| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Biology Department, Marquette University, Milwaukee, Wisconsin 53233
b Department of Surgery, Division of Transplant Surgery, Medical College of Wisconsin, Milwaukee, Wisconsin 53226
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cumulus cells, gamete biology, gametogenesis, meiosis, oocyte development
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Numerous studies have determined that the diabetic state is detrimental to both pre- and postimplantation embryo development in rodents. Postimplantation embryos cultured at high glucose levels in vitro exhibit growth retardation and physical abnormalities, the most common being failure of anterior neuropore closure [29]. Congenital malformations are approximately 3–4 times more frequent in infants from diabetic mothers than in nondiabetic women [2–4]. In addition to specific malformations, the overall size of 21-day-old fetuses from diabetic rats is smaller than fetuses of the same age isolated from control rats [30]. In fact, many of the changes in postimplantation development may be the result of developmental delay rather than actual teratogenesis [29].
In both chemically induced and spontaneous diabetic models, significant delays in preimplantation embryo development have been observed [24, 31–33] as well as a high incidence of degenerate and fragmented embryos [24, 33–36]. Blastocysts retrieved from diabetic animals typically contain fewer cells than those from nondiabetic control animals [31, 32, 35]. The damage to the embryo likely occurs early since almost 50% of two-cell embryos isolated from subdiabetic rats were unable to develop to the 8-cell stage, even in a nondiabetic tract [37]. High glucose concentrations in embryo culture medium lead to impaired embryo development [38, 39], and this effect may be due to downstream metabolic intermediates since developmental retardation in vitro is also observed with high concentrations of fructose, sorbitol, or ketone bodies [1, 40–43]. High glucose alone or combined with high ketone body levels inhibited further development of blastocysts in a dose-dependent manner [44, 45]. Despite the fact that development is compromised, blastocysts from diabetic rats and mice exhibit normal metabolism/uptake of glucose and pyruvate [46, 47]. Thus, the anomaly may not be a permanent metabolic aberration but rather an accumulated metabolic consequence of persistent hyperglycemia.
It is likely that exposure of the oocyte to diabetic conditions during folliculogenesis and meiotic maturation negatively affects its meiotic maturation and subsequent developmental potential. Indeed, Diamond et al. [24] have reported that meiotic resumption is attenuated in superovulated diabetic mice. In recent studies, we have demonstrated how alterations in energy substrate supplementation can profoundly influence meiotic regulation in vitro [48–50]. Varying the relative amounts of glucose and pyruvate can either induce or impede germinal vesicle breakdown (GVB) in cumulus cell-enclosed oocytes in the presence of meiotic inhibitors such as hypoxanthine or dibutyryl cAMP (dbcAMP); glucose provides an inhibitory influence when high levels of pyruvate are present yet is required for hormone-induced maturation [49–51]. Evidence suggests that the glycolytic pathway mediates the inhibitory action of glucose through the generation of ATP [50, 52], and the positive action of glucose requires the participation of the pentose phosphate pathway [53–55]. Both the meiosis-inducing and -suppressing effects of glucose on oocyte maturation appear to be mediated by the gap junctional communication pathway that metabolically couples the oocyte with the somatic compartment of the follicle [52, 56, 57].
Experiments with spontaneously maturing cumulus cell-enclosed mouse oocytes have indicated that glucose is required for optimal completion of nuclear maturation to metaphase II, but supraphysiological levels (27.8 mM) may be detrimental [58]. A similar need for glucose during bovine oocyte maturation has been demonstrated, but exposure of maturing oocytes to high levels of glucose (
20 mM) compromises subsequent developmental capacity [59, 60]. The oocyte can directly utilize only negligible amounts of glucose [55], but the glycolytic and pentose phosphate pathways are much more active in cumulus cells [61–64]. Hence, the somatic compartment is a critical mediator in the interaction of glucose in meiotic regulation.
