Protein kinase Bα (PKBα) is a key regulator of metabolism, proliferation and differentiation. We have explored the role of PKBα in adipogenesis using wild-type and PKBα-knockout mouse embryonic fibroblasts (MEFs) and show that lack of PKBα prevents MEF differentiation into adipocytes. Expression of ectopic PKBα in PKBα-deficient cells restores adipogenesis. We identified 80 genes whose expression was upregulated in wild-type MEFs during adipogenesis but whose expression was significantly reduced in PKBα-deficient MEFs under the same conditions. Significantly, the regulator of adipogenesis Krüppel-like transcription factor 15 gene expression was downregulated in PKBα-deficient MEFs but could be restored by expressing an active PKBα in the deficient cells. The level of lipocalin 2, renin 1 and receptor-activity-modifying protein 3 genes expressed by adipose cells was also decreased in PKBα-deficient MEFs, and are inhibited by LY294002 treatment during early adipocyte differentiation of 3T3-L1 cells. The results underscore an essential role for PKBα in the transcriptional program required for adipogenesis.
Adipogenesis is a complex process with multiple steps including growth arrest, clonal expansion, withdrawal from the cell cycle and terminal differentiation (Gregoire et al., 1998; Rosen et al., 2000). In parallel, adipose conversion is accompanied by temporally regulated expression of numerous genes. Some of these genes encode transcription factors, such as members of the CCAAT/enhancer binding protein (C/EBP) family and peroxisome proliferator-activated receptor γ (PPARγ). These are key regulators of this transcriptional program and promote the expression of most genes characterising fat cells, such as the adipocyte-specific fatty-acid-binding protein aP2 (Spiegelman et al., 1993).
Evidence has accumulated that insulin and insulin-like growth factor-1 (IGF-1) signalling pathways play a significant role in adipogenesis. Both are essential inducers of adipocyte differentiation (Smith et al., 1988). Insulin receptor-deficient mice display underdeveloped adipose tissue (Cinti et al., 1998) and mouse embryonic fibroblasts (MEFs) derived from these mutants fail to differentiate into adipocytes (Nakae et al., 2003). Furthermore, a double-knockout deficiency of insulin receptor substrate-1 (IRS-1) and IRS-2 results in failure of the adipocyte developmental program (Miki et al., 2001). Finally, phosphatidylinositol 3-kinase (PI 3-kinase) activity transiently increases during conversion of preadipocytes into adipocytes (Sakaue et al., 1998), and the presence of PI 3-kinase inhibitors completely blocks the differentiation process (Tomiyama et al., 1995; Christoffersen et al., 1998; Xia and Serrero, 1999), indicating that the PI 3-kinase pathway is also involved in adipogenesis.
The protein kinase B (PKB, also named Akt) is a downstream target of PI 3-kinase and plays a critical role in many cellular processes stimulated by insulin and growth factors (Brazil and Hemmings, 2001; Lawlor and Alessi, 2001; Whiteman et al., 2002). The three members of the PKB family, PKBα (Akt1), PKBβ (Akt2), and PKBγ (Akt3), are encoded by distinct genes but share a similar structural organisation (Datta et al., 1999; Scheid and Woodgett, 2001; Hill and Hemmings, 2002). It remains a matter of debate whether these isoforms have specific or redundant functions in vivo. This issue was addressed recently by disruption of different PKB genes in mice (Yang et al., 2004). PKBβ-deficient mice exhibit insulin resistance and a diabetes-like syndrome (Cho et al., 2001a; Garofalo et al., 2003). Depending on the genetic background, PKBβ-knockout mice are also characterised by moderate growth deficiency as well as age-dependent decrease in adipose tissue mass (Garofalo et al., 2003). However, MEFs derived from these mutants are not significantly different to wild-type MEFs in their ability to differentiate into adipocytes (Peng et al., 2003). Unlike PKBβ-deficient mice, PKBα-knockout mice have normal glucose homeostasis but display deficient placental development, which results in growth retardation and enhanced neonatal morbidity (Chen et al., 2001; Cho et al., 2001b; Yang et al., 2003). Subcutaneous fat is also moderately reduced in the absence of PKBα, implicating this isoform in adipogenesis (Yang et al., 2003). Finally, the major characteristic of PKBγ-knockout mice is reduction in brain size (Easton et al., 2005; Tschopp et al., 2005). The phenotypic differences between these PKB-knockout mice suggest that PKB isoforms participate jointly in the regulation of some cellular processes, such as growth, but also exert specific roles in vivo, for example in glucose metabolism.
