Search tips
Search criteria 


Logo of jbcThe Journal of Biological Chemistry
J Biol Chem. 2010 June 4; 285(23): 17673–17680.
Published online 2010 March 31. doi:  10.1074/jbc.M109.079343
PMCID: PMC2878531

Akt2 Modulates Glucose Availability and Downstream Apoptotic Pathways during Development*An external file that holds a picture, illustration, etc.
Object name is sbox.jpg


Glucose is the primary energy substrate for eukaryotic cells and the predominant substrate for the brain. Studies suggest that glucose serves an additional role in the regulation of cellular functions, including viability. Zebrafish is a tractable system for defining the cellular and molecular mechanisms perturbed by impaired glucose transport and metabolism. Previously, we demonstrated a critical role for the facilitative glucose transporter, Glut1, in the regulation of embryonic brain development. In this study, we aim to identify mediators in this Glut1-sensitive process by investigating the role of the antiapoptotic kinase, Akt2. Results show that abrogating expression of akt2 causes a phenotype strikingly similar to that observed when glut1 expression is inhibited. akt2-deficient embryos exhibit increased neuronal apoptosis, impaired glucose uptake, and death by 72 h postfertilization. Similar to what was observed in the glut1 morphants, inhibiting the expression of the proapoptotic protein, bad, in the context of impaired akt2 expression results in the inhibition of apoptosis and rescue of the morphant embryos. Intriguingly, overexpression of glut1 in the akt2 morphants was also able to rescue these embryos. Quantitative reverse transcription-PCR analysis revealed decreased glut1 transcript expression in akt2 morphant embryos. Taken together, these data suggest that Akt2 modulates glucose availability by regulating Glut1 expression at the transcript level. These data support a role for akt2 in an integrative pathway directly linking glucose, Glut1 expression, and activation of apoptosis and demonstrate the dependence of akt2 on glucose availability for the maintenance of cellular viability, particularly in the central nervous system.

Keywords: Apoptosis, Diseases/Metabolic, Metabolism, Metabolism/Glucose, Organisms/Zebrafish, Transport/Glucose, Akt2, GLUT1, bad


The acute metabolic transition that a neonate experiences at birth predisposes it to the development of hypoglycemia. The sudden switch from total reliance on maternal nutrients to dependence on endogenous fuel stores requires a rapid, coordinated, and integrated hormonal and enzymatic response by the neonate (1, 2). Proper homeostatic control is critical during this period because the consequences of repeated and/or prolonged episodes of hypoglycemia can result in neurodevelopmental impairment (3). Although it is clear that dysregulation of mechanisms controlling glucose homeostasis, including deficiencies in growth hormone or cortisol and hyperinsulinism, can cause hypoglycemia during this volatile period, there is virtually nothing known about the mechanisms underlying the irreversible neurologic sequelae. Although there are substantial data defining possible mechanisms underlying injury associated with hypoxia-ischemia, most of these studies have been performed in adult animals, making conclusions regarding neonates with isolated hypoglycemia indirect at best (4).

Investigators studying a variety of cell systems have established glucose uptake and metabolism as pivotal mediators of cell survival. Cancer cells are dependent on glycolysis for the generation of ATP even in the presence of sufficient oxygen, a phenomenon known as the Warburg effect (5, 6). Recent elegant studies describe a novel mechanism to explain the Warburg effect by which defects in mitochondrial respiration lead to activation of the Akt survival pathway, resulting in a selective advantage for the utilization of glucose in malignant cells (7). In hematopoietic systems, studies show that lymphocytes rapidly down-regulate the expression of Glut1 in response to growth factor depletion accompanied by reduction in mitochondrial potential and cellular ATP levels, ultimately leading to apoptosis (8). In models of glucose deprivation, neuronal protection is modulated by the IGF-1 system and mediated by the regulation of glucose availability through the regulation of Glut1 expression (9). Finally, mammalian models of preimplantation development have shown that aberrant glucose metabolism results in impaired expression of glucose transporters and increased apoptosis (10,12).

Studies show that the ability of activated Akt to inhibit apoptosis is dependent on glucose availability and is coupled to cellular metabolism. Akt affects glucose uptake and glycolysis by regulating the translocation of glucose transporters to the plasma membrane (13, 14). Additionally, activated Akt has been shown to stimulate the activity of hexokinase, the first enzyme in the glycolytic pathway (15, 16). Akt activation increases the overall rate of glycolysis, presumably through its effects on hexokinase activity, and promotes translocation of hexokinase I and II to the mitochondrial outer membrane, where high concentrations of ATP favor enzymatic phosphorylation of glucose (17). Thus, Akt is able to regulate glycolysis by controlling glucose entry into the cell via the expression of glucose transporters at the cell surface and by regulating the activity of enzymes involved in the glycolytic pathway. In addition to its role in regulating cellular metabolism, which we hypothesize is linked to its antiapoptotic functions, activated Akt promotes cell survival by influencing the activity of downstream apoptotic proteins (18). For example, Akt promotes cell survival by phosphorylating the proapoptotic protein Bad, keeping it sequestered and preventing translocation of Bax to the mitochondrial membrane, cytochrome c release, and eventually cell death (19,21).

