Developmental switch in brain nutrient transporter expression in the rat

Am J Physiol Endocrinol Metab 285: E1127–E1134, 2003; 10.1152/ajpendo.00187.2003.

Developmental switch in brain nutrient transporter expression in the rat Susan J. Vannucci1 and Ian A. Simpson2 1

Department of Pediatrics, Columbia University, New York, New York 10032; and 2Department of Neuroscience and Anatomy, Milton S. Hershey Medical Center, Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033 Submitted 25 April 2003; accepted in final form 18 July 2003

Vannucci, Susan J., and Ian A. Simpson. Developmental switch in brain nutrient transporter expression in the rat. Am J Physiol Endocrinol Metab 285: E1127–E1134, 2003; 10.1152/ajpendo.00187.2003.—Normal development of both human and rat brain is associated with a switch in metabolic fuel from a combination of glucose and ketone bodies in the immature brain to a nearly total reliance on glucose in the adult. The delivery of glucose, lactate, and ketone bodies from the blood to the brain requires specific transporter proteins, glucose and monocarboxylic acid transporter proteins (GLUTs and MCTs), respectively. Developmental expression of the GLUTs in rat brain, i.e., 55-kDa GLUT1 in the blood-brain barrier (BBB), 45-kDa GLUT1 and GLUT3 in vascular-free brain, corresponds to maturational increases in cerebral glucose uptake and utilization. It has been suggested that MCT expression peaks during suckling and sharply declines thereafter, although a comparable detailed study has not been done. This study investigated the temporal and regional expression of MCT1 and MCT2 mRNA and protein in the BBB and the nonvascular brain during postnatal development in the rat. The results confirmed maximal MCT1 mRNA and protein expression in the BBB during suckling and a decline with maturation, coincident with the switch to glucose as the predominant cerebral fuel. However, nonvascular MCT1 and MCT2 levels do not reflect changes in cerebral energy metabolism, suggesting a more complex regulation. Although MCT1 assumes a predominantly glial expression in postweanling brain, the concentration remains fairly constant, as does that of MCT2 in neurons. The maintenance of nonvascular MCT levels in the adult brain implies a major role for these proteins, in concert with the GLUTs in both neurons and astrocytes, to transfer glycolytic intermediates during cerebral energy metabolism. cerebral metabolism; blood-brain barrier; ketogenesis; lactate

are dependent on delivery of an adequate supply of metabolizable substrate to the brain during the perinatal period. Glucose is the primary energy substrate for the adult brain, and studies in experimental animals indicate that glucose is also an essential cerebral fuel for fetal and newborn animals under normal physiological con-

NORMAL CEREBRAL DEVELOPMENT AND FUNCTION

Address for reprint requests and other correspondence: S. J. Vannucci, Research Director, Pediatric Critical Care Medicine, Morgan Stanley Children’s Hospital of New York, Columbia Univ. College of Physicians & Surgeons, 3959 Broadway, BHN 10-24, New York, NY 10032 (E-mail: [email protected]). http://www.ajpendo.org

ditions, with additional contributions from lactate in the immediate postnatal period (see reviews in Refs. 6 and 37 and references therein). However, it is well known that during the normal development of both human and rodent brain, there is a switch in fuel utilization from a combination of glucose, lactate, and the ketone bodies (␤-hydroxybutyrate and acetoacetate) in the immature brain to a total reliance on glucose in the adult (8, 23); only under conditions of starvation, diabetes, and hypoglycemia can these alternate substrates support cerebral metabolism in adults (13). The high fat content of both human milk and rodent milk generates circulating ketone body concentrations sufficient to contribute to the energetic and synthetic needs of early cerebral maturation (10, 23, 26, 29). The similarities between human and rodent nutrition support the value of studying the developmental pattern of cerebral ketone body uptake and utilization in suckling rodents. With the onset of suckling and the ingestion of the high-fat milk diet, rat pups become hyperketonemic and remain so during the suckling period (see review in Ref. 26). Whereas cerebral glucose utilization is relatively low in the newborn rat brain (25, 42), rates of ␤-hydroxybutyrate uptake and utilization actually peak during the second and third postnatal weeks before declining to low adult levels after weaning (8, 14, 24). The delivery of both glucose and the monocarboxylic acids from blood to brain requires specific transporter proteins, the facilitative glucose transporter proteins (GLUTs) and the proton-coupled monocarboxylic acid transporter proteins (MCTs), respectively. The GLUT family has recently been expanded and now comprises 12 distinct integral membrane proteins (16). Despite the description of the new family members, GLUT1 and GLUT3 remain the primary isoforms responsible for cerebral glucose uptake and utilization (9, 41). GLUT1 is present in brain as two molecular mass forms of the same protein, with the more heavily glycosylated 55-kDa form present at high concentrations in the microvascular endothelial cells that comprise the blood-brain barrier (BBB) and the less glycosylated 45-kDa form in the choroid plexus and nonvascular The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