Previous studies have suggested a role for PKB in adipogenesis, because expression of a constitutively activated form of PKB in 3T3-L1 pre-adipose cells caused spontaneous differentiation into adipocytes (Kohn et al., 1996; Magun et al., 1996). In the current study, we investigated the role of the α isoform of PKB in adipose differentiation. We show that MEFs derived from PKBα-deficient mice fail to differentiate into adipocytes in vitro and identified target genes regulated by PKBα during the early steps of adipocyte conversion.
Defective adipogenesis in the absence of PKBα
To investigate the role of PKBα in adipocyte differentiation, MEFs were derived from wild-type and PKBα–/– embryos from the same litter and thus have an identical genetic background. They were induced to differentiate into adipocytes in vitro using a standard adipogenic induction cocktail of the phosphodiesterase inhibitor isobutylmethylxanthine (IBMX), the synthetic glucocorticoid dexamethasone and insulin, followed by addition of insulin every 2 days (see Materials and Methods for details). After 8 days of differentiation, Oil Red O staining was performed to monitor intracellular lipid accumulation. Wild-type MEFs cultured in the presence of differentiating agents accumulated fat droplets but no obvious lipid accumulation was observed in PKBα–/– MEFs (Fig. 1A). This failure to support lipid storage in PKBα–/– MEFs persisted for at least 16 days following induction of differentiation (data not shown). As the adipogenic process is characterised by expression of specific adipocyte markers such as PPARγ and aP2, we analysed expression of these two transcripts in both cell types by semi-quantitative RT-PCR. Whereas treatment of wild-type MEFs with adipogenic inducers led to the induction of Pparg and aP2 mRNA, no change in mRNA levels for either marker was detected in the absence of PKBα (Fig. 1B).
The three PKB isoforms (α, β, γ) are expressed in wild-type MEFs (Peng et al., 2003). Western blot analysis using isoform-specific antibodies was performed to determine the expression profile of the three isoforms during differentiation in wild-type and PKBα–/– MEFs. The data presented in Fig. 2A confirm the absence of PKBα protein in PKBα–/– MEFs and also showed that the amounts of PKBβ and γ were not significantly different in PKBα–/– and wild-type MEFs, precluding any compensatory upregulation of PKBβ and/or γ in the absence of the α isoform. In addition, the activation status of PKB was monitored during adipose conversion using phospho-specific PKB antibodies that recognise the two crucial phosphorylation sites of PKB present in all three isoforms: Thr308 (in PKBα) in the activation loop of the kinase domain and Ser473 (in PKBα) in the hydrophobic motif (Alessi et al., 1996; Meier et al., 1997; Brodbeck et al., 1999). PKB activity increased during differentiation of wild-type MEFs whereas this activity is reduced in the absence of PKBα expression (Fig. 2B). Taken together, these results show that the PKBα isoform is necessary for the adipogenic differentiation of MEFs.
Ectopic expression of PKBα restores adipogenesis in PKBα–/– MEFs
To examine whether the impairment of adipocyte differentiation observed in PKBα–/– MEFs is specifically due to the absence of PKBα, the cells were infected with a retrovirus expressing HA-PKBα (Fig. 3A). Ectopic expression of PKBα in PKBα–/– MEFs resulted in a marked increase in intracellular lipid accumulation after 15 days of differentiation as shown by Oil Red O staining (Fig. 3B). Moreover, PKBα–/– MEFs expressing a constitutively active form of PKBα (m/p-HA-PKBα) were able to spontaneously differentiate into adipocytes in the absence of adipogenic treatment (data not shown). Finally, retroviral introduction of PKBα promoted Pparg and aP2 gene expression during the adipose conversion process (Fig. 3C).