In the present study, we extend our analysis of a previously described, Glut1-sensitive apoptotic pathway describing the molecular mechanism for this effect (22). Based on its role in the regulation of apoptosis and its dependence on glucose availability, we hypothesized that Akt2 is a critical component in this pathway. Our findings reveal that this proto-oncogene is probably upstream in this pathway and show for the first time in a whole animal model system the importance of glucose homeostasis in the regulation of cellular viability during development. These data support the premise that regulation of cell survival by glucose goes beyond its role as a nutrient substrate. We propose that cells “sense” the level of glucose and thereby regulate critical cellular processes, including viability.


Zebrafish Maintenance and General Procedures

Zebrafish were maintained and staged as described previously (23, 24). Wild-type zebrafish strains AB or AB/WIK were used throughout this study. Fish were maintained in a photoperiod of 14:10 (light/dark) in egg water at a constant temperature of 28.5 °C and fed three times daily. For all experiments, embryos were obtained by in vitro fertilization according to standard procedures (24). Embryos were dechorionated with watchmaker forceps. To inhibit pigmentation, egg water was supplemented with 0.003% (w/v) 2-phenylthiourea. In glucose uptake experiments, embryos were anesthetized with tricaine.2 A 0.4% tricaine solution (pH 7.0) was prepared in water and stored at room temperature. For anesthetization, this solution was diluted in fresh tank water (filtered tap water) to a final concentration of 0.02%, and fish were incubated for ~5 min prior to the procedure. All chemicals were from Sigma unless otherwise specified.

Statistical Analysis

Differences between control values and experimental values were compared by Student's t test. All data are expressed as means ± S.E. All experiments were performed at least three times. For mRNA rescue experiments, percentage of normal phenotype was assessed on at least 25 embryos/group/experiment. For quantitative real-time reverse transcription-PCR (qRT-PCR), amplification was performed in triplicate and repeated at least three times using mRNA from three separate morpholino (MO) injections. Significance was defined as p < 0.005.

Microscopy and Image Analysis

All embryos were examined daily with an Olympus SZX12 stereomicroscope. To image zebrafish, embryos were positioned in a drop of 3% methylcellulose on microscope slides. Differential interference contrast images were obtained using an Olympus IX71 microscope fitted with a Nomarski objective, and images were acquired with a TH4-100 camera (Olympus) and Olympus Microsuite software. Fluorescently labeled embryos were visualized utilizing a laser-scanning confocal microscope (BX61WI FV500; Olympus, Melville, NY) utilizing the ×10 objective. Representative fish were imaged using identical confocal settings, and serial Z stacks were acquired using a pinhole aperture of 150 μm. Images were collected with Fluoview software (Olympus).

Morpholino Design and Knockdown

To inhibit the expression of Akt2, antisense MOs targeting the translational start site (5′-CTCTGACGACGCTGATCTCGTTCAT-3′) or the exon splice donor site of exon 10 (5′-AACTTACCGCAAACAGAAACGTCGA-3′) were synthesized (Gene Tools LLC, Philomath, OR). As an injection control, we designed a 5-bp mismatch (5′-CTgTcACGACGCTcATCTCcTTgAT-3′) MO. This MO was identical to the Akt2 translational start site MO except for a 5-bp mismatch, indicated by lowercase letters. The inhibition of glut1 and bad was accomplished using MOs and conditions previously described (22). For knockdown experiments, 5 ng of each MO was microinjected into embryos at the 1–2-cell stage using pulled glass microcapillary pipettes attached to a micromanipulator (model P-97, Sutter Instrument Co.). Injection was driven by compressed N2 gas using the picopump PLI-100 (Harvard Instruments).

Criteria for Morphological Assessment of Embryos

MO-treated embryos were assessed microscopically at 24 hpf. Analysis was blinded and performed by two independent evaluators. akt2 morphant embryos are characterized by increased tissue opacity in the head, resulting in the loss of key morphological markers in the CNS (Fig. 1 and supplemental Fig. S1). The objective criteria used in scoring embryos was the presence or absence of the midbrain/hindbrain boundary and the degree of tissue opacity in the head (see supplemental Fig. S1).

Abrogation of akt2 results in severe morphant phenotype similar to the glut1 morphant. A–D, phenotype of 24-hpf embryos injected with a MO targeting the akt2 splice site (akt2 splice MO) or the translational start site (akt2 start MO) compared ...