0193-1849/03 $5.00 Copyright © 2003 the American Physiological Society

E1127

E1128

BRAIN GLUTS AND MCTS IN DEVELOPMENT

brain; GLUT3 is the predominant neuronal glucose transporter. The levels of both GLUT1 and GLUT3 are low in the immature rat brain, and they increase with cerebral maturation and synaptogenesis after the second postnatal week (39). The proteins responsible for the transport of lactate, pyruvate, and the ketone bodies belong to a separate family of proton-coupled MCTs. A total of eight distinct MCT proteins have now been described, of which MCT1 and MCT2 are the predominant isoforms in newborn and adult rodent brain (1, 2, 11, 17, 18, 28, 32). MCT1 is the major, if not only, ketone body and lactate transporter of the BBB (17, 18, 28) and is also the primary isoform in astrocytes, whereas MCT2 has a more neuronal localization (7, 12, 17, 30, 31). Because of the importance of lactate and ketone body uptake to normal cerebral development, early studies examined MCT1 and MCT2 mRNA (28) and protein (11, 18) in newborn, suckling (postnatal days 14–17), and adult rat brain and reported very high levels of both isoforms during suckling and reduced levels in adult brain. Rates of cerebral energy metabolism, met by either glucose or monocarboxylates, vary regionally (13, 26), as do developmental patterns of GLUT expression (39). The purpose of this study was to investigate more thoroughly the developmental pattern of regional MCT expression in rat brain in vivo during the transition from monocarboxylate metabolism to glucose metabolism. MATERIALS AND METHODS

Animals. Dated pregnant Wistar rats were purchased from Charles River Laboratories, housed individually, and allowed to deliver normally. On the day of birth (postnatal day 1; P1), litters were randomized and adjusted to 10 pups/litter to allow for uniform nutrition during development. At specific ages, pups were randomly removed from the litters for analysis, and remaining litters were readjusted to 10/litter with same-aged pups, until time of weaning at P28. Thereafter, rats were divided by gender and housed in groups of six. Animals were allowed free access to normal rat chow and water throughout. At the time designated for animals to be killed, unsexed animals were anesthetized with halothane and decapitated. Trunk blood was collected and extracted in 0.5 M perchloric acid for the determination of metabolites, as previously described (35). Brains were removed and either frozen in ⫺30°C isopentane or processed for the isolation of cerebral microvessels and vascular-free membranes. Isolation of microvessels and vascular-free membranes. Microvessels were prepared from pooled cortical shells of unsexed pups at P14 and P27, as previously described (39). Briefly, cortical shells were stripped of meninges and surface vessels, pooled to obtain 2 g of tissue, and homogenized in microvessel buffer (MVB, in mM: 15 HEPES, 147 NaCl, 4 KCl, 3 CaCl2, 1.2 MgCl2, 5 glucose, and 0.5% BSA, pH 7.4) with protease inhibitors (1 ␮g/ml each of aprotinin, leupeptin, and pepstatin) and 50 ␮M 4-(2-aminoethyl)-benzenesulfonyl fluoride. Homogenates were centrifuged at 1,000 g and 4°C for 10 min. Pelleted vessels were separated from myelin by centrifugation through 17% Dextran (Sigma) at 3,600 g and 4°C for 20 min and filtered through 125-␮m nylon mesh onto a column of glass beads supported by 40-␮m nylon mesh. Columns were washed extensively with MVB, and microvesAJP-Endocrinol Metab • VOL

sels were recovered from the beads in BSA-free MVB, resuspended in Tris-EDTA-sucrose (TES; 20:1:255 mM, pH 7.4), and sonicated. Vascular-free membranes were obtained by centrifugation of the initial supernatant at 300,000 g and 4°C for 20 min. Membranes were washed to remove BSA from the original buffer and resuspended with sonication in TES. For the preparation of regional vascular-free membranes, sections of cortex, hippocampus, thalamus, and cerebellum were dissected from five rats of each age, pooled, homogenized in MVB without BSA, and centrifuged at 1,000 g to remove vessels and myelin. The resulting supernatant was centrifuged to obtain total vascular-free membranes. Sonicated microvessels and vascular-free membranes were analyzed for protein content with bicinchoninic acid reagent (Sigma Chemicals, St. Louis, MO). Western blot analysis. Microvessel and membrane samples were solubilized and separated on 10% SDS-polyacrylamide gels, as previously described (39), and proteins were transferred electrophoretically to nitrovinyl membranes for Western blot analysis by chemiluminescence (Renaissance ECL, DuPont). All gels included an adult rat brain microsomal brain standard for purposes of normalization and quantitation, as previously described (39). All antisera were antipeptide, rabbit polyclonal antibodies. GLUT1 and GLUT3 antisera have been previously described (21). Antibodies to MCT1 and MCT2 were raised against their respective COOH-terminal sequences [MCT1 (C)NTHNPPSDRDKESS; MCT2 (C)LQNSSGDPAEEESP] and commercially prepared (Zymed Laboratories, San Francisco, CA). For purposes of quantitation as well as presentation, multiple exposures of each blot were obtained, and the optical density measurements were made on the lightest exposure to ensure that all samples were within the linear response range of the film. The values for each experimental sample was expressed relative to the brain standard, in arbitrary standard units (39). In situ hybridization. Cryosections (16 ␮m) were analyzed for specific mRNA expression by in situ hybridization using 35 S-labeled riboprobes, as previously described (17, 41). RESULTS