Gene expression profiles after adipogenic treatment of wild-type and PKBα–/– MEFs
To identify genes regulated by PKBα during the early stages of adipogenesis, total RNA was extracted from wild-type and PKBα–/– MEFs at 2 days postconfluence (time 0) and 2, 6, 24, 48 hours after induction of differentiation. Transcript levels were determined by microarray with MOE430A Affymetrix gene chips, which detect 22,690 distinct mouse genes and ESTs. Adipogenesis is characterised by the expression of key regulator genes, such as members of the C/EBP family. Yeh and collaborators found that mRNA levels of the transcription factors C/EBPδ and C/EBPβ were transiently induced during adipose conversion of 3T3-L1 cells (Yeh et al., 1995). In our microarray experiment, we also observed a transient increase in mRNA levels for both C/EBPδ and C/EBPβ 2 hours after adipogenic treatment of wild-type MEFs, validating the microarray results (data not shown). Similar induction of these two genes was also observed with PKBα–/– MEFs, indicating that PKBα is not necessary for the expression of C/EBPδ and C/EBPβ (data not shown). To identify genes whose expression during the first steps of adipocyte differentiation is PKBα-dependent, we first selected genes that displayed an at least twofold change relative to time 0 of wild-type MEFs in at least one of the time points and a significant difference (P<0.05) in a one-way ANOVA. Only genes expressed at similar levels at time 0 in wild-type and PKBα–/– MEFs were further analysed. Of the 1103 genes satisfying these criteria, only those showing at least a twofold difference in expression between wild-type and PKBα–/– MEFs in at least one time point were selected. This yielded 80 genes that were significantly upregulated during the differentiation of wild-type MEFs but did not respond in the absence of PKBα. The genes were classified into ten clusters based upon the temporal expression profiles of adipogenic induction of wild-type MEFs (Table 1). Of these, 24 genes were strongly upregulated after 2 hours and then declined rapidly (clusters 1, 2) or declined slowly (cluster 3) to the basal levels, or showed a plateau at 6 hours before reduction of expression (cluster 4) (Fig. 4). Twenty-three genes increased after 6 hours and then declined to basal values (cluster 5) or maintained a moderate or high level of expression (clusters 6, 7) (Fig. 4). Clusters 8 and 9 comprise genes whose expression progressively increased in wild-type MEFs following induction of differentiation. Expression profiles of cluster 8 were not modified in the absence of PKBα, whereas expression of genes in cluster 9 was enhanced in PKBα–/– MEFs after 24 hours of adipogenic treatment and then declined to basal values (Fig. 4). Cluster 10 includes all otherwise unclassified genes. To validate the microarray results, the expression profiles of some genes with a marked difference between wild-type and PKBα–/– MEFs were determined by semi-quantitative RT-PCR (data not shown). The results obtained by this method were similar to the microarrays.
PKBα is required for Krüppel-like factor 15 (Klf15) expression during adipose differentiation of MEFs
In the microarray experiment, the gene encoding transcription factor KLF15 was induced by the adipogenic cocktail in wild-type MEFs but was only basally expressed in PKBα–/– MEFs (Fig. 5A). It was recently demonstrated that KLF15 increases during adipose conversion of 3T3-L1 cells and plays an important role in this process (Gray et al., 2002; Mori et al., 2005). We further analysed the expression pattern of Klf15 in wild-type and PKBα–/– MEFs by semi-quantitative RT-PCR after 0, 1, 2, 5 and 8 days of adipose differentiation. As expected, treatment of wild-type MEFs with the adipogenic cocktail resulted in the rapid induction and maintenance of Klf15 mRNA throughout the 8 days of differentiation. In contrast with PKBα–/– MEFs, this transcription factor is only modestly expressed with a transient and delayed induction after 2 days which could be due to the other PKB isoforms on Klf15 expression (Fig. 5B). We also observed that after 24 hours of adipogenic treatment, retroviral-mediated PKBα expression in PKBα–/– MEFs reversed decreased expression of Klf15 (Fig. 5C). To further investigate regulation of Klf15 by PKBα, we assessed Klf15 expression in PKBα–/– MEFs expressing a constitutively active form of PKBα (Fig. 5D). In these cells, Klf15 mRNA was markedly induced 3 days postconfluence compared with control vector-infected cells (Fig. 5E). The increase in Klf15 gene expression was accompanied by an augmentation of Pparg expression (Fig. 5E). These data thus suggest that PKBα is involved in the regulation of Klf15, itself a key regulator in the transcription factor network controlling adipogenesis.
