Title: Dual regulation of the AMP-activated protein kinase provides a novel mechanism for the control of creatine kinase in skeletal muscle
Abstract: Article16 March 1998free access Dual regulation of the AMP-activated protein kinase provides a novel mechanism for the control of creatine kinase in skeletal muscle Markella Ponticos Markella Ponticos MRC Molecular Medicine, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author Qi Long Lu Qi Long Lu Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author Jennifer E. Morgan Jennifer E. Morgan Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author D.Grahame Hardie D.Grahame Hardie Department of Biochemistry, University of Dundee, Dundee, DD1 4HN GB Search for more papers by this author Terence A. Partridge Terence A. Partridge Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author David Carling Corresponding Author David Carling MRC Molecular Medicine, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author Markella Ponticos Markella Ponticos MRC Molecular Medicine, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author Qi Long Lu Qi Long Lu Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author Jennifer E. Morgan Jennifer E. Morgan Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author D.Grahame Hardie D.Grahame Hardie Department of Biochemistry, University of Dundee, Dundee, DD1 4HN GB Search for more papers by this author Terence A. Partridge Terence A. Partridge Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author David Carling Corresponding Author David Carling MRC Molecular Medicine, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB Search for more papers by this author Author Information Markella Ponticos1, Qi Long Lu2, Jennifer E. Morgan2, D.Grahame Hardie3, Terence A. Partridge2 and David Carling 1 1MRC Molecular Medicine, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB 2Muscle Cell Biology Groups, Clinical Sciences Centre, Imperial College School of Medicine, Hammersmith Hospital, DuCane Road, London, W12 0NN GB 3Department of Biochemistry, University of Dundee, Dundee, DD1 4HN GB *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:1688-1699https://doi.org/10.1093/emboj/17.6.1688 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The AMP-activated protein kinase (AMPK) is activated by a fall in the ATP:AMP ratio within the cell in response to metabolic stresses. Once activated, it phosphorylates and inhibits key enzymes in energy-consuming biosynthetic pathways, thereby conserving cellular ATP. The creatine kinase–phosphocreatine system plays a key role in the control of ATP levels in tissues that have a high and rapidly fluctuating energy requirement. In this study, we provide direct evidence that these two energy-regulating systems are linked in skeletal muscle. We show that the AMPK inhibits creatine kinase by phosphorylation in vitro and in differentiated muscle cells. AMPK is itself regulated by a novel mechanism involving phosphocreatine, creatine and pH. Our findings provide an explanation for the high expression, yet apparently low activity, of AMPK in skeletal muscle, and reveal a potential mechanism for the co-ordinated regulation of energy metabolism in this tissue. Previous evidence suggests that AMPK activates fatty acid oxidation, which provides a source of ATP, following continued muscle contraction. The novel regulation of AMPK described here provides a mechanism by which energy supply can meet energy demand following the utilization of the immediate energy reserve provided by the creatine kinase–phosphocreatine system. Introduction Mammalian AMP-activated protein kinase (AMPK) belongs to a family of protein kinases that play a major role in response to metabolic stress (Corton et al., 1994; Hardie, 1994; Hardie et al., 1994; Hardie and Carling, 1997). This family of kinases has been highly conserved throughout evolution, and homologues have been identified in yeast and plants (Carling et al., 1994; Hardie and Carling, 1997). AMPK was first identified in rat liver through its role in the phosphorylation and inactivation of a number of enzymes involved in lipid metabolism (Carling et al., 1987; Hardie et al., 1989; Hardie, 1992), and subsequently was shown to phosphorylate enzymes in other metabolic pathways (Carling and Hardie, 1989; Hardie et al., 1989). AMPK is regulated by reversible phosphorylation, being phosphorylated and activated by a distinct AMPK kinase (AMPKK), thereby forming a protein kinase cascade (Carling et al., 1987; Hawley et al., 1995, 1996). In addition, as its name suggests, AMPK is activated by AMP via a number of independent mechanisms. AMPK is activated allosterically by micromolar concentrations of AMP (Carling et al., 