Live Imaging Reveals the Link Between Decreased Glucose Uptake in Ovarian Cumulus Cells and Impaired Oocyte Quality in Female Diabetic Mice

Abstract

Maternal diabetes has been demonstrated to adversely affect preimplantation embryo development and pregnancy outcomes. Emerging data suggest that these effects are associated with compromised oocyte quality. However, direct evidence of a pathway by which maternal diabetes exerts its effects on the oocyte is still lacking. Cumulus cells are metabolically coupled to oocytes, and bidirectional communication between them is essential for the development and functions of both compartments. The primary focus of this work was to evaluate the connection between glucose uptake in cumulus cells and oocyte quality in diabetic mice. This experiment has been difficult, because cumulus cells need to be separated from oocytes and labeled with isotope in the process of measuring glucose uptake. Here, we report a method for live imaging glucose transport in single cumulus-oocyte complexes using a fluorescent glucose analog (6-(N-(7-nitrobenz-2-oxa-1,3-diazol- 4-yl)amino)-6-deoxyglucose). By tracking the ATP content and spindle/chromosome status in individual oocytes surrounded by cumulus cells with differing glucose uptake activity, we reveal that compromised oocyte quality in diabetic mice is linked to decreased glucose uptake in cumulus cells.

Women with poorly controlled diabetes often suffer from reproductive problems, such as miscarriage, embryonic developmental abnormalities, and congenital malformations (1, 2). Maternal diabetes in rodents influences embryonic and fetal development in a very similar manner to that of humans (3). Emerging evidence has shown that oocytes from diabetic mice experience delayed maturation, abnormal cellular metabolism, mitochondrial dysfunction, and meiotic defects (4–6), which may be transmitted to the embryo and manifested later as the reproductive defects mentioned above. However, direct evidence of a pathway by which maternal diabetes exerts its effects on the oocyte is still lacking.

In the growing ovarian follicle, multiple layers of granulosa cells completely envelope the oocyte, and only the innermost layers (also termed cumulus cells) are in direct contact with the oocyte via gap junctions forming cumulus-oocyte complex (COC), through which the oocyte and cumulus cells are metabolically coupled (7). On one hand, oocytes carry out glycolysis poorly and require cumulus cells to metabolize glucose into products to support their maturation (8, 9). On the other hand, glucose is a major substrate for the synthesis of hyaluronic acid by cumulus cells through hexosamine biosynthesis pathway (10, 11). Substantial hyaluronan is produced during cumulus expansion, which has significant effects on oocyte quality (12, 13). Hence, glucose uptake in cumulus cells is one of the very first and critical steps for oocyte development. Particularly, we recently found a significant decrease in glucose uptake of cumulus cells from diabetic mice by radiolabeled deoxyglucose uptake assay (14). These findings prompted us to hypothesize that glucose deprivation in cumulus cells may contribute to the impaired oocyte quality in diabetic mice. To test this hypothesis, we first used 6-(N-(7-nitrobenz-2-oxa-1,3-diazol- 4-yl)amino)-6-deoxyglucose (6-NBDG), a nonradiolabeled, nonmetabolizable fluorescent glucose analog (11, 15), to monitor glucose uptake in cumulus cells of single live COC and then tracked ATP content and spindle/chromosome status in the surrounded oocyte. Our data demonstrated, for the first time, that disturbed metabolism and meiotic defects in diabetic oocytes are linked to impaired glucose uptake in the adjacent cumulus cells.

Materials and Methods

All mouse studies were approved by the Animal Studies Committee at Washington University School of Medicine and conform to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health.

Generation of diabetic mice

To generate a diabetic model, female B6SJLF1 mice (20–24 d) received a single injection of streptozotocin at a dose of 190 mg/kg. Four days after injection, a tail-blood sample was measured for glucose concentrations. If glucose levels were greater than 300 mg/dl, the animal was selected for use as a diabetic model. In this study, totally, 85 mice were injected with streptozotocin, and 51 diabetic mice were obtained. Age-matched control mice were randomly selected.

