a fluorometric assay (Table 1, step 5). All metabolites except fructose 1,6-bisphosphate (FBP) are expressed as millimoles per kilogram wet weight based on the wet weight of 160 pg per embryo. FBP levels were extremely low in individual embryos and are expressed as micromoles per kilogram wet weight. Absolute concentrations of metabolites can be calculated in picomoles by multiplying by 0.16. ATP was used as a marker of viability, and if any embryos had abnormally low levels, the entire set of experiments was discarded. Previous studies have shown that cells with low ATP undergo necrosis rather then apoptosis, because apoptosis requires energy (6, 15). ATP levels are placed at the outset of each table or figure to demonstrate the equality of the embryos tested.
All assays for a particular metabolite in a given experiment were made at the same time and conducted at ambient temperatures. Metabolites were measured according to the protocols of Table 1 and reagents of Table 2. Standards were carried throughout the entire procedure. Most of the enzymes were purchased in a suspension of ammonia sulfate, spun down, and reconstituted with 20 mM imidazole HCl (pH 7.0) and 0.02% BSA. All experiments were completed a minimum of three times. For each metabolite measurement under each condition, at least 16 individual embryos were used. The NADP+ cycling reagent, used only for the ATP assay, contains 100 mM imidazole HCl (pH 7.0), 7.5 mM α-ketoglutarate, 5 mM glucose 6-phosphate, 25 mM NH4Ac, 0.02% BSA, 100 μM ADP, 100 μg/ml beef liver glutamate dehydrogenase, and 10 μg/ml Leuconostoc mesenteroidesglucose 6-phosphate dehydrogenase. This enzyme ratio gives ∼100,000-fold amplification, as detailed in Table 1 for ATP. The indicator reaction for the NADP+ cycling step, orstep 5 in Table 1, involves adding 10 μl of the reaction to 1 ml of the indicator, 6-phosphogluconate reagent, containing 50 mM imidazole HAc (pH 7.0), 1 mM EDTA, 30 mM NH4Ac, 5 mM MgCl2, 100 μM NADP+, and 2 μg/ml yeast 6-phosphogluconate dehydrogenase. The NAD+ cycling reagent, which is used for all other metabolite assays in Table 1, contains 100 mM Tris · HCl (pH 8.1), 2 mM β-mercaptoethanol, 2 mM oxaloacetate, 300 mM ethanol, 0.02% BSA, 15 μg/ml alcohol dehydrogenase, and 1.5 μg/ml heart malate dehydrogenase. This enzyme ratio gives ∼15,000-fold amplification overnight at room temperature and is what was used to measure citrate and pyruvate. This ratio was altered accordingly to achieve the number of cycles detailed in Table1. The indicator reaction for the NAD+ cycling step, orstep 5 in Table 1, involves adding 10 μl of the reaction to 1 ml of the indicator, malate reagent, containing 20 mM 2-amino-2-methylpropanol HCl (pH 9.9), 10 mM l-glutamate, 200 μM NAD+, 5 μg/ml malate dehydrogenase, and 2 μg/ml glutamic-oxaloacetic transaminase.
Data are expressed as means ± SE. Differences in metabolites between 2-cell and blastocyst stage embryos recovered either in vivo or in vitro were analyzed using Student's t-test. FBP and pyruvate measurements were the exception. Because of the large number of embryos required for the measurement of these two metabolites, the 2-cell embryos, in vivo blastocyst, and in vitro blastocyst assays were grouped together and the results analyzed using ANOVA with the Bonferroni/Dunn post hoc test. Differences in metabolites from embryos cultured in different concentrations of glucose were analyzed using ANOVA with the Bonferroni/Dunn post hoc analysis. Differences were considered significant at P < 0.05. StatView 4.5 (Abacus Concepts, Berkeley, CA) was used for statistical analyses.
Progression from 2-cell to blastocyst stage in vivo or in vitro results in an accumulation of TCA cycle metabolites and a decrease in glycolytic metabolites.
