Molecular and Cellular Neuroscience 19, 125–137 (2002)

ry.co

Electrical Muscle Activity Pattern andTranscriptional and PosttranscriptionalMechanisms Regulate PKA Subunit Expressionin Rat Skeletal Muscle

Frank Hoover,* ,1 John M. Kalhovde,* Maria Krudtaa Dahle,†

Bjørn Skålhegg,‡ Kjetil Tasken,† and Terje Lømo**Department of Physiology, †Department of Medical Biochemistry, Department of NutritionResearch, and ‡Institute of Basic Medical Sciences, University of Oslo, 0317 Oslo, Norway

We have examined protein kinase A (PKA) subunit expres-sion in adult rat skeletal muscles. Northern blots identifiedPKA catalytic � and regulatory (R) I � and RII � subunits asthe major subunits expressed in slowly contracting soleus(SOL) and rapidly contracting extensor digitorum longus(EDL) muscles. In addition, the steady-state RNA levels ofPKA subunit mRNAs and activities of RI � and RII � pro-moters are similar in SOL and EDL. These data indicatethat posttranscriptional mechanisms account for the two-fold differences in PKA subunit protein levels reportedearlier. Electrical stimulation of denervated SOL with anEDL-like activity pattern (fast pattern) transformed SOLinto an EDL-like muscle with regard to PKA protein levels.These experiments suggest that the posttranscriptionalregulation is activity pattern-dependent. Denervation spe-cifically increased RI � promoter activity and RI � mRNAlevels in SOL and EDL. Further experiments indicated thatthe RI � 1a upstream sequences were activated followingdenervation. Direct electrical stimulation prevented therise in RI � mRNA levels following denervation, demon-strating that electrical muscle activity regulates transcrip-tion.

INTRODUCTION

Motor neurons in adult rats activate skeletal muscleswith distinctive patterns of activity that govern musclefiber phenotype and function. For example, fast motor

1 To whom correspondence should be addressed at the Departmentof Oncology, Oncology Research Laboratory, Haukeland University

Hospital, 5021 Bergen, Norway. Fax: (�47) 55 97 20 46. E-mail:frank.hoover@vir.uib.no.

1044-7431/02 $35.00© 2002 Elsevier Science (USA)All rights reserved.

neurons activate muscle fibers in fast extensor digito-rum longus (EDL) muscles with infrequent short burstsof impulses at frequencies of up to 200 Hz; slow motorneurons activate muscle fibers in slow soleus (SOL)muscles with long trains of impulses at average fre-quencies of about 20 Hz (Hennig and Lømo, 1985).Experiments involving cross-reinnervation or electricalstimulations show that such fast or slow patterns ofactivity turn on or off genes resulting in fast or slowcontractile properties, respectively (Ausoni et al., 1990;Calvo et al., 1996; Close, 1972; Eken and Gundersen,1988; Pette and Vrbova, 1992; Windisch et al., 1998). Itfollows that some signaling pathways which regulatespecific gene expression programs in skeletal musclecan be controlled by activity patterns (Buonanno et al.,1998; Hughes, 1998).

The signaling pathways that transduce electricalmuscle activity into distinct muscle gene expressionprograms are not well understood. Several observationssuggest that the broad-specificity serine–threonine pro-tein kinase A (PKA) is involved. For example, levels ofcAMP and PKA activity in skeletal muscles increaseafter denervation (Carlsen, 1975; Chahine et al., 1993;Hoover et al., 2001). Furthermore, cAMP induces ex-pression of the acetylcholine receptor (AChR) and so-dium channel subunits in cultured muscle cells (Cha-hine et al., 1993; Sherman et al., 1985) and counteractsthe destabilizing effect of denervation on junctionalAChRs (Xu and Salpeter, 1995). We hypothesize that ifPKA were to be involved in neural signal transduction,then PKA subunit expression might be expected to

doi:10.1006/mcne.2001.1053, available online at http://www.idealibra

m on MCN

exhibit properties that can be regulated by electricalactivity.

125

The PKA holoenzyme comprises different isoforms ofregulatory (R) (RI �, RII �, RI �, and RII � and catalytic(C) (C �, C �, and C �) subunits that are encoded bydifferent genes and splice variants (Skalhegg andTasken, 2000). In skeletal muscle, the PKA subunits areconcentrated at the neuromuscular junction, but arealso expressed outside of the end-plate region (Hooveret al., 2001). RI � subunits tend to predominate in thecytosolic fraction, while the RII � subunits are found inboth the cytosolic and the particulate fraction (Hooveret al., 2001). SOL contains approximately twice as muchPKA protein as EDL and SOL displays larger increasesin PKA phosphotransferase activity and cAMP bindingthan EDL after denervation (Hoover et al., 2001). Themolecular mechanisms underlying these differences inPKA subunit expression are unknown. They could betranscriptional and/or posttranscriptional in nature.Differences in electrical activity or trophic influencescould be responsible for the differences between inner-vated SOL and EDL muscles and the effects of dener-vation on them. Here we have designed experiments todistinguish between these possibilities. We assayed Rand C subunit mRNA levels in innervated and dener-vated EDL and SOL muscles. We transfected SOL andEDL using in vivo electroporation of PKA subunit pro-moter constructs, to assay transcription more directlyand precisely. We assessed the contribution of RI pro-moters in innervated and denervated muscles usingspecific probes for leader exons encoding the two RI �promoters. Finally, we stimulated denervated muscleswith electrical stimulus patterns to examine whetherelectrical activity alone could regulate PKA subunitexpression levels.

RESULTS

Expression of PKA Subunit Transcripts in SOL andEDL Muscles

Our previous work has demonstrated that the slowSOL contains approximately twice as much PKA C, RI�, and RII � protein as the EDL and that the level of RI� protein increases in both muscles following denerva-tion (Hoover et al., 2001). We therefore expected thatSOL and EDL would contain different levels of PKAsubunit mRNAs and that denervation would cause up-regulation of PKA subunit mRNAs. To test this, weexamined PKA subunit mRNA levels on Northern blotscontaining RNA from four to six muscles and usingprobes specific for C (� and �), RI (� and �), and RII (�and �) subunits. Contrary to our expectation, we found

that EDL and SOL contained similar levels of steady-state PKA mRNA (Fig. 1). We determined that C � wasthe major catalytic PKA subunit expressed in skeletalmuscle. C � appeared as a single species hybridizing at

FIG. 1. Northern blot analysis of PKA and CamK subunits in nor-mal and denervated soleus and EDL muscles. Total RNA was isolatedfrom 3-day denervated (Den) and contralateral innervated (Inn) mus-cles, using CsCl-gradient centrifugation. Twenty micrograms of RNAwas fractionated in denaturing agarose gels, blotted onto nylon filters,and hybridized with probes corresponding to the indicated subunitsof PKA and CamK. Sizes of the mRNAs in kb are shown to the right;asterisks indicate ribosomal RNA migration. Autoradiograms arerepresentative of three experiments.