COC retrieval

Control and diabetic mice were superovulated with 10 IU of pregnant mare's serum gonadotropin (Sigma, St. Louis, MO) by ip injection; 48 h later, the ovaries were removed and placed in M2 medium (Sigma). COC were obtained by rupturing antral ovarian follicles.

Visualization of glucose uptake

A fluorescent glucose analog, 6-NBDG (Molecular Probes, Eugene, OR), was used to track glucose uptake in cumulus cells of single COC (see figure 2 below). Briefly, isolated COC from diabetic mice were incubated in M2 medium containing 100 μm 6-NBDG for 20 min at 37 C. After three washes in 30-μl microdrops of M2 medium (2 min per wash), individual COC were loaded on a glass slide (five COC per slide) with 20-μl microdrops of M2 medium each and imaged at 488 nm using a fluorescence microscope (Zeiss Axioskop, Gottingen, Germany). NIH ImageJ software was used to quantify average NBDG fluorescence of cumulus cells as a measure of glucose uptake. Fluorescence intensity was randomly measured in two to three regions of interest strictly limited to the cumulus cells from each COC images.

Fig. 2.

Cartoon summarizing the experimental protocol to investigate the relationship between glucose uptake in cumulus cells and the enclosed oocyte quality. COC collected from control and diabetic mice were incubated with 6-NBDG. Individual live COC were imaged for fluorescence quantification in cumulus cells as a measure of glucose uptake. A, Single enclosed internal GV oocytes were measured for ATP levels. B, After in vitro maturation of single COC, the enclosed internal MII oocyte was tracked for spindle and chromosome status by immunofluorescence microscopy.

In preliminary experiments, NBDG uptake by cumulus cells was standardized and found to be linear between 0 and 30 min of incubation and from 50 to 200 μm concentration. Incubation in 100 μm NBDG for 20 min was chosen because the fluorescence signals were adequate and unsaturated, and significant difference in fluorescence intensity between control and diabetic cumulus cells was observed.

Measurement of ATP content in single oocytes

Single oocytes freed of cumulus cells were frozen on a glass slide. After freeze drying overnight under vacuum at −35 C, each single oocyte was extracted in nanoliter volume under oil. ATP assay was designed to link reactions ending with nicotinamide adenine dinucleotide phosphate/reduced nicotinamide adenine dinucleotide phosphate, which then were enzymatically amplified in a cycling reaction, and a byproduct of the amplification step was measured in a fluorometric assay. ATP levels are expressed as millimoles per kilogram wet weight based on the wet weight of 160 pg per oocyte. Absolute ATP concentrations can be calculated in picomoles by multiplying by 0.16. The detailed assay protocol has been described in Chi et al. (16).

In vitro oocyte maturation

After 6-NBDG incubation and recording fluorescence images, intact COC were washed out and then cultured individually in 20-μl microdrops of M2 medium supplemented with 10% fetal bovine serum (HyClone Laboratories, Inc., Logan, UT) and 10 IU/ml pregnant mare's serum gonadotropin for 14–16 h (17). At the end of maturation, cumulus cells were removed by incubating in M2 medium with 0.1% hyaluronidase.

Staining of spindle and chromosome in oocytes

Single denuded metaphase II (MII) oocytes were fixed with 4% paraformaldehyde for 30 min and then permeabilized with 0.5% Triton X-100 for 20 min. Followed by blocking in 1% BSA-supplemented PBS for 1 h, samples were incubated overnight at 4 C with fluorescein isothiocyanate-conjugated β-tubulin antibody (1:100; Sigma) to visualize the spindle. Chromosomes were evaluated by costaining with 4′,6-diamidino-2-phenylindole for 5 min. Oocytes examined under a fluorescence microscope.