The majority of TCA cycle substrates demonstrated a significant increase in concentration as the embryo developed from a 2-cell- to a blastocyst-stage embryo (see Table 3 and Fig. 1). This increase, comparing 2-cell to blastocyst, occurred both in embryos cultured in vitro and in embryos obtained directly in vivo. Similarly, the FBP increased significantly, whereas glycerol 3-phosphate, a glycolytic intermediate, dropped during the same period in both conditions. This may reflect the embryo's adaptation to an anoxic peri-implantation existence, as well as the embryo's switch at the blastocyst stage to a predominantly glycolytic metabolism. The low level of FBP in early cleavage-stage embryos agrees with previous studies (1).
Blastocysts exposed to moderate glucose concentrations (5.6 mM) compared with normal glucose conditions (0.2 mM) experience a decrease in TCA cycle metabolites and an increase in glycolytic metabolites.
Except for citrate and malate, the TCA cycle metabolites in embryos cultured in 5.6 mM glucose were only slightly lower than those in embryos cultured in a normal glucose concentration of 0.2 mM. In addition, only glycerol 3-phosphate levels were significantly higher in the blastocysts in 5.6 mM glucose (See Table4 and Fig.2), and pyruvate and FBP were only slightly higher. Because the prior 2-cell to blastocyst studies suggested that a slowing of flux via the TCA cycle corresponds to an increase in TCA metabolites and a decrease in glycerol 6-phosphate levels, these results of 5.6 mM glucose compared with 0.2 mM glucose suggest the opposite. The mildly elevated glucose concentration appears to be causing increased flux through the TCA cycle with lowering of the levels, and perhaps a saturation of the glycolytic pathway, and with an increase in one of the byproducts used to make glycolipids, glycerol 3-phosphate.
Blastocysts exposed to high glucose concentrations (50 mM) compared with normal glucose concentrations (0.2 mM) experience an increase in TCA cycle metabolites, increased pyruvate, and a decrease in glycolytic metabolites.
Blastocysts cultured in 50 mM glucose compared with 0.2 mM glucose had a predominantly higher number of TCA cycle metabolites, significantly higher pyruvate and glycerol 6-phosphate levels, and significantly lower FBP levels (See Table 4 and Fig. 2). The discrepancy between citrate and pyruvate levels suggests strongly that a block to pyruvate oxidation via the TCA cycle exists in these embryos. Moreover, significantly decreased FBP levels suggest that glycolysis is also blocked, perhaps due to the low free glucose levels that are known to be present in these embryos exposed to high glucose conditions (21). The elevated pyruvate levels may occur because of increased pyruvate uptake, which occurs in response to glucose deprivation (7, 8).
Progression from the 2-cell to the blastocyst stage was associated with a significant increase in TCA cycle metabolites, a dramatic increase in FBP, and a decrease in glycerol 3-phosphate, as seen in this analysis. These results give a profile of a normal transition from a metabolism based on pyruvate oxidation via TCA cycle to a metabolism based on glucose metabolism via glycolysis. TCA cycle metabolites may accumulate because of blastocyst reliance on glycolytic metabolism, slowing of the TCA cycle, and accumulation of substrates. Less than 1% of glucose consumed at a blastocyst stage is oxidized via the TCA cycle (5), and this finding of an increase in TCA cycle metabolites corresponds with previous reports that have used techniques similar to those used in this study (1).
Comparing blastocysts cultured in 0.2 mM glucose with those in 5.6 mM glucose demonstrates the effect of a moderate increase in glucose on the metabolic products. The decrease in TCA cycle metabolites suggests increased flux through this pathway, with increased substrate utilization. Regulation of TCA cycle flux depends in part on increased availability of pyruvate and NAD+. Increasing glucose availability by incubating in 5.6 mM glucose would increase pyruvate production and decrease NADH levels as increased pyruvate is converted to lactate. Moreover, increased FBP levels would also stimulate TCA cycle flux and perhaps slow glycolysis. Previous studies have shown that glucose consumption rate is saturated at 0.29 mM but that a further 10% increase is obtained at 3 mM, which is similar to the 5.6 mM used in this study (5). All of our findings are consistent with a slight but measurable increase in glucose consumption. Importantly, none of these metabolic changes at this moderate glucose concentration induce apoptosis in the embryo, as shown in previous studies (3, 22, 24).
