INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE FACILITATIVE GLUCOSE transporters (GLUT) are a family of proteins that transport glucose and coordinate this transport with intracellular use of glucose (1, 7, 10). The distribution of GLUT isoforms in the tissues is relatively specific. For example, cardiac and skeletal muscles primarily express GLUT-4 (1, 7, 10, 12) but the liver expresses GLUT-2 (26). In most mammalian cells, the entry of glucose into metabolic pathways initially depends on facilitative diffusion of glucose across cellular plasma membranes via the transporters.

Several lines of evidence have shown that during hypoxia or ischemia, oxidative metabolism decreases greatly, while anaerobic glycolysis increases to partially compensate for ATP generation (9, 14, 16, 20-22, 27). Such increase in anaerobic glycolysis is an important strategy of cell adaptation to hypoxia/ischemia, especially in excitable tissues such as the brain, heart, and skeletal muscle because their ionic homeostasis and functional integrity are mainly dependent on a constant and relatively large supply of energy. Because GLUT protein kinetics are saturable and thus rate limiting (11), glucose transport becomes potentially limited to energy supply during hypoxia/ischemia. Therefore, the regulation of membrane GLUT protein number would be an important mechanism for tissue adaptation and possible survival in hypoxic/ischemic stress (2). Indeed, GLUT proteins are regulatable in response to a number of environmental and intrinsic factors, including hypoxia (1, 3, 6-8, 10, 15, 17, 24). Because the balance of energy production and use is a key factor for cell survival during energy deprivation, differences in GLUT modulation may be at the basis of the variance in hypoxic/ischemic tolerance or susceptibility among various tissues, cell types, or extent of maturation.

So far, however, little is known regarding the regulation of GLUT mRNA and protein in the heart and skeletal muscles with prolonged hypoxia in both the adult and immature animals. This information is important for designing functional studies related to GLUT transport and glucose uptake and use during prolonged hypoxia because the responses to chronic hypoxia vary among tissues and ages (28, 30). Therefore, the purpose of this work is to study whether 1) chronic hypoxia alters GLUT mRNA and protein levels in heart and skeletal muscles, 2) there is a difference between the heart and limb skeletal muscle in response to stress, and 3) there are age-related differences, especially between postnatal developing rats and adult rats (7, 23, 30, 31). Our hypothesis is that GLUT mRNA and protein level are increased by prolonged O₂ deprivation, but this increase may depend on GLUT isoform, tissue, and age.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals and reagents. EcoR I and BamH I restriction endonucleases were bought from New England BioLabs (Beverly, MA), and [-³²P]deoxycytidine 5'-triphosphate was from Amersham (Arlington Heights, IL). PCR II vectors and TA OneShot INVF' competent Escherichia coli cells were purchased from Invitrogen (San Diego, CA). Total RNA isolation reagent (TRIzol), random primers DNA labeling kits, sodium dodecyl sulfate solutions (SDS, 10%), and 0.5 M EDTA solution were bought from GIBCO BRL (Grand Island, NY). Agarose was purchased from FMC BioProducts (Rockland, ME), and 3-(N-morpholino)propanesulfonic acid (MOPS) and formamide were from American Bioanalytical (Natick, MA). RNA-size ladder (0.24-9.5 kb), 50× Denhardt's solution, salmon or herring testis DNA, E. coli tRNA, Triton X-100, sodium deoxycholate, and protease inhibitors (leupeptin, etc.) for Western blots were purchased from Sigma Chemical (St. Louis, MO). All other chemicals and reagents for Western blots (broad-range prestained protein standard, dithiothreitol, N,N,N',N',-tetramethylethylenediamine ammonium persulfate, bromophenol blue, etc) were purchased from Bio-Rad (Hercules, CA) and the enhanced chemiluminescence (ECL) Western blotting system was purchased from Amersham (Little Chalfont, Buckinghamshire, UK).

