The structure and activity of cyclic AMP-dependent protein kinase A

Recent data concerning the activity of protein kinase A (cyclic adenosine 3\5 ' monophosphatedependent protein kinase) in processes associated with gene expression and regulation, as well as general information concerning protein kinase A structure, synthesis and activity are presented in this article. Protein kinase A is a tetrameric protein comprising two regulatory and two catalytic subunits. The enzyme is activated by binding cAMP to the regulatory subunit, dissociation of the holoenzyme, and liberation of free catalytic subunits. Cyclic AMP is synthesized after adenylyl cyclase is activated by extracellular stimuli. Two types of protein kinase A are known, protein kinase A I and protein kinase A I I . They differ in their regulatory subunits, R I and R I I . Each regulatory subunit may occur in two isoforms, R I a, R I (3, R I I a and R I I (3. Three variants of catalytic subunits exist, C a, C (3 and C y ; the C (3 subunit may occur in three subisoforms, C (31, C p2 and C P3. The R I a, R I I a, and C a subunits are expressed ubiquitously, while the R I p, R I I p, C p and C y subunits are expressed in endocrine, neuroendocrine, neural and leukemic tissues. The activity of protein kinase A depends on cellular localization, which determines the access of the enzyme to cAMP and substrate, and on the proportions of the regulatory and catalytic subunits. Protein kinase A I I is associated with cellular structures through the regulatory subunit R I I , which is bound to protein kinase A anchor proteins (AKAP) or to microtubule-associated protein (MAP-2). Protein kinase A I is soluble in cellular cytosol. The protein kinase A subunits, R I a, R I p, R I I a, R I I p C a, C p, and C y are encoded by separate genes whose promoters are activated by cAMP and bind Sp 1; they lack TATA and CAAT sequences and have several transcription start sites. Protein kinase A is involved in the expression of various proteins through the regulation of the activity and synthesis of transcription factors, cycline A, cycline-dependent protein kinase inhibitors, and phosphorylation of microtubule-associated proteins. Protein kinase A is involved in hormone synthesis and secretion. Basal cellular metabolism is affected by protein kinase A through phosphorylation and regulation of the activity of protein kinases, phosphorylases, phosphatases, protease inhibitors, and through 214 STRUCTURE AND ACTIVITY OF PROTEIN KINASE A influencing their synthesis. The cellular compartmentalization of diverse isoforms of protein kinase A, differing in activity and substrate specificity, brings about a variety of cAMP-mediated cellular responses to different hormonal stimuli.


INTRODUCTION
Adenosine 3',5'-monophosphate-dependent protein kinase (protein kinase A, ATP-protein phosphotransferase; EC 2.7.1.37) is known to play an important regulatory role in many biochemical processes.The enzyme is activated by cAMP binding to its regulatory (R) subunit, which results in the dissociation of protein kinase A and liberation of active, catalytic (C) subunit.Cyclic AMP is accumu-lated following stimulation of membrane receptor adenylyl cyclase coupled to the stimulatory G protein a subunit or (3y subunits of membrane receptors (Taussing and Gilman, 1995;Daaka et al., 1997;Dessauer and Gilman, 1997;Dessauer et al., 1997;Bayewitch et al., 1998).Protein kinase A mediates the effects of cAMP through phosphorylation of specific protein substrates.A general review of protein phosphorylation was presented earlier (Ostrowska, 1987), and new data concerning cAMP-mediated phosphorylation and regulation of transcriptional processes have recently been published in Polish literature (Ostrowska, 1999).