Since glucose can play different roles in the meiotic maturation of isolated oocytes, it was important to assess how a diabetic environment would influence oocyte maturation. In this study, we compared meiotic regulation in cumulus cell-enclosed oocytes from diabetic and nondiabetic immature mice under a variety of conditions. We have also examined effects on ovulation and preimplantation embryo development. The results of this study demonstrate that meiotic resumption is adversely affected in oocytes from diabetic animals, which may impact negatively on subsequent developmental capacity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To establish a protocol that ensured diabetic conditions, a preliminary study of streptozotocin (STZ)-induced diabetes was carried out. Except where otherwise noted, C57BL/6J x SJL/J F1 female mice, 20–23 days old, were used for all experiments. The mice were weighed and an initial tail blood sample was taken prior to injection. A commercial glucometer and test strips were used to assess glucose levels in tail-vein blood samples. These blood samples indicated control blood glucose concentrations of 80–160 mg/dl. The mice were then separated into 2 groups. Mice weighing 11.0 g were administered a single i.p. injection of streptozotocin (STZ; Sigma Chemical Corporation, St. Louis, MO) at a dose of 200 mg/kg. Control mice received an equal volume of the sodium citrate vehicle buffer (0.01 mM, pH 4.5). Tail blood samples were taken daily for 5 days postinjection and the glucose levels determined. Injection of the vehicle did not affect the glucose levels, as all tail blood samples revealed a glucose level no higher than 160 mg/dl over the course of the 5 days. Mice receiving STZ, however, exhibited a significant increase in blood glucose levels, usually within 1 day of injection, and levels exceeding 300 mg/dl were observed after 3 days. In this initial exercise, approximately 33% of the mice failed to respond to treatment with sizable increases in blood glucose, but in subsequent experiments, this number averaged less than 10%. The diabetic state was established prior to hormonal priming to ensure follicle development under hyperglycemic conditions.
The 2 general protocol schemes for most experiments are presented in Figure 1. Mice exhibiting a glucose level 300 mg/dl on Day 3 were considered to be diabetic and were used in subsequent experiments. Mice were primed with 5 IU equine chorionic gonadotropin (eCG; Sigma) 3 days after the initial injection of STZ or vehicle buffer (Fig. 1). For some experiments, diabetic mice were administered daily s.c. injections of 0.5 IU/kg slow-releasing insulin (Humulin Ultralente; Eli Lilly and Co., Indianapolis, IN) for 3 successive days beginning 2 days after STZ administration, with the final injection 1 day post-eCG. Glucose levels were monitored 4 h after the last insulin injection to ensure that blood glucose returned to control levels.
|
Culture Conditions for In Vitro Maturation
For in vitro oocyte maturation, mice were killed 2 days after eCG priming and ovaries were removed and placed into culture medium. Antral follicles were punctured with sterile needles to release cumulus cell-enclosed oocytes, which were then washed through 2–3 changes of medium and allocated to the appropriate treatment groups. For 1 of the experiments, denuded oocytes were obtained by repeated pipetting with a Pasteur pipet. Oocytes were transferred to plastic tubes containing 1 ml of culture medium, gassed with a humidified mixture of 5% O2, 5% CO2, and 90% N2, and placed in a water bath at 37°C for varying periods of time. The culture medium used for oocyte maturation experiments was Eagle minimal essential medium (MEM) containing Earle salts, antibiotics (50 µg/ml streptomycin sulfate, 75 µg/ml penicillin G), and 3 mg/ml lyophilized crystallized bovine serum albumin (ICN ImmunoBiologicals, Lisle, IL). The medium was supplemented with pyruvate (0.23 mM) plus varying concentrations of glucose or mannose, meiotic inhibitors, and/or FSH as described for each experiment. For cumulus cell expansion, medium contained 5% fetal bovine serum (FBS).
hCG-Induced Oocyte Maturation and Ovulation
For in vivo oocyte maturation, mice received an i.p. injection of 5 IU hCG 2 days after eCG priming (on Day 5 of protocol; see Fig. 1). Mice were then killed at various times from 1.5 to 10 h after hCG, and cell-enclosed oocytes (CEO) were isolated as described above and assessed for maturation. For oocytes retrieved 3 h post-hCG, 0.3 mg/ml hyaluronidase was included in the isolation medium to prevent adherence of the complexes to ovarian components and to aid in removal of the cumulus cells. Twelve and 14 h after hCG injection, ovulated oocyte-cumulus cell masses were retrieved from the oviducts in medium containing 0.3 mg/ml hyaluronidase to remove the cumulus cells. The number of ovulated ova was determined and ova were then processed for chromatin staining to determine meiotic status. Follicular oocytes obtained 10 h post-hCG were also fixed and stained to determine meiotic status.