Decreased expression levels of Lcn2, Ramp3, Ren1 and Klf15 by inhibition of PI 3-kinase in 3T3-L1 cells
Of the 80 genes identified by microarray analysis, several genes, such as lipocalin 2 (Lcn2), receptor-activity-modifying protein 3 (Ramp3) and renin 1 (Ren1), were robustly expressed in wild-type MEFs after adipogenic treatment but did not respond in the mutant MEFs (Fig. 6A,B). Using semi-quantitative RT-PCR, we further studied the expression profile of these genes during the first steps of adipocyte differentiation of 3T3-L1 cells, which is the most widely used model for investigating mechanisms of adipogenesis in vitro (Green and Meuth, 1974). Transcript levels of Lcn2, Ramp3 and Ren1 increased after a few hours in the presence of the adipogenic cocktail, with a pattern similar to that observed in wild-type MEFs (Fig. 6D). Addition of LY294002, an inhibitor of PI 3-kinase that leads to a decrease in PKB activity (Fig. 6C), reduced the mRNA levels of these three genes (Fig. 6D), as well as the mRNA level of Klf15 (Fig. 6E). Taken together, the results indicate that the expression of Lcn2, Ramp3, Ren1 and Klf15 is dependent on PI 3-kinase and PKB and these genes are good candidates to be involved in the early stages of adipogenesis.
In this study, we have demonstrated that the α isoform of PKB plays a crucial role in the process of MEF differentiation into adipocytes. We have shown that adipose conversion is impaired in the absence of PKBα, whereas ectopic expression of PKBα in PKBα-deficient cells leads to accumulation of lipid droplets and expression of adipocyte-specific genes, such as Pparg and aP2. We did not observe any upregulation of PKBβ and PKBγ compensating for the absence of PKBα in PKBα–/– MEFs. Thus, PKBα is the major isoform of PKB involved in adipocyte differentiation of MEFs.
Furthermore, we have identified several genes whose expression is significantly upregulated during the early steps of adipocyte differentiation in wild-type MEFs compared with PKBα–/– MEFs. These include the transcription factor KLF15, which was recently identified as an essential regulator of adipogenesis. We have shown that PKBα–/– MEFs overexpressing a wild-type PKBα or a constitutively active membrane-targeted PKBα exhibit increased levels of Klf15. Changes in expression of genes such as Lcn2, Ramp3 and Ren1 have been demonstrated in wild-type MEFs as well as 3T3-L1 cells after the initiation of adipocyte conversion. In addition, inhibition of PKB activity in the presence of LY294002 leads to decreased levels of these three genes as well as Klf15 in 3T3-L1 cells. These data suggest the involvement of the PI 3-kinase/PKB pathway in the regulation of Klf15, Lcn2, Ramp3 and Ren1 expression during the early phase of adipocyte differentiation.
It was reported recently that mice deficient for both PKBα and PKBβ display dwarfism and die immediately after birth (Peng et al., 2003). The mutant mice exhibit severe developmental defects in adipose tissue as well as in bone, skin and muscle. Moreover, MEFs derived from these double-knockout mice are unable to differentiate into adipocytes or to express high levels of PPARγ. The PKBα/PKBβ-deficient MEFs partially recover adipocyte differentiation after infection with retrovirus expressing PPARγ (Peng et al., 2003). Here, we have demonstrated that MEFs devoid of the PKBα isoform display defective adipocyte differentiation and that reintroduction of PKBα into these cells can restore lipid accumulation and expression of Pparg and aP2. Interestingly, the ability of PKBβ–/– MEFs to differentiate into adipocytes is not significantly different to that of wild-type MEFs (Peng et al., 2003). Thus, it is likely that PKBα is the major PKB isoform involved in adipogenesis in MEFs. Using small double-strand RNA interference, Xu and Liao have also observed that a decrease in PKBα in 3T3-L1 cells prevents adipocyte differentiation (Xu and Liao, 2004). This indicates that PKBα involvement in adipogenesis is not restricted to our cell models.