126 Hoover et al.

2.4 kb, and denervation for 3 days had no effect on thelevel of C � mRNA. We did not detect any positivesignals for the C � subunit, which is consistent with theobservation that PKA � forms are expressed primarilyin the nervous system (Brandon et al., 1997). We de-tected three distinct RI � mRNAs (3.2, 2.9, and 1.7 kb)(Skålhegg and Tasken, 1997), which results from thepresence of three different polyadenylation sites. De-nervation increased the levels of all three RI � mRNAs,the 1.7-kb message increasing more than the other RI �transcripts. We detected a single species of RII � mRNAin SOL and EDL, and, like the C � mRNA, we did notdetect any change in RII � mRNA levels followingdenervation. We did not detect any transcripts for theRI � subunit. A weak band at 3.2 kb (a size consistentwith RII �) was observed after long exposure times (3weeks) when filters were probed with a RII � cDNA.The intensity of this band was similar in SOL and EDLand was not altered by denervation. These results werespecific for the PKA subunits, since reprobing of theblots with calcium–calmodulin kinase I and II subunitprobes showed no clear differences in fast and slowmuscles or following denervation.

To identify the distribution of PKA mRNA in skeletalmuscle, we analyzed end-plate-containing (n � 6) andend-plate-free regions (n � 6) of the SOL by Northernblotting (Fig. 2). Expression of C � was similar in bothregions at the mRNA level (Fig. 2). RI � was expressed

in both end-plate-containing and end-plate-free regionsand this expression increased considerably after dener-vation, affecting all three RI � mRNAs. Like the othersubunits, RII � mRNA was expressed in both end-plate-containing and end-plate-free regions of the SOL. De-nervation had little, if any, effect on RII � mRNA levels(Fig. 2). These results show, in contrast to reports byothers (Imaizumi-Scherrer et al., 1996; Morita et al.,1995), that PKA subunits are expressed both near andfar from neuromuscular junctions (NMJs). They are alsoconsistent with the distribution of the PKA proteins inskeletal muscle (Hoover et al., 2001).

Activity Pattern-Dependent PosttranscriptionalRegulation of PKA Subunit Protein Levels

The observation that SOL and EDL express similarsteady-state amounts of PKA subunit mRNAs but dif-ferent amounts of PKA subunit proteins suggested thatthe two muscles possess different mechanisms for post-transcriptional regulation. To explore the transcrip-tional activity in SOL and EDL, we transfected bothmuscles with plasmid constructs containing human RI� and RII � 5� upstream regions linked to a chloram-phenicol acetyltransferase (CAT) reporter gene. The RI� and RII � promoter constructs, shown schematicallyin Fig. 3, contain sufficient sequences to direct andregulate reporter gene expression in vitro (Foss et al.,1997; Solberg et al., 1997). In preliminary experiments,we were not able to detect CAT expression using directinjection methods (Wolff et al., 1990). We suspected thatthe low transfection efficiency of the direct injectiontechnique combined with weaker cis-acting elements ofthe PKA promoters could account for this result. Toimprove transfection efficiency, we therefore stimu-lated the SOL (n � 3) and EDL (n � 3) directly afterthe DNA injection (electroporation; Mathiesen, 1999).To correct for differences in transfection efficiency weco-injected the muscle creatine kinase (MCK) promoterlinked to the luciferase reporter gene (Walke et al.,1996), since this construct does not appear to be regu-lated by electrical activity (not shown). Three days afterDNA injection and electroporation, muscles were iso-lated and assayed for CAT activity, using the luciferaseactivity in the same samples as reference. Figure 3shows that both the RI � and the RII � 5� flankingregions directed CAT expression to similar levels ininnervated SOL and EDL, providing further evidencethat the different protein levels in SOL and EDL are dueto different posttranscriptional regulation mechanisms.In addition, the results show that the human PKA pro-moter constructs function in vivo in the rat and that the

FIG. 2. Distribution of PKA subunit mRNAs in normal and dener-vated SOL muscle. PKA subunit mRNA levels were assayed byNorthern blotting in junctionally enriched and extrajunctional re-gions. RNA blots were hybridized with radiolabeled DNA probesencoding the indicated subunits of PKA. Abbreviations: I, innervated;d, denervated; J, junctionally enriched, and XJ, extrajunctional.

127Regulation of PKA in Skeletal Muscle

relative CAT expression is higher for the RII � promoterthan for the RI � promoter.

To determine the contribution of electrical pattern onPKA protein levels, we denervated and directly stimu-lated SOL muscles (n � 6) with a fast pattern for up to1 month. We analyzed these muscles for phosphotrans-ferase activity and C and RI � subunit levels and com-pared them to contralateral SOL and EDL control mus-cles (n � 6, Fig. 4). Stimulation significantly reducedtotal phosphotransferase activity of denervated SOL to

levels comparable to that of EDL (P � 0.0277, Wil-coxon signed rank for paired comparisons). Accord-ingly, Western blots showed that the fast pattern alsoreduced SOL C � and RI � subunits to levels typical ofthe fast EDL.

Denervation Activates the RI � 5� FlankingSequences

We next asked whether increased mRNA stability ortranscriptional activation was responsible for the den-

FIG. 4. Pattern of electrical activity alters PKA subunit levels. (A)Levels PKA catalytic subunit in normal and denervated plus stimu-lated SOL and normal EDL muscle. Total phosphotransferase activitywas assayed in muscle homogenates from normal SOL muscles (SOL,n � 6), denervated SOL muscles stimulated with the fast pattern for4 weeks (DEN�STIM, n � 6), and normal EDL muscles (n � 6).Molar amounts of C were calculated from the raw data to generate thebox plot. Briefly, the rectangle represents 50% of the data and themean is indicated by the horizontal line dissecting the box. The linesabove and below the box establish the ranges of the data. Virtually allof the phosphotransferase activity was inhibited by the addition ofprotein kinase inhibitor (not shown). (B) Immunoblot analysis of C �and RI � subunits in normal SOL, denervated plus fast pattern-stimulated SOL (DEN�STIM), and normal EDL muscles obtainedfrom the same animal. Immunoblot experiments represent at leastthree trials, and blots were stained with Ponceau S prior to probing toconfirm the Bradford protein analysis.

FIG. 3. Basal promoter activities of the 5� flanking regions of thehuman RI � (A) and RII � (B) genes in innervated SOL and EDL.Constructs contained 1058 bp of RI � 5� or 1100 bp of RII � 5� flankingregion ligated in front of a CAT reporter gene in the pCAT Basicvector (Promega). The RI � vector included the 1a and 1b leader exons(solid boxes) and the RII � vector the single leader exon, both vectorscontaining upstream promoter regions (thick line). The vectors wereelectroporated into normal SOL or EDL muscles along with MCK/luciferase vector to control for transfection efficiency. Three days latermuscles were isolated and assayed for CAT and luciferase activities.Results were normalized by dividing CAT activities directed fromPKA promoters by MCK/luciferase activity. The bars on the columnsrepresent the standard deviations of the mean. The results are fromthree experiments using SOL and EDL from the same animals. Emptyvector sequences did not give any detectable CAT activity (notshown).

128 Hoover et al.

ervation-induced increase in amount of RI � mRNA.Evidence supporting both mechanisms has been ob-tained in other systems (Amieux et al., 1997; Knutsen etal., 1991, 1992a,b; Tasken et al., 1990). In a time coursestudy ranging from 3 h to 4 days after denervation, weobserved that RI � mRNA was highest at 2 days afterdenervation and then remained elevated (four to sixmuscles were used for each time point; Fig. 5A). Theseresults were consistent with a role for cAMP in theregulation of the RI � transcriptional unit (Solberg et al.,1991). Denervation for 3 days had little or no effect onendogenous RII � mRNA from whole muscles (Fig. 1)or from end-plate-free regions in SOL (Fig. 2), as pre-viously observed for the RII � protein (Hoover et al.,2001). To examine if changes could be observed at othertime points, we prepared filters for Northern blottingthat contained total RNA from normal muscles andmuscles denervated from 3 h to 4 days (four to sixmuscles were pooled for each time-point, Fig. 5A). Wedid not detect any significant changes in RII � mRNAlevels.