The levels of circulating metabolites (glucose, lactate, and ␤-hydroxybutyrate) in the rat pups used in this study are presented in Table 1 and are in good agreement with previous studies in developing rats (26). Relative to normal adult values, levels of the predominant ketone body, ␤-hydroxybutyrate, are high

Table 1. Circulating metabolites in postnatal rats Age, days

Glucose

Lactate

␤-Hydroxybutyrate

7 10 14 17 21 27

5.74 ⫾ 0.18 7.45 ⫾ 0.20 7.14 ⫾ 0.74 8.21 ⫾ 0.35 7.76 ⫾ 0.54 7.02 ⫾ 0.34

1.57 ⫾ 0.34 1.45 ⫾ 0.36 1.48 ⫾ 0.26 1.45 ⫾ 0.36 1.37 ⫾ 0.16 1.38 ⫾ 0.24

0.62 ⫾ 0.17 0.80 ⫾ 0.11 0.91 ⫾ 0.24* 0.55 ⫾ 0.08 0.32 ⫾ 0.19 0.15 ⫾ 0.02

Trunk blood was obtained from postnatal rats when they were killed and extracted in 0.5 M perchloric acid. Glucose, lactate, and ␤-hydroxybutyrate were analyzed fluorometrically, as described in MATERIALS AND METHODS. Values are means ⫾ SE for 6–10 animals/ age expressed in mM. * Significantly different from postnatal days (P)17-P27, P ⬍ 0.05.

285 • NOVEMBER 2003 •

www.ajpendo.org

BRAIN GLUTS AND MCTS IN DEVELOPMENT

at P7, peak at P14, and then decline to adult levels by P27, just before weaning, as reported by others (26). The developmental pattern of ketogenesis is mirrored in the extent of expression of the monocarboxylate transporter MCT1, but not of MCT2, as shown in Fig. 1. MCT1 mRNA is abundantly expressed at both P7 and P14, with a widespread but heterogeneous pattern suggesting expression by multiple cell types, including the pyramidal cell layer of the hippocampus. A large component of the MCT1 expression is punctate in appearance, similar to that routinely observed for GLUT1 mRNA in the endothelial cells of cerebral microvessels (41). As the circulating levels of ketone bodies decline with development, so does the level of MCT1 expression; aspects of the punctate vascular expression are still apparent at P21 but are essentially gone at P27, characteristic of the adult situation (17). Interestingly, MCT2 mRNA expression does not appear to be as developmentally regulated. We (17) and others (30, 31) have previously reported that MCT2 mRNA and

Fig. 1. Developmental expression of monocarboxylic acid transporter (MCT)1 and MCT2 mRNA in rat brain. Cryosections of rat brains (16 ␮m) from postnatal day 7 (P7) through P27 were analyzed for MCT1 and MCT2 mRNA expression by in situ hybridization with 35Slabeled riboprobes, as described in MATERIALS AND METHODS. Autoradiograms of representative adjacent coronal sections at the level of the anterior hippocampus/thalamus are shown. T, thalamus; DG, dentate gyrus. AJP-Endocrinol Metab • VOL

E1129

protein in brain are primarily neuronal, and the autoradiographic images presented in Fig. 1 support this conclusion. MCT2 mRNA expression is pronounced in the cortex of the P7 rat, especially in layers I and II, and this pattern is maintained throughout development, in agreement with patterns of ␤-hydroxybutyrate uptake in the adult (13). Developmental alterations in expression are not as clearly seen, with the exception of the dentate gyrus of the hippocampus (arrows), where MCT2 mRNA expression coincides temporally with neuronal maturation, analogous to the neuronal glucose transporter GLUT3 (40). The expression of MCT1 mRNA at P14 is especially intense in the thalamus (see T in Fig. 1). To determine whether this intensity is due to one, or multiple, cell types at this stage of development, slides were dipped in photographic emulsion, and the signal was visualized with epiluminescence, which allows for simultaneous visualization of hematoxylin-stained tissue. Results obtained for MCT1 mRNA were compared with in situ hybridization analysis of GLUT1 and GLUT3 on adjacent coronal sections, and the epiluminescent, high-power bright-field micrographs are depicted in Fig. 2. At P14, MCT1 mRNA is highly concentrated in the microvessels in this region, with expression in other neural cells, especially cells with large, round, pale-staining nuclei typical of neurons. GLUT3 mRNA is confined to neurons, and GLUT1 expression is generally low at this age, as we have reported before (39, 40). This is in sharp contrast to the situation observed at P27, when GLUT1 is highly expressed in the cerebral microvessels and other cell types, especially those with small, irregular nuclei more indicative of glia, and the intensity of MCT1 mRNA expression is reduced in the microvessels as well as other cell types. To determine whether the dramatic switch in nutrient transporter expression in the BBB is reflected in levels of the protein, isolated microvessels and vascular-free cortical membranes were prepared at P14 and P27 and analyzed for MCT1 and GLUT1 content by Western blot, as shown in Fig. 3. The developmental increase in GLUT1 expression in microvessels, detected as the more highly glycosylated 55-kDa form as well as the 45-kDa nonvascular form, is in agreement with our previous reports. The high level of MCT1 in P14 microvessels, in association with low GLUT1, and the reverse relationship at P27 are consistent with the in situ hybridization data for mRNA. However, the observation of equivalent concentrations of MCT1 protein in nonvascular membranes (i.e., primarily neurons and glia), despite the decline in mRNA, was surprising and different from other studies (18). MCT2 was not detected in the isolated microvessels (data not shown). To investigate the level of nonvascular MCT1 and MCT2, especially as they relate to the developmental and regional changes in GLUT1 and -3, vascular-free membranes were prepared from frontal cortex, hippocampus thalamus, and cerebellum from P7 to P35. Figure 4 illustrates representative Western blots of thalamus, as well as the relative quantitation of non-