PROTEIN KINASE A STRUCTURE AND ISOENZYMES
Protein kinase A is a tetrameric holoenzyme comprising two regulatory and two catalytic subunits (Gill and Garren, 1970;Hofmann et al., 1975;McKnight et al., 1988;Taylor et al., 1990).A diagrammatic representation of protein kinase A structure is presented in Figure 1.Each of the four known R subunit retains the general features of the molecule.The amino-terminal (N-terminal) region of the R subunit contains a dimer interaction site encompassing approximately eighty amino acid residues, rich in hydrophilic and charged amino acids.These are followed by two antigenic sites separated by a "hinge", a pseudosubstrate region.The hinge region is an auto-inhibitor substrate-like site in which the R subunit binds to the C subunit.The carboxyl-terminal (C-terminal) part of the R subunit contains two tandem cAMP-binding sites, A and B. Site A is located inside of the molecule, site B is situated at the C-terminal.The spatial structure of the cAMP-binding sites is similar to catabolite gene activator protein (CAP) from Escherichia coli (CAP binds to the lac promoter in the presence of cAMP, thus promoting the transcription of the lac gene).The model of cAMP binding sites in the R subunit was built by fitting the R subunit amino acid sequence into the crystallographic coordinates of CAP.The cAMP-binding site is a (3-barrel composed of eight (3-strands and environed by three oc-helices.Cyclic AMP is bound inside of the (3-barrel due to interactions between glycine, glutamic acid and arginine (see section cAMP-binding sites), located on the strands of (3-barrel and exocyclic oxygen of the phosphate moiety and 2-OH' of the ribose of cAMP (Taylor et al., 1990;Ostrowska, 1999).Recently, a crystallographic study of the fragment of the R subunit of protein kinase A revealed a similar structure for the A cAMP binding site.Site B, located at the C-terminal, is less shielded by (3-strands and more accessible to cAMP (Gibson and Taylor, 1997).
The catalytic subunit is a slightly elongated molecule comprising two lobes, smaller and larger, separated by a deep cleft.The smaller lobe is associated primarily with the binding of MgATP.It corresponds to the N-terminal segment Figure 1.The scheme of protein kinase A structure.R, regulatory subunit, C, catalytic subunit, D, dimerization domain, H, hinge region, A and B, the sites binding cAMP (according to data presented by McKnight et al., 1988;Scott et al., 1990;Taylor et al., 1990;Gibson and Taylor, 1997 and as presented in Ostrowska, 1999) (residues 15-127).This domain is dominated by a (3-sheet consisting of five antiparallel (3-strands.The only helical element is inserted between (3-strands 3 and 4. The N-terminal of the C subunit begins with an amphipathic a helix that lies primarily along the surface of the larger lobe and is not visible in the crystal structure.In mammalian species, the N-terminal glycine is myristoylated.The myristoyl group stabilizes the C subunit (Taylor et al, 1990).
The larger lobe is associated with peptide substrate binding and catalysis.It is predominantly helical and has seven a-helices.The only (3 structure region is located on the surface of the cleft at the interface between the two lobes where four anti-parallel strands form a sheet.The regions important for recognition of the peptide and catalysis are located within this large lobe (from Glu 127 to Glu 331).The seventy C-terminal amino acids (281-350) extend over a large portion of the surface of the enzyme from the bottom of the larger lobe to the top of the smaller one.A part of this region appears to participate in the recognition of both the protein substrate and ATP.The catalytic subunit contains several phosphorylation sites Ser 14,Ser 139,Thr 197,Ser 338.The interaction between R and C subunits occurs through amino acids located in the pseudo-substrate region and cAMPbinding site A of the regulatory subunit and His 87, Trp 196, Lys 213 of the cata-lytic subunit (Knighton et al, 1991;Taylor et al, 1992;Gibson et al., 1997;Gibson and Taylor, 1997).
Ion exchange chromatography on DEAE cellulose revealed the presence of two isoforms of protein kinase A that were named type I and type II.The type I was eluted at 0.05-0.1 M NaCl and type II at 0.15-0.25 M NaCl.Protein kinase A isoforms vary in their regulatory subunits, which differ in molecular weight, amino acid composition and immunological determinants.These dissimilarities result in holoenzymes with different properties such as the high affinity for MgATP of protein kinase A I or the ability of protein kinase A II to catalyze intramolecular phosphorylation, and differences in catalytic subunit recognition and substrate specificity (McKnight et al., 1988;Taylor et al., 1990;Gibson et al, 1997;Gibson and Taylor, 1997).
A study involving cloning of cDNA from different tissues showed further heterogeneity of protein kinase A subunits.In mammals, four distinct isoforms of the R subunit have been demonstrated, R I a, R I [3, R II a and R II (3 and three isoforms of C subunit C a, C (3 and C y. Additionally, in the mouse brain, three isoforms of the catalytic C (3 subunit were distinguished: C (31, C (32 and C (33.The isoforms RI a, RII a and C a are expressed ubiquitously, the expression of others RI (3, RII (3, C (3 and C y is restricted to tissues having specific functions (Clegg et al., 1988;McKnight et al., 1988;Garrel et al., 1993;Guthrie et al., 1997).

THE GENES AND PROTEINS OF PROTEIN KINASE A SUBUNITS
Mouse regulatory RI a and RI (3 subunit nucleotide and amino acid sequences are presented in Figure 2.Each of the coding regions of mouse regulatory R I a and RI (3 subunits contains 1140 nucleotides encoding 380 amino acids (Clegg et al, 1988).Sixty-seven (17.6%) amino acid residues are unconsented between RI a and RI (3 subunits.The majority of unconserved amino acids are located in the in N-terminal eighty amino acid region of regulatory R I subunits.The coding regions of R I a and R I (3 subunit genes differ in 295 (25.9%) nucleotides.The diversity in codon usage is considerably higher than amino acid and nucleotide diversity.
The rat regulatory R II a and R II (3 subunit nucleotide and amino acid sequence is presented in Figure 3.The subunits are 400 amino acid residue molecules.The subunits differ in N-terminal dimerization domains.The rat N-terminal part of the regulatory RII a subunit was not sequenced (Scott et al., 1987), so the nucleotide and amino acid sequences are not presented.The majority of amino acid sequences of regulatory R II a and R II (3 subunits are conserved, although certain parts of conserved sequences are shifted.The diversity in coding triplets is higher than in amino acid sequence.