RT-PCR Analysis

To verify the efficiency and specificity of our akt2 splice site-targeted MO knockdown, total RNA was extracted from MO-injected (control, akt2 start site, or akt2 splice site) embryos at 24 hpf using the RNeasy minikit (Qiagen, Inc., Valencia, CA). One microgram of total RNA from each treatment group was treated first with DNase (Fisher) to eliminate residual genomic DNA and then reverse-transcribed with oligo(dT) primers using Superscript III (Invitrogen) to generate cDNA for further analysis. Primers (forward, 5′-TCATGGAGTACGCAAATGGA-3′; reverse, 5′-CTCTGTTTGGGGTCTTT-3′) located in exons flanking the MO-targeted region (exon 10) were designed to amplify differentially processed fragments. PCR products were subsequently gel-purified and sequenced. The following PCR parameters were used for amplification: 94 °C for 2 min, 35 cycles of 94 °C for 30 s, 55 °C for 1 min, and 72 °C for 2 min, utilizing an MJ Research PTC-100 thermocycler.

RNA Rescue Experiments of akt2 Morphants

mRNA was generated from constructs containing either zebrafish glut1 or akt2 by in vitro transcription using mMESSAGE mMACHINE (Ambion, Inc.), utilizing the vector's intrinsic T7 promoter. Resulting mRNA was purified using the RNeasy minikit (Qiagen), analyzed by gel electrophoresis, and quantitated to produce a working concentration of 200 ng/μl. In mRNA rescue experiments, 1 ng/embryo of glut1, akt2, or control mRNA was co-injected with 5 ng/embryo of the akt2 morpholino targeting the translational start methionine. For rescue experiments with bad, MO was co-injected with control or akt2 MO at 5 ng/embryo.

Glucose Uptake

The yolk sac of 24-hpf MO-treated embryos was injected with 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose (2-NBDG), a fluorescent glucose analog (Molecular Probes, Inc., Eugene, OR) and incubated at room temperature. At 5 min after 2-NBDG injection, embryos were anesthetized with tricaine and subjected to confocal immunofluorescent microscopy to visualize the transport of glucose into the embryos.

Detection of Apoptotic Cell Death

Apoptosis was detected with the vital dye acridine orange (acridinium chloride hemi-zinc chloride, Sigma). MO-treated embryos were dechorionated and incubated with 5 μg/μl acridine orange for 30 min at room temperature in the dark. The embryos were then washed three times for 5 min with egg water. Embryos were immediately visualized using confocal microscopy.

Quantitative Analysis of 2-NBDG Uptake and Apoptosis in akt2 Morphant Embryos

NIH Image version 1.63 was utilized to quantify mean fluorescence as a measure of 2-NBDG uptake and apoptosis in morpholino-injected embryos. Confocal images were converted to grayscale and inverted, and mean fluorescence was measured per unit area (cm2). Specifically, using the capture tool, the entire image, excluding the yolk sac, was outlined, and total fluorescence was calculated. At least 6 images/treatment group were analyzed, the average total fluorescence intensity was calculated (mean fluorescence/cm2), and results were plotted as a percentage of control-treated embryos.

qRT-PCR Analysis

Total RNA was extracted from 24-hpf embryos injected with control MO, glut1 start MO, or akt2 start MO, and cDNA was synthesized as described above. cDNA from each group was analyzed for glut1 expression using qRT-PCR and normalized to the housekeeping gene, ribosomal protein L13a (Rpl13α), which has been validated for use as a reference gene during zebrafish embryonic development (25). An aliquot of cDNA was added to a reaction mixture containing 1× iQ SYBR Green Supermix (Bio-Rad) and gene-specific forward and reverse primers (glut1, forward primer (5′-GTGATTGGGTCCTTGCAGTT-3′) and reverse primer (5′-CTGAGAAGGAGCCGAGAATG-3′); Rpl13α, forward primer (5′-CGCAAGAGAATGAACACCAA-3′) and reverse primer (5′-ACAACCATGCGCTTTCTCTT-3′)). PCR primers were designed using the Primer3 software to ensure that pairs had optimal characteristics for amplification (26, 27). Amplification was carried out in triplicate using a real-time PCR machine (Bio-Rad) and the following amplification conditions: 94 °C for 30 min, 58 °C for 30 min, and 68 °C for 1 h. Quantification of cDNA was based on monitoring increased SYBR fluorescence during exponential phase amplification and determination of the PCR cycle number at which the amplified product exceeded a defined threshold (the “crossing threshold”). These data were also standardized to the reference gene, Rpl13α. Specificity of the amplification was verified by confirmation of predicted product size and uniformity using melt curves and agarose electrophoresis of the PCR products. Additionally, specificity was confirmed by simultaneous analysis of a “no RT” reaction mixture containing all components except reverse transcriptase.


Identification of Zebrafish akt2 Homolog

To identify the zebrafish akt2 homolog, we employed the “reciprocal best hit” method as described previously (28, 29). Briefly, the human AKT2 protein sequence was used in a bioinformatic search resulting in the identification of a Danio rerio sequence having 86.1% sequence identity with human AKT2 at the amino acid level (GenBank™ AY056465). Further analysis revealed that the identified D. rerio akt2 gene maps to chromosome 18 40.20, which was found to be syntenic with human chromosome 19q13.2. In addition to akt2, the following genes are also located in both of these regions: vasp, chrna5, snrpd2, and akt3. Synteny information was based on the location of genes available from the Ensembl Genome Browser. Taken together, these data strongly suggest that this gene is not only the homolog but very likely the zebrafish orthologue of human AKT2.