285 • NOVEMBER 2003 •

www.ajpendo.org

E1130

BRAIN GLUTS AND MCTS IN DEVELOPMENT

Fig. 2. Glucose transporter (GLUT)1, GLUT3, and MCT1 mRNA in thalamus of P14 and P27 rats. Coronal cryosections (16 ␮m) from P14 and P27 rat brains were analyzed for GLUT1, GLUT3, and MCT1 mRNA expression by in situ hybridization with 35S-riboprobes, exposed to photographic emulsion, and counterstained with hematoxylin as described in MATERIALS AND METHODS. High-power bright-field micrographs of thalamus are visualized with epiluminesence; positive hybridization is in green. mv, Microvessel; asterisks, neuronal nuclei.

vascular MCT1 and MCT2 in each region from P7 through P35. Each antibody detected a single prominent band at 45 kDa (MCT1) or 40 kDa (MCT2), which was completely blocked by preabsorption with the respective peptide (data not shown). Because each sample was prepared by combining tissue from five postnatal rats, the concentration represents the average

Fig. 3. MCT1 and GLUT1 protein expression in microvessels and vascular-free cortical membranes prepared from P14 and P27 rat brain. Microvessels (5 ␮g protein) and vascular-free cortical membranes (20 ␮g protein) were analyzed by Western blot for levels of GLUT1 and MCT1. AJP-Endocrinol Metab • VOL

value per milligram of membrane protein, which incorporates the normal deviation. Concentration of MCT1 protein in the parenchyma of cerebellum and thalamus is 50–100% greater than that observed in either cortex or hippocampus, with no appreciable decline throughout development. MCT2 protein is detected as a prominent band of ⬃40 kDa. Interestingly and in contrast to the mRNA, concentrations of MCT2 protein are highest at P7, especially in frontal cortex and cerebellum, with progressive declines even during suckling. The exception is the hippocampus, where MCT2 protein increases over the period of synaptogenesis during the second and third postnatal weeks and plateaus at the adult level. It is important to remember that the measured values are concentrations, and the overall brain is expanding such that a lack of increase in MCT1 protein concentrations likely represents an increase in the total protein amount, and total MCT2 amounts may remain unchanged in the adult brain. We have previously described the developmental patterns of GLUT expression during this period of development in the rat (39, 40). To evaluate the observations of MCT expression in the context of total energy substrate utilization, we further analyzed the

285 • NOVEMBER 2003 •

www.ajpendo.org

BRAIN GLUTS AND MCTS IN DEVELOPMENT

E1131

Fig. 4. Regional MCT1 and MCT2 protein levels in nonvascular brain with development. Vascular-free membranes were prepared from pooled samples (5 animals/age) of cortex, hippocampus, thalamus, and cerebellum isolated from P7 to P35 rat brains, as described in MATERIALS AND METHODS, and analyzed for MCT1 and MCT2 by Western blot. Top: representative Western blots of MCT1 and MCT2 in thalamic samples (20 ␮g protein/lane) from P7-P35 rat brains. S, rat microsomal brain standard (20 ␮g). Bottom: Western blots for each region were analyzed relative to the standard, and values are expressed in arbitrary standard units presented graphically. Each value represents an average of 5 samples/age, since sample was prepared by pooling 5 regions.

regional vascular-free samples depicted in Fig. 4 for GLUT1 and GLUT3 content. A direct comparison of the Western blot analyses for all four proteins is presented for the hippocampal samples, and the relative quantitation for all regions is presented graphically in Fig. 5. The developmental patterns for both GLUT1 and GLUT3 are as previously reported (39), with low levels of both proteins during the first postnatal week, GLUT3 increasing sharply with neuronal maturation (P14-P21) accompanied by gradual, steady increases in nonvascular, 45-kDa GLUT1 with gliogenesis, brain growth, and the increased reliance on glucose to fuel cerebral metabolism. What is striking, and different from what has been reported, is that the decreased utilization of ketone bodies during the latter stages of development is not accompanied by corresponding declines in the levels of MCT1 and MCT2 protein in the parenchyma. DISCUSSION