Dimerization domains
The amino acid sequence of regulatory subunits RI a and RI (3 and the secondary structure of the dimerization region of R I a are presented in Figure 2. In the dimerizing fragment of the regulatory RI a subunit, two inter-chain disulfide bonds are formed between Cys 18 and Cys 39 of the anti-parallel N-terminal chains (Zick and Taylor, 1982;Bubis et al., 1987).The R I (3 subunit dimerization region contains four cysteine residues, Cys 6, Cys 18, Cys 34 and Cys 39, and more disulfide bonds may be formed.The disulfide bonds are resistant to reduction.Mutagenesis of Cys 39 to His does not disturb the dimerization of R I a regulatory subunits, thus, it seems that the cysteine bonds are not necessary for dimers to form.Circular dichroism studies and algorithm data predict an approximately 40-50 % helical structure of the dimerizing fragment of RI a subunit (Leon et al., 1997).The small helical fragments of Ser 14 -Glu 17 and Ile 27 -Leu 31 are separated with small (3-strands Cys 18 -Gin 23 and Asp 33 -Pro 43, which are followed by a larger ahelix containing the fragment from Met 47 to Glu 60.The eighty amino acid dimerization regions of regulatory R I a and R I (3 subunits differ in thirty-five (45%) amino acids, however, all cysteine residues are located on P-strands, and the majority of leucine residues are conserved in a-helices.
The N-terminal fragment of the regulatory R II subunit also plays the role of a cellular protein anchor.The first fourteen amino acids are essential for regulatory RII subunit dimerization and for protein anchoring.A RII a regulatory subunit mutant lacking these amino acids is unable to dimerize and to bind MAP 2. The first thirty N-terminal amino acids seem to be sufficient for regulatory RII subunit dimerization but not for cellular protein binding.The protein fusion of thirty Nterminal amino acids and protein carrier is able to form a dimer, but it can not bind MAP 2 (Scott et al., 1990).Additionally, Phe 36 is thought to be necessary for dimer formation (Li and Rubin, 1995).
The amino acid sequence of bovine regulatory RII a and RIIP subunit dimerization /anchoring domains is presented in Figure 4 A, and the secondary and tertiary structure of this region of the RII a subunit is presented in Figure 4 B. The region is constructed of P-strand (residues 1-5), p-turn I (residues 6-9), a-helix I (residues 10-23), P-turn II (residues 24-28) and a-helix II (residues 29-42).The dimerization domain is maintained by intra-and inter-subunit interactions of the aromatic ring of phenylalanine residues, and the hydrogens of residues of valines and leucines.The intra-subunit interactions are between Val 20 of a-helix I and Phe 31 of a-helix II.The inter-subunit interactions are between Phe 36 of a-helix II of one of the R II a subunit molecule and Val 20 and Leu 13 of the a-helix I of other R II a subunit molecule (Hausken et al., 1994(Hausken et al., , 1996;;Newlon et al., 1997).
The anchor protein binding domain is supposed to be located in a larger regulatory R II subunit N-terminal fragment, comprising forty-four (Lou et al., 1990;Hausken et al., 1994) or eighty-two (Scott et al., 1990) amino acid residues.Isoleucine residues 3 and 5 are required for binding anchoring proteins (Hausken et al., 1994(Hausken et al., , 1996)).The amino acids of P-turns, Pro 7 and Pro 26 in regulatory RII a subunit and Ala 7 and Ala 26 in R II P subunit are thought to be responsible for the specificity of R II a and R II P subunit binding to anchor proteins.The amino acids known to be necessary for dimers to form, Leu 9, Leu 13, Val 20, Phe 31 and  The amino acid sequences of the dimerization regions of bovine R I a and R II a and mouse R I P and R II P subunits are compared in Figures 5 A and B, respectively.The amino acid sequence is not conserved in the dimerization regions between the cytoplasmic soluble RI and cellular structure-bound RII subunits in both a and P isoenzymes.The regulatory R I a subunit does not contain phenylalanine residues in their forty amino acid N-terminal region, which maintain the dimer structure in regulatory RII subunits.The leucine residues separated by charged amino acids occur in both RI and RII subunits and may form leucine zipper-like structures one with another, or with other cellular proteins.The isoleucine residues (3 and 5), required for target protein anchoring in the regulatory R II a and R II P subunits are absent in R I a and R I P subunits.All isoforms of regulatory subunits, RI a, R I P, R II a and RII P contain of serine and threonine residues that may be potential phosphorylation sites that may influence subunitsubunit and subunit-cellular protein interactions.
Two antigenic sites separated by a "hinge", pseudo-substrate region are located next.et al, 1988) and R II p (Scott et al., 1987) subunits

The hinge region
The hinge region is an auto-inhibitor substrate-like site in which the regulatory subunit binds to the catalytic subunit.The hinge region is more sensitive to protease than other domains of the regulatory subunit of protein kinase A. The amino acid sequences of the hinge regions of mouse RI a and RI (3; rat RI (3 and rat R II a and R II (3 subunits are presented below: The hinge regions of regulatory R II a and R II P subunits may be phosphorylated by protein kinase A at Ser 96.The hinge regions of the regulatory subunits R I a and R I P do not possess a phosphorylation site.

cAMP -binding sites
The C-terminal part of the regulatory subunit contains two tandem cAMP-binding sites, A and B. Site A is located inside of the molecule, in a region of approximately 140 to 260 amino acids; site B is situated at the C-terminal.The fifteen amino acid region of the cAMP binding sites is highly conserved in all cAMP binding proteins.This region is thought to be the core of the cAMP binding site (Shabb and Corbin, 1992).The amino acid sequences of the core A and B cAMP binding sites of bovine R I a (Titani et al., 1984), mouse R I a and R I P (Clegg et al., 1988), bovine RII a (Takio et al., 1982) and bovine RII P (Luo et al, 1990)  The amino acid residues that interact with exocyclic phosphate and ribose hydroxyl group of cAMP, Gly, Glu and Arg (bolded) are conserved in both A and B cAMP binding sites in RI a, R I P, R II a and R II P subunits.The amino acid sequence is conserved in cAMP binding site A between a and P isoforms of the same, R I or R II, regulatory subunit.The differences between site A cores of R I and R II subunits are in two amino acids (lie/Met and Gly/Asn, underlined).The differences between site B cores of R I and R II subunits are in five amino acids (underlined).Sites A and B of all isoforms of regulatory subunits contain threonine or serine residues that may be phosphorylation sites.The site B core of R I a and R I P subunits contains one threonine residue, while other cAMP binding site cores (sites A and B of R II a and R II p, and site A of R I a and R I P) contain two amino acid residues that may be phosphorylated.The pattern of phosphorylation of cytosol soluble R I subunit may differ from that of membrane-bound R II subunit.
The conserved amino acid sequence in cAMP binding sites A and B may suggest that the sites A and B originate from a gene duplication.The amino acids known to be involved in cAMP binding are encoded by the same or different nucleotide triplets in sites A and B. The triplets coding for Gly (GGG) and Glu (GAG) are the same in sites A and B of the mR I P, and the triplets coding for Glu (GAA) are conserved in sites A and B of the bR II P subunit.However, site A Arg is encoded by AGA in mR I a , mR IP and bR IIP and site B Arg is encoded by CGT in mR I a, CGG in mR I P and CGA in bR II p.The coding triplets are partially conserved, partially changed.Gene duplication of cAMP binding sites in the regulatory subunit of protein kinase A during evolution can not be excluded.