Inhibiting akt2 Results in a Phenotype That Recapitulates the glut1 Morphant

The Akt family of protein kinases has pleiotropic effects on cell size, metabolism, and survival (18). Based on studies using Akt null mice, it is hypothesized that Akt1 and Akt2 have overlapping as well as distinct functions (30,32). Akt1 is believed to regulate processes affecting cell size, whereas Akt2 has been implicated in mechanisms regulating cellular metabolism. With these differences in mind, we sought to define the role of Akt2 in embryonic development and test our hypothesis that this kinase is involved in activation of the previously described Glut1-sensitive apoptotic cascade (22).

Targeted knockdown experiments in which Akt2 expression was abrogated utilizing specialized antisense oligonucleotides, MOs, were performed. Two MOs, one that inhibits translation at the start methionine (akt2 start MO) and another that interferes with processing of mature transcripts by targeting the intron/exon boundary of exons 9 and 10 (akt2 splice MO), were injected into one-cell stage embryos. As shown in Fig. 1, inhibiting the expression of akt2 with either MO resulted in a similar and dramatic phenotype (Fig. 1, B and D). At 24 hpf, akt2 morphant embryos were characterized by increased tissue opacity in the head, resulting in loss of key morphological markers in the CNS. Additionally, the akt2 morphant is strikingly similar to the glut1 morphant, supporting our hypothesis that Akt2 is involved in the Glut1-sensitive apoptotic pathway (Fig. 1, C versus B and D). Overall, treatment with either akt2 MO resulted in a death by 72 hpf. The morphant phenotype was observed in 92.4 ± 1.5% of embryos injected with the akt2 start MO and 93.9 ± 2.6% of embryos injected with the akt2 splice MO compared with 12.2 ± 3.1% (p < 0.001) of embryos injected with the control MO.

To illustrate the functional impact and specificity of the akt2 splice MO, analysis of akt2 transcripts from MO-treated embryos was performed. As illustrated in Fig. 1E, analysis of mRNA from embryos injected with either the control MO or the akt2 start MO resulted in generation of a single product having a predicted size of ~500 bp. In contrast, analysis of mRNA from embryos treated with the akt2 splice MO revealed two additional transcripts. Nucleotide sequence analysis of these variants showed that the smaller transcript resulted from skipping exon 10, whereas the larger fragment was generated due to the retention of the intron preceding exon 10. Sequence analysis of PCR fragments running at the predicted size for all three groups confirmed the wild-type sequence of Akt2 (data not shown). The specificity of the phenotype was further confirmed in experiments in which overexpression of zebrafish akt2 mRNA was able to significantly rescue the morphant phenotype (Fig. 2 and supplemental Table S1).

Rescue of akt2 morphants with D. rerio akt2 mRNA. A–C, lateral view of 24-hpf embryos injected with control 5-mispair MO and control (con) mRNA, the akt2 start site MO and control mRNA, or the akt2 start site MO and D. rerio akt2 mRNA reveals ...

Inhibiting akt2 Results in Increased Apoptosis and Impaired Glucose Uptake

The increased areas of tissue opacity in the head of the akt2 morphants are similar to those observed in the glut1 morphants (22) and suggest increased apoptosis in this region. To assess programmed cell death in the CNS of akt2 morphants, embryos were stained with the vital dye acridine orange (22). Inhibition of akt2 results in increased cell death in the developing embryo (Fig. 3, A versus B) and is similar to what was observed in the glut1 morphants (22). Quantitative analysis revealed a greater than 5-fold increase in apoptosis in akt2 morphants compared with control embryos (Fig. 3C).

Abrogation of akt2 results in enhanced apoptotic cell death. A and B, embryos injected with either control or akt2 start site MO were stained with the vital dye acridine orange, which preferentially stains apoptotic cells. At 24 hpf, akt2 morphant embryos ...

Akt is implicated in the regulation of glucose metabolism based on its regulation of hexokinase activity and translocation of Glut1 to the plasma membrane. Given the similarity in phenotypes between akt2 and glut1 morphants, we hypothesized that inhibiting the expression of Akt2 may result in impaired expression of Glut1 and result in decreased glucose transport into the developing embryo. By measuring uptake of the fluorescently labeled non-metabolizable glucose analog, 2-NBDG, we observed that inhibiting the expression of akt2 results in a decrease in glucose uptake. Quantitative analysis utilizing NIH Image software revealed a 54% decrease in 2-NDBG uptake in akt2 morphant embryos relative to control. This was in contrast to the 90% decrease in uptake in embryos treated with the glut1 MO and the 80% inhibition observed in the presence of the glucose transport inhibitor cytochalasin B (Fig. 4). Although we cannot completely rule out the possibility that the decrease in glucose uptake is merely a secondary response to the increased apoptosis observed in the Akt2 morphants, we hypothesize that this inhibition is due to decreased glucose transporter expression. If the decrease in glucose uptake were secondary to apoptosis, we believe the impact on glucose uptake would have been much more global.