To determine if transcriptional activation or mRNAstability was responsible for the increase in RI � mRNA,we denervated muscles and electroporated the RI �promoter constructs into them at 1 and 2 days postden-ervation and assayed muscle extracts 3 days later (Fig.5B). We observed a significant increase in CAT activitywhen DNA was transfected into 2-day denervated com-pared to innervated muscles (P � 0.033, Mann–Whitneyrank comparison), suggesting that mRNA transcriptionalactivation was responsible for the upregulation of the RI �mRNA. No difference was observed when DNA wastransfected into 1-day denervated muscles.

Electrical Muscle Activity Regulates RI �Transcription and Translation

Denervation specifically increased the rate of tran-scription of RI � and not of the other PKA subunits inskeletal muscle. To examine if lack of nerve-derivedtrophic factor or nerve-evoked electrical muscle activitycaused this increase, we denervated SOL muscles andimplanted intramuscular electrodes to directly stimu-late the muscle in vivo. We performed Northern blottingexperiments on treated muscles to assay mRNA levels(Fig. 6A). After 3 days of denervation, RI � mRNAlevels had increased substantially (lane 2). In contrast, 3days of denervation and direct stimulation of SOL mus-cles with a slow (SOL-like) pattern of electrical activitysuppressed the levels of RI � mRNA (lane 3). We as-sayed RI � subunit protein levels since denervationalone causes a clear increase (Fig. 6B and Hoover et al.,

2001). Western blotting assays on muscles that had beendenervated and stimulated with a fast pattern for 3days showed no clear increase in RI � levels (Fig. 6B). Inaddition, we performed R-subunit binding assays,which also showed normal levels of R subunit follow-ing stimulation (data not shown).

Nonmuscle cells around muscle fibers were includedin our assays. In innervated muscle, the volume ofnonmuscle cells is small compared to that of the musclefibers. In denervated muscles, Schwann cells, immune

FIG. 5. Time-dependent increase in RI � mRNA after denervation.(A) Northern blots were hybridized with probes encoding PKA sub-units 3 h to 4 days after denervation, as indicated. Autoradiogramshows upregulation of PKA RI � subunit mRNA relative to inner-vated control (I) and no effect on RII � mRNA. Similar results wereobtained in two additional experiments. (B) Promoter activities of the5� flanking regions of the human RI � in innervated and denervatedSOL muscle. A construct containing 1058 bp of the RI � 5� flankingregion ligated to a CAT reporter gene was electroporated togetherwith an activity-independent MCK/luciferase vector into SOL mus-cles that were either left innervated (I) or denervated 1 (1d) or 2 (2d)days later. The construct included the 1a and 1b leader exons withupstream promoter regions. Three days later muscles were isolatedand assayed for CAT and luciferase activities. Results from threeanimals were normalized by dividing PKA promoter-driven CATactivity by MCK/luciferase activity. The bars on the columns repre-sent the standard deviations of the mean. Empty vector sequences didnot give any detectable CAT activity (not shown). Asterisk denotessignificance as assessed by Mann–Whitney rank test (P � 0.033).

129Regulation of PKA in Skeletal Muscle

cells, and fibroblasts accumulate and proliferate (Con-nor and McMahan, 1987; Murray and Robbins, 1982).These processes occur predominantly in the end-plateregion and did not appear to affect our assays, since RI� expression in end-plate-containing and end-plate-freeregions was similar. Finally, the electrical stimulationshould affect directly predominantly excitable cellssuch as the muscle fibers. For these reasons, we con-

clude that the results we obtained reflect predominantlyprocesses in the muscle fibers.

Muscle Activity Regulates RI � Transcription byActing on the 1a Promoter Exon

The two RI � leader exons from human (Solberg et al.,1997) and rat (Dahle et al., 2001) have been cloned.Further ranges of alternative leader exons have beenidentified in mouse (1a to 1e; Barradeau et al., 2000) andthe rat (1a to 1c). However, mRNAs containing 1c, 1d,and 1e have not been detected in muscle, although RI �1e is expressed in myoblasts. To identify the contribu-tion of different leader exons to RI � gene expression inSOL, we prepared probes from these sequences andperformed Northern blotting experiments. Figure 7shows that the RI � 1a exon contributed to the three RI� messages. In addition, denervation increased its ex-pression, while electrical muscle stimulation kept it atnormal levels. In contrast, the RI � 1b gene was ex-pressed at lower levels and neither denervation norelectrical stimulation had clear effects (Fig. 7).

DISCUSSION

Motor neurons activate skeletal muscle fibers in waysthat strongly affect their contractile and metabolic prop-erties. PKA activity also affects these fiber properties(see below). How innervation and electrical muscle ac-

FIG. 7. Expression and regulation of RI � leader exons in SOLmuscle. Total RNA was isolated from normal (Inn), denervated (Den),or denervated plus stimulated (Den�Stim) SOL muscles, using CsCl-gradient centrifugation. Twenty micrograms of RNA was fractionatedin denaturing agarose gels and blotted onto nylon filters. Resultingblots were hybridized with oligonucleotide probes specific for RI � 1aand 1b. Arrows denote the relative positions of the RI � bands. Theexperiment was repeated several times with similar results.

FIG. 6. Effects of muscle stimulation on RI � subunits mRNA andprotein levels. Total RNA was isolated from normal, denervated, ordenervated plus directly stimulated SOL muscles using CsCl-gradientcentrifugation. Twenty micrograms of RNA was fractionated in de-naturing agarose gels and blotted onto nylon filters. Resulting blotswere hybridized with RI � probes. Lanes: (1) SOL—innervated, (2)SOL—denervated, (3) SOL—denervated plus stimulated with slowpattern (20 Hz) for 3 days. Arrows denote the sizes of the three RI �mRNAs. Autoradiograms are representative of multiple experiments.(B) Immunoblot analysis of C � and RI � subunits in normal (SOL),denervated (DEN SOL), and denervated plus fast pattern stimulated(DEN�STIM) SOL muscles. Immunoblot experiments represent atleast three trials, and blots were stained with Ponceau S prior toprobing to confirm the Bradford protein analysis.

130 Hoover et al.

tivity affect PKA activity is largely unknown and is thefocus of this work. We first identified the PKA subunitsexpressed in adult rat fast EDL and slow SOL muscles,which previously have not been well characterized. Wethen showed that posttranscriptional regulation ac-counts for the higher levels of PKA subunit protein andcatalytic activity in SOL compared to EDL and that thisregulation is affected by the pattern of electrical activityin the muscle. Finally, we showed that denervationcauses increased transcription of RI � subunits by acti-vating the RI � 1a leader exon and that this activationdepends on electrical muscle activity.