Catalytic subunit
The nucleotide and amino acid sequences of mouse catalytic subunits C a and C P and their isoforms, C p2 and C p3 are presented in Figure 6.In the 240nucleotide promoter region of the C a subunit gene, five transcription start sites and three DNA sequences binding Sp 1 transcription factor have been found.In the same promoter region of the C P subunit gene, four transcription start sites have been found and one Sp 1 binding sequence.Three additional Sp 1 regulated sequences are located in the region extended to -390 nucleotides of the 5'-untransalated region.
The exon-intron composition of the mouse catalytic C a subunit is presented in Table 1, and a diagram of the gene is shown in Figure 7.The C a subunit gene comprises ten exons and nine introns.The first exon also contains 350 nucleotides of the 5'-untranslated promoter sequence (Chrivia et al., 1988).
Two polyadenylation AATAAA signals have been found in the 3'-untranslated region at 2057-2062 and 2076 -2081 nucleotides in the C a subunit gene (Chrivia et al., 1988), however, the polyadenylation signals are absent in the 3'-untranslated region of the C P subunit gene (Uhler et al., 1986a).
The amino acid homology between C a and C P is 91%, the weakest homology is in the N-terminal part of the molecule, 70% identity over 70 N-terminal amino acids (Uhler et al., 1986a,b).Mouse brain C P catalytic isoforms C p2 and C P3   (Chrivia et al., 1988) differ from C pi (C P) in the length of the protein chain and N-terminal amino acid composition.The C P2 and C P3 proteins are not myristoylated, unlike the other catalytic subunits (Guthrie et al., 1997).

SYNTHESIS OF PROTEIN KINASE A SUBUNITS
The regulatory RI a, RIP, RII a, RIIP and catalytic C a, C P and C y subunits of protein kinase A are expressed from single genes (Uhler and McKnight, 1987;Clegg et al., 1988;McKnight et al., 1988;Oyen et al., 1988;Landmark et al., 1991;Garrel et al., 1993).The isoforms of mouse C (3 subunit, C pi, C P2 and C P3 originate from the same gene (Guthrie et al., 1997).The genes of protein kinase A share some common characteristics: they belong to the class of genes whose promoters are GC-rich, lack TATA boxes, and initiate transcription at multiple sites.Most of them are activated by cAMR The expression of the R II P regulatory subunit is induced hormonally and limited to specific tissues such as endocrine, neuroendocrine, neural and leukemic.The DNA for the R II P subunit was isolated from genomic libraries of the mouse (Singh et al., 1991) and rat (Kurten et al., 1992) and sequenced.The 5'-flanking, 4530 nucleotide non-coding region of the rat liver R II p gene has been sequenced and characterized.The core of the promoter region starting from nucleotide -400 is presented below (Kurten et al., 1992).The promoter region is GC-rich and does not contain TATA nor CAAT boxes.Nucleotide sequence analysis shows two tandem AP2 consensus sequences (boxed) and five Spl binding sites (underlined).RNase protection experiments indicated the presence of 11 transcription initiation sites, indicated by arrows.A transfection study of rat granulosa cells using different constructs of R II P promoter and mobility shift assay carried out on nuclear extracts of rat brain, ovary and liver showed that the region induced by cAMP was localized between 394 and 176 nucleotides of the promoter region.The sequences distal and proximal to this region are involved in the basal gene expression (Kurten et al., 1992).The AP2 binding sites confers cAMP inducibility.The GC-rich sequences are activated by Spl protein and may be activated by cAMP (Ahlgren et al., 1999).In the mouse, the AP-1 sequence located 1180 bases upstream from the ATG initiation codon has been identified.This site may be additionally regulated by cAMP (Singh et al., 1991).The polyadenylation signals AATAAA were found in the regions: 1508-1513; 1761-1766; 3041-3046; and 3058-3064 nucleotides in the 3'-end of the rat ovary R II P subunit cDNA (Jahnsen et al., 1986).
In the anterior pituitary gland, mRNAs of all known regulatory subunits, RI a, R IP, RII a and RIIP and two catalytic subunits, C a and C P of protein kinase A are expressed.The expression of mRNAs of RIIP and C a subunits is activated by cAMP.This activation is supposed to be mediated via cAMP induced protein(s) that are involved in the induction of R II P and C a genes (Garrel et al., 1993).

ACTIVATION AND REGULATION OF PROTEIN KINASE A ACTIVITY
Protein kinase A is activated by cAMP binding to its R subunit followed by dissociation of the C subunit.Since the protein kinase A complex is composed of two regulatory and two catalytic subunits and each of them contains two cAMP binding sites, the activation of one molecule of the enzyme requires four molecules of cAMP.The kinetic study of cAMP binding to regulatory subunits mutated in the A or B cAMP-binding sites has shown that the cyclic nucleotide binds first to site B, causing conformational changes that make site A accessible to cAMP.The liberation of the catalytic subunit results in uncovering its active center that had been blocked by an auto-inhibitory region of the regulatory subunit, which has an amino acid sequence similar to the substrate sequence recognized by catalytic subunit (McKnight et al., 1988;Taylor et al, 1990).