Statistical analyses

Data are presented as mean ± sd. Group differences were evaluated using unpaired Student's t test. Correlations between two parameters were obtained via linear regression analysis (GraphPad Prism 5, San Diego, CA).

Results

Decreased 6-NBDG uptake in cumulus cells of diabetic mice

To test whether 6-NBDG transport can be used to measure glucose uptake activity in cumulus cells as other radiolabeled tracers do, live COC isolated from control and diabetic mice were incubated with 6-NBDG and examined by fluorescence microscopy. Generally, cumulus cells of most control COC (∼90%, n = 115) were uniformly and brightly incorporated the 6-NBDG substrate (Fig. 1A). In contrast, cumulus cells within approximately 60% of diabetic COC (n = 108) display seemingly normal NBDG uptake. The remaining approximately 40% of diabetic COC shows comparatively dim fluorescence, and sometimes, only a few cumulus cells per COC accumulated 6-NBDG (Fig. 1B, arrowheads). Quantification of fluorescent intensity demonstrated a overall 34% decrease in 6-NBDG uptake by diabetic cumulus cells in comparison with controls (average intensity, 35.2 ± 13.6 vs. 53.3 ± 8.1 control) (Fig. 1C), which is consistent with our previous results derived from ³H-2-deoxyglucose uptake assay (14). On one hand, these results suggest that 6-NBDG transport can be used as a visible indicator for glucose uptake in live COC. On the other hand, such a heterogeneous population with differing glucose uptake activity in cumulus cells makes diabetic COC a unique in vivo model, in this study, to determine the relationship between glucose uptake in cumulus cells and developmental competence of the enclosed oocyte.

Fig. 1.

Decreased glucose uptake in cumulus cells of diabetic mice compared with controls. 6-NBDG, a fluorescent glucose analog, was used to track glucose uptake in cumulus cells of single live COC. A, Representative image showing 6-NBDG uptake of cumulus cells within about 90% of control COC. B, Representative image showing 6-NBDG uptake of cumulus cells within about 40% of diabetic COC. A diabetic COC with dim fluorescence and only few 6-NBDG-labeled cumulus cells is indicated by arrowheads. C, Quantitative analysis of fluorescence intensity reveals the significantly decreased glucose uptake in diabetic cumulus cells compared with controls (control, n = 115; diabetic, n = 108). Error bars indicate ± sd. P < 0.05 was considered to be statistically significant. Scale bars, 40 μm.

Decreased glucose uptake in cumulus cells is correlated with low ATP content in oocytes of diabetic mice

It is well known that cumulus cells support oocyte development through provision of essential nutrients, metabolic precursors, and information molecules (18). Previous data from our lab have shown the lower ATP levels in germinal vesicle (GV) stage oocytes from diabetic mice than controls (5, 6, 19). Variation in ATP content have been suggested to significantly affect oocyte quality, embryonic development, and even implantation process (20–22). Hence, we decided to determine whether the decreased glucose uptake in cumulus cells is correlated with this low ATP levels in diabetic oocytes that we observed. To do this, the experimental protocol outlined in Fig. 2A was conducted. After incubation with 6-NBDG, individual live COC were imaged for fluorescence quantification in cumulus cells as a measure of glucose uptake, and then, the single enclosed internal oocytes were measured for ATP content. As shown in Fig. 3, compared with control group, increased proportion of COC from diabetic mice showed the decreased NBDG uptake in cumulus cells and lowered ATP levels in the oocyte. Linear regression analysis demonstrates a strong and significant correlation between them (R = 0.55, P < 0.0001) (Fig. 3B), indicating that glucose deprivation in cumulus cells is responsible, at least in part, for the reduced ATP content in oocytes of diabetic mice.

Fig. 3.