Probes. A 2.086-kb rat GLUT-4 cDNA (containing the full coding region of GLUT-4) in EcoR I sites of Pbluescript vector, namely pIRGT, was a gift from Dr. M. Mueckler (13) at Washington University. A 2.6-kb rat GLUT-1 cDNA clone in pBS⁺ vector, namely prGT3, was a gift from Dr. M. Birnbaum (4) at Harvard University. The GLUT-1 cDNA was subcloned into EcoR I sites of PCR II vector. Both cDNAs were excised from the plasmids by digestion with EcoR I and then purified with PCR Prep DNA purification system of Promega (Madison, WI) and used for random probe labeling. A 1.25-kb rat glyceraldehyde 3-phosphate dehydrogenase (GAPDH) cDNA probe was used for assaying the response of the housekeeping gene.

Animals. Sprague-Dawley rats at postnatal day 0 (P0), day 3 (P3), day 30 (P30), and day 90 (P90) were purchased from Camm Research, Wayer, NJ. Each studied group included four to eleven rats that came from three to five litters. Fetal tissues were pooled because of the small size of their organs.

Hypoxia induction. The level and duration of hypoxia were chosen on the basis of the following two considerations: 1) the hypoxic stress should be severe enough to produce a significant effect without jeopardizing the animal's life and 2) the resulting PO₂ level is such that it is clinically relevant and not infrequently encountered. On the basis of the literature (2, 3, 5, 6, 17, 24, 25) and our previous work (28, 30), we have chosen to expose rats to a fractional inspired O₂ of 9 ± 0.5%. This hypoxic level leads generally to a fall of arterial PO₂ to ~26-30 Torr (25), which is equal to moderate to severe hypoxia. Because we (28, 30) and others (25) have observed that 3-6 wk of hypoxia could cause significant changes in membrane protein expression, we exposed the rats in an isobaric chamber for 1 mo in the present studies. Two age groups, i.e., immature (or developing) and adult rats, were used and placed in an isobaric chamber with controlled O₂ level. The former group was exposed to hypoxia starting from P3 and the latter group starting from P90. Control litters were kept in room air. After a 30-day exposure, the animals were decapitated with age-matched controls after inhalation anesthesia with methoxyflurane. Heart and skeletal muscles (right hindlegs) were rapidly removed and stored at 80°C until use.

At the end of the chronic hypoxic period, body weights (Table 1) and skeletal muscle size (25) were significantly smaller than control animals. However, despite the decrease in body weight in the chronically hypoxic animals, their hearts increased in size and weight. For example, the weight of the hearts in the immature rats increased by >50% after chronic hypoxia (Table 1).

RNA extraction. Total tissue RNA was extracted as previously described (30). In brief, the frozen tissues were homogenized with a polytron (Brinkmann, Westbury, NY) in TRIzol solution (1 ml/100 mg of tissue). After the addition of chloroform (0.2 ml/1 ml of TRIzol reagent), sample solutions were centrifuged for phase separation. The aqueous phase was then transferred, and RNA was precipitated by adding isopropanol. The RNA pellet was washed with 75% ethanol twice and redissolved in diethyl pyrocarbonate (DEPC) water with 0.25% SDS. RNA was quantitated by measuring its absorbance at 260 nm with a spectrophotometer (DU-70, Beckman, Fullerton, CA). The ratio of RNA absorbance at 260/280 nm and RNA pattern on gel electrophoresis were used to evaluate RNA quality. The yield of RNA per milligram tissue was within the range indicated by the manufacturer for TRIzol solution, i.e., 1-2 µg/mg tissue for heart and skeletal muscle. There was no statistical difference between hypoxia and control tissues in the yield of RNA per tissue weight (P > 0.05).