Protein kinase A anchoring proteins
The activity of protein kinase A is compartmentalized to specific regions of the cell; compartmentalization is determined by subcellular localization of the regulatory subunit.The regulatory R1 subunit is found primarily in cytoplasm, while the regulatory R II subunit binds to membrane structures such as plasma membranes, Golgi complexes, centrosomes, mitotic spindle poles and nuclear proteins (Nigg et al., 1985;Meinkoth et al., 1990;Podesta et al, 1991;Coghlan et al, 1994;Faux and Scott, 1996;Dell'Acqua and Scott, 1997;Colledge and Scott, 1999).
Two types of regulatory R II subunit anchoring proteins are known, microtubule-associated protein 2 (MAP-2) and protein kinase A anchor protein (AKAP).The MAP-2 (270-300 kDa) copurifies with brain microtubules.It appears as a projection on the microtubule surface and also has the property of promoting microtubule assembly in vitro (Vallee, 1980;Majewska, 1995).
At the beginning, AKAPs were isolated from bovine (p75) and rat (pi50) brains and thereafter from other animal tissues and species.These proteins bind calmo-dulin.More than forty regulatory R II a subunit binding bands, ranging in size from 25 to 300 kDa have been detected electrophoretically in various tissues (Carr et al, 1992).Several AKAPs have been cloned and well characterized, AKAP-75 from bovine brain (Hirsch et al, 1992), Ht31 from human thyroid (Carr et al, 1992), AKAP-84 from hepatic cells (Chen et al., 1997), AKAP-550 from Drosophila melanogaster (Han et al, 1997) and AKAP-95 from rat pituitary GH 4 Cj cells (Coghlan et al, 1994).The latter AKAP has been shown to be an ubiquitous nuclear regulatory RII subunit anchoring protein also having a DNA binding domain.It has been suggested that AKAP-95 plays a role in promoting the binding of transcriptional factors to DNA, affecting chromatin structure and activating transcription.
In the regulatory RII subunit, the domain binding MAP-2 and AKAPs is located within the eighty amino acid N-terminal.Dimerization of the regulatory RII (3 subunit is required for binding anchoring protein (Hausken et al, 1994;Dell'Acqua and Scott, 1997).The MAP-2 protein binds the regulatory R II subunit at a site near its N-terminal and, in AKAPs, the RII binding site is located at the C-terminal part of the molecule (Carr etal, 1992;Hirsch etal., 1992;Coghlan etal, 1994;Chen et al, 1997).The regions binding the RII subunit in AKAPs and MAP-2 do not have a conserved primary structure.Studies of the secondary and tertiary structures of the regulatory R II subunit and its anchoring proteins, and assay of mutated R II subunit and mutated proteins anchoring the R II subunit revealed that hydrophobic surfaces at the tethering sites of AKAPs and complementary surfaces on R II subunit dimers are essential for formation of stable AKAP-protein kinase A complexes (Hausken et al, 1994(Hausken et al, , 1996;;Li and Rubin, 1995;Newlon et al, 1997).

Proportions of regulatory and catalytic subunits
The endogenous activity of protein kinase A depends on the proportions of the regulatory and catalytic subunits and combination of their isoforms.Maintaining cAMP-dependent control of protein kinase A activity is due to a certain excess of the regulatory subunit in relation to the catalytic subunit.In an excess of catalytic subunit, the cell would lose the effect of cAMP on kinase activation and would have a continuously active protein kinase A. The cell is protected against an excess of catalytic subunit via a compensation mechanism of the regulatory subunit.This mechanism was observed in S49 cells overproducing the catalytic subunit.In these cells, the level of regulatory R I a subunit increased exceeding the catalytic subunit level (the level of R II subunit remained constant).It has been postulated that this compensation results from an increase in the stability and rate of translation of regulatory R I a subunit mRNA and stabilization of R I a protein (McKnight et al, 1988;Knutsen et al, 1991).On the other hand, the associated holoenzyme is protected against proteolysis.The catalytic subunit dissociated from the regulatory subunit easily undergoes proteolysis (Hemmings, 1986).Thus, the excess of regulatory over catalytic subunits provides a threshold for cAMP concentration-dependent protein kinase A activation and establishes protection of protein kinase A activity.

THE ROLE OF PROTEIN KINASE A IN CELLULAR METABOLISM
Protein kinase A is involved in various cellular processes.Protein synthesis is affected by protein kinase A at the transcription level.Genes are regulated via phosphorylation and activation of cAMP response element binding protein (CREB) interacting with cAMP response element (CRE).The CREB is a 45 kDa protein which dimerizes after cAMP dependent phosphorylation and binds CRE located in the promoter regulatory regions of cAMP-activated genes (Montminy and Bilezikijan, 1987;Gonzalez et al, 1989).It recruits basal transcriptional factors TFIIB, TFIID, TATA-box binding protein (TBP) to the promoter (Ferreri et al, 1994;Kwok et al, 1994;Parker et al, 1996;Nakajima et al, 1997a).Phosphorylated CREB dimerizes by a leucine zipper structure, forming a homodimer, or, if it dimerizes with another transcriptional factor, a heterodimer is formed.Phosphorylation and activation of CREB promote binding a large 265 kDa protein, CREB binding protein (CBP).The CBP binds histone acetyltransferase and itself possesses intrinsic histone acetyltransferase and helicase activity (Bannister and Kouzarides, 1996;Nakajima et al, 1997b).This complex recruits RNA polymerase II to the promoter and activates transcription (Ferreri etal, 1994;Kwok etal, 1994;Liang and Hai, 1997;Nakajima etal, 1997a,b).It is suggested that CRE is also activated by the R II P regulatory subunit of protein kinase A (Srivastava et al, 1998).Another cAMP-regulated element is the AP-2 sequence (CCCCAGGC consensus) activated by AP-2 protein (Johnson et al., 1997;Wang et al., 1997).
Recently, a role for protein kinase A in cell cycle regulation has been reported.Protein kinase A functions as a positive regulator of the cycline A gene promoter.Cycline A associates with cycline-dependent protein kinases (cdks) during phases of the cell cycle and is required for DNA replication in the S phase (Desdouets et al., 1995).Protein kinase A-dependent phosphorylation interferes with the tyrosine kinase phosphorylation of growth receptors and Ras-stimulated activation of Raf-1 kinase, resulting in the inhibition of MAP kinase and alteration of cell proliferation (Giasson et al., 1997;York et al, 1998).Protein kinase A downregulates oncoprotein Op 18 (metablastin or stathmin), the protein destabilizing microtubules, which diminishes oncoprotein activity (Gradin et al., 1998).Furthermore, protein kinase A is involved in the arrest of virus multiplication and cell transformation by induction of p27 and p21, proteins that inhibit viral promoter activation and tumor development (Chinery et al., 1997;Deleu et al., 1998).Protein kinase A phosphorylates cytoplasmic and membrane protein substrates that influence the movement of proteins in the cytoplasm and ion channel permeability, and is involved in the secretion processes (Vallee, 1980;Carr et al., 1992;Hirsch et al., 1992;Kurashima et al., 1997).
Changes in the phosphorylation of key enzymes of metabolic pathways affect basal cellular metabolism leading to intensification of catabolic processes (Cohen and Hardie, 1991;Jakubowicz and G^sior, 1993).Phosphorylation of glycogen phosphorylase activates the enzyme, while phosphorylation of glycogen synthase inhibits it, leading to a decrease in glycogen synthesis.The phosphorylation of acetyl CoA carboxylase and 3-hydroxy-3-methyl-glutamyl CoA reductase diminishes the synthesis of fatty acids and cholesterol, respectively (Cohen and Hardie, 1991;Mounier et al., 1997).Increased phosphorylation of these enzymes is not associated with their direct enzymatic phosphorylation, but is a consequence of the inhibition of the activity of phosphoprotein phosphatases by cAMP-dependent phosphorylation (Cohen and Hardie, 1991).The activity of proteases is regulated by protein kinase A through phosphorylation and activation of calpastatin, a protease inhibitor, and cAMP-dependent activation of the calpastatin gene (Salamino et al., 1994;Cong et al., 1998).