Expression of PKA Subunits in Normal Muscles

We detected RI �, RII �, C �, and minor amounts ofRII � transcripts in normal EDL and SOL muscles,suggesting that the predominant forms of PKA in adultrat muscles are RI �2–C �2 and RII �2–C �2 (Tasken et al.,1993a). These transcripts were expressed throughoutthe muscle fiber, in agreement with reports that PKAaffects both junctional and extrajunctional muscle prop-erties. For example, at the neuromuscular junction, PKAactivation restores normal stability to the nicotinicAChRs that are destabilized by denervation (Xu andSalpeter, 1995). PKA-dependent phosphorylation ofAChRs rescues the receptors from long-term desensiti-zation (Paradiso and Brehm, 1998) and thus may mod-ulate postsynaptic sensitivity to acetylcholine and effi-ciency of transmission. Calcitonin gene-related peptide(CGRP), which motor nerve terminals release, appar-ently acts through PKA when it increases AChR andacetylcholine esterase expression in chick muscle fibers(Choi et al., 2000; Laufer and Changeux, 1987). Outsidethe junction, phosphorylation of calcium channels bycAMP-dependent PKA appears to play a critical role involtage-dependent potentiation and regulation of con-traction (Gray et al., 1997; Igami et al., 1999; Johnson etal., 1997; Liu et al., 1997; Rotman et al., 1995). Inhibitionof PKA causes the rat diaphragm to produce more forceand fatigue faster during low-frequency stimulation(Supinski et al., 2000). Phosphorylation of dystrophin byPKA causes a threefold increase in dystrophin bindingto actin in vitro (Senter et al., 1995) with possibly impor-tant effects on the mechanical integrity of muscles. In-sulin lowers PKA activity in skeletal muscles from nor-mal monkeys but not from prediabetic or diabeticmonkeys (Ortmeyer, 1997). Finally, skeletal musclesfrom hyper- or hypothyroid rats show lower cAMP-dependent phosphodiesterase activity than normal(Mano et al., 1995). These many results indicate thatPKA affects multiple junctional and nonjunctional mus-

cle properties in both health and disease and our find-ings are consistent with these results.

Posttranscriptional Regulation of PKA Activity

Normal SOL and EDL muscles contained similar lev-els of RI �, RII �, and C � mRNAs even though thelevels of PKA protein are significantly higher in SOLthan in EDL (Hoover et al., 2001). Furthermore, musclesexpressed similar amounts of reporter gene activityafter transfection with the promoters for RI � and RII �and showed similar steady-state RNA levels. These re-sults indicate that RI � and RII � are transcribed atsimilar rates in SOL and EDL. Accordingly, posttran-scriptional mechanisms account for the higher PKAactivity in SOL.

Little is known about posttranscriptional regulationof protein function in muscle. In cultured myotubes,cAMP lowers the levels of ryanodin receptor mRNAand the number of ryanodin receptors without affectingtranscriptional activity (Ray et al., 1995). In vivo, chroniclow-frequency stimulation leads to both increased tran-scription and translation of mitochondrial citrate syn-thase activity, although transcription dominates (Seed-orf et al., 1986). In general, however, adaptation inskeletal muscle fibers appears to be mostly controlled atthe level of transcription (Buonanno and Fields, 1999),as in the case of myosin heavy and light chains (Petteand Vrbova, 1992).

We found that fast pattern stimulation transformedSOL into an EDL-like muscle with respect to PKA pro-tein and catalytic activity. Accordingly, the posttran-scription regulation appears activity pattern-dependentsuch that fast and slow patterns drive PKA proteinlevels toward those typical of fast and slow muscles,respectively. Fast and slow types of myosin heavy orlight chains show similar activity pattern-dependentplasticity at the level of transcription (see Pette andVrbova, 1992).

Transcriptional Regulation of PKA Activity

We first examined the effects of denervation andfound substantial and similar increases in levels of RI �mRNA in end-plate-containing and end-plate-free re-gions. This increase reflected, at least in part, increasedrate of transcription because denervated muscles trans-fected with RI �-promoter–CAT constructs showedthreefold higher CAT activity than normal. Levels of RII� and C � were not affected by denervation, suggestingthat the RI �–C � isozyme mediates PKA-dependenteffects of denervation in SOL and EDL.

131Regulation of PKA in Skeletal Muscle

The effects of denervation do not reveal whether theyarise from lack of nerve-derived trophic factors or elec-trical muscle activity. We therefore stimulated the de-nervated muscle electrically with patterns that resem-bled the natural activity of slow motor units in SOL(Hennig and Lømo, 1985). We found that that electricalactivity prevented the rise in RI � mRNA and proteinlevels induced by denervation. This result indicates thatelectrical muscle activity, as normally imposed by mo-tor neurons, regulates RI � expression outside the neu-romuscular junction. For RI � expression at the neuro-muscular junction, the present results offer noinformation because the large amount of extrajunc-tional tissue in our assays would probably mask anysuch changes. Nerve-derived trophic factors, such asneural agrin, neuregulin, or CGRP, may therefore haveaffected junctional properties through PKA signalingwithout being detected in this work. Recently, we haveobtained evidence that neural agrin controls the meta-bolic stability of AChRs at neuromuscular junctions(Bezakova et al., 2001), which earlier work has shownmay involve PKA signaling (Xu and Salpeter, 1995).

Activity-dependent RI � mRNA expression resem-bles more that of extrajunctional AChR expression,which is repressed by both fast and slow stimulationpatterns in denervated muscles (Goldman et al., 1988;Lømo and Westgaard, 1975). Quantitative differencesexist, however, since AChRs disappear faster from thesarcolemma during fast than during slow pattern(Lømo and Westgaard, 1975). PKA may therefore pri-marily modulate properties that fast and slow musclefibers share rather than properties that are unique toone or another fiber type.

Similarities between the genes for RI � and AChRsare consistent with this interpretation. The 5� region ofthe RI � gene contains two leader exons controlled bydifferent promoter regions (Solberg et al., 1997). Onlythe RI � mRNA containing the upstream exon 1a in-creases following denervation. Examination of 1900 bpof the 5� flanking sequences of the RI � gene revealedthat a cluster of consensus elements known to conferelectrical gene regulation was present immediately up-stream of the multiple transcription starts of promoter1a. Specifically, a consensus E box, a CAGG elementoverlapping an SP1 site, and an NFAT binding site werepresent within a 32-bp region in the proximal RI � 1apromoter (Fig. 8). In addition, an Ap2 site at approxi-mately �100 is present. A similar organization of ele-ments is seen in the 47-bp enhancer of the rat nAChR�-subunit promoter (Walke et al., 1996), which is alsoactivated by denervation (Fig. 8). On the other hand,some similarly clustered elements are present farther

upstream in the SURE element of the troponin I slowgene (Calvo et al., 1996), which confers fiber-type-spe-cific gene expression in slow fibers.

Properties within the neuromuscular system are reg-ulated by numerous gene products, some of which areessential for survival, while others have modulatoryroles that other products can compensate for. Thusneural agrin is essential for NMJ formation and sur-vival, whereas elimination of some growth factors mayaffect the phenotype of motor neurons and muscle fi-bers very little, although they may modulate neuromus-cular transmission. Where the proteins in the PKA sys-tem are placed on this scale remains to be established.Experiments that can conditionally eliminate a givenPKA subunit in muscle should provide some answers.

In conclusion, this work provides evidence that elec-trical muscle activity, as normally imposed by motorneurons, regulates PKA activity in adult fast and slowskeletal muscles of the rat through both transcriptionaland posttranscriptional mechanisms. Electrical muscleactivity alone regulates transcription of PKA RI � sub-units. AChR expression in muscle is similarly regu-lated. Muscle activity also affects posttranscriptionalregulation of PKA proteins but in an activity pattern-dependent way such that a fast pattern results in lowerprotein levels than a slow pattern.

