Phosphoinositide Metabolism in the Developing Conceptus Effects of Hyperglycemia and scy//o-lnositol in Rat Embryo Culture P.J. STRIELEMAN, M.A. CONNORS, B.E. METZGER

Culture of the postimplantation rat conceptus in hyperglycemic medium causes developmental abnormalities and is associated with a diminished water-soluble myo-inositol content. We investigated the effect myo-inositol depletion has on lipid-soluble phosphoinositides, precursors, and water-soluble inositol phosphates. Rat conceptuses were cultured from gestational day 9.5 (presomite, early head fold) to day 10.5 (7-15 somites) in 6.7-73.3 mM D-glucose. Significant decreases in the phosphoinositides of the embryo were observed with increased culture D-glucose concentrations. PI was reduced 15-34%, PIP 18-46%, and PIP2 26-46%. Yolk sac phosphoinositides also were reduced but to a lesser degree. Culture in hyperglycemic media also mediated significant reductions of conceptus inositol phosphates. To investigate whether effects similar to those induced by D-glucose could be mediated by another agent capable of decreasing myo-inositol content, we used scy//o-inositol, a transported but nonmetabolized isomer of myo-inositol. Conceptuses cultured in medium containing scy//o-inositol (0.06-16.7 mM) had dose-dependent decreases of myo-[3H]inositol in water-soluble and lipid-soluble fractions. Incorporation of myo-[3H]inositol into phosphoinositides and inositol phosphates was decreased concomitantly. Developmental effects of D-glucose and scy//o-inositol were assessed in rat conceptuses cultured from day 9.5 (presomite, early head fold) to day 11.5 (22-28 somites). Culture in 40.0-73.3 mM glucose and 0.06-33.3 mM

From the Center for Endocrinology, Metabolism, and Nutrition, Department of Medicine, Northwestern University Medical School, Chicago, Illinois. Address correspondence and reprint requests to Boyd E. Metzger, MD, Northwestern University, Searle 10-555,303 E. Chicago Avenue, Chicago, IL 60611. Received for publication 18 October 1991 and accepted in revised form 30 March 1992. PI, phosphatidylinositol; PIP, phosphatidylinositol 4 monophosphate; PIP2, phosphatidylinositol 4, 5 bisphosphate; TCA, trichloroacetic acid; IP,, inositol monophosphate; IP2, inositol bisphosphate; IP3, inositol triphosphate; ANOVA, analysis of variance; df, degree of freedom; NS, no significance; HPLC, high-performance liquid chromatography; STZ, streptozocin.

DIABETES, VOL 41, AUGUST 1992

scy//o-inositol impaired growth while increasing dysmorphogenesis in a dose-dependent manner. The results suggest that decreases in conceptus myo-inositol and associated diminution of phosphoinositides, which are the inositol/lipid cycle precursors, are dysmorphogenic and may contribute to the etiology of diabetic embryopathy. Diabetes 41:989-97,1992

A

major complication in diabetic pregnancy is an increased incidence of congenital abnormalities, which are initiated early in pregnancy (1,2). The malformations result from altered diabetic maternal fuels and fuel-related products that may act alone or synergistically as teratogens to modify normal cell growth and differentiation (1,2). Our laboratory has been interested in understanding the mechanisms underlying alterations in mammalian development that arise as a result of chronic hyperglycemia. We used the rodent embryo culture technique of New (3), as modified in our laboratory (4), to investigate this question. A predominant characteristic of growth in hyperglycemic media is the failure of the developing conceptus to accumulate sufficient myo-inositol, with the level of tissue myo-inositol consistently correlated with the degree of dysmorphogenesis (5-7). myo-inositol tissue/plasma gradients are low in the early postimplantation conceptus, suggesting that minimal reserve is available for a period of rapid development (5,8). myoinositol is essential for normal in vitro embryonic development, as evidenced when a culture of rat conceptuses in myo-inositol-depleted serum (9) or tissue depletion by inhibition of myo-inositol metabolism with lithium chloride (10,11) results in concomitant dysmorphogenic effects. Similar to other tissues affected by hyperglycemia-related myo-inositol depletion (12-15), disruption of myoinositol transport (8) from exogenous sources is a primary effector of myo-inositol depletion in rat conceptuses,

989

PHOSPHOINOSITIDES IN THE RAT CONCEPTUS

whereas, in contrast, sorbitol accumulation does not appear to be causally linked (5-7). The rat conceptus can transport myo-inositol from the culture media, but the process is neither rapid nor efficient (8). Hyperglycemic conditions further compromise this condition as D-glucose competitively inhibits myo-inositol transport (8), thus diminishing the normal increase in the level of free myo-inositol in postimplantation rat conceptus (5,6,8), embryo (7), or embryonic neural tissue (16) during organogenesis. This failure to amass tissue myo-inositol and the associated dysmorphogenic affects can be overcome by supplementing with myo-inositol (6,7,17). Although all intracellular functions of myo-inositol have not been elucidated, sufficient myo-inositol is required for synthesis of phosphoinositides, which are important in the inositol/lipid cycle intracellular signaling system (18). Several intracellular second messengers (inositol triphosphate, 1,2 diacylglycerol, arachidonic acid) are derived from the phosphoinositides. Roles for each of these second messengers in normal cellular growth and differentiation have been postulated (18,19). Whether reported decreases in conceptus water-soluble myo-inositol (5-8) also deplete lipid-soluble phosphoinositides has not been examined. We investigated the consequences of hyperglycemia-mediated myoinositol depletion on the phosphoinositides and inositol phosphates of the 10.5 day rat conceptus. We also have characterized the effect of hyperglycemia on myo-inositol distribution between the embryo and yolk sac membranes. To clarify whether specific disruption of myoinositol homeostasis is dysmorphogenic, we investigated whether scy//o-inositol, a nonmetabolized isomer of myoinositol, causes effects similar to those observed with D-glucose. Preliminary reports of these data have appeared previously (20,21).