CONCLUSIONS
Protein kinase A mediates the regulation of a great variety of cAMP-dependent processes.The enzyme is involved in growth and development, cell proliferation and basal cellular metabolism.Protein kinase A regulates transcription factors by direct phosphorylation or by stimulation of its synthesis or synthesis of proteins activating the expression of transcription factors, influences the activity of cycline-dependent protein kinases by activating cycline A and cdk inhibitor genes, phosphorylates proteins that are involved in microtubule formation, and affects the activity of mitogen-activated protein kinase.
Protein kinase A occurs as several isoenzymes due to the presence of diverse isoforms of its subunits.The isoenzymes of protein kinase A differ in the substrate specificity and activity.The expression of protein kinase A subunits is tissue-and hormone-specific.The tissue specificity of protein kinase A isoforms and compartmentalization in the cell, which determine access to cAMP and substrate, is the reason why protein kinase A may be involved in many cellular processes and its activation leads to various cellular responses to stimulation mediated by cAMP .

Figure 4 .
Figure 4. A. Amino acid sequence of the dimerization domains of bovine cardiac muscle regulatory R II a (Takio et al., 1982) and bovine brain RII P (Luo et al, 1990) subunits.The conserved amino acid sequence is indicated by grey filled bars

Figure 5 B
Figure 5 B. The amino acid sequence of the dimerization domains of mouse regulatory R I P (Clegget al, 1988)  and R II p(Scott et al., 1987) subunits

Figure 6 .Figure 7 .
Figure 6.The nucleotide sequence and primary structure of the mouse catalytic C a and C (3 (Uhleret al., 1986 A and B;Chrivia et al., 1988) subunits and C P2 and C (33 subunit isoforms(Guthrie et al., 1997) of protein kinase A. The conserved amino acid sequence is shown by grey filled bars.The Spl consensus sequences are underlined.The transcription start sites are indicated by arrows.The gaps indicate the sites of intron insertions