EXPERIMENTAL METHODS

Animals and Surgical Procedures

Adult male Long Evans or Wistar rats (150–300g/body weight) were used in this study. Surgical oper-

FIG. 8. Cartoon comparing the upstream regions of the PKA RI � 1a,nicotinic acetylcholine receptor subunit �, and troponin I slow genes.Arrows represent transcriptional start sites in front of the codingregions (represented by large labeled boxes), and the locations ofvarious transcriptional elements are shown.

132 Hoover et al.

ations were performed under Equithesin anesthesia(42.5 mg chloral hydrate and 9.7 mg pentobarbital in 1ml solution, 0.4 ml/100 g weight, ip). Unilateral dener-vations were performed by reflection of a 5-mm seg-ment of the sciatic nerve in the thigh. These experi-ments were approved under the guidelines establishedby the Norwegian Experimental Board and EthicalCommittee for Animal Experiments and were overseenby the veterinarian responsible for the animal house. Atthe respective time point, the SOL and EDL muscles onboth the denervated and the innervated contralateralside were removed from deeply anesthetized animals.The muscles were frozen in liquid nitrogen and storedat �70°C for further use. Following muscle extractiondeeply anesthetized animals were sacrificed by cervicaldislocation.

Protein Extracts

Fresh or frozen muscle whole extracts from muscleswere prepared by mechanical disruption in 10 volumesof ice-cold homogenization buffer containing 10 mMpotassium phosphate (pH 6.8), 1 mM EDTA, 250 mMsucrose, 0.5% Triton X-100, and a cocktail of proteaseinhibitors (chymostatin, leupeptin, antipain, and pep-statin A; Peninsula Laboratories). The homogenate wascentrifuged at 4°C for 30 min at 15,000g. Followingcentrifugation, the supernatant was decanted and col-lected and the protein concentration was determined byusing a modified Bradford assay (Sedmak and Gross-berg, 1977).

Phosphotransferase Activity

Catalytic activity of PKA in muscle extracts was as-sayed by phosphorylating a PKA-specific substrate(Kemp et al., 1977) (Kemptide; Peninsula Laboratories,Inc.) using [�-32P]ATP (sp act 6000 Ci/mmol; Amer-sham) in an assay mixture described by Roskoski(1983). Calculation of the molar concentration of C wasbased on the specific activity of homogeneous bovineheart C subunit (15 �mol/min/mg). Phosphotransfer-ase activity was measured both in the presence and inthe absence of cAMP (5 �M) and PKI (1 �M). Theaddition of PKI virtually abolished all phosphotransfer-ase activity.

Western Blotting

An anti-human C � polyclonal antibody (Cat. No.SC-905; Santa Cruz Biotechnology, Inc.) was used at aconcentration of 0.1 �g/ml for immunoblot analysis.

Monoclonal antibodies directed against human RI �(Cat. Nos. P53620 and P55120, K.T. in collaborationwith Transduction Laboratories; Collas et al., 1999; Eideet al., 1998; Keryer et al., 1999) were used at a concen-tration of 1.0 �g/ml for immunoblot analysis.

Proteins were separated by SDS–PAGE and trans-ferred by electroblotting to nitrocellulose membranes;immunoblot analysis was performed as described else-where (Skålhegg et al., 1992). Primary antibodies weredetected by horseradish peroxidase-labeled protein A(dilution 1/25 000; Amersham; for polyclonal antibod-ies) or anti-mouse IgG (dilution 1/5000; TransductionLaboratories; for monoclonal antibodies) in the secondlayer and developed using ECL (Amersham). Gels werestained with Coomassie blue and blots were stainedwith Ponceau S prior to probing to confirm equal load-ing of the protein lanes. Each experiment was per-formed at least in triplicate.

Labeling of cDNA Probes

Human cDNA probes for RI � (Sandberg et al., 1987),RI � (Solberg et al., 1991), RII � (Øyen et al., 1988), RII �(Levy et al., 1988), and C � and � (Beebe et al., 1990) andrat probes for CamK II � and CamK I were labeled tospecific activities of 0.8–1.2 � 109 cpm/�g DNA using[�-32P]dCTP (sp act 3000 Ci/mmol; Amersham) and astandard random-prime kit (Amersham). Specific oligo-nucleotide probes for rat RI � 1a (96 bp) and RI � 1b(150 bp) were labeled with T4 polynucleotide kinaseand [�-32P]dCTP.

Northern Blot Analysis

Extraction of total RNA was performed using cesiumchloride centrifugation (Tasken et al., 1993b). RNA blotswere prepared essentially as described (Hoover et al.,1998). Briefly, total RNA (20 �g) from each sample wasdenatured in 50% (v/v) formamide, 6.0% formaldehydeat 50°C for 15 min. Total RNA (20 �g) from each samplewas denatured in 50% (v/v) formamide, 6.0% formal-dehyde at 50°C for 15 min. The RNA samples wereresolved in 1.5% agarose gels containing 6.7% formal-dehyde (v/v) in a 20 mM sodium phosphate buffer, pH7.0. Quality, migration, and lane loading of RNA wereassessed by ethidium bromide staining of the gels be-fore transfer by capillary blotting technique to nylonmembranes (Biotrans, ICN). The membranes werebaked (80°C, 1 h), prehybridized, and hybridized asdescribed by ICN. The filters were hybridized at 42°C in50% formamide with a labeled cDNA probe (1 � 106

cpm/ml). The membranes were washed four times in

133Regulation of PKA in Skeletal Muscle

2� SSC, 0.1% SDS at room temperature for 5–10 min,followed by two washes at 50°C in 0.1� SSC, 0.1% SDS.Autoradiography was performed using Hyperfilm MP(Amersham) and exposures varied for up to 2 weeks at�70°C. Results were evaluated using visual inspection.The autoradiograms were digitized into a computer,manipulated for brightness and contrast in AdobePhotoShop (V4.0; Adobe Systems), and imported intoAdobe Pagemaker (V6.5; Adobe Systems) to generatethe final figures. Before rehybridizations, filters werewashed in 50% formamide, 10 mM sodium phosphate,pH 6.5, at 65°C for 1 h.

In Vivo Transfection of Skeletal Muscle

We used a novel electropermeabilization technique tostudy the regulation of PKA promoter constructs in vivo(Mathiesen, 1999). Adult rat SOL or EDL were exposedsurgically and 50–100 �g of CAT reporter vector withPKA RI � or RII � promoter regions plus 25 �g of avector containing the mouse MCK promoter drivingluciferase (Walke et al., 1996) was injected into the mus-cles. The MCK–luciferase construct was used to controlfor transfection efficiency (Walke et al., 1996).

Following injection, whole muscle electroporationwas performed using two 1-cm-long, uninsulated silverwires (diameter 0.6 mm), 2.5 mm apart and connectedto a Haer 6 BP-AS stimulator (Frederick Haer). Thestimulator was used in the constant voltage mode. Theoutput voltage was limited by the maximal currentoutput (50 mA) and varied between 25 and 45 V. Thestimulation consisted of 30,000 bipolar square-wavepulses delivered in 30 trains (with 2-s intervals) of 1000pulses at 1000 Hz. Each pulse lasted 400 �s. After 3days, the muscles were removed and assayed for CATand luciferase activities.

We determined that the MCK promoter was not reg-ulated by innervation, since in a set of independentexperiments we cotransfected the MCK–luciferase con-struct into innervated and denervated SOL along with aminimal enkephalin promoter linked to CAT (Huggen-vik et al., 1991). Three days after in vivo electroporation,we did not detect any changes in reporter gene activityin response to either denervation or accompanying risesin cAMP (not shown).