Northern blots. The methods are the same as in our previous publication (30). After denaturation by heating at 70°C for 15 min in a denaturing solution (20 mM MOPS, 5 mM sodium acetate, 1 mM EDTA, 2.2 M formaldehyde, and 50% formamide, pH 7.2), RNA was quickly chilled and then loaded on 1% agarose gel prepared in MOPS buffer with 2.2 M formaldehyde and ethidium bromide (0.15 µg/ml). Two RNA samples from any given pair of tissues (hypoxia vs. control) were always equally loaded on the same gel for simultaneous hybridization. The electrophoresed gels were inspected under ultraviolet light to evaluate RNA quality. Equal RNA loading was verified by comparison of the 28S and 18S ribosomal RNA bands in all lanes. The RNAs were then transferred from gel to nylon membranes (nylon-1 from GIBCO BRL or maximum strength NYTRAN from Schleicher & Schuell) by capillary blotting overnight in 10× saline sodium citrate (SSC; 1× SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7.0), and immobilized by ultraviolet crosslinking (UV Stratalinker, Stratagene, La Jolla, CA). After transfer, both the gel and nylon membrane were examined under ultraviolet light to confirm complete transfer.

Prehybridization was performed at 42°C for 2-4 h in buffer consisting of 50% formamide, 5× Denhardt's solution, 5× SSC, 0.1% SDS, and 200 µg/ml denatured herring or salmon testis DNA. The blots were then hybridized at 42°C for 18 h in the same buffer containing ³²P-labeled cDNA probe [1-3 × 10⁶ counts/min (cpm)/ml for GLUT probes and 0.5-1 × 10⁶ cpm/ml for GAPDH probe]. The probes were produced with random-primer DNA labeling kits and were denatured by boiling for 5 min before use. The hybridized blots were rinsed for 5 min at room temperature in 2× SSC plus 0.1% SDS and then serially washed at 42°C as needed. The wet blots were then sealed in plastic bags and exposed to X-ray film (Kodak X-Omat AR or XRP-5, Eastman Kodak, Rochester, NY) with one or two intensifying screens (DuPont Cronex, Lighting plus-T). After development of the autoradiographs with GLUT-1 probe, blots were stripped and checked after 24-h film exposure to determine whether there was any signal left. The stripped blots were then used for hybridization with GLUT-4 and GAPDH probes. In this study, we reused the blots for a maximum of three times. Because of larger abundance of GLUT-4 mRNA than GLUT-1 (see below and Refs. 7, 10), films were exposed for 12-16 h for detection of GLUT-4 mRNA but 32-48 h for GLUT-1.

Western blots. The detergent protein samples were prepared in triple-detergent buffer (8, 24). The tissues were weighed, rinsed with ice-cold phosphate-buffered saline and then homogenized with a Polytron homogenizer in cold triple-detergent buffer [50 mM tris(hydroxymethyl)aminomethane (Tris), pH 7.4, 150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, and 0.1% SDS] plus protease inhibitor cocktail [0.2 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 1 µM pepstatin, 20 µM leupeptin, 1 µg/ml aprotinin, and 1 mM EDTA]. The homogenates were spun at 1,000 g for 5 min, and the supernatant was agitated gently at 4°C for 30 min. After spinning at 27,000 g for 30 min, the supernatant was stored at 80°C until used. Protein concentration was assayed by Lowry's method with a Bio-Rad kit. The samples were mixed with SDS (2%), dithiothreitol (100 mM), glycerol (10%), and bromophenol blue (<0.1%) and boiled for 5 min before electrophoresis. Fifty micrograms of protein was resolved in each lane on 10% SDS-polyacrylamide gel in a gel box (Protean II, Bio-Rad). The fractioned protein was electrically transferred to a nitrocellulose membrane (PROTRAN, Schleicher & Schuell) using the Trans-Blot cell (Bio-Rad).