Cerebral metabolism during postnatal development and neurological maturation was extensively described during the 1970s and 1980s. These early studies, which included measurements of glucose, lactate, and ketone body uptake and utilization, coupled with analyses of metabolic enzyme activities (for review see Ref. 26), support the developmental switch in cerebral metabolic fuels from a combination of glucose, lactate, and ketone bodies in the very young human and rat brain to the obligatory requirement for glucose in the adult. We previously described the relationship between maturational changes in cerebral glucose utilization and the expression of the glucose transporter proteins GLUT1 and GLUT3 in the rat (42). With the more recent cloning of the MCTs, there has been some description of their developmental expression in mouse and rat brain (11, 18, 28). The present study greatly extends these initial observations to include specific determinations of mRNA and protein in the vascular endothelial cells of the BBB, as well as the regional AJP-Endocrinol Metab • VOL

expression of nonvascular MCT1 and MCT2 mRNA and protein in the postnatal rat brain. The major findings of the study are that 1) the developmental switch in cerebral metabolic fuels is most closely reflected in the expression of MCT1 mRNA and protein in the BBB, but not in expression of MCT2; 2) early in development (through P14), MCT1 mRNA is expressed in both neurons and glia and becomes predominantly glial in the mature brain; and 3) nonvascular concentrations of both MCT1 and MCT2 proteins remain elevated in the mature brain. The in situ hybridization studies depicted in Fig. 1 demonstrate an early and widespread expression of MCT1 mRNA, apparent in all cell types (BBB, neurons, glia) through the second postnatal week. Thus MCT1 may be the isoform responsible for ketone body transport into all cells at this stage of maturation. Thereafter, MCT1 mRNA levels decline in both BBB and neurons, and the expression appears more restricted to glial components, as has been previously described in rat and mouse brain (11, 17, 28). MCT2 mRNA has a far more neuronal pattern of expression, even in the P7 rat brain, and maintains this pattern throughout development. The neuronal pattern of MCT2 expression is in agreement with our previous report in adult mouse brain (17) and reports of others (7, 28, 30). However, the disparate developmental patterns for MCT1 and MCT2 mRNAs are in contrast to the results reported by Pellerin et al. (28). These authors measured MCT1 and MCT2 mRNA by Northern blot analysis of postnatal rat brain and reported that both isoforms dramatically increase 10-fold to their maximal levels (100%) between P0 and P15 and just as sharply decline thereafter, approximating 20% (MCT1) and 5% (MCT2) of maximal at P30. Thus, although our data clearly support their additional description of specific alterations in MCT1 mRNA in BBB (Fig. 2), with declines after P14, there appears to be no significant reduction in MCT2 mRNA with cerebral maturation. More striking is the observation that levels of

285 • NOVEMBER 2003 •

www.ajpendo.org

E1132

BRAIN GLUTS AND MCTS IN DEVELOPMENT

Fig. 5. Regional developmental expression of brain nutrient transporters. The vascular-free membranes described in Fig. 4 were analyzed for MCT1 and MCT2 protein levels by Western blot. Representative Western blots of GLUT1 and GLUT3 in hippocampus are presented (top) in comparison with those for MCT1 and MCT2 and depicted graphically (bottom). Values were analyzed relative to the standard in arbitrary standard units, and developmental pattern for each transporter within each region is illustrated graphically.

nonvascular MCT1 and MCT2 protein do not change dramatically with development (Fig. 4). The combined analysis of MCT1 and MCT2 proteins with that of both GLUT1 and GLUT3 in the nonvascular compartment of brain during postnatal development (Fig. 5) allows for a more comprehensive description of the regulation of both sets of nutrient transporters with changing cerebral metabolism. Cerebral metabolism in the postnatal brain requires the integration of increasing metabolic demand associated with neuronal maturation and synaptic activity, as well as lipid/myelin biosynthesis. This complex metabolic relationship is further compounded by a change in substrate availability from a combination of glucose, lactate, and ketone bodies to one based solely on carAJP-Endocrinol Metab • VOL

bohydrate metabolism after the third postnatal week. The newborn rat brain is ideally equipped to utilize all three substrates, and each substrate exerts effects on the utilization of the others. Ketone bodies inhibit the oxidation of glucose to CO2 and cause the accumulation of glycolytic intermediates, with the resultant export of lactate out of the brain, also mediated by the MCTs. The export of lactate from the immature brain is believed to result from a combination of the relatively delayed developmental expression of pyruvate dehydrogenase and its inhibition by the acetyl-CoA generated from the degradation of ␤-hydroxybutyrate and acetoacetate, and from the potential competition for import into the mitochondria between pyruvate and these substrates (27). Lactate can also be taken up and oxidized by the immature brain, as has been demonstrated during hypoglycemia (15, 34) and hypoxia-ischemia (38). It has been shown that any condition, pathological or physiological, that results in an elevated blood lactate will promote its uptake and utilization by the developing brain, with a resultant sparing of brain glucose utilization (22). Glucose conversely promotes the incorporation of the ketone bodies into amino acids and lipids. In the latter case, this is due to its metabolism through the pentose phosphate pathway generating high levels of NADPH, which are necessary for lipid biosynthesis, which peaks at P20-P21 in the rat (3). There is clearly a close association between the availability of ketone bodies in the circulation and the expression of MCT1 in the BBB, as well as the enzymes required for ketone body oxidation in the parenchyma. The developmental switch in the utilization of the available circulating fuels is accommodated by the transition from the abundant expression of MCT1 in the endothelial cells of the BBB early in postnatal life to the predominant expression of GLUT1 after P21P27. Equally, the high concentration of MCT1 in the BBB establishes a situation in which the levels of lactate in the brain are directly related to the blood levels, as has been observed in a variety of in vivo paradigms in the immature rat (22, 38). However, the prevailing assumption has been that the expression of both MCT1 and MCT2 covaries with metabolism and is a reflection of the switch in respiratory fuels and thus that they follow the same time line. The results of this study indicate that this is not the case and, when the developmental expression of these two transporters is examined in the parenchyma, each isoform displays unique regional and cellular characteristics. Although levels of MCT1 mRNA clearly decline with age, as depicted in Figs. 1 and 2, BBB levels of the protein decline by 50%, whereas nonvascular concentrations of the protein remain essentially constant. However, the membrane samples analyzed by Western blot comprise all nonvascular cell types, primarily neurons and glia, and as the brain matures and grows there is a greater contribution of glial components. Thus the fairly stable patterns of MCT1 protein concentrations, interpreted in the context of the change in the cell-specific expression apparent in Fig.