CAT and Luciferase Assays

Individual transfected muscles (stored at �70°C)were removed and immediately homogenized with aPolytron in 1 ml of Reporter Lysis Buffer (Promega).Extracts were incubated for 15 min at room temperature

with gentle shaking and vortexed prior to centrifuga-tion at 5000 rpm in a table-top centrifuge. The superna-tants were divided into two tubes, one for the CATassay and one for the luciferase assay. The extract forthe CAT assay was heated at 60°C for 10 min and spunat 16,000g at 4°C. The supernatant was transferred to anew tube. Duplicate assays were prepared in a mi-crofuge tube containing 50–100 �l cell extract, 0.5 �Ci[3H]chloramphenicol (NEN), 2.5 �g n-butyryl-CoA(Sigma), and water to a total of 125 �l. Reactions pro-ceeded at 37°C for 3–20 h. To ensure that enzyme ac-tivities were within the linear range, dilutions of puri-fied CAT enzyme were assayed to produce a standardcurve. Reactions were terminated with 300 �l of mixedxylenes to the samples. Samples were vortexed andspun at 16,000g in a microfuge. The upper xylene phasewas transferred into a new tube and back-extractedwith the addition of 100 �l 0.25 M Tris, pH 8.0. Thesolution was vortexed and spun in a microfuge to sep-arate the phases and 100 �l was removed from thexylene phase and placed into a scintillation vial. Thesamples were analyzed in a scintillation counter; finalresults were obtained by subtracting background val-ues (no extract). The extract for the luciferase assay(noted above) was equilibrated to room temperature, 20�l of the extract was mixed with 100 �l of LuciferaseAssay Reagent (Promega) and placed in a TD-20/20Luminometer (Turner Design), and luciferase-gener-ated luminescence was integrated over a 10-s period.Each sample was assayed in duplicate. Negative con-trols containing untransfected muscle extracts, emptyvector sequences, or no extract were subtracted fromthe experimental values to obtain final results. To con-trol for differences in transfection efficiency among dif-ferent samples, CAT values were divided by the lucif-erase values.

Direct Stimulation of Skeletal Muscle in Vivo

This technique was performed as previously de-scribed in Windisch et al. (1998). Briefly, the sciaticnerve was cut in the thigh and its proximal end re-flected and sutured to the subcutis to prevent reinner-vation. A longitudinal opening through skin and fasciawas made to expose the SOL. The ends of two steelwires with their Teflon insulation removed for the dis-tal 20–25 mm were placed across the SOL, making surethat the uninsulated wires touched muscle fibers andnot tendons. Each wire was fixed to the tissue a fewmillimeters away from the SOL with a thin supramidthread. Through openings in the skin on the back andhead the insulated wires were pulled under the skin to

134 Hoover et al.

the head and from there through protective plastictubes (o.d. 0.6 mm) to rotating contacts fixed above therat. On the head, the connective tissue over the bonewas removed and three stainless steel bone screws wereinserted such that there was just room for the protectivetube between them. The exposed bone was coveredwith a thin film of dental glue. Dental cement (SimplexRapid; Austenal Dental Products Ltd., Harrow HA12HG, England) was then placed on the bone with thescrews and the protective tube pushed into the cementbefore it cured. Keeping the protective tube relativelyslack allowed the rat to move freely within the cagewhile being unable to reach the tube and break theconnection to the stimulator placed on a shelf above.Stimulation started within 24 h of denervation andlasted for 3 days to 4 weeks. The stimulus was bipolarwith a duration of 0.2 ms and an intensity of 10–15 mAin each direction. The stimulation pattern was 20 Hz for10 s every 20 s (slow pattern) or 100 Hz for 1 s every 60 s(fast pattern).

ACKNOWLEDGMENTS

We are appreciative to Drs. Daniel Goldman (University of Mich-igan, Ann Arbor, MI) for his gift of the MCK-Luciferase vector,Howard Schulman (Stanford University, Stanford, CA) for CamK IIcDNAs, and Kaare Gautvik (University of Oslo, Norway) for theCamK I cDNA. We thank Sigrid Schaller, Guri Oppsahl, and GladysJosefsen for superb technical assistance. This work was supported bythe Anders Jahres Foundation (F.H., K.T., T.L), the Nansen Fund(F.H.), the Norwegian Research Council (T.L., K.T.), the NorwegianCancer Society (K.T., M.K.D.), the Novo Nordisk Foundation (K.T.),and a European Union Biotechnology grant (BIO4 CT96 0216, T.L.).

REFERENCES

Amieux, P. S., Cummings, D. E., Motamed, K., Brandon, E. P., Wailes,L. A., Le, K., Idzerda, R. L., and McKnight, G. S. (1997). Compen-satory regulation of RIa protein levels in protein kinase A mutantmice. J. Biol. Chem. 272: 3993–3998.

Ausoni, S., Gorza, L., Schiaffino, S., Gundersen, K., and Lømo, T.(1990). Expression of myosin heavy chain isoforms in stimulatedfast and slow rat muscles. J. Neurosci. 10: 153–160.

Barradeau, S., Imaizumi-Scherrer, T., Weiss, M. C., and Faust, D. M.(2000). Alternative 5�-exons of the mouse cAMP-dependent proteinkinase subunit RI � gene are conserved and expressed in both aubiquitous and tissue-restricted fashion. FEBS Lett. 476: 272–276.

Beebe, S. J., Øyen, O., Sandberg, M., Frøysa, A., Hansson, V., andJahnsen, T. (1990). Molecular cloning of a tissue-specific proteinkinase (C gamma) from human testis—Representing a third iso-form for the catalytic subunit of cAMP-dependent protein kinase.Mol. Endocrinol. 4: 465–475.

Bezakova, G., Rabben, I., Sefland, I., Fumagalli, G., and Lømo, T.

(2001). Neural agrin controls acetylcholine receptor stability in skel-etal muscle fibers. Proc. Natl. Acad. Sci. USA 98: 9924–9929.

Brandon, E. P., Idzerda, R. L., and McKnight, G. S. (1997). PKAisoforms, neural pathways, and behaviour: Making the connection.Curr. Opin. Neurobiol. 7: 397–403.

Buonanno, A., Cheng, J., Venepally, P., Weis, J., and Calvo, S. (1998).Activity-dependent regulation of muscle genes: Repressive andstimulatory effects of innervation. Acta Physiol. Scand. 163: 217–226.

Buonanno, A., and Fields, R. D. (1999). Gene regulation by patternedelectrical activity during neural and skeletal muscle development.Curr. Opin. Neurobiol. 9: 110–120.

Calvo, S., Stauffer, J., Nakayama, M., and Buonanno, A. (1996). Tran-scriptional control of muscle plasticity: Differential regulation oftroponin I genes by electrical activity. Dev. Gen. 19: 169–181.

Carlsen, R. C. (1975). The possible role of cyclic AMP in the neuro-trophic control of skeletal muscle. J. Physiol. 247: 343–361.

Chahine, K. G., Baracchini, E., and Goldman, D. (1993). Couplingmuscle electrical activity to gene expression via a cAMP-dependentsecond messenger system. J. Biol. Chem. 268: 2893–2898.

Choi, R. C., Siow, N. L., Zhu, S. Q., and Tsim, K. W. (2000). ThecAMP-dependent protein kinase mediates the expression of AChEin chick myotubes. NeuroReport 11: 801–806.

Close, R. I. (1972). Dynamic properties of mammalian skeletal mus-cles. Physiol. Rev. 52: 129–197.