rRJI{3 Phe Thr Val Glu Val Leu Arg His Gin TTC ACG GTG GAG GTG CTG AGG CAC CAG 10 20 Pro Ala Asp Leu Leu Glu Phe Ala Leu Gin His CCC GCC GAC CTGCTG GAG TTC GCG CTG CAG CAC 30 40 rRJla Phe Ala Val Gly Tyr Phe Thr Arg Leu Arg TTC GCG GTG GAG TAC TTC ACA CGC CCT CGC rJRJlp Phe Thr Arg Leu Gin Gin Glu Asn Glu Arg Lys Gly Ala Ala Arg Ser Ala Met Arg Ala TTC ACG CGG CTG CAG CAG GAG AAG GAG CGC AAG GGC GCC GCG CGT TCG GCC ATG AGG GCA 50 60 rRJla Glu Ala Arg Arg Gin Glu Ser Asp Ser Phe Ile Ala Pro Pro Thr Thr Phe His Ala Gin GAG GCC CGC CGC CAG GAA TCA GAC TCG TTC ATC GCC CCC CCG ACG ACC TTT CAC GCG CAG rRIl p Gly Pro Gly Gly Thr Arg Ala Gin Pro Arg Ala Glu Glu Pro Ser Lys Gly Val Asn Phe GGT CCT GGG GGG ACG CGG GCG CAG CCG CGG GGG CGA ACC CCC AGT AAG GGT GTC AAC TTC 70 80 rRJla Glu Ser Ser Gly Val Pro Val lie Glu Glu Asp Gly Gin Ser Glu Ser Pro Ser Asp Asp GAG TCC AGC GGG GTC CCC GTC ATC GAG GAG GAC GGG CAG AGT GAA TCG CAC TCG GAC GAT • B • rRJip Ala Glu Glu Pro Met Arg Ser Asp Ser Glu Asn Gly Glu Glu Glu Glu Ala Ala Glu Ala GCC GAG GAG CCC ATG CGC TCC GAT TCC GAG AAC GGC GAA GAG GAG GAG GCC GCG GAA GCA 90 rRH a Glu Asp Leu Glu Val Pro Ile Pro Ala Lys GAG GAT CTG GAA GTT CCG ATT CCA GCA AAA ma 100 Phe Thr Arg Arg Va! Ser Val Cys Ala Glu TTT ACT AGA CGA GTA TCA GTC TGT GCA GAA rRJl p Gly Ala Phe Asn Ala Pro Val Ile Asn Arg Phe Thr Arg Arg Ala Ser Val Cys Ala Glu GGG GCG TTC AAC GCT CCA GTT ATA ACC CGG TTC ACA AGG CGT GCC TCG GTA TGT GCA GAA 110 120 rRIIaThr Phe Asn Pro Asp Glu Glu Glu Asp Asn Asp Pro Arg Val Val His Pro Lys Thr Asp AAG TTT AAC CCT GAT GAA GAA GAA GAT AAT GAT CCA AGG GTG GTT CAC CCA AAA GAC GAC i \ zrrzr."= rRIl Ala Tyr Asn Pro Asp Glu Glu Glu Asp Asp Ala Glu Ser Arg Ile Ile His Pro Lys Thr p GCT TAT AAT CCT GAT GAA GAA GAA GAT GAT GCA GAG TCC AGG ATA ATA CAT CCC AAA ACT 130 rRJI a Glu Gin Arg Cys Arg Leu Gin Gin Ala Cys GAG CAG AGG TGC AGA CTT CAG GAA GCC TGT rRIl P Asp Asp Gin Arg Asn Arg Leu Gin Glu Ala GAC GAT CAA AGA AAC AGA TTG CAA GAA GCC 150 rRIl a Gin Glu Gin Leu Ser Gin Val Leu Asp Ala CAG GAA CAG CTT TCT CAA GTT CGT GAC GCC rRIl P Asp Pro Glu Gin Met Ser Gin Val Leu Asp GAT CCA GAA CAG ATG TCT CAA GTA TTA GAT 170 rRIl a His Val Ile Asp Gin Gly Asp Asp Gly Asp CAT GTC ATT GAC GAA GGA CAT GAT GGA GAC rRHPGlu His Val lie Asp Gin Gly Asp Asp Gly GAA CAC GTA ATC GAT CAA GGT GAT GAT GGT 140 Lys Asp Ile Leu Leu Phe Lys Asn Leu Asp AAA GAC ATT CTG CTG TTC AAA AAC CTG GAT Cys Lys Asp lie Leu Leu Phe Lys Asn Leu TGC AAA GAC ATC CTG CTG TTT AAG AAC CTG 160 Met Phe Lys Arg Ile Val Lys Thr Asp Glu ATG TTC AAA AGG ATA GTC AAA ACT GAC GAG Ala Met Phe Glu Lys Leu Val Lys Glu Gly GCC ATG TTT GAA AAA TTG GTC AAA GAA GGG 180 Asn Phe Tyr Val Ile Glu Arg Gly Thr Tyr AAC TTT TAT GTC ATA GAA AGG GGA ACC TAT Asp Asn Phe Tyr Val Ile Asp Arg Gly Thr iAC AAC TTT TAC GTC ATC GAC AGA GGA AGA 190 rRIl a Asp lie Leu Val Thr Lys Asp Asn Gin Thr GAC ATT TTA GTA ACA AAG GAT AAT CAA ACA 200 Arg Ser Val Gly Gin Tyr Ala Asn Arg Gly CGA TCT GTT GGT CAG TAT GCA AAC CGT GGC rRIl 0 Phe Asp ile Tyr Val Lys Cys Asp Gly Val Gly Arg Cys Val Gly Am Tyr Asp Asn Arg TTT GAT ATT TAT GTA AAA TGT GAT GGC GTT GGA AGA TGC GTT GGT AAC TAT GAC AAT CGT 210 220 rRJla Ser Phe Glv Glu Lett Ala Leu Met Tvr Asn Thr Pro Are Ala Ala Thr Ile Val Ala Thr AGTTTT GGA GAA CTA GCC CTG ATG TAC AAT ACC CCG AGA GCT GCT ACC ATT GTG GCC ACC rRIl P Gly Ser Phe Glv Glu Leu Ala Leu Met Tvr GGG AGT TTT GGA GAA CTG GCC TTA ATG TAC Asn Thr Pro Arg Ala Ala Thr Thr, He Ala AAT ACA CCC AGA GCA GCT ACA ACT ATC GCT rRIl a Ser Asp Gly Ser Leu Trp Gly Leu Asp Arg TCA GAC GGC TCC CTT TGG GGA TTG GAC CGG 240 Val Thr Phe Arg Arg He lie Val Lys Asn GTG ACT TTT AGG AGA ATC ATA GTG AAG AAC rRJI P Thr Ser Pro Gly Ala Leu Trp Gly Leu Asp Arg Val Thr Pro Arg Arg lie He Val Lys ACC TCT CCT GGT GCT CTG TGG GGT TTG GAC AGG GTG ACC TTC AGG AGA ATA ATA GTA AAA 250 260 rRIl a Asn Ala Lys Lys Arg Lys Met Phe Glu Ser Phe Ik Glu Ser Val Pro Leu Phe Lys Ser AAT
GCA AAG AAG AGG AAG ATG TTC GAA TCG TTTATT GAG TCT GTA CCG CTC TTT AAA TCA r-^ :z:iz::z::-;^ \ m "a rRl p I