285 • NOVEMBER 2003 •

www.ajpendo.org

BRAIN GLUTS AND MCTS IN DEVELOPMENT

1, indicate a reduction in the neuronal MCT1 component after P14, replaced by the increased MCT1 due to the expanding glial compartment. The impact of gliogenesis is perhaps seen to be greater in the comparison of MCT2 mRNA expression with the protein patterns of Fig. 4. MCT2 is not expressed in the BBB, and all of the MCT2 mRNA is nonvascular, neuronal expression. The intensity of the mRNA signal is not reduced with maturation, although the level of MCT2 protein per milligram membrane protein in cortex, thalamus, and cerebellum clearly is. This reduction could be simply explained as due to a greater amount of glial membrane in the samples, creating a “dilution” of the MCT2 concentration as well as a reduction in MCT2 protein per neuron. The exception to this is the hippocampus, where the MCT2 protein levels clearly increase after the second postnatal week to their adult levels. The hippocampus is a densely neuronal structure and may not be subject to the same degree of glial expansion as seen in other regions, such as the cortex. Or the values may indeed represent a greater amount of MCT2 protein in both the pyramidal cell layer and dentate gyrus with maturation. In the hippocampus, the developmental pattern of MCT2 expression closely mirrors that of the neuronal glucose transporter GLUT3 (Fig. 5), with increases in both. The high levels of these two neuronal nutrient transporter proteins attained after the period of active synaptogenesis support a coordinate role for glucose transport and lactate transport in neuronal metabolism in the mature brain. We have previously reported the postnatal rise in regional GLUT3 protein (39, 40), and when the rise is considered in the context of a possible “dilution” of neuronal membrane, it further underscores the major importance of glucose as a metabolic fuel to the neuron. The purpose of this study was to investigate specifically that period of postnatal development characterized by the switch in circulating fuels, and the data do not extend into the adult. However, the P35 rat brain approximates the adult situation, and it is apparent from the results of Figs. 4 and 5 that the mature rodent brain has a clear capacity to transport both glucose and lactate across the BBB as well as intracellularly. Such capacity could certainly facilitate the metabolic trafficking recently proposed by Magistretti et al. (20) as the lactate shuttle; i.e., GLUT1 levels are maximal in the BBB as well as in glia, along with MCT1, and MCT2 in neurons. However, neuronal GLUT3 is also maximal, strongly supporting a major role for glucose transport directly into neurons in the adult brain. The ultimate question that arises from these as well as previous studies relates to the mechanisms underlying the regulation of these transporter proteins during development, as well as in response to altered metabolic situations such as hypoglycemia, diabetes, and the administration of a ketogenic diet. The results presented here, in conjunction with several previous studies, clearly support a unique mode of regulation of nutrient transporters in the BBB as the first adaptation to a change in available substrate. Thus changes in the expression of MCT1 mRNA and protein in the AJP-Endocrinol Metab • VOL