Western blot detections were performed with ECL Western blot system (Amersham). In brief, the membrane was blocked by incubation in Tween-Tris-buffered saline (TTBS; 0.05% Tween 20, 20 mM Tris, 500 mM NaCl, pH 7.5) containing 5% nonfat milk for 1-2 h. After rinse with TTBS, the membrane was incubated in TTBS containing rabbit polyclonal antibodies against GLUT-4 (1:200-400) or GLUT-1 (1:500-1,000) for 2-4 h at room temperature with gentle agitation. The GLUT antibodies were purchased from Charles River East Acres Biologicals (Southbridge, MA). The blot was washed in TTBS and then incubated in horseradish-peroxidase-linked whole antibody against rabbit immunoglobulin from donkey (1:1,000) for 1 h. Finally, GLUT protein signal was detected by incubating the membrane in a mixture of detection reagents 1 and 2 for 1 min and then exposing it to Kodak X-ray film for autoradiography.

Data analysis. Exposed films from either Northern or Western blots were developed by a Kodak M35A X-OMAT processor. Autoradiographs of Northern blots were quantitated by an imaging analyzer (Molecular Dynamics). The relative optical density of signal bands was used as an index of mRNA amount. Because relative optical densities from different blots cannot be directly compared, we treated our data as follows: for any given pair (hypoxia vs. control), signal density obtained from control was considered 100% and then the density value from hypoxia was converted to percent change from control. Two or more values from the same pair (but from different experiments) were averaged. With respect to the maturational data, adult mRNA levels were considered 100% and signal densities at the other ages on the same blot were compared with the adult level. Finally, a two-tail paired t-test was applied. Values were expressed as means ± SE and means were considered significantly different when P values were <0.05. Data from Western blots were processed in the same way.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Developmental profiles of heart and skeletal muscle GLUT subtypes. GLUT-4 and GLUT-1 mRNA levels were measured at various ages with respective cDNA probes. As shown in Figs. 2-4, GLUT-1 and four probes specifically hybridized each to a single band of RNA transcript, which was localized between 28S and 18S with a size of 2.8-2.9 kb.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Development of GLUT-1 (A) and GLUT-4 (B) mRNA in rat heart. Quantification of GLUT mRNA signal density was made from Northern blots. Adult mRNA levels were considered 100% and signal densities at other ages on the same blot were compared with the adult level. E19, embryonic day 19; P0, postnatal day 0; P30, postnatal day 30; A, adult. * P < 0.05 compared with adult level. Note high density of GLUT-1 mRNA in fetuses and low density in adult, with the opposite pattern for GLUT-4 mRNA.

The density of GLUT-4 mRNA was much higher in the adult than in the newborn and the opposite was true for GLUT-1 mRNA. During development, GLUT-4 and GLUT-1 displayed very different patterns of change in mRNA (Fig. 1). For example, GLUT-4 mRNA level was very low in the heart of fetuses (embryonic day 19) and increased with age. The density of fetal heart GLUT-4 mRNA was only 10% of the adult level and doubled by the neonatal period with continued increase after birth. At P30, GLUT-4 mRNA density was very similar to that in the adult. In sharp contrast to GLUT-4, the density of GLUT-1 mRNA was highest in the heart of fetuses at embryonic day 19, ~350% of the adult level. By birth, the density had decreased to ~270% of the adult level and most of the decrease in GLUT-1 occurred in the first 4 postnatal wk, with GLUT-1 mRNA level reaching adult levels by P30 (Fig. 1).

In the skeletal muscles, GLUT mRNAs show a similar pattern during development, i.e., there is a low density of GLUT-4 and a high density of GLUT-1 in early life and an opposite expression in the adult. As shown in Fig. 2, GLUT-4 mRNA density in the embryonic skeletal muscles was much lower than that in the adult. At the age of 30 days, levels of both GLUT-1 and GLUT-4 mRNA were close to those at the age of 120 days.

View larger version (89K): [in this window] [in a new window]   Fig. 2.   Major difference in GLUT-4 mRNA expression between embryonic (E) and adult (A) skeletal muscles. Note that the signal was very low in embryonic muscles but >10-fold higher in the adult muscles.