The Krüppel-like family of transcription factors play important regulatory roles in cellular proliferation, development and differentiation. We have demonstrated that Klf15 is upregulated during the first steps of adipocyte differentiation of wild-type MEFs but not in the absence of PKBα. Several studies have shown recently that members of the KLF family are essential for adipogenesis. KLF2 decreases rapidly upon adipocyte differentiation of 3T3-L1 cells and has been described as a negative regulator of adipogenesis through inhibition of Pparg (Banerjee et al., 2003). By contrast, KLF5 was shown to be required for adipocyte conversion of 3T3-L1 cells (Oishi et al., 2005). KLF15 is highly expressed in adipose tissue and also during differentiation of 3T3-L1 cells into adipocytes (Gray et al., 2002). It was demonstrated recently by RNA interference and expression of a dominant-negative mutant of KLF15 that this transcription factor acts in adipogenesis through the regulation of Pparg gene expression (Mori et al., 2005). In this report, we have also shown that Klf15, as well as Pparg, are enhanced in the absence of hormonal stimulation in PKBα–/– MEFs overexpressing active PKBα, which can spontaneously differentiate into adipocytes. It was shown previously that PKB affects Pparg gene expression, at least in part, through the Forkhead transcription factor FOXO1 (Peng et al., 2003). Our findings now suggest that the effect of PKBα on Pparg expression is also achieved through KLF15. Further work is now required to determine the mechanism of how PKBα can upregulate Klf15 gene expression.
The results presented here also show a strong and transient induction of Ramp3 transcripts within 48 hours after induction of adipogenesis. The single transmembrane domain protein Ramp3, which is expressed in rat fat tissue (Nagae et al., 2000), is an accessory protein for the G-protein-coupled receptor (GPCR) and is involved in trafficking and folding (Morfis et al., 2003). Several studies have shown that Ramp3, associated with calcitonin-receptor-like receptor, can act as a receptor for adrenomedullin, but it appears that Ramp3 may also regulate other receptors. Interestingly, it was reported recently that expression and secretion of adrenomedullin are more obvious in 3T3-L1 preadipocytes than in adipocytes (Li et al., 2003). These observations are consistent with evidence presented here that Ramp3 is a good candidate to be involved in the first hours of adipocyte differentiation.
Renin is the rate-limiting protease of the renin-angiotensin system (RAS) involved in the production of angiotensin II. The RAS is widely known for its essential role in blood pressure regulation. It has been reported recently that all the main components of this system are also expressed in adipose tissue (Engeli et al., 2000; Ailhaud et al., 2002) but the involvement of the RAS in adipogenesis is not yet clearly established. Angiotensin II may be a potent inhibitor of adipose conversion (Schling and Loffler, 2001) or it may indirectly stimulate adipocyte differentiation (Saint-Marc et al., 2001). In this study, we found that induction of adipose conversion rapidly leads to an increase in Ren1 gene transcript in both wild-type MEFs and 3T3-L1 cells, whereas Ren1 mRNA is almost undetectable in PKBα–/– MEFs or in the presence of an inhibitor of PI 3-kinase. These results suggest that the PI 3-kinase/PKB pathway regulates Ren1 expression and underscore a possible role for the RAS in the process of adipogenesis.
Lcn2, a member of the lipocalin family of small secreted proteins, has also been shown in this study to be strongly upregulated during the early stages of adipose differentiation. Lipocalins bind small hydrophobic molecules and specific cell surface receptors and form macromolecular complexes (Flower, 1996). Interestingly, it was shown by a proteomic approach that the Lcn2 protein is secreted by 3T3-L1 adipocytes (Kratchmarovat et al., 2002). Furthermore, Lcn2 gene expression was found by microarray analysis to increase more than fivefold at 24 hours, 4 and 7 days after initiation of adipocyte conversion of 3T3-L1 cells (Jessen and Stevens, 2002). Here, we have demonstrated that the Lcn2 transcript is induced very quickly, from 6 hours after induction of adipose differentiation, and that the absence of PKBα or addition of a PI 3-kinase inhibitor prevents increase of Lcn2 mRNA. Thus, it seems clear that Lcn2 is regulated through a PI 3-kinase/PKB pathway during adipogenesis, but its precise role remains to be determined.