DNA Staining of Oocytes
Ovulated ova were fixed for 30–45 min in 3% formaldehyde in phosphate-buffered saline (PBS), transferred in a small volume to a microscope slide and dried on a warming tray. A 10-µl drop of glycerol:PBS (1:1) containing 1 µg/ml Hoechst 33342 (Polysciences, Warrington, PA) was applied over the ova and a cover slip was then added and sealed with nail polish. Oocytes were examined under a Leitz Orthoflux II fluorescence microscope for meiotic status.
Nonobese Diabetic Mice
Nonobese diabetic (NOD) mice (Taconic Labs, Germantown, NY) develop diabetes spontaneously at approximately 6–8 wk of age. Mice exhibiting blood glucose levels 200 mg/dl were considered to be diabetic. We received these mice when they were 3–4 mo old and after they had been used in another study in which the acute antinociceptive response and G protein activation by opioid receptor agonists was evaluated [65]. The mice had previously received an intracerebroventricular injection of opioid agonist. Two days after this injection, NOD diabetic mice and nondiabetic controls were primed with 5 IU eCG and immature cumulus cell-enclosed oocytes were obtained 2 days later, as described above.
Cumulus Expansion
Oocyte-cumulus cell complexes were isolated from diabetic and control mice and cultured 17–18 h in 3 ml MEM/FBS in plastic Petri dishes. Dishes were placed in a modular incubator (Billups-Rothenberg, Del Mar, CA) gassed with the humidified 5:5:90 gas mixture and placed in a water-jacketed incubator at 37°C. At the end of the culture, complexes were assessed for expansion and a cumulus expansion index was calculated using a subjective numerical scale (0 representing no expansion and 4 representing complete expansion) as previously described [56].
Preimplantation Embryo Culture
To obtain pronuclear-stage embryos, eCG-treated control and diabetic mice received an injection of hCG and were mated. Twenty-five hours later, pronuclear-stage embryos were isolated from the oviduct and cultured for 5 days in 3 ml KSOM/AA in Petri dishes. The dishes were placed in modular incubators, gassed with humidified 5:5:90 gas mixture, and placed in a water-jacketed incubator at 37°C. Cultures were examined 1 day later, and those embryos cleaving successfully to the two-cell stage were transferred to a new dish with fresh medium of the same type and cultured for an additional 4 days. At the end of the culture period, the percentage of embryos that reached the blastocyst stage was determined.
Chemicals
All medium components, hypoxanthine, isobutylmethylxanthine (IBMX), dibutyryl cyclic AMP, hyaluronidase, and streptozotocin were purchased from Sigma. Biological-grade ovine FSH-17 (o-FSH; 20 U/mg) was generously provided by the National Hormone and Pituitary Program of NIDDK (Bethesda, MD). A stock solution of FSH, prepared in phosphate-buffered saline containing 3 mg/ml BSA and stored at -20°C, was diluted to a final concentration of 0.1 µg/ml.
Statistical Analysis
Oocyte in vitro maturation experiments were conducted at least 3 times with at least 30 oocytes per group per experiment. The experiments with NOD mice were carried out with 17–35 oocytes per group per experiment. At least 4 mice were used for each data point in experiments involving ovulation or oocyte maturation in vivo. For in vitro germinal vesicle breakdown kinetics, the range in number of oocytes retrieved per mouse was 21–65. For metaphase II assessment of oocytes matured in vivo, oocyte numbers varied widely (0–75) due to the effect of treatment on ovulation. For metaphase II assessment of oocytes matured in vitro, the range was 18–40 per treatment group. Preimplantation embryo cultures were carried out at least 7 times, with at least 17 embryos per group per experiment. Data are reported as the mean ± SEM. Frequencies were subjected to arcsin transformation and groups of 3 or more were analyzed statistically by ANOVA followed by Duncan multiple range test. Paired comparisons were made by Student t-test. A P value less than 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cumulus cell-enclosed oocytes (CEO) from diabetic and control mice were cultured in control, inhibitor-free medium for 45, 75, or 105 min in varying concentrations of glucose (0, 0.055, 5.5, and 22 mM) to assess the effects of glucose levels on spontaneous maturation in vitro. The CEO from each mouse were divided into 3 equal groups (8–26 oocytes per group) so the oocytes for each replicate of a particular treatment group originated from the same mouse. At all glucose concentrations, the kinetics of maturation were accelerated in CEO from diabetic mice when compared with those from control mice (Fig. 2). This difference was most apparent at the earliest time point (45 min), with differences in GVB of 12%–24%. In addition, the maturation kinetics were generally slower at the two higher glucose concentrations for both control and diabetic groups (Fig. 2).