1989; Mitchelhill et al., 1994; J.E.Sullivan et al., 1994a). AMP also activates the upstream kinase, i.e. AMPKK is itself an AMP-activated protein kinase (Moore et al., 1991; Hawley et al., 1995). Finally, binding of AMP to AMPK makes it a better substrate for phosphorylation by AMPKK and a worse substrate for dephosphorylation by protein phosphatases (Davies et al., 1995; Hawley et al., 1995). The effects of AMP are antagonized by high concentrations of ATP so it appears that the kinase responds to the ATP:AMP ratio, rather than AMP itself (Hardie and Carling, 1997). Once activated, AMPK phosphorylates a number of enzymes involved in biosynthetic pathways, causing their inactivation and preventing further ATP utilization (Hardie and Carling, 1997). These findings have led to the hypothesis that the AMPK cascade has evolved to monitor the energy status, or fuel supply, within the cell and to initiate appropriate energy-conserving mechanisms in response to ATP depletion (Hardie, 1994; Hardie et al., 1994; Hardie and Carling, 1997). There is compelling evidence to support the hypothesis that AMPK is regulated in response to the ATP:AMP ratio in liver. It remains possible, however, that in other tissues, e.g. skeletal muscle, alternative mechanisms are involved in the regulation of AMPK. Paradoxically, although AMPK is highly expressed in skeletal muscle, very little activity can be detected, making it difficult to define a role for AMPK in this tissue (Verhoeven et al., 1995). Several recent reports, however, describe a potential role for AMPK in the activation of fatty acid oxidation in skeletal muscle, which provides an additional source of ATP required during contraction (Winder and Hardie, 1996; Hutber et al., 1997; Vavvas et al., 1997). This proposed function is perfectly consistent with the overall hypothesis for AMPK in energy conservation. A key feature of energy metabolism in skeletal muscle, but not in liver, is the presence of the creatine kinase (CK)–phosphocreatine (PCr) system (reviewed in Bessman and Carpenter, 1985; Wallimann et al., 1992). CK catalyses the interconversion of creatine (Cr) and ATP with PCr and ADP. PCr can be viewed as a temporary energy buffer providing the immediate fuel source during periods of contraction when ATP is used up faster than it is produced (Wallimann, 1994). It is generally accepted that the CK–PCr system is involved in maintaining the delicately balanced ATP:ADP ratio within the muscle cell. A number of CK isoforms exist, and their subcellular compartmentalization has led to the hypothesis that PCr and Cr act as shuttle molecules between the sites of ATP production and utilization (Wallimann, 1994). In skeletal muscle, this shuttle mechanism has the advantage of targeting a proportion of ATP to specific sites, e.g. myosin ATPase, which have high energy demands during contraction (Wallimann, 1994). This process also prevents the overall depletion of ATP, and consequently the build up of ADP, which would be detrimental to other pathways (van Deursen et al., 1993). The well established role for CK in energy metabolism, combined with the reported role for AMPK in the regulation of fatty acid oxidation in skeletal muscle, led us to examine whether these two systems are linked. Here we show, for the first time, that AMPK phosphorylates and inactivates the muscle-specific isoform of CK (MM-CK). Furthermore, we demonstrate that, in addition to regulation by the ATP:AMP ratio, AMPK activity is modulated by the PCr:Cr ratio. Together, these results imply that AMPK regulates the activity of CK by responding to the level of CK substrates. The generally accepted role for CK is to maintain the high levels of ATP necessary to satisfy the immediate energy demands required during contraction. Although one would imagine that this process would be tightly regulated, until now there has been no evidence for such control. We propose that it is the AMPK which provides the control mechanism required for the efficient function of CK and co-ordinates the energy status of the cell with the regulation of other metabolic pathways. Results Phosphorylation of creatine kinase by AMPK As a first step to determine whether CK is a physiological substrate for AMPK, we used purified MM-CK in an in vitro phosphorylation assay with AMPK (Figure 1A and B). SDS–PAGE analysis of the phosphorylation reaction revealed that a single polypeptide of 43 kDa, corresponding to the MM-CK monomer, is phosphorylated by AMPK. The stoichiometry of phosphorylation by AMPK in vitro is ∼0.5 mol of phosphate per mol of CK monomer (data not shown). During the time course of the reaction, MM-CK was not phosphorylated in the absence of AMPK. The phosphorylation recognition