Effects of chronic hypoxia on heart GLUT at different ages. Chronic hypoxia decreased GLUT-4 mRNA level in the adult heart (Fig. 3) by an average of 25 ± 9% (n = 11) and this was significantly different from control (P < 0.02). GLUT-4 protein, however, did not show a parallel decrease, and the average level of GLUT-4 protein in the hypoxic heart was 95 ± 3.7% of control (P > 0.05, n = 4).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Effect of chronic hypoxia on GLUT in adult rat heart. Top 2 panels are results from Northern blots and bottom 2 are from Western blots. C, control; H, hypoxia. High background in left panel of Northern blot (GLUT-1 mRNA) resulted from a relatively long-term film exposure because of relatively low signals. Note that chronic hypoxia decreased both GLUT-1 and GLUT-4 mRNA density but had little effect on the protein levels.

Chronic hypoxia caused a qualitatively similar effect on GLUT-1 mRNA. GLUT-1 mRNA level decreased (Fig. 2) in four of five hearts studied; the fifth had a similar level as control. The average GLUT-1 mRNA of all five adult hypoxic hearts was 72 ± 13% of control. GLUT-1 protein level was 93.7 ± 2.6% of control, and the difference between hypoxia and control was not statistically significant (P > 0.05, n = 4).

In contrast to the adult heart, the immature heart had a different response to chronic hypoxia. As shown in Fig. 4, both GLUT-4 and GLUT-1 mRNAs showed no appreciable decrease with the hypoxic stress. The change with hypoxia in GLUT-4 and GLUT-1 mRNA in the immature hearts was small (95.3-104.3% of control, P > 0.05, n = 9). However, GLUT proteins showed a different response. Both GLUT-4 and GLUT-1 proteins increased in immature hearts by an average of 86.8 ± 5.5 and 61.1 ± 7.9% of control, respectively (P < 0.01, n = 4).

View larger version (41K): [in this window] [in a new window]   Fig. 4.   Effect of chronic hypoxia on GLUT in the immature rat heart. Top 2 panels are results from Northern blots, and bottom 2 are from Western blots. Note that chronic hypoxia did not change GLUT-1 and GLUT-4 mRNA levels in any major way, but increased GLUT-1 and GLUT-4 protein levels.

Effects of chronic hypoxia on skeletal muscle GLUT in the immature and adult rat. Although GLUT-4 mRNA in skeletal muscles of adult rats exposed to chronic hypoxia was not different from control (Figs. 5 and 6), GLUT-4 protein was upregulated. On average, GLUT-4 mRNA in hypoxic skeletal muscles was close to control level (P > 0.05, n = 7), whereas GLUT-4 protein level increased by 20.3 ± 3.4% (P < 0.05, n = 4).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Regulation of GLUT-4 mRNA in heart and skeletal muscles of immature and adult rats in response to chronic hypoxia. In each group, GLUT-4 mRNA level of control rats is considered 100% and is compared with the level in hypoxic rats. Note that GLUT-4 mRNA increased in the immature skeletal muscles but did not change in the adult ones after chronic hypoxia. Heart GLUT-4 mRNA decreased in the adult but did not show an appreciable change in the immature rats. * P < 0.05 compared with their control levels.

View larger version (59K): [in this window] [in a new window]   Fig. 6.   GLUT-4 mRNA regulation in skeletal muscles of immature and adult rats. P30, immature group (hypoxia exposure from newborn to P30). P120, adult group (hypoxia exposure from P90 to P120). Note that GLUT-4 mRNA increased in skeletal muscle of immature rats with little change in the adult in response to chronic hypoxia. GAPDH mRNA did not show any change at both ages.

The regulation of GLUT-1 mRNA and protein was similar to that of GLUT-4. GLUT-1 protein increased with the hypoxic stress (116.9 ± 5.4 vs. 100%, P = 0.05, n = 4) but there was no significant change at the mRNA level (P > 0.1, n = 5).