|
FSH-Induced Transient Arrest of Spontaneous Maturation In Vitro
To determine how diabetes and different glucose concentrations affect FSH-induced transient arrest during short-term cultures, CEO from diabetic and control mice were exposed to 0.1 µg/ml FSH plus increasing concentrations of glucose for 1.5, 2.25, or 3 h before assessment of maturation. As in the previous experiment, oocytes from each mouse were divided into 3 equal groups. When medium contained 0, 0.055, or 5.5 mM glucose, FSH suppressed maturation more effectively in control CEO than in CEO from diabetic mice (Fig. 3, A–C). The effect was temporary, as differences were no longer observed by 3 h of culture. There was no significant difference in maturation between control and diabetic CEO at any of the time points when culture medium contained 22 mM glucose (Fig. 3D). Maturation kinetics were not influenced by the different glucose concentrations.
|
Hormone-Induced Maturation In Vitro
In this series of experiments, the ability of meiotically arrested CEO to undergo FSH-induced maturation was assessed. To test the influence of the diabetic state and glucose on meiotic induction in vitro, CEO from control and diabetic mice were maintained in meiotic arrest for 17–18 h with either hypoxanthine or dibutyryl cAMP (dbcAMP), and FSH was included to stimulate maturation. Glucose was added in increasing concentrations (0.055, 0.55, 5.5, 11, and 22 mM).
Hypoxanthine-arrested CEO When CEO from control and diabetic mice were maintained in meiotic arrest with 4 mM hypoxanthine, glucose dose-dependently increased the maturation percentage, with CEO from control mice consistently exhibiting higher GVB frequencies than CEO from diabetic mice (Fig. 4). The addition of FSH produced effective stimulation of maturation in CEO from control mice, with an increase of 39% GVB at 0.055 mM glucose and an increase of 27% at 22 mM. FSH was less effective in CEO from diabetic mice at lower glucose concentrations (an increase of 23% at 0.055 mM), but comparable stimulation was achieved at 22 mM (an increase of 38%).
|
Dibutyryl cAMP-arrested CEO CEO isolated from both control and diabetic mice and cultured without FSH in the presence of 300 µM dbcAMP (Fig. 5A) exhibited a limited dose-dependent increase in maturation in response to glucose. This effect was not observed at 500 µM dbcAMP (Fig. 5B). The suppressed response to FSH in CEO from diabetic mice was more pronounced in dbcAMP-supplemented medium than in hypoxanthine-supplemented medium. In medium containing 300 µM dbcAMP, a modest increase in GVB (30%–32%) was observed in CEO from control mice upon FSH stimulation at the 2 lower glucose concentrations, but this meiotic response increased significantly with higher glucose levels such that, at 22 mM, 88% of FSH-treated oocytes resumed maturation compared with 31% in the absence of FSH. In CEO from diabetic mice, meiotic maturation in response to FSH was suppressed at all glucose concentrations but most notably at 11 and 22 mM (an increase of 18%–22%; Fig. 5A). The compromised meiotic induction in CEO from diabetic mice was more pronounced when the dbcAMP concentration was increased to 500 µM (Fig. 5B). Under these conditions, FSH triggered a 39%–43% increase in GVB in control CEO, but no significant stimulation of maturation occurred in oocytes from diabetic mice (increases of only 9%–20%).