motif for AMPK is φ(X,β)XXS/TXXXφ, where φ is a hydrophobic residue, β a basic residue and the order within parentheses is not critical (Dale et al., 1995; Woods et al., 1996b). The amino acid sequence of the muscle CK isoform contains three potential phosphorylation sites which fit this motif (corresponding to Ser136, Ser199 and Ser373 in the human sequence, see Table I) (Perryman et al., 1986). We have found that the brain isoform (BB-CK) is also phosphorylated by AMPK (data not shown). Two of the potential phosphorylation sites identified in the muscle isoform are conserved in the brain isoform (Ser136 and Ser199, Table I) (Kaye et al., 1987). Figure 1.Phosphorylation and inactivation of MM-CK by AMPK in vitro. MM-CK (4 μg) was incubated with [γ-32P]ATP and AMP in the presence or absence of AMPK. Samples were taken at 0, 15 and 30 min and analysed by SDS–PAGE followed by autoradiography. The Coomassie-stained gel (A) and corresponding autoradiogram (B) are shown. The migration of molecular mass standards and MM-CK are indicated. (C) The effect of phosphorylation on MM-CK activity was determined. MM-CK was phosphorylated by AMPK (●). After 45 min (marked by an arrow), the phosphorylation mixture was divided into two. EDTA (10 mM) was added to one half to prevent further phosphorylation (○). In order to dephosphorylate MM-CK, purified PP2C was added to the remaining half (▪). A control reaction in which MM-CK was incubated with [γ-32P]ATP and AMP, but in the absence of either AMPK or PP2C, is shown (□). During the incubations, aliquots were removed and MM-CK activity determined spectrophotometrically. Activities are expressed as a percentage of MM-CK activity at the beginning of the control incubation. (D) The phosphorylation of MM-CK was determined by autoradiography following SDS–PAGE analysis of samples taken during the time course of the reactions. The data shown are representative of four independent experiments. Download figure Download PowerPoint Table 1. Potential AMPK phosphorylation sites within creatine kinase isoforms Enzyme Site Peptide sequence Human MM-CK (Ser136) VRTGRSIKGY (Ser199) FDKPVSPLLL (Ser373) LEKGQSIDDM Human BB-CK (Ser136) VRTGRSIRGF (Ser199) FDKPVSPLLL Rat acetyl-CoA carboxylase (Ser79) MRSSMSGLHL Consensus φ(X,β)XXSXXXφ The amino acid sequence surrounding potential sites (shown in bold) within MM-CK and BB-CK phosphorylated by AMPK are shown. Studies using synthetic peptides based on the sequence surrounding the known phosphorylation site within acetyl-CoA carboxylase (Ser79) have revealed a consensus motif for phosphorylation by AMPK (Dale et al., 1995). Amino acids are shown using the single letter code, with X representing any amino acid, φ a hydrophobic amino acid and β a basic amino acid. The effect of phosphorylation by AMPK on MM-CK activity was determined (Figure 1C and D). In the absence of AMPK, the activity of MM-CK remained constant throughout the incubation. Phosphorylation by AMPK caused ∼60% inhibition of MM-CK activity. This effect was almost completely reversed by dephosphorylation of MM-CK by protein phosphatase 2C (PP2C). This is the first study, to our knowledge, which demonstrates unequivocally that phosphorylation by a defined protein kinase regulates MM-CK activity. Effect of creatine/phosphocreatine on AMPK Having established the regulation of MM-CK by AMPK, we decided to examine the effect of PCr and Cr, the substrates of CK, on the activity of AMPK in vitro (Figure 2). We found that PCr inhibits AMPK in a dose-dependent manner. In the presence of 20 mM PCr, which is within the physiological range in muscle (5–40 mM) (M.J. Sullivan et al., 1994; Wallimann, 1994; Steeghs et al., 1997), the activity of AMPK was <50% of that measured in the absence of PCr (Figure 2A). In contrast, Cr had no direct effect on AMPK activity (Figure 2B), even at 20 mM Cr, a concentration well above the physiological range (5–10 mM) (M.J.Sullivan et al., 1994; Wallimann, 1994; Steeghs et al., 1997). In the presence of 10 mM Cr, however, inhibition of AMPK by 20 mM PCr was totally abolished (Figure 2B). This antagonistic effect is dependent not only on Cr concentration but also on the concentration of PCr (Figure 2C). This shows that the inhibition of AMPK is dependent on the ratio of PCr:Cr, rather than on the concentration of PCr alone. Although we used AMPK purified from rat liver for these studies, the effect of PCr and Cr was identical if kinase isolated from rat skeletal muscle was used (data not shown). A fall in the PCr:Cr ratio within the cell would be expected, therefore, to cause activation of AMPK as a result of the release of inhibition exerted by PCr. This effect mirrors the situation in hepatocytes, where a fall in the ATP:AMP