RESEARCH DESIGN AND METHODS

Whole embryo culture. Charles River virgin female rats (Crl:CDR(SD)BR) were housed with a light/dark cycle of 14/10 h and mated with males from the same strain between 1800 and 0800. Pregnancy was confirmed by the presence of sperm in a vaginal smear the following morning. The preceding midnight was considered day 0 of gestation. Pregnant rats were killed between day 9.5 and 9.7 of gestation. The uterus was opened; the conceptuses were separated from the surrounding decidua; and Reichert's membrane was ruptured. Whole-embryo culture was performed by the methods of New (3), as modified in our laboratory (4). Culture was conducted for 24 or 48 h in 4 - 6 ml culture media with 1 conceptus/ml in an atmosphere described previously (4). Each culture tube contained conceptuses from at least two separate pregnant females. The culture medium consisted of immediately centrifuged, heat-inactivated serum from normal female rats combined in the proportion of 3:1 (vol/vol) with isotonic (0.85% wt/vol) saline (control), saline plus 5% D-glucose (high D-glucose media), or saline plus 1.5% scy//o-inositol (scy//o-inositol media). Under these culture conditions, the D-glucose concentration of the control media was 6.7 mM D-glucose and

990

the high D-glucose media 40.0, 56.6, or 73.3 mM D-glucose. These D-glucose concentrations produce frequent, predictable developmental changes during in vitro rat embryo culture (4-7,22). Scy//o-inositol was supplemented at 0.06, 0.55, 5.5, 16.7, and 33.3 mM. myo[3H]inositol (0.5-0.8 |o,Ci/ml) was included in tubes containing conceptuses used for biochemical analysis. Conceptuses used for all biochemical analysis were cultured from day 9.5 (presomite, early head fold) to day 10.5 (7-15 somites) of gestation. The dysmorphogenic effects of D-glucose observed at day 11.5 were engendered during this period (22,23). Morphological analysis was performed on conceptuses cultured from day 9.5 (presomite, early head fold) to day 11.5 (22-28 somites) in media containing increasing concentrations of D-glucose or scy//o-inositol. Tissue preparation and extraction. At day 10.5-10.7, conceptuses were removed from the culture media, rinsed in ice-cold 0.85% saline, and placed in a small petri dish containing saline maintained at 4-6°C. Under a dissecting microscope, the conceptus was separated into embryo and yolk sac membranes (yolk sac tissue and allantois); the tissue was rinsed in cold 0.85% saline and immediately placed in 3 ml ice-cold 1:2 chloroform/ methanol. Tissue was disrupted further by sonication. Partition of tissues from embryo or yolk sac into lipidand water-soluble components was performed by acid extraction (24). Samples in 3 ml of 1:2 chloroform/ methanol were mixed with 1.8 ml 1.2 N HCI, 1.8 ml chloroform, and 0.6 ml 100 mM KCI; and the phases were separated by centrifugation. The chloroform phase was collected, and the water-soluble phase re-extracted twice with 1.5 ml of chloroform. The water-soluble phase was dried in a vacuum evaporator (Savant Speedvac), and the residue was redissolved in 1 ml of water and used to determine tissue levels of water-soluble myo [3H]inositol. The chloroform phase was washed with 3 ml methanol/1 N HCI (1:1) and taken to dryness under nitrogen. The dried lipid was stored at -20°C until analysis of lipid-soluble myo-[3H]inositol or phosphoinositides was conducted. Recovery of [3H]PIP2 standard was >95% using this method. The insoluble tissue residue was dried and dissolved in 1 ml 0.5 N NaOH for estimation of tissue protein and DNA (25,26). Separation of phosphoinositides. Phosphoinositides were separated from the acid extracted lipid-soluble fraction by thin layer chromatography on oxalate-treated Silica gel 60 plates using a (60:20:23:18:12) chloroform/ methanol/acetone/acetic acid/water solvent system (27). The dried plates were treated with ENHANCE (Du PontNEN, Boston, MA), and the phosphoinositides visualized by fluorography. Bands corresponding to the phosphoinositides were scraped into scintillation vials containing 1 ml of 1:1 methanol/water; scintillation fluid was added; and radioactivity was determined. myo-[3H]Inositol incorporated into the phosphoinositides was calculated in terms of pmol of myo-inositol/|xg DNA on the basis of myo-[3H]inositol-specific radioactivity in the initial incubation media (8,28). myo-inositol content of culture me-