E1133

BBB are observed in concert with alterations in plasma levels of ␤-hydroxybutyrate. Similarly, administration of a ketogenic diet to adult rats for a 5-wk period resulted in increased levels of BBB MCT1 protein (19). However, as observed during normal cerebral development, parallel changes were not observed in the nonvascular glial compartment. An analogous situation was observed in the chronically hypoglycemic rat, which resulted in significant increases in BBB GLUT1 mRNA and protein, with no effect on the nonvascular 45-kDa GLUT1 (33). There has been convincing evidence that endothelial cell regulation of GLUT1 involves alterations in mRNA stability through the action of cytoplasmic binding proteins (4–6). It is tempting to speculate that there may exist a coordinate regulation mediated by these binding proteins, such that a change in available substrate would trigger an increase in one transporter with a decrease in the other. Clearly, further studies are required to delineate such mechanisms of regulation in the BBB, as well as alternative pathways in neurons and glia. Lactate and the ketone bodies are important cerebral substrates in pathological conditions such as diabetes and hypoxia or ischemia. An understanding of the normal regulation of expression of these transporters in the developing brain should contribute to the design of strategies to increase expression in adult brain as potential therapeutic interventions. We thank WeiWei Wang, Lisa Willing, and Robert Brucklacher for expert technical assistance. DISCLOSURES This work was supported by National Institute of Child Health and Human Development PO1-HD-30704 (to S. J. Vannucci). REFERENCES 1. Bergersen L, Rafiki A, and Ottersen OP. Immunogold cytochemistry identifies specialized membrane domains for monocarboxylate transport in the central nervous system. Neurochem Res 27: 89–96, 2002. 2. Bergersen L, Waerhaug O, Helm J, Thomas M, Laake P, Davies AJ, Wilson MC, Halestrap AP, and Ottersen OP. A novel postsynaptic density protein: the monocarboxylate transporter MCT2 is co-localized with delta-glutamate receptors in postsynaptic densities of parallel fiber-Purkinje cell synapses. Exp Brain Res 136: 523–534, 2001. 3. Bilger A and Nehlig A. Quantitative histochemical changes in enzymes involved in energy metabolism in the rat brain during postnatal development. II. Glucose-6-phosphate dehydrogenase and beta-hydroxybutyrate dehydrogenase. Int J Dev Neurosci 10: 143–152, 1992. 4. Boado RJ and Pardridge WM. The 5⬘-untranslated region of GLUT1 glucose transporter mRNA causes differential regulation of the translational rate in plant and animal systems. Comp Biochem Physiol B Biochem Mol Biol 118: 309–312, 1997. 5. Boado RJ and Pardridge WM. Ten nucleotide cis element in the 3⬘-untranslated region of the GLUT1 glucose transporter mRNA increases gene expression via mRNA stabilization. Brain Res Mol Brain Res 59: 109–113, 1998. 6. Boado RJ, Tsukamoto H, and Pardridge WM. Evidence for translational control elements within the 5⬘-untranslated region of GLUT1 glucose transporter mRNA. J Neurochem 67: 1335– 1343, 1996. 7. Broer S, Rahman B, Pellegri G, Pellerin L, Martin JL, Verleysdonk S, Hamprecht B, and Magistretti PJ. Compar-

285 • NOVEMBER 2003 •

www.ajpendo.org

E1134

8. 9. 10. 11.

12.

13. 14. 15. 16.

17.

18.

19. 20. 21.

22. 23. 24.

25.

BRAIN GLUTS AND MCTS IN DEVELOPMENT

ison of lactate transport in astroglial cells and monocarboxylate transporter 1 (MCT 1) expressing Xenopus laevis oocytes. Expression of two different monocarboxylate transporters in astroglial cells and neurons. J Biol Chem 272: 30096–30102, 1997. Cremer JE. Substrate utilization and brain development. J Cereb Blood Flow Metab 2: 394–407, 1982. Dwyer DS, Vannucci SJ, and Simpson IA. Expression, regulation, and functional role of glucose transporters (GLUTs) in brain. Int Rev Neurobiol 51: 159–188, 2002. Edmond J, Auestad N, Robbins RA, and Bergstrom JD. Ketone body metabolism in the neonate: development and the effect of diet. Fed Proc 44: 2359–2364, 1985. Gerhart DZ, Enerson BE, Zhdankina OY, Leino RL, and Drewes LR. Expression of monocarboxylate transporter MCT1 by brain endothelium and glia in adult and suckling rats. Am J Physiol Endocrinol Metab 273: E207–E213, 1997. Hanu R, McKenna M, O’Neill A, Resneck WG, and Bloch RJ. Monocarboxylic acid transporters, MCT1 and MCT2, in cortical vitro and in vivo. Am J Physiol Cell Physiol 278: C921– C930, 2000. Hawkins RA, Mans AM, and Davis DW. Regional ketone body utilization by rat brain in starvation and diabetes. Am J Physiol Endocrinol Metab 250: E169–E178, 1986. Hawkins RA, Williamson DH, and Krebs HA. Ketone-body utilization by adult and suckling rat brain in vivo. Biochem J 122: 13–18, 1971. Hellmann J, Vannucci RC, and Nardis EE. Blood-brain barrier permeability to lactic acid in the newborn dog: lactate as a cerebral metabolic fuel. Pediatr Res 16: 40–44, 1982. Joost HG and Thorens B. The extended GLUT-family of sugar/polyol transport facilitators: nomenclature, sequence characteristics, and potential function of its novel members. Mol Membr Biol 18: 247–256, 2002. Koehler-Stec EM, Simpson IA, Vannucci SJ, Landschulz KT, and Landschulz WH. Monocarboxylate transporter expression in mouse brain. Am J Physiol Endocrinol Metab 275: E516–E524, 1998. Leino RL, Gerhart DZ, and Drewes LR. Monocarboxylate transporter (MCT1) abundance in brains of suckling and adult rats: a quantitative electron microscopic immunogold study. Brain Res Dev Brain Res 113: 47–54, 1999. Leino RL, Gerhart DZ, Duelli R, Enerson BE, and Drewes LR. Diet-induced ketosis increases monocarboxylate transporter (MCT1) levels in rat brain. Neurochem Int 38: 519–527, 2001. Magistretti PJ, Pellerin L, Rothman DL, and Shulman RG. Energy on demand. Science 283: 496–497, 1999. Maher F, Vannucci S, Takeda J, and Simpson IA. Expression of mouse-GLUT3 and human-GLUT3 glucose transporter proteins in brain. Biochem Biophys Res Commun 182: 703–711, 1992. Miller AL, Kiney CA, and Staton DM. Effects of lactate on glucose metabolism of developing rat brain. Brain Res 316: 33–40, 1984. Nehlig A. Cerebral energy metabolism, glucose transport and blood flow: changes with maturation and adaptation to hypoglycaemia. Diabetes Metab 23: 18–29, 1997. Nehlig A, Boyet S, and Pereira de Vasconcelos A. Autoradiographic measurement of local cerebral beta-hydroxybutyrate uptake in the rat during postnatal development. Neuroscience 40: 871–878, 1991. Nehlig A, de Vasconcelos AP, and Boyet S. Quantitative autoradiographic measurement of local cerebral glucose utilization in freely moving rats during postnatal development. J Neurosci 8: 2321–2333, 1988.