ratio leads to stimulation of AMPK activity (Corton et al., 1994). In this case, however, activation is brought about by an increase in the allosteric activator, AMP. The effect of PCr/Cr on AMPK activity in vitro occurs in the absence or presence of AMP (data not shown). These results suggest that the regulation of AMPK by the PCr:Cr ratio and the ATP:AMP ratio may act in parallel within the cell. Since a synthetic peptide (SAMS) was used to assay for AMPK activity, it seems likely that inhibition is a direct effect of PCr on AMPK, rather than being substrate mediated. Figure 2.Effect of PCr and Cr on AMPK activity in vitro. (A) Dose-dependent inhibition of AMPK activity by PCr. Results are the mean of three independent experiments ± SEM. (B) AMPK activity was measured in the absence or presence of PCr (20 mM), Cr (10 mM) or both PCr and Cr. Results are the mean of four independent experiments ± SEM. (C) AMPK activity was measured in the presence of 20 mM PCr (●) or 40 mM PCr (○) at varying concentrations of Cr. Results are the mean of three independent experiments ± SEM. (D) Effect of pH on regulation of AMPK by PCr and Cr. AMPK activity was measured at varying pH in the absence or presence of 20 mM PCr (●) or both 20 mM PCr and 10 mM Cr (□). Results are the mean of four independent experiments ± SEM. For all the above experiments, AMPK activity was determined by phosphorylation of the SAMS peptide. In (A), (B) and (C), AMPK activity is expressed as a percentage of activity measured in the absence of additions. For (D), activity is expressed as a percentage of the activity in the absence of additions at each pH. (E) Effect of PCr and Cr on phosphorylation of MM-CK by AMPK. MM-CK (4 μg) was phosphorylated by AMPK in the absence or presence of PCr (20 mM) or Cr (10 mM). The reactions were analysed by autoradiography following SDS–PAGE. Download figure Download PowerPoint Intracellular pH plays an important role in metabolism in muscle (Dudley and Terjung, 1985; Wallimann et al., 1992; M.J.Sullivan et al., 1994), therefore we next investigated whether there was any effect of pH on the regulation of AMPK by the PCr:Cr ratio. AMPK was assayed at varying pH in the absence or presence of PCr (20 mM) or both PCr (20 mM) and Cr (10 mM) (Figure 2D). In the presence of PCr, the degree of inhibition of AMPK increases with increasing pH. At pH 6.6, PCr inhibits kinase activity by ∼30%, whereas at pH 7.3 the inhibition rises to ∼50%. A greater effect of pH was seen when AMPK was assayed in the presence of both PCr and Cr. At pH 7.3, Cr did not protect AMPK significantly against the inhibition by PCr, whereas at pH 6.6, Cr afforded almost complete protection. As anticipated, PCr also inhibits the phosphorylation of MM-CK by AMPK (Figure 2E), although in this case we cannot rule out the possibility that PCr alters the rate of phosphorylation by acting on both AMPK and the substrate, CK. Similarly, as found with the SAMS peptide assay, Cr had no direct effect on the phosphorylation of MM-CK. However, we were unable to determine whether Cr antagonized the PCr inhibition under these conditions. AMPK activity could not be assayed either by phosphorylation of MM-CK or by the SAMS peptide assay in the presence of MM-CK, PCr and Cr. Presumably, this is because the presence of MM-CK and both its substrates in the assay will effectively dilute out the [γ-32P]ATP with unlabelled ATP due to phosphate cycling. Our results suggest that the activity of AMPK in muscle is determined by several physiological parameters (Figure 3). It would be predicted that in resting muscle these parameters will maintain AMPK in an inactive state and that activation of AMPK occurs following muscle stimulation. Figure 3.A simplified model for regulation of AMPK in muscle. A number of factors affect AMPK activity. In resting muscle, AMPK activity is low and this coincides with high concentrations of PCr and ATP. Following muscle contraction, PCr and ATP are utilized and Cr and AMP are produced and, combined with a fall in pH, AMPK is activated. Despite this simplified model, it is likely that other factors are involved in the control of AMPK in vivo. Download figure Download PowerPoint Association of AMPK and creatine kinase AMPK is a heterotrimer of a catalytic subunit, α (Mr ∼63 kDa), and two regulatory subunits, β (30 kDa) and γ (36 kDa) (Carling et al., 1994; Mitchelhill et al., 1994; Stapleton et al., 1994; Dyck et al., 1996; Woods et al., 1996a). Isoforms for each of the subunits have been identified recently (Stapleton et al., 1996), but the precise physiological role of the isoforms remains unclear. In order to phosphorylate MM-CK in vitro, we used AMPK that had been immunoprecipitated from rat liver