DIABETES, VOL. 41, AUGUST 1992

P.J. STRIELEMAN, M A CONNORS. AND B.E. METZGER

dium was determined by bioassay with Kloeckera brevis yeast as described previously (5,8). Inositol phosphate determination. Intact 10.5 day conceptuses were immersed rapidly and rinsed free of culture media in three changes of 3 0 - 3 5 ml ice-cold saline. Two conceptuses and 50-75 |xl saline were transferred to 0.5 ml 12% TCA containing 20 n-g of phytate digest (29), and immediately were sonicated to disrupt the tissue. Acid-precipitated tissue suspensions were centrifuged, the supernatant was removed, and the residue pellet was rinsed with 0.5 ml 12% TCA. The combined extracts were washed with water-saturated diethyl ether and neutralized with 1 M NaCO 3 . Extracted samples were stored at -20°C until analysis. The [3H] inositol phosphates—IP-,, IP2, and IP 3 —were separated by ion-exchange chromatography over Dowex 1-X8 (30). Validity of the system and the fractionation pattern of inositol phosphates were confirmed with [3H]inositol phosphate standards with recoveries >95%. Morphological analysis. At the end of 48 h culture, conceptuses were assessed for heartbeat and visceral yolk sac circulation, separated into yolk sac membranes and embryo, and visually examined for abnormalities as described previously (4). Tissue protein and DNA content were determined in 0.5 N NaOH digests (25,26). Statistical analysis. Data were calculated as mean ± SE. Statistical difference was analyzed by one-way ANOVA or Student's t test for mean values and Fisher's exact test for differences between groups of morphological lesions. P < 0.05 was considered significant. Serum for culture was obtained from retired breeder female rats and frozen at -20°C until use (4). Reagent grade chemicals purchased for experiments included D-glucose, PI, PIP, PIP2 from Sigma (St. Louis, MO); scy//o-inositol from Cal Biochem (San Diego, CA); silica gel 60 thin layer chromatography plates from Whatman (Maidstone, England); HPLC grade organic solvents from Mallinckrodt (St. Louis, MO); /7?yo-[3H]inositol with PT6-271 polymer from Amersham (Arlington Heights, IL); and ENHANCE from DuPont-NEN.

RESULTS

Inositol distribution. The relative content of myo-[ 3 H] inositol in yolk sac and embryo, and the partition between water-soluble and lipid-soluble fractions are summarized in Table 1. Preliminary experiments established that after 24 h culture, tissue myoinositol mass determined by direct assay agreed with that calculated from incorporated tissue myo[ 3 H] inositol and culture media specific activity. Thus, apparent reductions of myo[ 3 H] inositol reflected equivalent depressions in unlabeled myoinositol. Following 24 h culture in control media (6.7 mM D-glucose), incorporation of myo[ 3 H] inositol into the water-soluble fraction of embryo and yolk sac tissue was comparable (1215 ± 53 vs. 1792 ± 234 dpm/n-g DNA). Culture in increasing concentrations of D-glucose diminished incorporation of myo[ 3 H] inositol into the watersoluble fraction by 55-70% in the embryo, whereas smaller decreases of 3 3 - 4 8 % were observed in yolk sac tissue. One-way ANOVA indicated that culture in high

DIABETES, VOL. 41, AUGUST 1992

TABLE 1 Effect of D-glucose on net myo[3H] inositol uptake in 10.5 day rat embryo and yolk sac myoinositol

Embryo

Yolk sac

Media glucose (mM) 6.7 40.0 56.6 73.3 6.7 40.0 56.6 73.3

Water-soluble

Lipid-soluble

n

% of control

% of control

4 4 4 4 4 4 4 4

100 ± 4 45 ± 9 * 35±6t 30±4f 100 ± 13 67 ± 5 53 ± 7 * 52 ± 11*

100 67 58 60 100 90 76 72

± 10 ± 12 ±6* ±7* ± 12 ± 12 ±8 ± 16

Conceptuses were cultured in the presence of 0.5 jxCi/ml myo-[3H]inositol and increasing concentrations of D-glucose for 24 h. Subsequently, they were rinsed free of media, separated into embryo and yolk sac membranes, extracted into watersoluble and lipid-soluble fractions, and analyzed for myo[3H]inositol. Values are means ± SE. Control values (DPM/p,g DNA): embryo, water-soluble 1215 ± 53, lipid-soluble 2549 ± 10; yolk sac, water-soluble 1792 ± 234, lipid-soluble 6894 ± 816. n, number of individual extractions performed for each condition. Data from two separate experiments. *P < 0.01; t P < 0.001; tP < 0.05, compared with values from 6.7 mM D-glucose.