rRJI P Val Glu Ik Ala Arg Cys Leu Arg Gly Gin Tyr Phe ply GJu j-eu Ala |-eu Val Thr A?P GTG GAA ATC GCT CGG TGT CTC CGG GGA CAG TAT TTT GGA GAG CTT GCC CTG GTC ACT AAC 350 360 rRIl aLvs Pro Arg Ala Ala Ser Ala Tyr Ala Val Gly Asp Val Lys Cys Leu Val Met Asp Val AAG CCA AGA GCT GCT TCT GCT TAT GCG GTT GGA GAC GTC AAA TGC TTA GTC ATG GAT GTT rR
11 p Lvs Pro Are Ala Ala Ser Ala His Ala Ik Gly Thr Val Lys Cys Leu Ala Met Asp Val AAG CCA AGA GCA GCA TCT GCA CAC GCC ATT GGG ACT GTC AAA TGC TTA GCC ATG GAT TGT CAA GCA TTT (IM i AGG CTT CTG GGC CCC TG( \ I (, GAC A[C ATG AAG AGG AAC ATC TCA CAT AAGCAAAAG TGTGGGCjAAGAAAGCGCGCG'nTAGCGlTAAGTGAAGCAAG'lTACATAGCAGTGG'lTAG.1260 nt Figure 3.The nucleotide sequence and predicted primary structure of rat R II a (Scott et al., 1987) and R II (3 (Jahnsen et al, 1987) subunits The hinge region is bolded.The cores of cAMP binding sites A and B are underlined.The conserved amino acid sequence is shown by grey filled bars.Shifted, conserved, sequence is shown by unfilled bars.The lines shown single, conserved, shifted amino acids 20 bRII a AcSer His Ile Glu Ile Pro Pro Gly Leu Thr Glu Leu Leu Gin Gly Tyr Thr Val Glu Val bRII p Met Ser lie Glu Ile Pro Ala Gly Leu Thr Glu Leu Leu Gin Gly Phe Thr Val Glu Val 40 bRII a Leu Arg Gin Arg Pro Pro Asp Leu Val Asp Phe Ala Val Asp Tyr Phe Thr Arg Leu Arg Leu Arg His Gin Pro Ala Asp Leu Leu Glu Phe Ala Leu Gin His Phe Thr Arg Leu Gin 60 bRII a Glu Ala Arg Ser Arg Ala Ser Thr Pro Pro Ala Ala Pro Pro Ser Gly Ser Gin Asp Phe bRII p Glu Glu Asn Glu Arg Lys Gly Thr AlaArg Phe Gly His Glu Gly Arg Thr Trp Gly Asp 80 bRII a Asp Pro Gly Ala Gly Leu Val Ala Asp Ala Val Ala Asp Ser Glu Ser Glu Asp Glu Glu bRII P Ala Gly Ala Ala Ala Gly Gly Gly Thr Pro Ser Lys Gly Val Asn Phe Ala Glu Glu Pro bRII a Asp Leu Asp Val Pro Ile Pro Gly Arg Phe bRII P Arg His Ser Asp Ser Glu Asn Gly Glu Glu Pro Glu Arg Pro Met Ala Phe Leu Arg Glu Glu Tyr Phe Glu Lys Glu Lys Glu Glu Ala Met Ala Ser Pro Ser Cys Phe His Ser Glu Asp Glu Asp Ser Leu Lys Gly Cys Glu Met mRII p Met Ser His Ile Gin Ile Pro Pro Gly Leu Thr Glu Leu Leu Gin Gly Tyr Thr Val Glu 30 40 mR I P Tyr Val Gin Lys His Gly Ile Gin Gin Val Leu Lys Glu Cys Ile Val His Leu Cys Val 0 ® mR II p Val Gly Gin Gin Pro Pro Asp Leu Val Asp Phe Ala Val Glu Tyr Phe Thr Arg Leu Arg 50 60 mR I p Ala Lys Pro Asp Arg Pro Leu Arg Phe Leu Arg Glu His Phe Glu Lys Leu Glu Lys Glu m mR II P Glu Ala Arg Arg Gin Glu Ser Asp Thr Phe Ile Val Ser Pro Thr Thr Phe His Thr Gin Glu Asn Arg Gin Ile Leu Ala Arg Gin Lys Ser Asn Ser Gin Cys Asp Ser His Asp Glu EO mR II P Glu Ser Ser Ala Val Pro Val Ile Glu Glu Asp Gly Glu Ser Asp Ser Asp Ser Glu Asp 90 mR I p Glu lie Ser Pro Thr Pro Pro Asn Pro Val B mR II P Ala Asp Leu Glu Val Pro Val Pro Ser Lys are presented below.
mR I P Phe Gly Glu Leu Ala Leu Ik Tyr Gly Thr Pro Arg Ala Ala Thr

TABLE 1
Chrivia et al., 1988)ture of catalytic C a subunit of protein kinase A (according toChrivia et al., 1988)

TTG mC a Lys Ala Lys Glu Asp Phe Leu Lys Lys Trp Glu // Asp Pro Ser Gin Asn Thr Ala Gin Leu AAA
GCC A^^GAA GAT TTC ('1G AAA AAA 1GG GAA//GAC CCC TCT CAG AAT ACA GCC CAG TTG mC P Lys Ala Lys Glu Asp Phe Leu Arg Lys Trp Glu // Asn Pro Pro Pro Ser Asn Ala Gly Leu AAA GCC AAA GAA GAC TTT CTG AGG AAA TGG GAG//AAC CCT CCC CCG AGT AAT GCT GGG CTT STRUCTURE AND ACTIVITY OF PROTEIN KINASE

~i mC P Thr Pro Glu Tyr Leu Ala Pro Glu Ile Ile Leu Ser Lys// Gly Tyr Asn Lys Ala Val Asp AAC CCA GAG TAC CTG GCCCCG GAG ATC ATC CTC AGC AAG //GGT TAC AAT AAG GCG GTG GAC 230 240 mCaTrp Trp Ala Leu Gly Val Leu lie Tyr Glu Met Ala Ala Gly Tyr Pro Pro Phe Phe Ala TGG TGG GCT CTC GGA TTC CTC ATC TAC GAG ATG GCT GCT GGT TAC CCA CCC TTC TTC GCT
mCpTrp Trp