Abrogation of akt2 results in impaired glucose uptake. A and B, glucose uptake was assayed as previously described (22). The fluorescently labeled glucose analog, 2-NBDG, was injected into the yolk sac of embryos treated with control MO (A), akt2 start ...

Abrogation of Bad Phenotypically Rescues the akt2 Morphant

Bad, a well described proapoptotic protein, is a known substrate of Akt. Recent studies from our laboratory provide evidence supporting a critical role for this molecule in a glut1-sensitive apoptotic pathway (22). To further test our hypothesis that akt2 is also a central mediator in this pathway, we sought to determine if abrogating expression of bad in the context of impaired akt2 expression could restore viability of morphant embryos. Attenuating bad expression utilizing a MO targeting the translational start site resulted in the complete rescue of the akt2 morphant embryos (Fig. 5 and supplemental Table S2). 86.3 ± 6.1% of embryos co-injected with the akt2 and bad MO had a normal phenotype compared with only 7.4 ± 1.5% of embryos injected with the akt2 MO alone. The characteristic CNS morphology that defines this stage in development was observed in embryos simultaneously injected with both the akt2 and bad MO compared with those injected with akt2 MO alone.

Abrogation of bad rescues the akt2 neurodegenerative phenotype. Embryos were injected with control MO (A), akt2 MO + control MO (B), or akt2 MO + bad MO (C). D, akt2 morphant embryos co-injected with bad MO revealed that inhibition of bad reversed the ...

Glut1 Rescues the akt2 Morphant

It is clear from our previous study that glucose metabolism plays a critical role in the regulation of cell survival beyond that as a nutrient substrate. Additionally, akt2 has been implicated in facilitating cell survival by regulating cellular metabolism as well as regulating downstream proteins critical to the activation of the mitochondrial death pathway. Based on our previous study, we originally hypothesized that akt2 is downstream of glut1, and the striking similarity between the glut1 and akt2 morphants supports this hypothesis (Fig. 1). Intriguingly, our findings illustrate that inhibiting the expression of akt2 results in impaired glucose transport (Fig. 4), supporting a role for akt2 in regulating glucose availability probably via modulation of glut1 expression and/or function. To further elucidate how these molecules are interacting, we performed mRNA rescue experiments. Results reveal that overexpression of akt2 does not rescue the glut1 morphants, whereas overexpression of glut1 does rescue the akt2 morphants, as illustrated in Fig. 6. Specifically, co-injection of akt2 mRNA with the glut1 MO resulted in a normal phenotype in only 10.9 ± 2.8%, which is not significantly different from embryos injected with control mRNA, 6.0 ± 1.8% (p = 0.07). This is in contrast to embryos injected with the akt2 MO in which co-injection with either akt2 mRNA (used as a positive control) or glut1 mRNA was able to rescue the embryos. A normal phenotype was observed in 61.2 ± 5.8% of embryos co-injected with akt2 mRNA and 50.4 ± 12.3% of embryos co-injected with glut1 mRNA compared with only 4.2 ± 0.8% of embryos injected with control mRNA (p < 0.001) (Fig. 6 and supplemental Table S3). Surprisingly, these data suggest that Glut1 is downstream of Akt2, not upstream as we originally hypothesized.

Overexpression of glut1 rescues akt2 morphant embryos. A, glut1 MO was co-injected with control mRNA (negative control), Drglut1 mRNA (positive control), and Drakt2 mRNA. The glut1 morphant phenotype was only rescued in embryos overexpressing glut1. A ...

Abrogation of akt2 Correlates with Decreased glut1 Transcript Levels

akt2 is implicated in the regulation of glucose availability via modulation of glut1 translocation to the plasma membrane as well as regulation of glut1 expression at the level of transcription (33, 34). We have shown that abrogation of akt2 results in impaired glucose uptake, and overexpression of Glut1 is able to rescue the akt2 morphant phenotype. Taken together, these data suggest that akt2 is regulating glucose availability by modulating glut1 expression and/or function. To determine if abrogation of akt2 impacts glut1 expression, we performed qRT-PCR analysis on mRNA generated from embryos treated with control MO, glut1 start MO, or akt2 start MO. Results from this analysis reveal a significant decrease in glut1 transcript levels in akt2 MO-injected embryos when compared with control or glut1 MO-treated embryos. Normalizing to the ribosomal protein, Rpl13α, glut1 expression in embryos treated with the glut1 MO was 92.1 ± 9.0% of control-treated embryos compared with 10.2 ± 6.8% for embryos injected with the akt2 MO (Fig. 7).