glucose effected a significant overall depression of myoinositol levels in the water-soluble fraction of embryo (F = 27.933, df = 3,12, P < 0.001) and yolk sac (F = 5.609, df = 3,12, P < 0.001). Incorporation of myo[3H]inositol into the lipid-soluble fraction of the embryo (2549 ± 10 dpm/|xg DNA) was less than that observed in the yolk sac (6894 ± 816dpm/^g DNA). Including D-glucose in the culture media effected decreases in myo[3H]inositol incorporation into the lipid-soluble fraction similar to those found in the water-soluble fraction. However, the magnitude of the decreases in lipid-soluble myoinositol of both embryo and yolk sac tissue was less than that observed in their respective water-soluble fractions. A 33-40% decrease was observed in the embryo, and a 10-28% decrease was observed in the yolk sac. The lipid-soluble fraction of the embryo was (F = 4.575, df = 3,12, P < 0.05) altered significantly by glucose culture conditions, whereas the decreases in yolk sac lipid-soluble fraction were not statistically significant. Effect of D-glucose on the phosphoinositides. The effects of culture in high D-glucose media on the embryo and yolk sac phosphoinositides are shown in Table 2. All detectable lipid-soluble myo[ 3 H]inositol radioactivity was recovered in the three phosphoinositides—PI, PIP, and PIP2. Culture in increased media D-glucose effected incremental decreases in the individual phosphoinositides of both embryo and yolk sac tissue. Embryo PI decreased 15-34%, PIP 18-45%, and PIP2 2 6 - 4 6 % . One-way ANOVA comparison of myo-[3H]inositol incorporation within the four glucose culture conditions indicated a highly significant overall effect of increasing medium D-glucose to depress all embryo phosphoinositides (PI, F = 16.099, df = 3,29, P < 0.001; PIP, F = 6.820, df = 3,29, P < 0.001; PIP2, F = 21.117, df = 3,29,

991

PHOSPHOINOSITIDES IN THE RAT CONCEPTUS TABLE 2 Effect of D-glucose on the phosphoinositides of the 10.5 day rat embryo and yolk sac Media glucose (mM)

PI % of control (n)

PIP % of control (n)

PIP2 % of control (n)

Embryo

6.7 40.0 56.6 73.3

100.0 85.3 66.4 65.9

± 3 . 8 (17) ± 2.8* (6) ± 1.4t (5) ± 2.8t (5)

100.0 82.1 66.7 54.5

± 6 . 5 (17) ± 8.1 (6) ± 5.7* (5) ± 4.1 * (5)

100.0 dt 3.8(17) 74.0 db 5.3t (6) 63.4 dt 0.8* (5) 54.2 dt 4.6t (5)

Yolk sac

6.7 40.0 56.6 73.3

100.0 ± 6 . 5 (18) 100.5 ± 7 . 2 (6) 73.6 ± 4 . 1 * (5) 64.3 ± 1.9* (5)

100.0 100.3 73.1 57.3

± 7.7(18) ± 8 . 7 (6) ± 4.2 (5) ± 4.5* (5)

100.0 dt 6.9 (18) 98.0 dt 6.9 (6) 76.7 db 4.1 (5) 58.8 db 3 . 1 * (5)

Conceptuses were cultured in the presence of 0.5 |xCi/ml [3H]-myo-inositol and increasing concentrations of D-glucose for 24 h to label the phosphoinosistides. Conceptuses were separated into embryo and yolk sac membranes, the lipids acid extracted, and individual phosphoinositides separated by thin layer chromatography. Control values (pmol myo[3H]inositol incorporated/|xg DNA: embryo, PI, 117 ± 4, PIP, 6.8 ± 0.4, PIP2, 7.3 ± 0.3; yolk sac, PI, 231 ± 15, PIP, 15.9 ± 1.2, PIP2,21.7± 1.5. Values are means ± SE. (n) number of observations. *, P < 0.05; fP < 0.001; *P < 0.01, compared with values from 6.7 mM D-glucose.

P < 0.001). Individually, PIP2 levels were affected most with a highly significant (P< 0.001) decrease at all increased-culture D-glucose concentrations. Similar to embryonic tissue, yolk sac phosphoinositides also were reduced by culture in high-glucose media. One-way ANOVA indicated a significant overall effect on all yolk sac phosphoinositides (PI, F = 4.788, df = 3,30, P < 0.01; PIP, F = 4.350, df = 3,30, P < 0.01; PIP2, F = 4.787, df = 3,30, P < 0.01). The extent of the decrease was variable and less than that observed in the embryo, and it was mediated by the 56.6 and 73.3 mM D-glucose concentrations. Individually, reduction in yolk sac PIP2 was statistically significant only at 73.3 mM D-glucose (P0.55 mM were necessary to elicit statistically significant neural tube lesions. Neural tube malformations included open posterior neural pore (47-83%), open posterior neural tube (4-17%), open anterior neural pore (10-67%), and open anterior neural pore (17%). Other lesions included anomalous brain development (8-100%), malformed optic and otic vesicles (13-67%), abnormal heart development (0-67%), abnormal or incomplete ventral axial rotation (0-83%), and somite irregularities (8-50%). In addition to embryonic defects, abnormal or incomplete development of yolk sac circulation was observed at scylloinositol concentrations of 0.55-33.3 mM.