Collas, P., Le Guellec, K., and Tasken, K. (1999). The A-kinase-anchor-ing protein AKAP95 is a multivalent protein with a key role inchromatin condensation at mitosis. J. Cell Biol. 147: 1167–1180.

Connor, E. A., and McMahan, U. J. (1987). Cell accumulation in thejunctional region of denervated muscle. J. Cell Biol. 104: 109–120.

Dahle, M. K., Reinton, N., Orstavik, S., Tasken, K. A., and Tasken, K.(2001). Novel alternatively spliced mRNA (1c) of the protein kinaseA R1 alpha subunit; subunit is implicated in haploid germ cellspecific expression. Mol. Reprod. Dev. 59: 6–11.

Eide, T., Coghlan, V., Orstavik, S., Holsve, C., Solberg, R., Skalhegg,B. S., Lamb, N. J., Langeberg, L., Fernandez, A., Scott, J. D., Jahnsen,T., and Tasken, K. (1998). Molecular cloning, chromosomal local-ization, and cell cycle-dependent subcellular distribution of theA-kinase anchoring protein, AKAP95. Exp. Cell Res. 238: 305–316.

Eken, T., and Gundersen, K. (1988). Electrical stimulation resemblingnormal motor-unit activity: Effects on denervated fast and slow ratmuscles. J. Physiol. 402: 651–669.

Foss, K. B., Solberg, R., Simard, J., Myklebust, F., Hansson, V.,Jahnsen, T., and Tasken, K. (1997). Molecular cloning, upstreamsequence and promoter studies of the human gene for the regula-tory subunit RII� of cAMP-dependent protein kinase. Biochem.Biophys. Acta 1350: 98–108.

Goldman, D., Brenner, H. R., and Heinemann, S. (1988). Acetylcholinereceptor �-, �-, �-, and �-subunit mRNA levels are regulated bymuscle activity. Neuron 1: 329–333.

Gray, P. C., Tibbs, V. C., Catterall, W. A., and Murphy, B. J. (1997).Identification of a 15-kDa cAMP-dependent protein kinase-anchor-ing protein associated with skeletal muscle L-type calcium chan-nels. J. Biol. Chem. 272: 6297–6302.

Hennig, R., and Lømo, T. (1985). Firing patterns of motor units innormal rats. Nature 314: 164–166.

Hoover, F., Mathiesen, I., Skålhegg, B., Tasken, K., and Lømo, T.(2001). Differential expression and regulation of the PKA signallingsystem in fast and slow skeletal muscle. Anat. Embryol. 203: 193–201.

Hoover, F., Seleiro, E., Kielland, A., Brickell, P., and Glover, J. (1998).Retinoid X Receptor gamma gene transcripts are expressed by a

135Regulation of PKA in Skeletal Muscle

subset of early generated retinal cells and eventually restricted tophotoreceptors. J. Comp. Neurol. 391: 204–213.

Huggenvik, J. I., Collard, M. W., Stofko, R. E., Seasholtz, A. F., andUhler, M. D. (1991). Regulation of the human enkephalin promoterby two isoforms of the catalytic subunit of cyclic adenosine 3�,5�-monophosphate-dependent protein kinase. Mol. Endocrinol. 5: 921–930.

Hughes, S. M. (1998). Muscle development: Electrical control of geneexpression. Curr. Biol. 8: R892–894.

Igami, K., Yamaguchi, N., and Kasai, M. (1999). Regulation of depo-larization-induced calcium release from skeletal muscle triads bycyclic AMP-dependent protein kinase. Jpn. J. Physiol. 49: 81–87.

Imaizumi-Scherrer, T., Faust, D. M., Benichou, J. C., Hellio, R., andWeiss, M. C. (1996). Accumulation in fetal muscle and localizationto the neuromuscular junction of cAMP-dependent protein kinaseA regulatory and catalytic subunits RI � and C �. J. Cell Biol. 134:1241–1254.

Johnson, B. D., Brousal, J. P., Peterson, B. Z., Gallombardo, P. A.,Hockerman, G. H., Lai, Y., Scheuer, T., and Catterall, W. A. (1997).Modulation of the cloned skeletal muscle L-type Ca2� channel byanchored cAMP-dependent protein kinase. J. Neurosci. 17: 1243–1255.

Kemp, B. E., Graves, D. J., Benjamini, E., and Krebs, E. G. (1977). Roleof multiple basic residues in determining the substrate specificity ofcyclic AMP-dependent protein kinase. J. Biol. Chem. 252: 4888–4894.

Keryer, G., Skalhegg, B. S., Landmark, B. F., Hansson, V., Jahnsen, T.,and Tasken, K. (1999). Differential localization of protein kinase Atype II isozymes in the Golgi–centrosomal area. Exp. Cell Res. 249:131–146.

Knutsen, H. K., Tasken, K. A., Eskild, W., and Hansson, V. (1992a).Inhibitors of RNA and protein synthesis stabilize messenger RNAfor the RII � subunit of protein kinase A in different cellularcompartments. Biochem. Biophys. Res. Commun. 183: 632–639.

Knutsen, H. K., Tasken, K. A., Eskild, W., Jahnsen, T., and Hansson,V. (1991). Adenosine 3�,5�-monophosphate-dependent stabilizationof messenger ribonucleic acids (mRNAs) for protein kinase-A(PKA) subunits in rat Sertoli cells: Rapid degradation of mRNAs forPKA subunits is dependent on ongoing RNA and protein synthesis.Endocrinology 129: 2496–2502.

Knutsen, H. K., Tasken, K. A., Eskild, W., Jahnsen, T., and Hansson,V. (1992b). Half-lives of different sized mRNAs for the PKA subunitRI � are regulated differently in response to inhibition of transcrip-tion and translation. Biochem. Biophys. Res. Commun. 184: 454–460.

Laufer, R., and Changeux, J. P. (1987). Calcitonin gene-related peptideelevates cyclic AMP levels of chick skeletal muscle: Possible neu-rotrophic role for a coexisting neuronal messenger. EMBO J. 6:901–906.

Levy, F. O., Øyen, O., Sandberg, M., Tasken, K., Eskild, W., Hansson,V., and Jahnsen, T. (1988). Molecular cloning, complementary de-oxyribonucleic acid structure and predicted full-length amino acidsequence of the hormone-inducible regulatory subunit of 3�–5�-cyclic adenosine monophosphate-dependent protein kinase fromhuman testis. Mol. Endocrinol. 2: 1364–1373.

Liu, Y., Kranias, E. G., and Schneider, M. F. (1997). Regulation of Ca2�

handling by phosphorylation status in mouse fast- and slow-twitchskeletal muscle fibers. Am. J. Physiol. 273: C1915–1924.

Lømo, T., and Westgaard, R. H. (1975). Further studies on the controlof ACh sensitivity by muscle activity in the rat. J. Physiol. 252:603–626.

Mano, T., Iwase, K., Yoshimochi, I., Sawai, Y., Oda, N., Nishida, Y.,Mokuno, T., Kotake, M., Nakai, A., Hayakawa, N., et al. (1995).Changes in calmodulin concentration and cyclic 3�,5�-nucleotide

phosphodiesterase activity in skeletal muscle of hyper- and hypo-thyroid rats. J. Endocrinol. 146: 287–292.

Mathiesen, I. (1999). Electropermeabilization of skeletal muscle en-hances gene transfer in vivo. Gene Ther. 6: 508–514.