|
Insulin-Reversal of STZ Effects on Meiotic Induction
To determine if the STZ effects are due to the diabetic condition and not some nonspecific effect of this drug, STZ-induced diabetic mice were treated with insulin on Days 2, 3, and 4 to normalize blood glucose levels (see Fig. 1). Measurement of blood glucose 4 h after administering insulin on Day 2 showed a decrease in glucose levels from 398.4 ± 8.6 to 94 ± 4.4 mg/dl (n = 27). Oocytes were cultured 17–18 h in medium containing 300 µM dbcAMP in the absence or presence of FSH at glucose concentrations of 5.5 and 22 mM. Consistent with earlier results, the diabetic condition compromised the ability of FSH to stimulate meiotic resumption at either glucose concentration; however, insulin treatment restored meiotic induction at 5.5 mM glucose and nearly so at 22 mM (Fig. 6).
|
Meiotic Induction In Vitro in CEO from NOD Mice
Nonobese diabetic (NOD) mice were used to determine if suppression of meiotic maturation in vitro would also occur in oocytes from genetically diabetic mice. Nonobese diabetic mice develop diabetes spontaneously without the injection of a toxin [66]. The CEO isolated from NOD control and diabetic mice were cultured 17–18 h in medium containing 300 µM dbcAMP in the absence or presence of FSH at glucose concentrations of 5.5 and 22 mM (Fig. 7). In the absence of FSH, control CEO demonstrated a maturation frequency of 11%–19% and, upon addition of FSH, maturation was increased by 44%–46%. The CEO from NOD mice, on the other hand, exhibited a maturation frequency of 7%–11% in the absence of FSH. In medium containing 22 mM glucose, 30% of FSH-stimulated CEO underwent GVB, which was more than a 50% reduction when compared with controls. At the lower glucose concentration, a similar inhibitory trend was evident but was not significantly different. It should be noted that, in 2 out of 3 experiments performed in 5.5 mM glucose, only 18% and 28% of CEO from diabetic mice underwent maturation in response to FSH, compared with 40% and 56% in controls, respectively.
|
Effect of Diabetes on Cumulus Expansion
To assess the effect of diabetes on the ability of oocyte-cumulus cell complexes to undergo cumulus expansion in vitro, complexes from both control and diabetic animals were cultured 17–18 h in serum-supplemented medium alone or in medium containing FSH or FSH plus 50 µM IBMX. The IBMX, a nonspecific phosphodiesterase inhibitor, was used to promote the cAMP response to FSH and further stimulate cumulus expansion. The results of this experiment are shown in Table 1. Complexes from control animals exhibited negligible expansion in the absence of FSH, with a cumulus expansion index (CEI) of 0.10. Expansion was evident in FSH-treated complexes, which exhibited a CEI of 1.68, and this was further increased to 3.28 by IBMX. None of the 3 treatment groups from diabetic mice had CEIs that were different from corresponding controls.
|
Osmolarity
To exclude the possibility that the higher osmolarity resulting from hyperglycemic conditions contributes to the altered meiotic regulation, the effects of glucose on osmolarity and meiotic induction were compared with another sugar, mannose. Figure 8 shows that mannose, like glucose, produced a dose-dependent increase in osmolarity. There was no difference between mannose and glucose at any of the concentrations, but there was a 20 mOsmol/kg rise in osmolarity as the concentration of either sugar increased from 5.5 to 22 mM. When the effects of mannose were tested on meiotic induction in dbcAMP-arrested CEO, higher mannose concentrations failed to increase the maturation frequency in control or diabetic oocytes in either the absence or presence of FSH (Fig. 8). However, a decrease in FSH-induced GVB in CEO from diabetic mice was still observed.
|
Diabetic Effects on Post-hCG Events In Vivo
GVB kinetics Meiosis resumption in vivo was initiated after 1.5 h post-hCG. In control mice, 40% of the oocytes had undergone GVB by 2 h and this value plateaued by 2.5 h at approximately 80% (Fig. 9A). The maturation kinetics from diabetic mice, however, were significantly slower, as only 14% of the oocytes had resumed maturation by 2 h, and the frequency of GVB at 5 h was still lower than that observed at 2.5 h in controls (82% of oocytes from control mice and 45% of oocytes from diabetic mice). At 10 h post-hCG, the maturation frequency in diabetic mice was no longer significantly different from controls (78.9% in oocytes from control mice and 67.3% in oocytes from diabetic mice), thus suggesting that the diabetic condition may delay oocyte maturation rather than prevent it. When diabetic mice were treated with Humulin Ultralente insulin for 3 days prior to oocyte removal to normalize glucose levels, normal kinetics of meiotic resumption were restored (Fig. 9A).