In contrast, maternal diabetes or in vitro hyperglycemia, at 50 mM glucose, does lead to apoptosis in the mouse embryo, requiring expression of the proapoptotic Bcl-2 family member BAX (13,22). Our previous studies show that this apoptotic event is triggered by decreased GLUT1 expression and glucose transport at the blastocyst stage (3, 21). The present study demonstrates that this drop in transport leads to a decrease in glycolysis, resulting in lower FBP levels in the individual blastocysts. Other studies have linked inhibition of glycolysis to initiation of an apoptotic cascade (18). Likewise, overexpression of glucose transporters prevents hypoxia-induced programmed cell death (16). Glucose metabolism and glucose uptake also exhibit a protective effect against apoptosis induced by growth factor withdrawal in different cell types (2, 12, 19). We postulate that the decrease in glycolysis in these blastocysts, as demonstrated by decreased FBP levels, and the embryos' attempts to compensate by increasing pyruvate uptake, lead to severe alteration in mitochondrial physiology that results in the triggering of the apoptotic cascade.
As shown in the growth factor withdrawal model (27), a depletion of glycolytic byproducts leads to decreased electron transport substrates and a resulting decrease in outer mitochondrial membrane potential (9) due to formation of an electron gradient and closure of the VDAC. VDAC is a large conductance channel that, when open, is the major pathway for metabolite transport across the outer mitochondrial membrane (28). Closure of this channel occurs under conditions of altered intracellular NADH and pyruvate resulting from glycolytic changes (17, 29) and has been shown to trigger apoptosis in growth factor withdrawal models due to hyperpolarization, loss of outer mitochondrial integrity, and cytochrome c release (25, 26).
Previous data show that VDAC closure leads to severe perturbation in mitochondrial physiology, leading to lack of mitochondrial availability of malate and ADP, slowing down of the TCA cycle and NAD+/NADH shuttles, and accumulation of TCA cycle metabolites as seen in this study (11). This lack of pyruvate oxidation would also support the elevated pyruvate levels. Although the citrate and α-ketoglutarate levels are lower in this report, unlike all the other TCA cycle components, these two metabolites can readily be converted to glutamate and glutamine and may serve as alternative energy sources under these stress conditions. Because of the anticipated problems with the mitochondrial matrix, the NADH shuttle would not be functioning and thus might result in the increased glycerol 3-phosphate levels. Most of the changes in TCA cycle metabolites seen here can be explained by perturbations in mitochondrial physiology due to a closure of VDAC.
In summary, hyperglycemia, by causing a decrease in glucose transport, results in a decrease in glycolysis and a decrease in FBP in the mouse blastocyst as measured in this study. This depletion of glycolytic substrates, in combination with the embryo's attempts to increase pyruvate uptake as reflected here, may be responsible for loss of integrity of the outer mitochondrial membrane as seen in other cell types. An inability of the mitochondria to successfully complete pyruvate oxidation would explain the increase in TCA cycle metabolites seen in these embryos. Recent studies provide strong evidence to support the mitochondria as the site of apoptosis initiation in response to growth factor withdrawal (25-27). These apoptotic effects are believed to be due to metabolic alterations linked to changes in glycolysis, as seen here with the blastocysts undergoing alterations in response to high-glucose conditions. Data from this study suggest that hyperglycemia-induced apoptosis in the mouse blastocyst may also involve metabolic alterations, leading to problems with outer mitochondrial membrane permeability.
This work was supported by grants to K. H. Moley from the National Institute of Child Health and Human Development (NICHD; RO1 HD-38061) and from NICHD and National Institute of Diabetes and Digestive and Kidney Diseases (RO1 HD/DK-40390) and a research grant from the American Diabetes Association.
Address for reprint requests and other correspondence: K. H. Moley, Dept. of OB/GYN, Washington Univ. School of Medicine, 4911 Barnes-Jewish Hospital Plaza, 2nd Floor Maternity, St. Louis, MO 63110 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
March 27, 2002;10.1152/ajpendo.00046.2002
- Copyright © 2002 the American Physiological Society