AJP-Endocrinol Metab • VOL

26. Nehlig A and Pereira de Vasconcelos A. Glucose and ketone body utilization by the brain of neonatal rats. Prog Neurobiol 40: 163–221, 1993. 27. Patel MS and Owen OE. Development and regulation of lipid synthesis from ketone bodies by rat brain. J Neurochem 28: 109–114, 1977. 28. Pellerin L, Pellegri G, Martin JL, and Magistretti PJ. Expression of monocarboxylate transporter mRNAs in mouse brain: support for a distinct role of lactate as an energy substrate for the neonatal vs. adult brain. Proc Natl Acad Sci USA 95: 3990–3995, 1998. 29. Persson B, Settergren G, and Dahlquist G. Cerebral arteriovenous difference of acetoacetate and D –hydroxybutyrate in children. Acta Paediatr Scand 61: 273–278, 1972. 30. Pierre K, Magistretti PJ, and Pellerin L. MCT2 is a major neuronal monocarboxylate transporter in the adult mouse brain. J Cereb Blood Flow Metab 22: 586–595, 2002. 31. Pierre K, Pellerin L, Debernardi R, Riederer BM, and Magistretti PJ. Cell-specific localization of monocarboxylate transporters, MCT1 and MCT2, in the adult mouse brain revealed by double immunohistochemical labeling and confocal microscopy. Neuroscience 100: 617–627, 2000. 32. Price NT, Jackson VN, and Halestrap AP. Cloning and sequencing of four new mammalian monocarboxylate transporter (MCT) homologues confirms the existence of a transporter family with an ancient past. Biochem J 329: 321–328, 1998. 33. Simpson IA, Appel NM, Hokari M, Oki J, Holman GD, Maher F, Koehler-Stec EM, Vannucci SJ, and Smith QR. Blood-brain barrier glucose transporter: effects of hypo- and hyperglycemia revisited. J Neurochem 72: 238–247, 1999. 34. Thurston JH, Hauhart RE, and Schiro JA. Lactate reverses insulin-induced hypoglycemic stupor in suckling-weanling mice: biochemical correlates in blood, liver, and brain. J Cereb Blood Flow Metab 3: 498–506, 1983. 35. Vannucci RC and Vannucci SJ. Cerebral carbohydrate metabolism during hypoglycemia and anoxia in newborn rats. Ann Neurol 4: 73–79, 1978. 36. Vannucci RC and Vannucci SJ. Hypoglycemic brain injury. In: Seminars in Neonatology, edited by Saliba E. Cambridge, UK: Harcourt Publishers, 2001, p. 147–155. 37. Vannucci RC and Vannucci SJ. Glucose, acidosis, and the developing brain. In: Birth Asphyxia and the Brain: Basic Science and Clinical Implications, edited by Donn SM, Sinha SK, and Chiswick ML. Armonk, NY: Futura, 2002, p. 89–119. 38. Vannucci RC and Yager JY. Glucose, lactic acid, and perinatal hypoxic-ischemic brain damage. Pediatr Neurol 8: 3–12, 1992. 39. Vannucci SJ. Developmental expression of GLUT1 and GLUT3 glucose transporters in rat brain. J Neurochem 62: 240–246, 1994. 40. Vannucci SJ, Clark RR, Koehler-Stec E, Li K, Smith CB, Davies P, Maher F, and Simpson IA. Glucose transporter expression in brain: relationship to cerebral glucose utilization. Dev Neurosc 20: 369–379, 1998. 41. Vannucci SJ, Gibbs EM, and Simpson IA. Glucose utilization and glucose transporter proteins GLUT-1 and GLUT-3 in brains of diabetic (db/db) mice. Am J Physiol Endocrinol Metab 272: E267–E274, 1997. 42. Vannucci SJ, Seaman LB, Brucklacher RM, and Vannucci RC. Glucose transport in developing rat brain: glucose transporter proteins, rate constants and cerebral glucose utilization. Mol Cell Biochem 140: 177–184, 1994.

285 • NOVEMBER 2003 •

www.ajpendo.org