Abrogation of akt2 decreases glut1 transcript levels. qRT-PCR analysis on mRNA generated from embryos treated with control MO, glut1 start MO, or akt2 start MO. Results from this analysis reveal a significant decrease in glut1 transcript levels in akt2 ...


Repeated episodes of hypoglycemia during the neonatal period can cause devastating and often permanent neurological sequelae (4, 35,37). The mechanisms underlying this significant health problem remain poorly understood (38,40). Glucose, the primary energy substrate for eukaryotes, plays a critical role in human neurodevelopment. For example, infants with GLUT1 deficiency syndrome present neonatally with epileptic encephalopathy and profound developmental delay (41,44). This syndrome results from hypoglycorrhachia (decreased cerebrospinal fluid glucose) caused by haploinsufficiency of GLUT1, the primary glucose transporter expressed at the blood brain barrier. Similarly, episodes of hypoglycemia during the neonatal period are known to cause permanent neurological impairment (3). Although these clinical scenarios suggest a causal relationship between glucose metabolism and neurodevelopment, the mechanisms integrating these processes remain poorly understood.

Previous studies from our laboratory established zebrafish as a tractable system for defining the cellular and molecular mechanisms perturbed by impaired Glut1-dependent glucose transport (22). In this work, interference with Glut1 expression induced a striking neurodevelopmental syndrome that could be prevented by simultaneous inhibition of the proapoptotic protein bad. This inhibition of apoptosis occurred without apparent restoration of glucose availability, suggesting that cell death in the absence of glut1 is driven not by the cell's need for glucose as a nutrient substrate but rather by induction of the mitochondrial cell death pathway in response to inadequate glucose availability. These data support our hypothesis that glucose or perhaps a downstream glycolytic metabolite is serving as a signaling molecule in the regulation of cellular viability. The concept that glucose uptake and/or metabolism are directly regulating the apoptotic cascade is supported by studies in Xenopus laevis oocytes illustrating a direct link between glucose metabolism and apoptosis via activation of caspase-2 (45). Additionally, it was recently reported that regulation of apoptosis in both neurons and cancer cells is directly dependent on glucose metabolism via the pentose phosphate shunt and its impact on the redox state of the cell (46).

In the present study, we provide compelling evidence supporting a role for the antiapoptotic kinase, akt2, in an integrative pathway directly linking glucose, Glut1 expression, and activation of apoptosis; furthermore, we demonstrate the dependence of akt2 on glucose availability for the maintenance of cellular viability, particularly in the CNS. Specifically, these data reveal that inhibition of akt2 results in a phenotype strikingly similar to that observed in embryos with impaired glut1 expression (Fig. 1). Significant neuronal apoptosis (Fig. 3), impaired glucose uptake (Fig. 4), and dependence on the proapoptotic protein, bad (Fig. 5), all support an integrated pathway involving akt2 and glut1.

The strongest evidence linking these molecules is provided in rescue experiments in which overexpression of glut1 in akt2 morphants results in embryo survival (Fig. 6). Regulation of glut1 expression by akt2 is supported by the observation that Glut1 transcript levels are dramatically decreased in akt2 morphant embryos, as shown by qRT-PCR analysis (Fig. 7).

Although the idea that nutrient availability is critical for cellular survival seems obvious, there is increasing evidence to suggest that glucose regulates cell survival beyond its role as a simple nutrient substrate. We believe that our data as well as previous studies suggest that the cell “senses” the level of glucose when deciding its fate. For example, studies utilizing IL-3-dependent lymphocytes were some of the first to suggest that the supply of energy in the form of glucose or perhaps one of its downstream metabolites can directly influence the rate of apoptosis. In these studies, induction of apoptosis under conditions of growth factor restriction was significantly augmented by inhibition of glucose uptake (47,49). More recent studies revealed that these cells rapidly down-regulate the expression of Glut1 in response to growth factor depletion accompanied by a reduction in mitochondrial potential and cellular ATP levels, ultimately leading to apoptosis (8).

Because it is the main energy substrate for most cells, sophisticated mechanisms regulating glucose sensing have evolved in a variety of organisms. For example, yeasts respond to fluctuations in glucose utilizing a complex signaling cascade in which gene expression is directly regulated by glucose, ultimately resulting in the regulation of glucose transport (50). Although analogous systems have not been identified in mammals, their remarkable ability to maintain a constant level of serum glucose despite divergent periods of food intake suggests that pathways regulated by glucose availability are present. Supporting this hypothesis, it was recently reported that d-glucose is a direct agonist for liver X receptor, a nuclear receptor that coordinates hepatic lipid metabolism (51). Intriguingly, the antiapoptotic protein Bcl-XL is a known target of liver X receptor, and activation of this receptor has been shown to promote survival in some cell types (52).

Perhaps the most convincing data supporting a regulatory role of glucose in the modulation of cell survival has been demonstrated in models of cancer biology. The Warburg affect was described over 70 years ago and describes how tumors preferentially utilize glucose even in the presence of sufficient oxygen. The recent observation that Akt mediates the selective advantage for this is intriguing and supports what we have shown in our studies (7).