GLUT mRNA in the immature skeletal muscle had a different response to chronic hypoxia. GLUT-4 mRNA level increased to 168 ± 17% of control (P < 0.01, n = 10) with the hypoxic stress (Figs. 5 and 6). Average level of GLUT-1 mRNA was 37% over the control, but the difference was not statistically different (P > 0.05, n = 5). At the protein level, both GLUT-4 and GLUT-1 increased, as seen in Fig. 7. On average, GLUT-4 increased by 59.7 ± 8.5% (P < 0.01, n = 5) and GLUT-1 by 83.4 ± 7.9% over control (P < 0.01, n = 4).

View larger version (35K): [in this window] [in a new window]   Fig. 7.   GLUT proteins increased in the immature skeletal muscle with chronic hypoxia. Note that both GLUT-1 and GLUT-4 protein density increased in the hypoxic muscles.

Stable level of GAPDH in heart and skeletal muscle with chronic hypoxia. We used GAPDH as an index of the response of a housekeeping gene. Northern blot analysis with GAPDH cDNA probe showed that GAPDH cDNA probe hybridized to a single band that was below 18S, with a size of ~1.4 kb. Chronic hypoxia produced little effect on GAPDH mRNA level. As shown in Figs. 3 and 6, there was little change in GAPDH mRNA levels in the immature hearts and in skeletal muscles in both immature and adult rats exposed to chronic hypoxia.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This is the first study on the effect of low and prolonged O₂ levels on GLUT mRNA and protein in immature and adult heart and skeletal muscle. One major finding in this work is that GLUT regulation depends on the specific tissue and age of animal (Table 2). Because GAPDH mRNA showed little change, the alterations in GLUT mRNA level are very likely to be a specific response to the chronic O₂ deprivation. Alternatively, it is possible that these changes are due, in part, to the effect of chronic hypoxia on homeostatic mechanisms that, in turn, influence GLUT mRNA and protein. For example, chronic hypoxia can alter the level of a number of hormones or neurotransmitters that have their own effects on GLUT biochemistry. At this moment, we cannot distinguish between the effect of hypoxia per se from those that hypoxia can cause indirectly.

Both GLUT-1 and GLUT-4 are regulatable proteins. Although GLUT-1 and GLUT-4 are both involved in glucose transport in heart and skeletal muscles and responsive to insulin stimulation (7, 10), their development and distribution are very different. As shown in this work, heart GLUT-1 mRNA had the highest density in the fetuses and decreased after birth, whereas GLUT-4 mRNA increased with age and peaked in the adult animal. GLUT-1 is most likely a nonspecific glucose transporter in early life because it is expressed in almost all fetal tissues examined (7), whereas GLUT-4 is a more specialized insulin-sensitive glucose transporter, specifically developed in the heart, skeletal muscle, and fat cells (7, 10). The developmental patterns of these two GLUTs suggest that GLUT-1 transports glucose in the heart and muscles of fetuses and its role is gradually replaced by GLUT-4. Similarly, this study also shows that both GLUT-1 and GLUT-4 are regulatable by hypoxic stress, although the direction and amplitude of the regulation depends on age and tissue.

Difference in GLUT regulation as a function of age. Our results show that chronic hypoxia induces a response in the immature rats that is different from that in the mature rats with respect to GLUT protein regulation: the immature rats markedly increase GLUT protein in response to low O₂ in both the heart and skeletal muscles, unlike the adult, which showed either no (heart) or slight (skeletal muscles) increase. Because GLUT functional activity depends on the amount of the protein available (6, 7, 10, 11), the differential regulation of GLUT protein between the two ages likely reflects a difference in glucose transport. We would therefore postulate that if glucose metabolism and ATP production are basic to survival mechanisms during O₂ deprivation, immature muscle tissues are more dependent on glucose transport and anaerobic glycolysis than mature ones. Such a difference between immature and mature tissues was also observed in our previous electrophysiological studies on the central nervous system (29). During anoxia only (no inhibition of glycolysis), the increase in extracellular K⁺ was as high as 10-fold in the adult brain but only by 0.8-fold in the newborn compared with age-matched controls (29); after blocking glycolysis with iodoacetic acid, anoxia increased extracellular K⁺ in the newborn brain by a much larger proportion than in the adult brain (29). Results from other laboratories provide additional evidence that the immature brain can increase its glucose use, whereas the mature one decreases it during hypoxia (5). Because 1) O₂ consumption in excitable tissues is lower in the newborn than in the adult and 2) anaerobic metabolites such as lactate can be used efficiently in early postnatal life (18), increasing anaerobic glucose use with enhanced glucose transport in immature tissues would be useful and may be an important factor that renders the immature tissues more tolerant to hypoxic stress than adult ones.