Correlation between glucose uptake in cumulus cells and ATP content in oocytes. COC collected from control (A) and diabetic (B) mice were incubated with 6-NBDG. Glucose uptake in cumulus cells within single live COC was measured by quantifying NBDG fluorescence intensity. The enclosed GV oocytes were measured for ATP levels individually by microanalytic assay as described in Materials and Methods. ATP levels are expressed as millimoles per kilogram wet weight (mmKw) based on the wet weight of 160 pg per oocyte. Linear regression analysis was used to test correlation between two parameters.

A link between impaired glucose uptake in cumulus cells and the meiotic defects in oocytes of diabetic mice

The spindle assembly and chromosome movement are crucial events during oocyte maturation. We previously observed the disorganized spindle and chromosomes in ovulated MII oocytes from diabetic mice (6). Here, we checked whether this impaired glucose uptake in diabetic cumulus cells contributes to the meiotic defects in oocytes. Experiments were performed according to the procedure outlined in Fig. 2B. In brief, COC collected from control and diabetic mice were incubated with 6-NBDG. Glucose uptake in cumulus cells within individual live COC was measured by quantifying NBDG fluorescence. After in vitro maturation of single COC, the enclosed internal oocyte was tracked for spindle/chromosome status by immunofluorescence microscopy. As noted in Fig. 4B, dot plot analysis of NBDG fluorescence intensity in cumulus cells surrounding different oocyte populations clearly revealed a link between meiotic defects in the oocyte and glucose uptake in the adjacent cumulus cells. The majority of diabetic oocytes surrounded by cumulus cells with relatively high glucose uptake (average intensity, 43.3 ± 13.0 pixels; n = 60) presented a typical barrel-shape spindle and well-aligned chromosomes on the metaphase plate (Fig. 4C). In contrast, a significant proportion of diabetic oocytes enveloped by cumulus cells with dramatically reduced glucose uptake (average intensity, 24.8 ± 11.2 pixels; n = 23) (Fig. 4D, red arrowheads) consistently displayed spindle disorganization and chromosome misalignment (Fig. 4D, white arrowheads). To control for possible nonspecific and toxic effects of NBDG on oocyte meiosis, COC from age-matched normal mice were also evaluated, and no deleterious effects were observed (Fig. 4A). Collectively, these results suggest that impaired glucose uptake in diabetic cumulus cells and meiotic defects in the enclosed oocyte are related.

Fig. 4.

Impaired glucose uptake in cumulus cells induces the meiotic defects in oocytes of diabetic mice. A, Dot plot analysis of 6-NBDG fluorescence intensity in cumulus cells surrounding the control MII oocytes with normal or abnormal spindle/chromosome morphology (n = 74 total). B, Dot plot analysis of 6-NBDG fluorescence intensity in cumulus cells surrounding the diabetic MII oocytes with normal or abnormal spindle/chromosome morphology (n = 83 total). Blue error bars indicate mean ± sd. C, Fluorescence images showing that the oocyte surrounded by cumulus cells brightly labeled with 6-NBDG presents a typical barrel-shape spindle and well-aligned chromosomes. D, Fluorescence images showing that the oocyte surrounded by cumulus cells with dramatically reduced 6-NBDG uptake (red arrowheads) presents spindle disorganization and chromosome misalignment (white arrowheads). Scale bars, 30 μm.

Discussion

Bidirectional communication between oocytes and the companion cumulus cells is essential for the development and functions of both compartments. For example, cumulus cells provide oocyte with metabolites and metabolic regulators, and oocyte in turn control cumulus cell metabolic activities by secretion of paracrine factors (23). Hence, examination of COC as a whole is necessary for understanding the nutritional and metabolic factors conferring oocyte developmental competence (24, 25). In previous studies, we showed the reduced ATP content and increased spindle/chromosome defects in denuded oocytes from diabetic mice (5, 6). Meanwhile, we also found that facilitative glucose transporter 1 protein expression is down-regulated, and glucose uptake parallels this decrease in diabetic cumulus cells (14). However, we were unable to track the quality of single oocytes surrounded by cumulus cells with differing glucose uptake activity to determine whether they are correlated. This is because, in the process of glucose uptake measurement, cumulus cells need to be removed from COC, separated from oocytes, pooled together, and then radiolabeled for scintillation counting (14, 19). The current study presents a method of live imaging glucose uptake in cumulus cells within single COC without using isotopes. This new approach not only keeps the integrity of COC when quantifying glucose uptake in cumulus cells but also allows for further in vitro culture experiments even after glucose uptake measurement. Using this method, we were able to determine that diabetic oocytes surrounded by cumulus cells with decreased glucose uptake frequently show low ATP levels and are prone to develop meiotic defects during maturation (Figs. 3 and 4). These data suggested that glucose deprivation in ovarian cumulus cells is related to compromised oocyte competence in diabetic mice.