Our unique whole animal approach offers a facile means for elucidating the role of glucose homeostasis in embryonic brain development and demonstrates the importance of this pathway in the regulation of fundamental developmental processes. Abrogation of Akt2 or Glut1 results in a phenotype of severely impaired neurodevelopment, and these data support mechanisms directly linking nutrient availability and activation of apoptotic mechanisms during embryonic brain development. More broadly, this work will increase our understanding of the biological role of glucose in human development and provide new insights into the complex interplay of genetics and environmental variability in the causation of the complex neurological disorders, including mental retardation and impaired cognition.

Supplementary Material

Supplemental Data:


We thank Jonathan Gitlin and Matthew Goldsmith for helpful discussions in the preparation of this manuscript. We also thank Drs. Gujun Bu, Paul Hruz, Zsolt Urban, and David Wilson for critical review of the manuscript.

*This work was supported, in whole or in part, by National Institutes of Health (NIH) Neuroscience Blueprint Core Grant NS057105 (to Washington University and the Bakewell Family Foundation) and NIH Grant HD01459 (to M. O. C.).

An external file that holds a picture, illustration, etc.
Object name is sbox.jpgThe on-line version of this article (available at contains supplemental Fig. S1 and Tables S1–S3.

2The abbreviations used are:

ethyl-3-aminobenzoate methanesulfonic acid
hours postfertilization
central nervous system.