In conclusion, we have demonstrated that PKBα is necessary to trigger adipocyte differentiation of MEFs and have identified potential target genes affected by PKBα during the early phase of adipogenesis. The mechanism of regulation of these candidates by PKBα and their contributions to the adipocyte differentiation pathway await further investigation.
Materials and Methods
Cell culture and adipocyte differentiation
Mouse embryonic fibroblasts (MEFs) were generated from 13.5-day-old embryos obtained from heterozygous PKBα mice intercrosses (Yang et al., 2003). Briefly, after dissection of head and visceral organs for genotyping, embryos were minced and trypsinised for 30 minutes at 37°C. Embryonic fibroblasts were then plated and maintained in Dulbecco's modified Eagle medium (DMEM) with 10% foetal calf serum (FCS) (Life Technologies), 100 U/ml penicillin and 100 μg/ml streptomycin at 37°C in an atmosphere of 5% CO2. Genotype analysis was performed to determine cells carrying an homozygous null mutation for PKBα (Yang et al., 2003). All experiments were performed with wild-type and PKBα–/– MEFs between 15-20 passages. 3T3-L1 preadipocytes were maintained in DMEM containing 10% FCS. For adipocyte differentiation, 2-day-postconfluent cells (day 0) were treated with DMEM supplemented with 10% FCS, 8 μg/ml biotin, 4 μg/ml pantothenate, 0.5 mM 3-isobutyl-1-methylxanthine, 1 μM dexamethasone and 10 μg/ml insulin for 2 days (all from Sigma). Medium was renewed every 2 days with DMEM containing 10% FCS, 8 μg/ml biotin, 4 μg/ml pantothenate and 10 μg/ml insulin until cells were used. LY294002 (Calbiochem) was prepared in dimethyl sulfoxide (DMSO) and added at a final concentration of 50 μM in the medium 20 minutes before, and every 12 hours, after addition of the adipose cocktail. To visualise lipid accumulation, cells were stained with Oil Red O (Ramirez-Zacarias et al., 1992). Briefly, cells were washed with phosphate-buffered saline (PBS), fixed with 3.7% formaldehyde solution for 1 hour and stained with Oil Red O for 1 hour using a 60:40 (v/v) dilution in water of a 0.5% stock solution (in isopropanol). Cells were then washed twice with PBS and twice with water.
Retrovirus construction, production and infection
The retroviral vectors expressing hemagglutinin-PKBα (HA-PKBα) or (m/p)-HA-PKBα (constitutively active membrane-targeted form of PKBα) were constructed by subcloning the BglII-EcoRI fragment from the pECE-HA-PKBα or pECE-(m/p)-HA-PKBα vectors (Andjelkovic et al., 1997) into the pBABEpuro vector. BOSC retrovirus packaging cells were cultured in DMEM with 10% FCS. To produce the retroviruses, BOSC cells were transiently transfected with 15 μg of the retroviral vector, HA-PKBα or (m/p)-HA-PKBα constructs by the calcium phosphate method. Forty-eight hours after transfection, viral supernatants were harvested, filtered through a 0.45 μm membrane and applied to MEFs in 10 cm dishes with 5 μg/ml polybrene (Sigma). A second infection of MEFs was performed after 8-12 hours. Twenty-four hours after infection with retrovirus, cells were selected with 3-5 μg/ml puromycin (Sigma) for 6-8 days and resistant clones were expanded.
Total RNA was extracted from cells using TRIzol (Invitrogen) according to the manufacturer's instructions. RNA was quantified by spectrometry and the quality confirmed by gel electrophoresis. cDNA was synthesised from 10 μg total RNA using the SuperScript cDNA system (Invitrogen) and used to generate biotin-labelled cRNA with the Enzo BioArray High Yield RNA transcript labelling kit (Enzo Diagnostics, USA). After fragmentation, 10 μg cRNA was hybridised with the mouse MOE430A GeneChips™ (Affymetrix, Santa Clara, USA) following the protocol recommended by Affymetrix. Genechips were then scanned in an Affymetrix 2500 scanner and gene expression analysed using the GeneSpring Software 7 (Silicon Genetics). Only genes with an expression value greater than 100 in at least one condition were further examined. Genes were then selected for an at least twofold change relative to the control condition (time 0 of wild-type MEFs) in at least one condition and the data submitted to one-way ANOVA (A value of P<0.05 was considered significant). The resulting gene data were finally subjected to a Benjamini and Hochberg false-discovery rate multiple-testing correction and a Tukey Post-hoc test.