How does decreased glucose uptake in cumulus cells impair oocyte quality in diabetic mice? Oocytes are deficient in their ability to use glucose as an energy substrate and require cumulus cell-provided glycolytic byproducts for their own development (9). Pyruvate and ATP, as related products of glycolysis by cumulus cells, can be transferred to oocytes via gap junctions (23, 26). Given these findings, lowered ATP levels in diabetic oocytes perhaps are caused, at least in part, by insufficient provision of pyruvate and ATP from the surrounding cumulus cells because of the decreased glucose uptake. Additionally, emerging data imply that glucose metabolism in cumulus cells may be associated with spindle defects in the oocyte. After the resumption of meiosis, oocytes become transcriptionally inactive. At this stage, ATP is only needed to sustain basal metabolism and spindle formation in oocytes (27). Loss of pyruvate dehydrogenase and ATP has been suggested to be able to disrupt spindle/chromosome morphology in the oocyte (28, 29). Maternal aging-associated spindle abnormalities in oocytes have been widely reported (30, 31); and notably, glucose metabolism in mouse cumulus cells can prevent oocyte aging by maintaining both energy supply and intracellular redox potential (32). Here, using diabetic COC as a model, we present the direct experimental evidence showing that oocytes enclosed by cumulus cells with decreased glucose uptake are prone to develop spindle defects and chromosome misalignment (Fig. 4). Any error in the spindle assembly and chromosome alignment could result in oocyte chromosomal imbalance, contributing to embryonic aneuploidy, spontaneous abortion, and genetic diseases (33). It is therefore possible that glucose limitation related with facilitative glucose transporter 1 deficiency in cumulus cells induces the meiotic defects in diabetic oocytes, producing an aneuploid germ cell and subsequent embryo and then contributing to the reproductive problems experienced by diabetic females. Additional studies tracking the relationship between glucose uptake in cumulus cells and reproductive outcome in diabetic mice need to be done to test this hypothesis. On the other hand, in vitro maturation of diabetic COC in M2 medium with pyruvate did not prevent the meiotic defects occurring in oocyte (Figs. 2B and 4). This phenomenon indicates that: 1) impaired glucose uptake in cumulus cells perhaps brings long-term or irreversible adverse effects to oocyte meiosis; or 2) instead of being used as an energy source, other pathways of glucose utilization may also be involved in this process. A significant proportion of glucose consumed by COC is directed toward extracellular matrix components, particularly hyaluronic acid synthesis, via the hexosamine biosynthesis pathway; and this pathway has been suggested to be critical for developmental potential of oocytes and embryos (10–13). In addition, cumulus cells can use glucose by pentose phosphate pathway, generating reduced nicotinamide adenine dinucleotide phosphate for the reduction of antioxidant glutathione and phosphoribosylpyrophosphate for de novo purine synthesis in oocytes (11, 34). Blocking the pentose phosphate pathway in cumulus cells has been demonstrated to promote oocyte aging (32).

In summary, this work suggests that decreased glucose uptake in cumulus cells contributes to impaired oocyte quality in diabetic mice. Moreover, by visualizing glucose transport in single live COC, it is the first study to link glucose availability in cumulus cells to the ability of the oocyte to properly undergo meiosis.