and, during the course of our experiments, we analysed the immune complexes following phosphorylation. Surprisingly, we found that MM-CK associated with AMPK in the immune complex (Figure 4A). In order to rule out non-specific binding of MM-CK, we performed a number of controls. MM-CK was detected only in immune complexes containing AMPK, and did not associate with AMPK antibodies alone (Figure 4B). The same results were obtained using a number of different AMPK subunit antibodies (data not shown). The association between AMPK and MM-CK is relatively strong, since it survived extensive washing with buffer containing high salt (1 M NaCl) and detergent (1% Triton X-100). Furthermore, we did not find any significant effect of PCr or Cr on the interaction (Figure 4B). Although these results demonstrate that AMPK and MM-CK interact in vitro, it is difficult to draw firm conclusions regarding the stoichiometry of the association. It is clear from Figure 4B that the AMPK–MM-CK complex is not in a 1:1 ratio, since the AMPK subunit polypeptides (apparent Mr on SDS–PAGE of 63, 38 and 36 kDa) are barely visible, compared with relatively strong staining of the MM-CK polypeptide (apparent Mr 43 kDa). MM-CK is a dimer (Dawson et al., 1965), and it is possible that higher-order complexes may form under the conditions of the in vitro assay and this may account, at least in part, for the excess amount of MM-CK relative to AMPK. This physical interaction was also detected in differentiated muscle fibres in the mouse H-2Kb muscle cell line (Morgan et al., 1994). MM-CK is co-precipitated with AMPK following immunoprecipitation of H-2Kb cell extracts with antibodies specific for either AMPK-α1 (Figure 4C, lane 2) or AMPK-α2 (Figure 4C, lane 3), demonstrating that both catalytic isoforms associate with MM-CK in H-2Kb cells. Following immunoprecipitation with MM-CK antibodies, AMPK-β was detected in the immune complex (Figure 4D, lane 1). Figure 4.Association of AMPK and CK-M in vitro and in vivo. (A) MM-CK was phosphorylated by AMPK present in an immune complex obtained by immunoprecipitating a partially purified rat liver extract with anti-AMPK-β antibodies coupled to protein A–Sepharose. Following incubation, the immune complex was recovered by centrifugation, washed thoroughly with PBS containing 1 M NaCl and 1% Triton X-100 and analysed by SDS–PAGE followed by autoradiography. The Coomassie-stained gel (lane 1) and the corresponding autoradiogram (lane 2) show MM-CK associated with the immune complex. (B) MM-CK was incubated with AMPK using the same conditions used for phosphorylation, treated as above, and the immune complexes analysed by SDS–PAGE and visualized by Coomassie staining. Lane 1, MM-CK alone (no antibody); lane 2, MM-CK with antibody–protein A–Sepharose alone; lanes 3–5, MM-CK with AMPK immune complex in the absence (lane 3) or presence of 20 mM PCr (lane 4) and 10 mM Cr (lane 5). (C and D) Co-precipitation of AMPK and CK from differentiated H-2Kb cells. Partially purified extracts, prepared from cells which had been induced to differentiate by growth in non-permissive conditions for 4–5 days (37°C in the absence of IFN-γ), were used for immunoprecipitation with antibodies against MM-CK (lane 1), AMPK-α1 (lane 2) or AMPK-α2 (lane 3). Following extensive washing, the immune complexes were resolved by SDS–PAGE, transferred to a polyvinylidene fluoride membrane and probed with antibodies against either MM-CK (C) or AMPK-β (D). Arrows indicate the positions of MM-CK and AMPK-β. The additional band in lane 1 (C and D) is due to a strong cross-reaction of the rabbit IgG heavy chains with the anti-rabbit secondary antibody. Download figure Download PowerPoint Having shown by immunoprecipitation that AMPK and MM-CK associate in vitro and in cultured myotubes, we examined the localization of the proteins in vivo. Single muscle fibres were isolated from adult rat extensor digitorum longus muscle and used to detect AMPK and MM-CK by fluorescence microscopy. A regular striated pattern was observed following labelling with antibodies against either AMPK-β2 or MM-CK (Figure 5A and B), and double staining of the fibres demonstrated that AMPK and MM-CK co-localize (Figure 5C). Identical results were obtained using a number of different antibodies from different species (sheep anti-AMPK-α, rabbit anti-AMPK-γ, goat and rabbit anti-MM-CK; data not shown). The primary antibodies were judged to be specific by Western blotting of crude tissue lysates (data not shown), and there was no detectable signal in the presence of the secondary antibody alone. It has been reported previously that a pool of MM-CK localizes to the M line of myofibrils (Wallimann et al., 1984) and, if this is the case, our