Total RNA was extracted as above and 1-2 μg reverse-transcribed into cDNA using the avian myeloblastosis virus reverse transcriptase (Promega) and oligo-dT primer. PCR amplification was performed in 25 μl containing 1-2 μl of cDNA and 1.25 U Taq DNA polymerase (Eppendorf). The oligonucleotides used in PCR analysis were as follows: Pparg, 5′-GGAAAGACAACGGACAAATCA-3′ (sense) and 5′-ATCCTTGGCCCTCTGAGATG-3′ (antisense) (325 bp) (Croissandeau et al., 2002); aP2, 5′-GGAACCTGGAAGCTTGTCTCC-3′ (sense) and 5′-ACCAGCTTGTCACCATCTCG-3′ (antisense) (325 bp) (Hansen et al., 2001); β-actin, 5′-ATGGATGACGATATCGCTGCGCTG-3′ (sense) and 5′-CTAGAAGCACTTGCGGTGCACGAT-3′ (antisense) (1127 bp) (NM_007393); Klf15, 5′-CCCAATGCCGCCAAACCTAT-3′ (sense) and GAGGTGGCTGCTCTTGGTGTACATC (antisense) (161 bp) (Teshigawara et al., 2005); Ramp3, 5′-GGATGAAGTACTCATCCCACTG-3′ (sense) and 5′-GAATCGTGACAGATCACAGAG-3′ (antisense) (665 bp) (NM_019511); Ren1, 5′-CCTCACCAACTACCTGAATAC-3′ (sense) and 5′-CACAGCCTTCTTCACATAGCA-3′ (antisense) (622 bp) (NM_031192); Lcn2, 5′-CGATGTACAGCACCATCTATGAG-3′ (sense) and 5′-CTCTCTGGCAACAGGAAAGATG-3′ (antisense) (490 bp) (X14607). β-actin was used as an internal standard. PCR reaction mixtures were denatured at 94°C for 2 minutes and cDNA templates amplified as followed: 20-30 cycles of denaturation at 94°C for 1 minute, annealing at 60-65°C for 1 minute, and extension at 72°C for 1 minute. At the end of the cycling, the samples were incubated at 72°C for 7 minutes. The amplified DNA products were visualised on 1-2% agarose gels and photographed under ultraviolet light.
Western blot analysis
Cells were solubilised in lysis buffer containing 50 mM Tris-HCl (pH 7.5), 1% Nonidet P-40, 1 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, 2 μM microcystin-LR (Alexis), 1 mM sodium pyrophosphate, 10 mM NaF and 0.1 mM sodium orthopervanadate. Lysates were centrifuged at 15,000 g for 10 minutes at 4°C and protein concentration determined using the Bradford reagent (Bio-Rad) with BSA as standard. Protein extracts were separated by 7.5% or 10% SDS-PAGE and transferred onto Immobilon P membranes (Millipore) by electroblotting. After blocking with 5% milk-TBST (50 mM Tris-HCl pH 7.5, 150 mM NaCl, and 0.1% Tween 20), membranes were incubated with rabbit polyclonal antibodies against isoforms of PKB (Yang et al., 2003), phospho-PKB (Ser473 or Thr308) (Cell Signaling Technologies) or with mouse monoclonal antibodies against pan-actin (NeoMarkers) or 12CA5 HA. Blots were then washed with TBST, incubated with horseradish peroxidase-conjugated anti-rabbit or anti-mouse secondary antibodies and detected using enhanced chemiluminescence reagents (Amersham Biosciences).
The authors thank E. Oakeley and H. Angliker for microarray analysis, and D. Hynx for help with the mice. The Friedrich Miescher Institute for Biomedical Research is part of the Novartis Research Foundation.
- Accepted November 14, 2005.
- © The Company of Biologists Limited 2006