|
Ovulation and meiotic progression to metaphase II As shown in Figure 9B, at both 12 and 14 h post-hCG, diabetic mice ovulated about one half as many ova as controls (17.9 vs. 41.4 at 12 h; 25.7 vs. 48.3 at 14 h). Insulin-treated diabetic mice ovulated at a rate similar to control mice (Fig. 9B). To determine meiotic status, ova collected 10, 12, and 14 h post-hCG were fixed and stained with Hoechst dye. At each time point, there were fewer ova from diabetic mice at metaphase II (MII) than ova from control mice (Fig. 9C), but this difference was eliminated by insulin treatment.
To determine if meiotic progression to MII was also compromised in oocytes from diabetic mice maturing in vitro, both cumulus cell-enclosed and denuded oocytes from control and diabetic animals were cultured for 14 h in MEM/BSA containing 5.5 mM glucose and then fixed and stained to assess meiotic status. When cultured as cumulus cell-enclosed, oocytes from diabetic mice reached MII with a lower frequency (69%) than those from control animals (89%; Fig. 10). However, if the cumulus cells were removed before culture, there was no difference in the number of oocytes that completed maturation; moreover, this percentage was higher in both denuded groups (98%) than in the cumulus cell-enclosed groups, indicating that oocytes from diabetic mice are fully capable of completing meiotic maturation.
|
Preimplantation Embryo Development
To assess the effects of diabetes and differing glucose concentrations on preimplantation embryo development in vitro, control and diabetic mice were superovulated and mated (see Fig. 1) and pronuclear stage embryos were cultured for 5 days in KSOM/AA at 3 glucose concentrations (0.18, 5.5, and 22 mM). Fewer embryos from diabetic mice developed to the blastocyst stage compared with control embryos in normal KSOM containing 0.18 mM glucose. Glucose reduced development to blastocyst in a dose-dependent fashion in both groups, and the blastocyst percentages were consistently lower in embryos from diabetic mice (Fig. 11).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The initial experiments in this study established how the diabetic condition affects spontaneous oocyte maturation in the presence of several different glucose concentrations. These results revealed an acceleration of GVB in oocytes from diabetic mice in the presence or absence of glucose. Interestingly, both groups of oocytes exhibited significantly slower maturation kinetics at 5.5 and 22 mM glucose. This reduction in the maturation rate in response to elevated glucose is transient, manifested during short-term cultures, and may be due to increased production of negative factors by the cumulus cells. For example, increased production of ATP in the cumulus cells due to transit of glucose through the glycolytic pathway could contribute to this inhibitory effect [50, 52].
Hormone signaling was compromised in the oocyte-cumulus cell complex, as FSH action on oocyte maturation in vitro was suppressed by the diabetic state. Both negative and positive signals are conveyed to the oocyte from the cumulus cells in response to FSH, and the gonadotropin effect on meiosis is dependent on culture time. Under short-term culture conditions, FSH is inhibitory, delaying maturation by 1–2 h compared with spontaneous maturation in the absence of FSH [67]. Under control conditions, spontaneous maturation in CEO is delayed by 1–2 h upon addition of FSH to culture medium, but this delay was less pronounced in oocytes from diabetic animals and was not dependent on the presence of glucose since identical kinetics curves were obtained at 0, 0.055, and 5.5 mM glucose, although the delay was lost at 22 mM glucose. This difference in maturation kinetics appears to reflect the physiological state of the oocyte-cumulus cell complexes at the time of isolation rather than a response to medium glucose supplementation.