results indicate that AMPK is also found in this location. Figure 5.Co-localization of AMPK and CK in single rat muscle fibres. Muscle fibres were immunostained with antibodies against AMPK-β2 (A, red) or CK (B, green). (C) An overlay of the two shows them to be in register. AMPK-β2 antibodies were detected with biotinylated swine anti-rabbit antibodies, followed by Texas red-conjugated streptavidin. CK antibodies were detected with FITC-conjugated goat anti-mouse antibodies. The scale bar represents 20 μm. Download figure Download PowerPoint AMPK regulates creatine kinase in muscle cells In order to determine the physiological significance of the phosphorylation of MM-CK by AMPK in muscle, we used H-2Kb cells as a model. Under defined conditions, these cells can be induced to differentiate into myotubes (Morgan et al., 1994). Previous studies have shown that the activity of AMPK in cultured cells can be stimulated with 5-amino-4-imidazolecarboxamide riboside (AICA riboside) (J.E.Sullivan et al., 1994b; Corton et al., 1995; Henin et al., 1995). AICA riboside is taken up by cells and phosphorylated to the monophosphate form, 5-amino-4-imidazolecarboxamide ribotide (ZMP), an analogue of AMP. In some cell types, ZMP can accumulate to millimolar concentrations and can activate AMPK both by an allosteric mechanism and by promoting phosphorylation by the upstream kinase, AMPKK. Differentiated H-2Kb cells were incubated in the presence or absence of AICA riboside and the activities of AMPK and MM-CK in cell lysates were measured in the presence or absence of protein phosphatase inhibitors (NaF and NaPPi). In the presence of phosphatase inhibitors, treatment with AICA riboside resulted in a 3-fold stimulation of AMPK activity compared with control cells (Figure 6A). Under the same conditions, the activity of MM-CK was reduced by ∼50% compared with control cells (Figure 6B). In the absence of protein phosphatase inhibitors, the effects of AICA riboside on AMPK and CK were abolished. This result suggests that the activation of AMPK and the inhibition of CK observed with AICA riboside are due to phosphorylation events, which can be reversed by the action of endogenous phosphatases present within the cell lysate. These results strongly suggest that AMPK phosphorylates and inactivates MM-CK in muscle cells. Figure 6.Effect of AICA riboside on AMPK and CK activity in myotubes. H-2Kb cells, which had been induced to differentiate by growth in non-permissive conditions for 4–5 days, were incubated in the absence or presence of 500 μM AICA riboside for 30 min prior to harvesting. Cells were lysed in buffer A or buffer B (buffer A lacking the protein phosphatase inhibitors, NaF and NaPPi), partially purified and the extracts used to measure the activities of AMPK (A) and CK (B). Activities are plotted as fold stimulation of AMPK, or percentage of CK activity, measured in the absence of AICA riboside in buffer A. Results are the average of four independent experiments ± SEM. Download figure Download PowerPoint Discussion Phosphorylation of MM-CK by AMPK Here we describe for the first time the finding that MM-CK is regulated by phosphorylation in skeletal muscle. We show that AMPK, a protein kinase previously shown to be involved in the cellular response to ATP depletion, phosphorylates and inactivates MM-CK in vitro. Since MM-CK catalyses a reversible reaction in vitro, it might be thought of as an unlikely target for regulation by phosphorylation. In vivo, however, the reaction catalysed by MM-CK is far from equilibrium, and the enzyme works in essentially one direction due to the spatially distinct sites of PCr production and utilization (Wallimann et al., 1992; Wallimann, 1994). Under these conditions, regulation of MM-CK by phosphorylation could have a physiologically significant role. Indeed, given the importance of the role of MM-CK in the control of energy supply in muscle, it would be surprising if it was not regulated within the cell. What is even more surprising is that, despite extensive study, no regulation of MM-CK has been identified to date. It has been reported, however, that the BB-CK (brain) isoform is phosphorylated by PKC, although the physiological relevance of this finding is unclear (Quest et al., 1990). We have found that AMPK phosphorylates BB-CK (data not shown) although we have not investigated the significance of this further. Previous studies in liver have revealed that AMPK plays an important role in ATP conservation (reviewed in Hardie and Carling, 1997). CK isoforms are involved in coupling ATP generation with ATP utilization in muscle (reviewed in Bessman and Carpenter, 1985; Wallimann et al., 1992). Two mitochondrial forms of CK (Mi-CK) exist which