DISCUSSION

Growth in hyperglycemic conditions results in failure of the mammalian embryo to amass normal myoinositol content (5-8,16). This decrease is attributable in part to inhibition by D-glucose of myoinositol transport into the developing conceptus and not to (nor is it secondary to) increased sorbitol concentration (5,7,8). Concurrent with the decreased conceptus myoinositol, an increase in the incidence of embryo dysmorphogenesis has been observed (5-7). This study was designed to gain insight

TABLE 5 Effect of scy//o-inositol on growth and development of the 11.5 day rat embryo Malformations Scylloinositol (mM) 0

0.06 0.55 5.55 16.7 33.3

Protein n 53 24 30 24 12

(M-g) 195 ± 7 167 ± 11*

151 154 131 6 93

±9* ± 11t ± 7* ±5*

DNA (ng) 20.3 15.9 14.0 15.3 9.9 5.6

dfc0.9 ifc1.1t ifc 1.1* dfc 1.0* dt 0 . 9 * dt 0 . 8 *

Crown-rump (mm) 3.50 3.18 3.04 2.98 2.88 2.37

± 0.04 ±0.05* ± 0.07* ± 0.09* ± 0.04* ±0.12*

Somites (number) 26.2 24.5 22.7 23.1 22.0 21.8

± 0.2 ± 0.2* ± 0.4* ± 0.4* ± 0.3* ±0.3*

Total 5.7% 25.0%* 56.7%* 62.5%* 75.0%* 100%*

Neural tube 5.7% 0%

46.7%* 50.0%* 41.7%* 83.3%*

Brain 1.9% 8.3% 43.3%* 41.7%* 58.3%* 100%*

Extraneural tissue 3.8% 20.8%t 50.0%* 58.3%* 75.0%* 83.3%*

Conceptuses were cultured from days 9.5-11.5 in serum/saline media (3:1) containing 0, 0.06, 0.55, 5.55, 16.7, and 33.3 mM scy//o-inositol. Embryos were freed of extra-embryonic membranes for examination. Values are means ± SE. n, number of embryos examined. *, P < 0.05; t, P < 0.01; *, P < 0.001, compared with values from control 0 mM scy//oinositol.

994

DIABETES, VOL. 41, AUGUST 1992

P.J. STRIELEMAN, M.A. CONNORS. AND B.E. METZGER

into the biochemical alterations in myo-inositol metabolism mediated by hyperglycemic growth and whether the changes observed are associated with dysmorphogenesis. Conceptus culture was carried out between day 9.5 and 10.5, which encompasses a critical period of vulnerability to the teratogenic effects of hyperglycemia (22,23). Our initial efforts confirmed that culture in the presence of elevated D-glucose is teratogenic (Table 4) and effects decreases in lipid-soluble myo-inositol similar to those previously reported for water-soluble myo-inositol (5-8). In addition, dissection of the conceptus into embryo and yolk sac membranes enabled us to examine the distribution of water- and lipid-soluble myo-inositol. Similar to the results with 11.5 day conceptus (5-7) and acute shortterm culture with 10.5 day conceptus (8), incremental decreases in water-soluble myo-inositol content were observed after culture in the presence of increasing concentrations of D-glucose. Both embryo and yolk sac lipid-soluble myo-inositol decreased parallel to the watersoluble pool but to a lesser extent. Lipid-soluble/watersoluble ratio increased with culture in hyperglycemic media, indicating that phospholipogenesis is preserved partially in the presence of a diminished myo-inositol precursor pool. The reason for this is unclear, but a similar response has been reported in the sciatic nerve, which also is susceptible to D-glucose-induced myoinositol depletion (15). In all experimental groups, a higher inositol lipid content was seen in the yolk sac compared with the embryo. These results are similar to the finding in 11.5 day conceptuses that yolk sac phospholipid fatty acid content is elevated compared with ' embryo phospholipid fatty acid content (33). Culture of rat conceptuses in myo-inositol-depleted serum (9) or lithium-induced myo-inositol depletion (10,11) is teratogenic. Similarly, glucose-mediated decreases in myo-inositol content are correlated consistently with dysmorphogenesis in the rodent embryo (5-7) and can be overcome by supplementing culture medium with myo-inositol (6,7). The biochemical basis by which altered myo-inositol homeostasis, mediated by hyperglycemia, affects rodent development has not been elucidated. In the STZ-diabetic rat, hyperglycemia depletes cellular myo-inositol in the sciatic nerve and is accompanied by modifications in phosphoinositide metabolism with concomitant alterations in nerve function (14,3438). Similarly, in the rat conceptus, decreased myoinositol levels and an altered phosphoinositide pool are direct consequences of hyperglycemic culture (Table 2). Moreover, conceptus inositol phosphates are reduced (Fig. 1), suggesting that the receptor sensitive phosphoinositide pool is decreased concomitantly with the global phosphoinositide decrease. Evidence that disruption of phosphoinositide hydrolysis interferes with cell growth and differentiation has been provided by studies in which antibodies (39) or neomycin (40) were used specifically to inhibit phosphoinositide turnover with concurrent limiting of cellular proliferation. A potential direct role for phosphoinositide signaling in cell proliferation has been suggested by the ability of inositol bisphosphate to activate a low affinity form of human DNA polymerase a,