Although FSH is inhibitory in short-term cultures, it is stimulatory in long-term cultures, presumably through generation of a positive meiosis-inducing factor by the somatic compartment [68]. In the present study, meiotic induction by FSH in vitro was suppressed in both hypoxanthine- and dbcAMP-arrested oocytes from diabetic mice, although the effect was more pronounced under the latter conditions. In hypoxanthine-supplemented medium, higher concentrations of glucose had a stimulatory effect on maturation in either the presence or absence of FSH. Unlike the inhibitory effect of elevated glucose during short-term cultures, such meiotic induction by high glucose levels is manifested during longer overnight cultures and has previously been observed [50]. Although the mechanism is unknown, a candidate mediator is protein kinase C (PKC) since this kinase is activated under diabetic conditions [69, 70] and is known to trigger GVB in mouse CEO [71]. The FSH-induced meiotic maturation was mildly suppressed in the presence of hypoxanthine in oocytes from diabetic animals, but this is unlikely due to PKC activation unless the kinase is selectively activated in the oocyte because PKC has a direct inhibitory action on the oocyte but indirectly stimulates GVB in CEO by acting through the cumulus cells [71].
In dbcAMP-arrested oocytes, increasing concentrations of glucose were again stimulatory to maturation in FSH-free cultures, but this effect was more subtle than in hypoxanthine-containing cultures. On the other hand, suppression of FSH-induced maturation was more pronounced in dbcAMP-arrested CEO from diabetic mice, and, in fact, no significant meiotic induction occurred in 500 µM dbcAMP at any of the glucose concentrations tested. The reason for the disparity between hypoxanthine- and dbcAMP-arrested oocytes is not clear, but recent experiments have suggested that hypoxanthine is a less effective inhibitor of meiotic maturation [72].
The effects of diabetes on spontaneous maturation and both the inhibitory and stimulatory actions of FSH suggest a lesion in the communication between the cumulus oophorus and the oocyte such that the cumulus cells have a decreased capacity to convey either inhibitory or stimulatory signals to the oocyte. Preventing a presumptive inhibitory cumulus cell influence in spontaneously maturing and short-term FSH-treated oocytes would result in higher maturation percentages, while preventing transmission of a positive cumulus cell stimulus would suppress meiotic resumption during the longer overnight cultures. Also, the reduced ability of oocyte-cumulus cell complexes from diabetic mice to respond to FSH indicates that the cumulus cells are less capable of transducing a hormone-induced signal. Consistent with this idea, decreased oxygen uptake in response to FSH has also been reported in granulosa cells from diabetic mice [28]. Nevertheless, FSH-stimulated cumulus expansion, either in the presence or absence of IBMX, was unaffected by the diabetic condition. Therefore, the defect is selective in that not all FSH-mediated responses are affected. In addition, meiotic induction was still compromised in complexes from diabetic mice compared with those from control mice when dbcAMP-arrested oocytes were pulsed 3 h with high levels of cAMP analogues (unpublished data). Thus, it is unlikely that the lesion is principally due to a reduced capacity to generate cAMP but, instead, exists further downstream in the signal cascade. More definitive conclusions must await further experimentation.
The results with hCG-injected mice indicate that the diabetes-mediated changes to oocyte maturation are not restricted to in vitro conditions. Suppression of meiotic resumption in vivo was evident in diabetic animals during the first 5 h post-hCG, and the maturation rate recovered to control levels 5 h later. Diamond et al. [24] reported that, after 6 h of hCG treatment, oocyte maturation in diabetic B6C5 F1 mice was significantly reduced when compared with nondiabetic controls, and the authors concluded that diabetes delayed meiotic maturation in vivo. This has been confirmed herein by kinetics experiments. Not only was germinal vesicle breakdown inhibited, but meiotic progression to metaphase II was restricted as well, and the effect was maintained for up to 14 h post-hCG. This maturation block was likely a consequence of compromised communication between the somatic and germ cell compartments and was not due to an inability of oocytes at the time of isolation to complete maturation, as evidenced by the ability of the denuded oocytes from diabetic mice cultured overnight to reach MII at control levels (98%). On the other hand, the lower percentage of CEO from diabetic mice completing meiotic maturation in vitro when compared with control CEO indicates that the cumulus cells are either less efficient in conveying a stimulatory signal to the oocyte or have a stronger inhibitory influence. These results are consistent with those of a recent study in which a detrimental effect of high glucose (27.5 mM) on the completion of maturation by CEO in vitro was demonstrated [58].
Completion of meiotic maturation is an essential prerequisite for optimal fertilization and embryonic development since oocytes that fail to progress to metaphase II are not readily activated by sperm under normal conditions [73, 74]. Premature fertilization may also lead to problems in chromosome segregation and ploidy [75, 76]. Impaired embryo development, which is a consequence of uncontrolled type I diabetes, could result, at least in part, from a timing problem between meiotic maturation and fertilization due to a failure in the metaphase I-to-metaphase II transition. A limited ability of cumulus cells to transmit factors promoting cytoplasmic maturation could also contribute to decreased developmental capacity.
Earlier work has shown that ovulation is compromised in diabetic animals [10, 17–22], and this has been confirmed in the present study. Only about one half as many ova were ovulated by diabetic mice compared with control mice following a superovulation hormone regimen. A study by Powers et al. [21] demonstrated a defect in the blood-follicle barrier in diabetic mice as indicated by a delay in the hCG-stimulated influx of the serum glycoprotein, inter-
-inhibitor, that is associated with a deficit in superoxide dismutase activity. The reduction in ovulation rate may be due to failure to protect nitric oxide, which is an important regulator of this process [77, 78]. Other possible explanations for ovulation failure include a defect in the hypothalamic-ovarian axis, a suppressed LH surge, altered ovarian steroidogenesis, or a decrease in hormone-binding responsiveness [19, 20, 23–26].
To establish that the effects observed in diabetic mice were caused by the diabetic condition and were not an unrelated side-effect of drug treatment, streptozotocin-induced diabetic mice were treated daily for 4 days with insulin to normalize glucose levels. Under these conditions, blood glucose was returned to control levels, and the detrimental effects of diabetes on meiotic induction (both in vivo and in vitro), completion of meiotic maturation, and ovulation were reversed. These results support the conclusion that such physiological lesions are due to the diabetic condition. This idea was further supported by the decreased FSH-induced maturation observed in CEO from NOD mice, which develop diabetes spontaneously in the absence of drug treatment. However, we are unable to determine from our experiments whether the effects observed are principally the result of elevated glucose levels per se or are a consequence of reduced circulatory insulin. Nonspecific effects such as increased osmolarity were apparently not the cause of the effects observed, as shown in the mannose experiment, in which osmolarity increases comparable with those achieved with glucose did not significantly influence meiotic maturation.
Previous studies have demonstrated reduced preimplantation development in vitro in embryos from spontaneously diabetic NOD mice [33] or chemically induced diabetic mice [24]. In addition, elevated glucose levels have been shown to compromise the normal preimplantation development of one- or two-cell embryos from normal control mice [38, 39]. However, these earlier studies were carried out with Ham F10 medium, which has been shown to cause a two-cell block in cultured mouse embryos due to the presence of hypoxanthine [79]. This effect of hypoxanthine depends on the presence of glucose [80] that helps promote purine salvage and depletion of phosphoribosylpyrophosphate [81]. In the present study, pronuclear-stage embryos from control and diabetic animals were cultured in KSOM/AA, an improved embryo culture medium that alleviates the two-cell block caused by suboptimal culture conditions [82, 83]. The results demonstrate that 1) embryos from diabetic mothers are compromised in their developmental capacity as early as the pronuclear stage and 2) elevated glucose levels suppress preimplantation development in both types of embryos. These data are therefore consistent with the idea that diabetic conditions during oogenesis and oocyte maturation have a detrimental impact on later development and that continual exposure to elevated glucose following fertilization further compounds this problem.
|
FOOTNOTES |
|---|
1 This work was supported by funds from NIH (HD25291 and HD 39172 to S.M.D.). 
2 Correspondence: Stephen M. Downs, Marquette University, Biology Department, P.O. Box 1881, Milwaukee, WI 53201-1881. FAX: 414 288 7357; stephen.downs{at}marquette.edu
Accepted: February 4, 2002.
Received: August 30, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylase activity and serum concentrations of luteinizing hormone, estradiol and progesterone in immature rats with diabetes mellitus induced by streptozotocin. Proc Soc Exp Biol Med 1982; 171:207-213[Medline]