Morita, N., Namikawa, K., and Kiyama, H. (1995). Up-regulation ofPKA RI� subunit mRNA in rat skeletal muscle after nerve injury.NeuroReport 6: 1050–1052.

Murray, M. A., and Robbins, N. (1982). Cell proliferation in dener-vated muscle: Time course, distribution, and relation to disuse.Neuroscience 7: 1817–1822.

Ortmeyer, H. K. (1997). Insulin decreases skeletal muscle cAMP-dependent protein kinase (PKA) activity in normal monkeys andincreases PKA activity in insulin-resistant rhesus monkeys. J. BasicClin. Physiol. Pharmacol. 8: 223–235.

Øyen, O., Sandberg, M., Levy, F. O., Tasken, K., Beebe, S., Hansson,V., and Jahnsen, T. (1988). Molecular cloning and cell-specific ex-pression of newly discovered subunits of cAMP-dependent proteinkinases. Implications for different cellular responses to cAMP. AP-MIS Suppl. 2: 238–250.

Paradiso, K., and Brehm, P. (1998). Long-term desensitization ofnicotinic acetylcholine receptors is regulated via protein kinaseA-mediated phosphorylation. J. Neurosci. 18: 9227–9237.

Pette, D., and Vrbova, G. (1992). Adaptation of mammalian skeletalmuscle fibers to chronic electrical stimulation. Rev. Physiol. Biochem.Pharmacol. 120: 115–202.

Ray, A., Kyselovic, J., Leddy, J. J., Wigle, J. T., Jasmin, B. J., and Tuana,B. S. (1995). Regulation of dihydropyridine and ryanodine receptorgene expression in skeletal muscle. Role of nerve, protein kinase C,and cAMP pathways. J. Biol. Chem. 270: 25837–25844.

Roskoski, R., Jr. (1983). Assays of protein kinase. Methods Enzymol. 99:3–6.

Rotman, E. I., Murphy, B. J., and Catterall, W. A. (1995). Sites ofselective cAMP-dependent phosphorylation of the L-type calciumchannel � 1 subunit from intact rabbit skeletal muscle myotubes.J. Biol. Chem. 270: 16371–16377.

Sandberg, M., Tasken, K., Øyen, O., Hansson, V., and Jahnsen, T.(1987). Molecular cloning, cDNA structure and deduced amino acidsequence for a type I regulatory subunit of cAMP-dependent pro-tein kinase from human testis. Biochem. Biophys. Res. Commun. 149:939–945.

Sedmak, J. J., and Grossberg, S. E. (1977). A rapid, sensitive, andversatile assay for protein using Coomassie Brilliant Blue G250.Anal. Biochem. 79: 544–552.

Seedorf, U., Leberer, E., and Pette, D. (1986). Neural control of geneexpression in skeletal muscle. Effects of chronic stimulation onlactate dehydrogenase isoenzymes and citrate synthase. Biochem. J.239: 115–120.

Senter, L., Ceoldo, S., Petrusa, M. M., and Salviati, G. (1995). Phos-phorylation of dystrophin: Effects on actin binding. Biochem. Bio-phys. Res. Commun. 206: 57–63.

Sherman, S. J., Chrivia, J., and Catterall, W. A. (1985). Cyclic adeno-sine 3�:5�-monophosphate and cytosolic calcium exert opposingeffects on biosynthesis of tetrodotoxin-sensitive sodium channels inrat muscle cells. J. Neurosci. 5: 1570–1576.

Skålhegg, B., and Tasken, K. (1997). Specificity in the cAMP/PKAsignaling pathway. Differential expression, regulation, and subcel-lular localization of subunits of PKA. Front. Biosci. 2: 331–342.

Skålhegg, B. S., Landmark, B. F., Døskeland, S. O., Hansson, V., Lea,T., and Jahnsen, T. (1992). Cyclic AMP-dependent protein kinasetype I mediates the inhibitory effects of 3�,5�-cyclic adenosinemonophosphate on cell replication in human T lymphocytes. J. Biol.Chem. 267: 15707–15714.

136 Hoover et al.

Skalhegg, B. S., and Tasken, K. (2000). Specificity in the cAMP/PKAsignaling pathway. Differential expression, regulation, and subcel-lular localization of subunits of PKA. Front. Biosci. 5: D678–693.

Solberg, R., Sandberg, M., Natarajan, V., Torjesen, P. A., Hansson, V.,Jahnsen, T., and Tasken, K. (1997). The human gene for the regu-latory subunit RI� of cyclic adenosine 3�,5�-monophosphate-depen-dent protein kinase: Two distinct promoters provide differentialregulation of alternately spliced messenger ribonucleic acids. En-docrinology 138: 169–181.

Solberg, R., Tasken, K., Keiserud, A., and Jahnsen, T. (1991). Molec-ular cloning, cDNA structure and tissue-specific expression of thehuman regulatory subunit RI � of cAMP-dependent protein ki-nases. Biochem. Biophys. Res. Commun. 176: 166–172.

Supinski, G., Stofan, D., Callahan, L. A., Nethery, D., and DiMarco, A.(2000). Effects of protein kinase A inhibition on rat diaphragm forcegeneration. Respir. Physiol. 120: 115–123.

Tasken, K., Andersson, K. B., Skålhegg, B. S., Tasken, K. A., Hansson,V., Jahnsen, T., and Blomhoff, H. K. (1993a). Reciprocal regulationof mRNA and protein for subunits of cAMP-dependent proteinkinase (RIa and Ca) by cAMP in a neoplastic B cell line (Reh). J. Biol.Chem. 268: 23483–23489.

Tasken, K., Skålhegg, B. S., Solberg, R., Andersson, K. B., Taylor, S. S.,Lea, T., Blomhoff, H. K., Jahnsen, T., and Hansson, V. (1993b).

Novel isozymes of cAMP-dependent protein kinase exist in humancells due to formation of RI�–RI� heterodimeric complexes. J. Biol.Chem. 268: 21276–21283.

Tasken, K., Kvale, D., Hansson, V., and Jahnsen, T. (1990). Proteinkinase C activation selectively increases mRNA levels for one of theregulatory subunits (RI�) of cAMP-dependent protein kinases inHT-29 cells. Biochem. Biophys. Res. Commun. 172: 409–414.

Walke, W., Xiao, G., and Goldman, D. (1996). Identification andcharacterization of a 47 base pair activity-dependent enhancer ofthe rat nicotinic acetylcholine receptor �-subunit promoter. J. Neu-rosci. 16: 3641–3651.

Windisch, A., Gundersen, K., Szabolcs, M. J., Gruber, H., and T, L. M.(1998). Fast to slow transformation of denervated and electricallystimulated rat muscle. J. Physiol. 510: 623–632.

Wolff, J. A., Malone, R. W., WIlliams, P., Chong, W., Ascardi, G., Jani,A., and Fengler, P. L. (1990). Direct gene transfer into mouse musclein vivo. Science 247: 1465–1468.

Xu, F., and Salpeter, M. M. (1995). Protein kinase A regulates thedegradation rate of Rs acetylcholine receptors. J. Cell. Physiol. 165:30–39.

Yaspelkis, B. B., 3rd, Castle, A. L., Farrar, R. P., and Ivy, J. L. (1997).Contraction-induced intracellular signals and their relationship tomuscle GLUT-4 concentration. Am. J. Physiol. 272: E118–125.

Received May 31, 2001Revised September 24, 2001

Accepted October 10, 2001

137Regulation of PKA in Skeletal Muscle