DIABETES, VOL. 41, AUGUST 1992

which is involved in cellular growth (41) and by studies linking nuclear phosphoinositide metabolism with the onset of DNA synthesis in Swiss 3T3 cells (42,43). Direct evidence for a role of decreased inositol/lipid cycle activity in teratogenesis has been provided by a study with Xenopus blastocysts (44). In this study, lithiuminduced depletion of myo-inositol was teratogenic to Xenopus blastocysts, but the effect was overcome by supplementing with myo-inositol or phorbol esters but not with metabolically inactive ep/-inositol (44). This study indicates that, analogous to Xenopus, altering inositol/ lipid cycle metabolism in the mammalian embryo may contribute to teratogenesis. We did not examine whether other biochemical changes induced by culture in hyperglycemic media—in addition to reduced myo-inositol levels—contribute to the reduced phosphoinositide pool. Altered embryonic tissue energy potential for sequential phosphorylation of the phosphoinositides appears unlikely as it is not decreased by hyperglycemic conditions in vivo (16) or in vitro (Passonneau and Freinkel, unpublished observations). Moreover, the primary alteration induced by D-glucose was at PI, with symmetrical decreases observed in PIP and PIP2. Thus, the synthetic pathway to PIP and PIP2 appears to be affected only by loss of PI precursor. Altered enzyme activity for PI synthesis also appears unlikely. STZ-diabetes reduces CDP-diacylglycerolinositol phosphatidyltransferase and phosphatidylinositol-4-phosphate kinase activity in the sciatic nerve and brain with concomitantly decreased incorporation of myo-[3H]inositol into phospholipids (38,45,46). However, insulin treatment normalizes enzyme activity without increasing apparent phosphoinositide synthesis, indicating that a depressed tissue myo-inositol content is the predominant factor affecting synthesis. In another study, diet-induced myo-inositol deficiency caused a 40% decline in liver CDP-diacylglycerol-inositol phosphatidyltransferase activity without affecting PI mass (46). Thus, although decreased conceptus CDP-diacylglycerolinositol phosphatidyltransferase activity could be contributing to a decrease in the phosphoinositides, the diminished myo-inositol tissue pool most likely is the predominant factor involved. Direct evidence that D-glucose-mediated decreases in conceptus myo-inositol content lower phosphoinositide levels and that these alterations are linked to dysmorphogenesis is supported by our experiments with scy//o-inositol. scy//o-lnositol, an epimer of myo-inositol, like D-glucose competes for myo-inositol transport, but is not a substrate for the CDP-diacylglyceride-inositol transferase enzyme in the phosphoinositide synthetic pathway (31,32,47-50). It can replace myo-inositol in the tissue pool without being metabolized to the phosphoinositides. In this study, scy//o-inositol effectively inhibited conceptus myo-[3H]inositol uptake (Fig. 2), analogous to that observed in other tissues (47-50). The competition for transport depleted conceptus water-soluble and lipidsoluble myo-[3H] inositol content similar to that observed for D-glucose. The concentration dependence of the scy//o-inositol inhibition indicated that a fraction (6070%) of the uptake is inhibited specifically by scyllo-

995

PHOSPHOINOSITIDES IN THE RAT CONCEPTUS

inositol, whereas the remaining uptake is not reduced. Thus, where 0.55 mM scy//o-inositol reduces myo-[3H] inositol uptake into the water-soluble pool by 41%, a 30-fold increase in scy//o-inositol only increased inhibition to 60%. These results are analogous to and support previous studies of D-glucose inhibition of myo-inositol transport in the conceptus that indicated that uptake occurs primarily by a saturable facilitated transport system and secondarily by simple diffusion (8). scy//o-lnositol has been reported to impair phosphoinositide synthesis or inositol phosphate generation in myo-inositol-depleted liver cells (47) and cerebral cortex slices (51). Similarly, culture of conceptuses in media containing scy//o-inositol yielded a dose-dependent global reduction of embryo and yolk sac phosphoinositides (Table 3) and decreased inositol phosphate levels (Fig. 3), suggesting that the receptor-sensitive pool also is decreased. These results parallel those observed with D-glucose (Table 2, Fig. 1). Concurrent with the alterations of myo-inositol metabolism, both D-glucose and scy//oinositol effected significant growth retardation and dysmorphogenesis in embryos examined on day 11.5 (Tables 4 and 5). Moreover, the extent of dysmorphogenesis and decrease in phosphoinositide levels are comparable (Tables 2-5). These results taken together suggest that D-glucose and scy//o-inositol act through a similar mechanism to disrupt normal growth. Both inhibit myo-inositol transport into the developing conceptus, decreasing intracellular concentrations. In addition, a diminished phosphoinositide synthesis ensues that may be compromising the inositol/lipid intracellular signaling pathway, which is important for cellular growth and differentiation. If a similar disruption of myo-inositol metabolism occurs in utero, it, in combination with other teratogenic factors, would jeopardize the ability of the embryo to undergo normal development.

6.

7.

8. 9. 10. 11. 12.

13. 14. 15. 16. 17. 18. 19. 20. 21.

ACKNOWLEDGMENTS This research was supported by National Institutes of Health Grant DK10669, Training Grant DK07169, in part by an American Diabetes Association Northern Illinois Affiliate Young Investigator Award, and a grant from Ronald McDonald Charities. We thank Dr. James McLeod for his review of the manuscript and helpful suggestions. The initial design of this study was conducted with the assistance of the late Dr. Norbert Frienkel.

REFERENCES 1. Freinkel N: Diabetic embryopathy and fuel-mediated organ teratogenesis: lessons from animal models. Horm Metab Res 20:463-75, 1988 2. Sadler TW, Hunter ES, Wynn RE, Phillips LS: Evidence for multifactorial origin of diabetes-induced embryopathies. Diabetes 38:7074, 1989 3. New DAT: Whole embryo culture and the study of mammalian embryos during organogenesis. Biol Rev 53:81-122, 1978 4. Freinkel N, Cockroft DL, Lewis NJ, Gorman L, Akazawa S, Phillips LS, Shambaugh III GE: The 1986 McCollum Award Lecture. Fuelmediated teratogenesis during early organogenesis: the effects of increased concentrations of glucose, ketones or somatomedin inhibitor during rat embryo culture. Am J Clin Nutr 44:986-95, 1986 5. Hod M, Star S, Passonneau JV, Unterman TG, Freinkel N: Effect of hyperglycemia on sorbitol and myo-inositol content of cultured rat conceptus: failure of aldose reductase inhibitors to modify myo-

996

22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32. 33.

inositol depletion and dysmorphogenesis. Biochem Biophys Res Commun 140:974-80, 1986 Hod M, Star S, Passonneau JV, Unterman TG, Freinkel N: Glucoseinduced dysmorphogenesis in the cultured rat conceptus: prevention by supplementation with myo-inositol. IsrJMed Sci26:541 -44, 1990 Hashimoto M, Akazawa S, Akazawa M, Akashi M, Yamamoto H, Maeda Y, Yamaguchi Y, Yamasaki H, Tahara D, Nakanishi T, Nagataki S: Effects of hyperglycemia on sorbitol and myo-inositol contents of cultured embryos: treatment with aldose reductase inhibitor and myo-inositol supplementation. Diabetologia 33:597602, 1990 Weigensberg MJ, Garcia-Palmer FJ, Freinkel N: Uptake of myoinositol by early-somite rat conceptus: transport kinetics and effects of hyperglycemia. Diabetes 39:575-82, 1990 Cockroft DL: Changes with gestational age in the nutritional requirements of post implantation rat embryos in culture. Teratology 38:281-90, 1988 Hansen DK, Walker RC, Grafton TF: Effect of lithium carbonate on mouse and rat embryos in vitro. Teratology 41:155-60, 1990 Garcia-Palmer F, Weigensberg MJ, Freinkel N: Lithium and the early post implantation embryo: embryotoxicity and changes in myoinositol metabolism (Abstract). Clin Res 36:482A, 1988 Haneda M, Kikkawa R, Arimura T, Ebata K, Togawa M, Maeda S, SawadaT, Horide N, Shigeta Y: Glucose inhibits myo-inositol uptake and reduces myo-inositol content in culture rat glomerular mesangial cells. Metabolism 39:40-45, 1990 Yorek MA, Dunlap JA: The effect of elevated glucose levels on myo-inositol metabolism in cultured bovine aortic endothelial cells. Metabolism 38:16-22, 1989 Greene DA, Lattimer SA, Sima AAF: Sorbitol, phosphoinositides and sodium potassium ATPase in the pathogenesis of diabetic complications N Engl J Med 316:599-605, 1987 Palmano KP, Whiting PH, Hawthorne JN: Free and lipid myo-inositol in tissue from rats with acute and less severe streptozotocin induced diabetes. Biochem J 167:229-35, 1977 Sussman I, Matschinsky FM: Diabetes affects sorbitol and myoinositol levels of neuroectodermal tissue during embryogenesis in rat. Diabetes 37:974-81, 1988 Baker L, Piddington R, Goldman A, Egler J, Moehring J: Myo-inositol and prostaglandin reverse the glucose inhibition of neural tube fusion in cultured mouse embryos. Diabetologia 33:593-96, 1990 Rana RS, Hokin LE: Role of phosphoinositides in transmembrane signaling. Physiol Rev 70:115-64, 1990 Berridge MJ: Inositol lipids and cell proliferation. Biochem Biophys Acta 907:33-45, 1987 Strieleman PJ, Connors MA, Freinkel N: Glucose decreases phosphoinositide levels in cultured rat embryo (Abstract). FASEB J 4:5221 A, 1990 Strieleman PJ, Connors MA, Metzger BE: Impaired phosphoinositide signaling secondary to myo-inositol depletion: a role in birth defects associated with diabetes (Abstract). Diabetes 40(Suppl. 1):37A, 1991 Cockroft DL: Abnormalities induced in cultured rat embryos by hyperglycaemia. Br J Exp Path 65:625-36, 1984 Garnham EA, Beck F, Clarke CA, FRS, Stanisstreet M: Effects of glucose on rat embryos in culture. Diabetologia 25:291-95, 1983 Schacht J: Extraction and purification of polyphosphoinositides. Methods Enzymol 72:626-31, 1981 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the folin phenol reagent. J Biol Chem 193:265-75, 1951 Hinegardner RT: An improved fluorometric assay for DNA. Anal Biochem 39:197-201, 1971 Andrews WV, Conn PM: Measurement of inositol phospholipid metabolites by one dimensional thin layer chromatography. Methods Enzymol 141:156-68, 1987 Zhu X, Eichberg J: A myo-inositol pool utilized for phosphatidylinositol synthesis is depleted in sciatic nerve from rats with streptozocin-induced diabetes. Proc Natl Acad Sci USA 87:9818-22, 1990 Wreggett KA, Howe LR, Moore JP, Irvine RF: Extraction and recovery of inositol phosphates from tissues. Biochem J 245:93334, 1987 Berridge MJ, Dawson RMC, Downes CP, Heslop JP, Irvine RF: Changes in the levels of inositol phosphates after agonist-dependent hydrolysis of membrane phosphoinositides. Biochem J 212: 473-82, 1983 Deshusses J, Reber G: Myo-inositol transport in Klebsiella aerogeneses: scy/fo-inositol a non-metabolizable substrate for the study of the myo-inositol transport system. Eur J Biochem 72:87-91, 1977 Benjamin JA, Agranoff BW: Distribution and properties of CDPDiglyceride:inositol transferase from brain. J Neurochem 16:513— 27, 1969 Pinter E, Reece A, Ogburn PL, Turner S, Hobbins JC, Mahoney MJ,

DIABETES, VOL. 41, AUGUST 1992

P.J. STRIELEMAN. M.A. CONNORS. AND B.E. METZGER

34. 35.

36. 37.

38. 39. 40. 41. 42.

Naftolin F: Fatty acid content of yolk sac and embryo in hyperglycemia-induced embryopathy and effect of arachidonic acid supplementation. Am J Obstet Gyneco/159:1484-90, 1988 Natarjan V, Dyck PJ, Schmid HHO: Alterations of inositol lipid metabolism of rat sciatic nerve in strepozotocin induced diabetes. J Neurochem 36:413-19, 1981 Bell M.E, Peterson RG, Eichberg J: Metabolism of phospholipids in peripheral nerve from rats with chronic streptozotocin induced diabetes: increased turnover of phosphatidylinositol 4,5 bisphosphate. J Neurochem 39:192-200, 1982 Bell ME, Eichberg J: Decreased incorporation of [3H]Inositol and [3H]Glycerol into glycerolipids of sciatic nerve from the streptozotocin diabetic rat. J Neurochem 45:465-69, 1985 Berti-Mattera L, Peterson R, Bell M, Eichberg J: Effect of hyperglycemia and its prevention by insulin treatment on the incorporation of 32 P into polyphosphoinositides and other phospholipids in peripheral nerve of the streptozotocin diabetic rat. J Neurochem 45:169298, 1985 Clements RS, Stockard CR: Abnormal sciatic nerve myo-inositol metabolism in the streptozotocin-diabetic rat. Diabetes 29: 227-35, 1980 Matuoka K, Fukami K, Nakainshio 0, Kawai S., Takenawa T: Mitogenesis in response to PDGF and bombesin abolished by micro injection of antibody to PIP2. Science 239:640-43, 1988 Carney DH, Scott DL, Gordon EA, LaBelle EF: Phosphoinositides in mitogenesis: neomycin inhibits thrombin-stimulated phosphoinostide turnover and initiation of cell proliferation. Cell42:479-88,1985 Sylvia V, Curtin G, Norman J, Stec J, Busbee D: Activation of a low specific activity form of DNA polymerase alpha by inositol 1,4 bisphosphate. Cell 54:651-58, 1988 Cocco L, Martelli AM, Gilmour RS, Ognibene A, Mauzoli FA, Irvine

DIABETES, VOL. 41, AUGUST 1992

43.

44. 45. 46. 47. 48. 49. 50. 51.

RF: Rapid changes in phospholipid metabolism in the nuclei of Swiss 3T3 cells induced by treatment of the cells with insulin-like growth factor I. Biochem Biophys Res Comm 154:1266-72, 1988 Martelli AM, Gilmour RS, Neri LM, Manzoli L, Coups AM, Cocco L: Mitogen-stimulated events in nuclei of Swiss 3T3 cells: evidence for a direct link between changes of inositol lipids, protein kinase C requirement and the onset of DNA synthesis. FEBS Lett 283:24346, 1991 Busa WB, Gimlich RL: Lithium-induced teratogenesis in frog embryos prevented by a polyphosphoinositide cycle intermediate or a diacylglycerol analog. Dev Biol 132:315-24, 1989 Whiting PH, Palmano KP, Hawthorne JN: Enzymes of myo-inositol and inositol lipid metabolism in rats with streptozotocin-induced diabetes. Biochem J 179:549-53, 1979 Anderson LL, Trinh M, Flick PK: Early effect of myo-inositol deficiency on phosphatidylinositol metabolism in rat liver. Biochem Biophys Acta 917:341-43, 1987 Prpic V, Blackmore PF, Exton JH: Myo-inositol uptake and metabolism in isolated rat liver cells. J Biol Chem 257:11315-22, 1982 Spector R: The specificity and sulfhydryl sensitivity of the inositol transport system of the central nervous system. J Neurochem 27:229-36, 1976 Takenawa T, Tsumita T: Myo-inositol transport in plasma membrane of rat kidney. Biochem Biophys Acta 373:106-14, 1974 Moyer JD, Malinowski N, Napier EA, Strong J: Uptake and metabolism of myo-inositol by L1210 leukaemia cells. Biochem J 254: 95-100, 1988 Kennedy ED, Challiss RAJ, Ragan Cl, Nahorski SR: Reduced inositol polyphosphate accumulation and inositol supply induced by lithium in stimulated cerebral cortex slices. Biochem J 267:781-86, 1990

997

Phosphoinositide metabolism in the developing conceptus. Effects of hyperglycemia and scyllo-inositol in rat embryo culture.

Culture of the postimplantation rat conceptus in hyperglycemic medium causes developmental abnormalities and is associated with a diminished water-sol...
1MB Sizes 0 Downloads 0 Views