Difference in GLUT regulation as a function of tissue. The present work demonstrates that there is a major difference in GLUT regulation (at both protein and mRNA levels) between heart and skeletal muscles. For example, the adult heart had a decrease in GLUT mRNA and little change in GLUT protein, but the adult skeletal muscle had an increase in GLUT protein without an appreciable change in GLUT mRNA. This finding in the heart is opposite to that of a previous report that has demonstrated an upregulation in GLUT-1 of adult rat heart (24). One potential reason for this is the duration of hypoxia. The hypoxic period used in that previous study (24) was shorter than ours in this report. In fact, that study showed that a longer period of hypoxia caused less increase in cardiac GLUT-1 mRNA than 2-day hypoxia (24). Our results raise the possibility that GLUT regulation depends on the duration of O₂ deprivation. Indeed, we have observed that 3- to 4-day hypoxia at the same O₂ level used for the 30-day exposure causes an increase in GLUT mRNA level in the adult heart (data not shown). We therefore speculate that during hypoxic stress the changes in GLUT mRNA are different in the early stage than after a longer period of hypoxia. It is possible that the differences between our results and those reported (24) are related to the difference in the length of the hypoxic duration.

Our results demonstrate that an upregulation of glucose transporter system takes place in the skeletal muscle but not in the heart in the adult with chronic hypoxia. Because it is known that adult skeletal muscles are more resistant to hypoxia than the heart, we raise the question again as to whether the tolerance of skeletal muscles is related to the upregulation of glucose protein and glucose transport. This metabolic adaptation would allow skeletal muscle to use glucose and rely on anaerobic glycolysis for energy compensation. In addition, our data suggest that the change in GLUT protein may not be necessarily coupled to that of GLUT mRNA in response to chronic hypoxia. For example, adult skeletal muscle increases in GLUT-4 protein but shows no change in GLUT-4 mRNA.

It is interesting that there was a major difference between the heart and skeletal muscles in terms of change in size with chronic hypoxia. In our previous (28, 30) and present studies and in other studies (25), both body and skeletal muscles have been shown to decrease in size (or weight) with prolonged hypoxia. As measured in this study, however, rat hearts enlarged with chronic hypoxia (Table 1). The reason for this is likely to be related to an increase in working load because hypoxic stress 1) increases systemic and pulmonary vascular resistance and 2) enhances cardiac output to provide more blood and glucose to peripheral organs (25). It is also interesting to note that the brain (30), like the heart but unlike skeletal muscles and diaphragm (data not shown), downregulates its GLUT mRNA in the adult. It is possible then that more oxidative organs, such as the brain, have a different strategy during hypoxic stress. This strategy depends on decreasing metabolic rate rather than increasing anaerobic glycolysis. An important consequence of this difference is that organs such as the brain will not pollute themselves with acidity because the need for generating ATP through anaerobic glycolysis is minimized.

Perspectives

In summary, we have studied GLUT regulation in the heart and skeletal muscles in immature and adult rats with chronic hypoxia. Our data suggest that, in response to chronic hypoxia, 1) GLUT regulation depends on tissue type and 2) there are qualitative differences in GLUT regulation between immature and mature animals. We believe that tissue GLUT regulation and anaerobic use of glucose in a given tissue is part of an overall strategy to adapt to long-term O₂ deprivation.