1. Sperling M. A., Menon R. K. (2004) Pediatr. Clin. North Am. 51, 703–723, x [PubMed]
2. Menon R. K., Sperling M. A. (1988) Semin. Perinatol 12, 157–162 [PubMed]
3. Yager J. Y. (2002) Clin. Perinatol. 29, 651–674, vi [PubMed]
4. Rozance P. J., Hay W. W. (2006) Biol. Neonate 90, 74–86 [PubMed]
5. Warburg O. (1956) Science 124, 269–270 [PubMed]
6. Warburg O. (1956) Science 123, 309–314 [PubMed]
7. Pelicano H., Martin D. S., Xu R. H., Huang P. (2006) Oncogene 25, 4633–4646 [PubMed]
8. Rathmell J. C., Vander Heiden M. G., Harris M. H., Frauwirth K. A., Thompson C. B. (2000) Mol. Cell 6, 683–692 [PubMed]
9. Russo V. C., Kobayashi K., Najdovska S., Baker N. L., Werther G. A. (2004) Brain. Res 1009, 40–53 [PubMed]
10. Moley K. (1999) Diabetes and preimplantation events of embryogenesis, [PubMed]
11. Moley K. H., Chi M. M., Knudson C. M., Korsmeyer S. J., Mueckler M. M. (1998) Nat. Med. 4, 1421–1424 [PubMed]
12. Moley K. H., Chi M. M., Mueckler M. M. (1998) Am. J. Physiol. 275, E38–47 [PubMed]
13. Kan O., Baldwin S. A., Whetton A. D. (1994) J. Exp. Med. 180, 917–923 [PMC free article] [PubMed]
14. Hill M. M., Clark S. F., Tucker D. F., Birnbaum M. J., James D. E., Macaulay S. L. (1999) Mol. Cell. Biol. 19, 7771–7781 [PMC free article] [PubMed]
15. Gottlob K., Majewski N., Kennedy S., Kandel E., Robey R. B., Hay N. (2001) Genes Dev. 15, 1406–1418 [PubMed]
16. Rathmell J. C., Fox C. J., Plas D. R., Hammerman P. S., Cinalli R. M., Thompson C. B. (2003) Mol. Cell. Biol. 23, 7315–7328 [PMC free article] [PubMed]
17. Majewski N., Nogueira V., Bhaskar P., Coy P. E., Skeen J. E., Gottlob K., Chandel N. S., Thompson C. B., Robey R. B., Hay N. (2004) Mol. Cell. 16, 819–830 [PubMed]
18. Datta S. R., Brunet A., Greenberg M. E. (1999) Genes Dev. 13, 2905–2927 [PubMed]
19. del Peso L., González-García M., Page C., Herrera R., Nuñez G. (1997) Science 278, 687–689 [PubMed]
20. Datta S. R., Dudek H., Tao X., Masters S., Fu H., Gotoh Y., Greenberg M. E. (1997) Cell 91, 231–241 [PubMed]
21. Yang E., Zha J., Jockel J., Boise L. H., Thompson C. B., Korsmeyer S. J. (1995) Cell 80, 285–291 [PubMed]
22. Jensen P. J., Gitlin J. D., Carayannopoulos M. O. (2006) J. Biol. Chem. 281, 13382–13387 [PubMed]
23. Kimmel C. B., Ballard W. W., Kimmel S. R., Ullmann B., Schilling T. F. (1995) Dev Dyn. 203, 253–310 [PubMed]
24. Westerfield M. (2000) The Zebrafish Book, University of Oregon Press, Eugene, OR
25. Tang R., Dodd A., Lai D., McNabb W. C., Love D. R. (2007) Acta. Biochim. Biophys. Sin. 39, 384–390 [PubMed]
26. Rozen S., Skaletsky H. (2000) Methods. Mol. Biol. 132, 365–386 [PubMed]
27. Sun C., Skaletsky H., Rozen S., Gromoll J., Nieschlag E., Oates R., Page D. C. (2000) Hum. Mol. Genet. 9, 2291–2296 [PubMed]
28. Barbazuk W. B., Korf I., Kadavi C., Heyen J., Tate S., Wun E., Bedell J. A., McPherson J. D., Johnson S. L. (2000) Genome Res. 10, 1351–1358 [PubMed]
29. Woods I. G., Kelly P. D., Chu F., Ngo-Hazelett P., Yan Y. L., Huang H., Postlethwait J. H., Talbot W. S. (2000) Genome Res. 10, 1903–1914 [PubMed]
30. Chen W. S., Xu P. Z., Gottlob K., Chen M. L., Sokol K., Shiyanova T., Roninson I., Weng W., Suzuki R., Tobe K., Kadowaki T., Hay N. (2001) Genes. Dev. 15, 2203–2208 [PubMed]
31. Cho H., Thorvaldsen J. L., Chu Q., Feng F., Birnbaum M. J. (2001) J. Biol. Chem. 276, 38349–38352 [PubMed]
32. Cho H., Mu J., Kim J. K., Thorvaldsen J. L., Chu Q., Crenshaw E. B., 3rd, Kaestner K. H., Bartolomei M. S., Shulman G. I., Birnbaum M. J. (2001) Science 292, 1728–1731 [PubMed]
33. Barthel A., Okino S. T., Liao J., Nakatani K., Li J., Whitlock J. P., Jr., Roth R. A. (1999) J. Biol. Chem. 274, 20281–20286 [PubMed]
34. Taha C., Liu Z., Jin J., Al-Hasani H., Sonenberg N., Klip A. (1999) J. Biol. Chem. 274, 33085–33091 [PubMed]
35. Duvanel C. B., Fawer C. L., Cotting J., Hohlfeld P., Matthieu J. M. (1999) J. Pediatr. 134, 492–498 [PubMed]
36. Auer R. N. (2004) Metab. Brain. Dis. 19, 169–175 [PubMed]
37. de Lonlay P., Giurgea I., Touati G., Saudubray J. M. (2004) Semin. Neonatol 9, 49–58 [PubMed]
38. Boluyt N., van Kempen A., Offringa M. (2006) Pediatrics 117, 2231–2243 [PubMed]
39. Dekelbab B. H., Sperling M. A. (2006) Adv. Pediatr. 53, 5–22 [PubMed]
40. Wirrell E., Farrell K., Whiting S. (2005) Can. J. Neurol. Sci. 32, 409–418 [PubMed]
41. De Vivo D. C., Leary L., Wang D. (2002) J. Child. Neurol. 17, Suppl. 3, 3S15–23; discussion 13S24–13S25 [PubMed]
42. Wang D., Kranz-Eble P., De Vivo D. C. (2000) Hum. Mutat. 16, 224–231 [PubMed]
43. Wang D., Pascual J. M., Iserovich P., Yang H., Ma L., Kuang K., Zuniga F. A., Sun R. P., Swaroop K. M., Fischbarg J., De Vivo D. C. (2003) J. Biol. Chem. 278, 49015–49021 [PubMed]
44. Wang D., Pascual J. M., Yang H., Engelstad K., Jhung S., Sun R. P., De Vivo D. C. (2005) Ann. Neurol. 57, 111–118 [PubMed]
45. Nutt L. K., Margolis S. S., Jensen M., Herman C. E., Dunphy W. G., Rathmell J. C., Kornbluth S. (2005) Cell 123, 89–103 [PMC free article] [PubMed]
46. Vaughn A. E., Deshmukh M. (2008) Nat. Cell Biol. 10, 1477–1483 [PMC free article] [PubMed]
47. Williams G. T., Smith C. A., Spooncer E., Dexter T. M., Taylor D. R. (1990) Nature 343, 76–79 [PubMed]
48. Koury M. J., Bondurant M. C. (1990) Science 248, 378–381 [PubMed]
49. Heyworth C. M., Whetton A. D., Nicholls S., Zsebo K., Dexter T. M. (1992) Blood 80, 2230–2236 [PubMed]
50. Johnston M., Kim J. H. (2005) Biochem. Soc. Trans 33, 247–252 [PubMed]
51. Mitro N., Mak P. A., Vargas L., Godio C., Hampton E., Molteni V., Kreusch A., Saez E. (2007) Nature 445, 219–223 [PubMed]
52. Valledor A. F., Hsu L. C., Ogawa S., Sawka-Verhelle D., Karin M., Glass C. K. (2004) Proc. Natl. Acad. Sci. U.S.A. 101, 17813–17818 [PubMed]

Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology