Interactions between regulatory and catalytic subunits of the Candida albicans cAMP-dependent protein kinase are modulated by autophosphorylation of the regulatory subunit

Biochimica et Biophysica Acta 1542 (2002) 73^81 www.bba-direct.com

Interactions between regulatory and catalytic subunits of the Candida albicans cAMP-dependent protein kinase are modulated by autophosphorylation of the regulatory subunit Alicia Zelada a , Roc|¨o Castilla a;1 , Susana Passeron a , Luc Giasson b , Mar|¨a L. Cantore a; * a

Ca¨tedra de Microbiolog|¨a, Facultad de Agronom|¨a, Universidad de Buenos Aires, IBYF-CONICET, Avda. San Mart|¨n 4453, 1417 Buenos Aires, Argentina b School of Dentistry, GREB, Laval University, Ste-Foy, QC, Canada G1K 7P4 Received 9 May 2001; received in revised form 9 August 2001; accepted 4 October 2001

Abstract The cAMP-dependent protein kinase (PKA) from Candida albicans is a tetramer composed of two catalytic subunits (C) and two type II regulatory subunits (R). To evaluate the role of a putative autophosphorylation site of the R subunit (Ser180 ) in the interaction with C, this site was mutated to an Ala residue. Recombinant wild-type and mutant forms of the R subunit were expressed in Escherichia coli and purified. The wild-type recombinant R subunit was fully phosphorylated by the purified C subunit, while the mutant form was not, confirming that Ser180 is the target for the autophosphorylation reaction. Association and dissociation experiments conducted with both recombinant R subunits and purified C subunit showed that intramolecular phosphorylation of the R subunit led to a decreased affinity for C. This diminished affinity was reflected by an 8-fold increase in the concentration of R subunit needed to reach half-maximal inhibition of the kinase activity and in a 5-fold decrease in the cAMP concentration necessary to obtain half-maximal dissociation of the reconstituted holoenzyme. Dissociation of the mutant holoenzyme by cAMP was not affected by the presence of MgATP. Metabolic labeling of yeast cells with [32 P]orthophosphate indicated that the R subunit exists as a serine phosphorylated protein. The possible involvement of R subunit autophosphorylation in modulating C. albicans PKA activity in vivo is discussed. ß 2002 Elsevier Science B.V. All rights reserved. Keywords: cAMP; cAMP-dependent protein kinase; Autophosphorylation; Candida albicans

1. Introduction Abbreviations: PKA, cAMP-dependent protein kinase; R, regulatory subunit; C, catalytic subunit; PVDF, polyvinylidene di£uoride; BSA, bovine serum albumin; IPTG, isopropyl L-Dthiogalactoside * Corresponding author. Fax: +54-11-4514-8741. E-mail address: [email protected] (M.L. Cantore). 1 The ¢rst two authors contributed equally to this work.

It is now well established that cAMP signaling mediates several morphogenetic processes in fungi [1,2]. In the yeast Saccharomyces cerevisiae, cAMP and the cAMP-dependent protein kinase (PKA) play an important role in ¢lamentous growth [3,4]. In the dimorphic pathogenic fungus Candida albicans, elevated cAMP levels correlate with ¢lamen-

0167-4889 / 02 / $ ^ see front matter ß 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 8 9 ( 0 1 ) 0 0 1 6 8 - 9

BBAMCR 14819 5-2-02

74

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

tous growth [5]. Moreover, culture conditions leading to high endogenous concentrations of the cyclic nucleotide stimulate germ tube formation from yeast cells [6,7]. We have shown that inhibition of PKA activity in vivo completely blocks hyphal development [8]. Recently, it has also been reported that deletion of both alleles of the gene encoding the PKA catalytic subunit (TPK2) results in a mutant strain unable to germinate in certain conditions (L. Giasson, unpublished results and [9]). These ¢ndings strongly suggest that PKA is involved in the control of the morphological transition in C. albicans. In a previous work, we have reported the puri¢cation and characterization of C. albicans PKA. It was found that, as for many of the PKAs characterized so far, the enzyme is composed of two catalytic subunits (C) and a dimer of type II regulatory subunit (R) arranged in a tetrameric R2 C2 inactive holoenzyme [10]. Upon binding of cAMP to the PKA regulatory subunits, the catalytic subunits are released as active monomers. Modulation of PKA activity through changes in the C/R dissociation rate is one of the proposed regulatory mechanisms explaining the variation in kinase activity observed in vivo. In mammalian systems, it has been shown that the phosphorylation status of R a¡ects the interaction between R and C in type II PKAs. Thus, phosphorylation of the mammalian R by C (autophosphorylation) or by the glycogen synthase kinase 3 results in a decrease in the reassociation rate between C and R to form the holoenzyme, while phosphorylation by a proline-directed protein kinase produces the opposite e¡ect, i.e. an increase in the a¤nity of C for R [11^13]. In order to determine whether the putative autophosphorylation site (Ser180 ) deduced from the C. albicans cloned R subunit (L. Giasson, unpublished results) is the target for the C mediated phosphorylation and if this phosphorylation changes the a¤nity of R for C, we generated a recombinant R protein in which Ser180 was replaced by an alanine residue. The results obtained indicate that Ser180 is indeed the unique site phosphorylated by C. Kinetic studies performed in vitro with recombinant wild-type and mutant R subunits revealed that autophosphoryla-

tion of Ser180 leads to a signi¢cant decrease in the a¤nity between R and puri¢ed C. This result, taken together with the fact that the R subunit exists in vivo as a phosphoprotein, suggests that phosphorylation of R could represent a physiological regulatory mechanism of PKA activity. 2. Materials and methods 2.1. Chemicals All chemicals were of analytical grade. Kemptide (LRRASLG), cAMP, protein A-Sepharose CL-4B and protease inhibitors were from Sigma. Phosphocellulose paper P-81 was from Whatman. Polyvinylidene di£uoride (PVDF) membranes (Immobilon-P) were from Millipore. The Immunostaining Vectastain kit was from Vector. Isopropyl L-D-thiogalactoside (IPTG) and restriction endonucleases were from Promega. [Q-32 P]ATP (6000 Ci/mmol) and [3 H]cAMP (29 Ci/mmol) were from Dupont-NEN. X-Ray ¢lms (Curix-Rp1) were from Agfa Gevaert. The His-Bind kit for puri¢cation of His-tagged proteins was from Novagen. 2.2. Strains and vectors C. albicans strain 1001 (also ATCC 64385) was maintained and cultured on YPD medium. Escherichia coli strains MV1190 and CJ236 were used as hosts for site-directed mutagenesis, BL21 (DE3) for heterologous protein expression and DH10B for plasmid selection and ampli¢cation. Vector pET28a-c (+) from Novagen was used for protein expression together with the E. coli strains recommended by the manufacturer. 2.3. Site-directed mutagenesis of the R subunit A cDNA encoding the C. albicans R subunit was subcloned into the HindIII-XbaI site of pTZ18U and oligonucleotide-directed mutagenesis was performed using the Muta-gene in vitro Mutagenesis kit from Bio-Rad. Due to the unusual usage of codon CUG in C. albicans, which is decoded as Ser instead of Leu, bases 287 and 288 of codon CTG were mutated to generate codon TCG (LeuCSer). The sole pur-

BBAMCR 14819 5-2-02

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

pose of this modi¢cation was to allow the expression in E. coli of an R subunit having the same peptide sequence as the one naturally occurring in C. albicans. The resulting recombinant protein is referred to as the wild-type R subunit. To obtain the autophosphorylation-de¢cient R subunit, the potential phosphorylation site Ser180 was mutated by changing base T to G at position 539, thus converting codon ATC into codon AGC (SerCAla). Transformant colonies were screened for the proper mutations by the appearance of extra PvuII (ATCCAGC) or NspV (CTGCTCG) restriction sites in the plasmid DNA. The mutations were con¢rmed by single-strand dideoxy DNA automated sequencing. Primers containing [5P] NdeI and [3P] BamHI restriction sites were used to PCR amplify the HindIIIXbaI inserts of the mutated plasmids. The ampli¢ed products were digested with NdeI and BamHI and subcloned into the pET28a-c (+) bacterial expression vector. The correctness of the sequences inserted in pET28a-c (+) was veri¢ed by automated dideoxy DNA sequencing. 2.4. Expression and puri¢cation of the R subunits pET28a-c (+) bearing wild-type or mutant R sequences were transformed into BL21(DE3) according to standard procedures [14]. Transformants were grown overnight at 37³C in 3 ml of LB medium containing 30 Wg/ml kanamycin. These pre-cultures were then used to inoculate 50 ml of fresh LB-kanamycin medium. Expression of the R subunit was induced by the addition of 1 mM IPTG. After a 3 h incubation period, cells were harvested and used for protein extraction following Novagen's recommendations. The His-Tag domain adjacent to the cloning site borne by the pET28a-c (+) vector allowed puri¢cation of the expressed fusion proteins using a Ni2‡ chelation resin. Puri¢cation of the recombinant Nterminal His-tagged R subunits and elimination of the His-Tag were performed according to the manufacturer's instructions. The puri¢ed R subunits were freed from traces of cAMP by stripping and renaturation steps performed according to Etchebehere et al. [15].

75

2.5. Puri¢cation of the C. albicans C subunit The C subunit was puri¢ed to near homogeneity from C. albicans yeast cells as described previously [10]. 2.6. cAMP-binding assay Recombinant R subunits were incubated in 100 Wl of 50 mM HEPES bu¡er (pH 7), 0.25 mg/ml bovine serum albumin (BSA), 5 mM L-mercaptoethanol, 5 mM MgCl2 , and various concentrations of [3 H]cAMP. After a 20 min incubation period at 30³C, cAMP binding was measured using the ammonium sulfate precipitation method of DÖskeland and Ògreid [16]. Nonspeci¢c binding of [3 H]cAMP to the ¢lters was assessed by incubating samples in the absence of R. 2.7. Polyacrylamide gel electrophoresis Samples were mixed with an equal volume of 2USDS sample bu¡er and boiled for 2 min before being submitted to a 15% SDS^PAGE according to Laemmli [17]. After electrophoresis, proteins were visualized by Coomassie brilliant blue staining. Molecular mass markers used were: myosin (208 kDa), L-galactosidase (144 kDa), bovine serum albumin (87 kDa), carbonic anhydrase (44.1 kDa), soybean trypsin inhibitor (32.7 kDa), lysozyme (17.1 kDa) and aprotinin (7.1 kDa). For detection of labeled proteins, dry gels or PVDF membranes were autoradiographed at 370³C using Dupont `Lightning Plus' intensifying screens. 2.8. Immunoblot analysis Electrophoretically separated proteins were transferred to PVDF membranes by semi-dry electroblotting. The blots were stained reversibly with Ponceau S, blocked with 5% nonfat dried milk and incubated overnight with anti-R antiserum as described previously [10]. The bound antibody was detected using the Immunostaining Vectastain kit according to the manufacturer's recommendations. For detecting im-

BBAMCR 14819 5-2-02

76

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

munoprecipitated proteins, the second antibody was replaced by biotinylated protein A. 2.9. Phosphorylation of recombinant R subunits Phosphorylation of the recombinant R subunits was performed at 30³C using catalytic amounts of puri¢ed C in a 40 Wl assay mixture containing 20 mM Tris^HCl bu¡er (pH 7.5), 10 mM MgCl2 , 10 WM cAMP, 0.1 mM ATP (1^5 Ci/mmol). Aliquots of 30 Wl were withdrawn and the reaction was stopped by the addition of 4USDS-sample bu¡er [17]. After boiling them for 2 min, the samples were submitted to a 15% SDS^PAGE. The gel was then stained with Coomassie blue, dried and autoradiographed. Labeled protein bands were excised from the gel and 32 P incorporation measured using a liquid scintillation counter. 2.10. Inhibition of C kinase activity by recombinant wild-type and mutant R subunits Interaction between C and R subunits was investigated by measuring the ability of the R subunits to inhibit kemptide phosphorylation by C. Puri¢ed C subunits (2 nM) were preincubated for 60 min at 30³C with various concentrations of R subunits in 20 mM Tris^HCl bu¡er (pH 7.5), 1 mM L-mercaptoethanol, in the presence or absence of 10 mM MgCl2 and 0.1 mM ATP in a ¢nal volume of 50 Wl. The phosphorylation reactions were started by the addition of 10 Wl of a mixture containing kemptide (50 WM ¢nal concentration) and [Q-32 P]ATP (1000^1500 cpm/pmol). Samples lacking MgCl2 and ATP were supplemented with both compounds before starting the phosphorylation reaction. Incubations were performed at 30³C for 5 min and the reactions were stopped by the addition of 10 Wl 75 mM phosphoric acid. Aliquots of 50 Wl were spotted onto phosphocellulose paper, dropped into 75 mM phosphoric acid and washed as described by Roskoski [18]. 2.11. Reconstitution and dissociation of the holoenzyme

mutant R subunits in 20 mM Tris^HCl bu¡er (pH 7.5), 1 mM L-mercaptoethanol, in the presence or absence of MgATP. Preincubation of the samples and measurement of kinase activity at various cAMP concentrations were performed as described above. 2.12. Metabolic labeling of C. albicans yeast cells and immunoprecipitation C. albicans cells (1001 strain) were grown to late log phase in 100 ml of phosphate depleted YPD liquid medium [19] containing carrier-free [32 P]orthophosphate (500 WCi/ml). Cells were harvested by centrifugation, washed twice in ice-cold bu¡er A (20 mM sodium phosphate bu¡er (pH 6.8), 10 mM Na2 P2 O7 , 10 mM NaF, 10 mM Na2 MO4 and 20 mM L-glycerophosphate) and broken by vortexing with glass beads in bu¡er A containing 1 mM phenylmethylsulfonyl £uoride, 100 Wg/ ml pepstatin and 1 WM okadaic acid. Unbroken cells and debris were removed by low-speed centrifugation and the homogenate was centrifuged for 1 h at 100 000Ug. The supernatant was subjected to immunoprecipitation with C. albicans anti-R antiserum as previously described [10] and the resulting pellet was suspended in Laemmli bu¡er and submitted to SDS^ PAGE. 2.13. Phosphoamino acid analysis The immunoprecipitated labeled R subunit was submitted to a 15% SDS^PAGE and electroblotted onto a PVDF membrane. The radiolabeled band was excised and subjected to partial hydrolysis in 6 N HCl at 110³C for 45 min. After hydrolysis, the sample was mixed with cold phosphoamino acids and electrophoresed on a cellulose sheet at pH 3.5 (water/acetic acid/pyridine, 945:50:5) at 1000 V as described by Kuwahara et al. [20]. The positions of standard phosphoamino acids were visualized with ninhydrin. 2.14. Protein determination

The PKA holoenzyme was reconstituted by combining 2 nM of C subunit with 2 WM of wild-type or

Protein concentration was determined by the method of Bradford [21] using BSA as standard.

BBAMCR 14819 5-2-02

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

77

3. Results 3.1. Site-directed mutagenesis of the C. albicans R subunit The C. albicans R gene encodes a putative protein of 459 amino acids, 53% identical to that of S. cerevisiae (GenBank accession No. AF 317472, L. Giasson, unpublished results). Analysis of the deduced amino acid sequence showed that the organization of the functional domains is similar to that of the yeast and mammalian homologues (Fig. 1A) [22,23]. It also revealed the presence of the autophosphorylation consensus sequence: RRTS180 V (Fig. 1B). It has been shown in other systems that the serine residue present in this sequence is phosphorylated by the C subunit. To determine if the putative autophosphorylation site (Ser180 ) observed in C. albicans R is really phosphorylated by C, and if the phosphorylation status of this site a¡ects the interaction between R and C, a recombinant R protein was generated in which Ser180 was replaced by an alanine residue.

Fig. 2. SDS^PAGE analysis of puri¢ed recombinant R subunits. (A) Equal amounts (1 Wg) of puri¢ed recombinant wildtype and mutant R subunits (lane 1 and 2, respectively) stained with Coomassie blue. The position of molecular mass markers is shown on the left. The arrow indicates the position of C. albicans native R subunit. (B) Immunoelectrophoretic blotting of the puri¢ed proteins. Proteins from a gel identical to that shown in A were transferred to a PVDF membrane. R subunits were revealed with an antiserum raised against C. albicans native R subunit as described in Section 2. Lane 1, recombinant wild-type R; lane 2, recombinant mutant R. (C) Phosphorylation of R subunits. Puri¢ed R subunits (200 ng) were phosphorylated in vitro by C as indicated in Section 2. The reaction was stopped with 10 Wl of 2USDS and the samples were subjected to SDS^PAGE and autoradiographed. Lane 1, recombinant wild-type R subunit incubated in the absence of C; lanes 2 and 3, recombinant wild-type and mutant R subunits, respectively, incubated in the presence of C.

3.2. Expression, puri¢cation and properties of the recombinant R subunits

Fig. 1. Schematic representation of domain organization in R subunits. (A) Organization of functional domains. DD, dimerization domain; IS, inhibitor consensus site; cA:A, cAMP-binding domain A; cA:B, cAMP-binding domain B. (B) Amino acid sequence alignment surrounding the inhibitor consensus site of bovine (RI and RII), yeast (BCY1) and C. albicans (CA) R subunits. The inhibitor consensus site is underlined and the potential autophosphorylation site S180 of C. albicans R is highlighted in bold.

Recombinant wild-type and mutant R subunits were puri¢ed as described in Section 2. The expressed proteins, recovered mainly in the soluble fraction, represented approx. 15^20% of total proteins. A typical SDS^PAGE analysis of puri¢ed recombinant proteins is depicted in Fig. 2. Coomassie blue staining (Fig. 2A) reveals that the R subunits elute from the resin practically free of contaminants. Western blotting (Fig. 2B) shows that both wild-type and mutant R recombinant proteins were recognized

BBAMCR 14819 5-2-02

78

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

by an antiserum raised against puri¢ed native R. As can be seen in Fig. 2C, wild-type recombinant R was e¤ciently phosphorylated by C in the presence of cAMP while mutant R, bearing the Ser180 to Ala change, was not phosphorylated. This result con¢rms that Ser180 is the autophosphorylation site as predicted by sequence analysis. A stoichiometric incorporation of 1 mol of phosphate per mol of wild-type R monomers was determined. One mol of wild-type or mutant recombinant R was able to bind 2 mol of cAMP, indicating that nucleotide binding was not a¡ected by the Ser to Ala mutation (data not shown). 3.3. Inhibition of C kinase activity by recombinant R subunits In order to evaluate the impact of the autophosphorylation status of R on its interaction with C, the ability of the recombinant wild-type R subunit to inhibit kemptide phosphorylation was measured for both the unphosphorylated and the Ser180 phosphorylated forms of R. As can be seen in Fig. 3, the presence of MgATP during the preincubation step greatly diminished the ability of wild-type R to associate with C. This is indicated by the shift of the

Fig. 3. Kinetic analysis of the e¡ect of R phosphorylation on C phosphotransferase activity. C subunit (2 nM) was preincubated for 60 min at 30³C with various amounts of recombinant wildtype (a,b) or mutant (E,F) R in the presence (b,F) or absence (a,E) of MgATP as described in Section 2. Measurement of the catalytic activity was initiated by the addition of kemptide and [Q-32 P]ATP. Each value represents the mean þ S.D. for three independent experiments. 100% phosphotransferase activity corresponded to 0.25 nmol/min.

Fig. 4. Dissociation of reconstituted PKA holoenzymes by cAMP. PKA catalytic subunit (0.1 pmol) was preincubated with 100 pmol of recombinant wild-type (a,b) or mutant (E,F) R subunits for 60 min at 30³C in the presence (b,F) or absence (a,E) of MgATP. PKA phosphotransferase activity was measured after an additional 5 min incubation period at 30³C in the presence of di¡erent amounts of cAMP. Each value represents the mean þ S.D. for three independent experiments.

curve to the right. Half-maximal inhibition of C kinase activity was observed at about 4 nM and 33 nM of wild-type R dimers, in the absence and presence of MgATP respectively. The phosphorylation status of wild-type recombinant R in the presence of MgATP was assessed using [Q-32 P]ATP. It was observed that in all cases, R was in a fully phosphorylated state at the end of the preincubation period (data not shown). Therefore, one must assume that it is the phosphorylated form of wild-type R which interacts with C when holoenzyme reconstitution experiments are performed in the presence of MgATP. To rule out the possibility that MgATP itself plays a role in the interaction between R and C, as has been suggested for type I R/C interaction [24], similar experiments were performed using the Ala180 R mutant. As can be seen in Fig. 3, the presence of MgATP did not modify the reassociation kinetics of the mutant R, indicating that the phosphorylation of R at Ser180 is responsible for the decrease in the a¤nity between R and C, thus impairing formation of the inactive holoenzyme. 3.4. Dissociation of reconstituted holoenzymes by cAMP To assess the e¡ect of R phosphorylation status on R/C dissociation kinetics, holoenzymes were pre-

BBAMCR 14819 5-2-02

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

79

R subunits were submitted to a 15% SDS^PAGE as described in Section 2. As can be seen in Fig. 5A, autoradiography of the membrane revealed the presence of a labeled protein (lane 1) corresponding to the band recognized by the speci¢c anti-R antiserum (lane 2). Phosphoamino acid analysis of the labeled protein showed that phosphoserine was the only phosphoamino acid in the hydrolysate (Fig. 5B).

Fig. 5. In vivo phosphorylation of R subunit and phosphoamino acid analysis. (A) Yeast cells grown in phosphate depleted YPD medium in the presence of [32 P]orthophosphate were harvested and broken as described in Section 2. The soluble extract was immunoprecipitated with anti-R antiserum and the pellet was subjected to a 15% SDS^PAGE. Proteins were transferred to a PVDF membrane. Lane 1, Western blot of the membrane revealed with anti-R antiserum; lane 2, autoradiogram of the membrane. (B) Partial hydrolysis of the 32 P labeled R was performed as described in Section 2. Released phosphoamino acids were resolved by 1D electrophoresis. Circles on the autoradiogram indicate the position of standards. P-S, phosphoserine; P-T, phosphothreonine; P-Y, phosphotyrosine. `O' corresponds to the origin.

pared by mixing 2 nM of C subunit with 2 WM of wild-type or mutant R subunits, in the presence or absence of MgATP. Phosphotransferase activity of the reconstituted holoenzymes was then measured in the presence of various cAMP concentrations. Fig. 4 shows that inclusion of MgATP in the preincubation step resulted in a shift of the cAMP dose^ response curve to the left, indicating that the holoenzyme formed was more readily dissociated by cAMP. The holoenzyme reconstituted with mutant R showed essentially identical curves in the presence or absence of MgATP in the preincubation step. Again, these results indicate that there is a signi¢cant decrease in the a¤nity between R and C subunits when R is in the phosphorylated form, and that the presence of MgATP does not a¡ect the a¤nity between R and C subunits when R Ser180 is absent. 3.5. In vivo studies To analyze the in vivo phosphorylation of R subunit, C. albicans yeast cells were metabolically labeled with [32 P]orthophosphate. Immunoprecipitated

4. Discussion This paper reports results from experiments conducted to determine the in£uence of C. albicans PKA regulatory subunit phosphorylation status on the interaction with PKA catalytic subunit using recombinant wild-type and the Ala180 mutant R subunits. Recombinant wild-type and mutant R subunits seem to have been correctly expressed in E. coli considering that their molecular masses, their immunological recognition by the anti-R antiserum (see Fig. 2A,B) and the stoichiometry of cAMP were similar to those observed with native R puri¢ed from C. albicans [10]. The fact that recombinant wildtype R incorporated 1 mol of phosphate per mol of protein, while the Ser180 CAla mutant was unable to be phosphorylated (Fig. 2C), con¢rms that Ser180 is the amino acid phosphorylated by C, thus implying that autophosphorylation can e¡ectively take place. To evaluate the e¡ect of autophosphorylation on R/C interactions, the dose^response curves of C phosphotransferase activity in the presence of recombinant wild-type or mutant R subunits, and the cAMP dependence of the dissociation kinetics for wild-type and mutant reconstituted holoenzymes were analyzed. The results from these experiments are consistent with a decrease in the a¤nity between subunits promoted by phosphorylation of R as has already been reported for mammalian RII [25,26]. Thus, half-maximal inhibition of C was attained with about 8-fold higher wild-type R subunit in the presence of MgATP and the holoenzyme reconstituted under this condition dissociated with about 5-fold less cAMP concentration. Replacement of Ser180 by Ala resulted in a mutant R whose a¤nity for C, both in association and dissociation experiments, was slightly lower than that of the unphosphorylated wild-type R

BBAMCR 14819 5-2-02

80

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

subunit, and was not overall in£uenced by the presence of MgATP. It has been reported that, in the case of S. cerevisiae R subunits, replacement of the autophosphorylatable Ser by an Ala residue results in an enhanced a¤nity for C in the presence of MgATP [27]. However, since similar experiments in the absence of MgATP were not performed by these authors, the behavior of unphosphorylated R subunits cannot be compared between the two organisms. The crystal structure of the recombinant bovine RIK subunit allowed a detailed molecular and structural analysis of the regulatory domains involved in the interactions with the catalytic subunit and cAMP [28,29]. These studies revealed that the high a¤nity observed between the subunits results from the interaction of C with two R domains, one of them containing the substrate consensus sequence and the other located at the C helix of the cAMP-binding domain A. The fact that the Ser180 to Ala mutation in the C. albicans R subunit was not su¤cient to enable the holoenzyme to be stabilized by MgATP (a characteristic of type I regulatory subunits) is consistent with the notion that additional sites play a role in the R/C interactions. The decreased a¤nity observed between the R and C subunits, upon phosphorylation of type II regulatory subunits, shows the important role that a negative charge may play in hindering the proper accommodation of distal domains involved in holoenzyme formation. Our results indicate that R is phosphorylated in vivo in C. albicans. Moreover, preliminary results indicated that the labeled R resolved in two well separated labeled isoelectric variants in 2D gels (data not shown). Although we cannot discard other posttranslational modi¢cations, very probably they arise from phosphorylation on more than one serine residue, since comparison of the deduced amino acid sequence with the Prosite data base [30] indicates putative phosphorylation sites for several protein kinases including PKA. Taking into account that in vitro intramolecular phosphorylation of the R subunit was shown to be rapid and highly e¤cient [10], it is possible that one or both of the in vivo labeled R variants represent autophosphorylated forms of R. If this proves to be the case, the presence of the autophosphorylated PKA holoenzyme would then allow

C. albicans cells to respond more easily to small changes in cAMP levels due to the weaker a¤nity between the R and C subunits. Previous results, from ourselves [8] and others [9], have demonstrated the involvement of the cAMPPKA signaling pathway in the C. albicans yeast-tohypha morphogenetic transition. The in vitro and in vivo results presented in this work emphasize the potential signi¢cance of R phosphorylation on the regulation of PKA activity in vivo and, consequently, on the control of C. albicans morphogenesis. Further work is in progress, focusing on the identi¢cation of the in vivo phosphorylated sites. Acknowledgements This work was supported by grants from the Consejo Nacional de Investigaciones Cient|¨¢cas y Te¨cnicas (CONICET), the Universidad de Buenos Aires, the Agencia Nacional de Promocio¨n Cient|¨¢ca y Tecnolo¨gica (ANPCyT) and the Fundacio¨n Antorchas to S.P. and M.L.C., and from the Medical Research è mile-Beaulieu to Council of Canada and the Fonds E L.G. S.P. and M.L.C. are research members from CONICET, A.Z. is a postdoctoral fellow and R.C. is a research fellow from the same institution.

References [1] J. Kronstad, A. De Maria, D. Funnel, R.D. Laidlaw, N. Lee, M. Moniz de Sa¨, M. Ramesh, Signaling via cAMP in fungi: interconnections with mitogen-activated protein kinase pathways, Arch. Microbiol. 170 (1998) 395^404. [2] H.D. Madhani, G.R. Fink, The control of ¢lamentous differentiation and virulence in fungi, Trends Cell Biol. 8 (1998) 348^353. [3] H.U. Mo«sch, R. Roberts, G.R. Fink, Ras2 signals via the Cdc4/Ste20/mitogen-activated-kinase module to induce ¢lamentous growth in Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA 93 (1996) 5352^5356. [4] X. Pan, J. Heitman, Cyclic AMP-dependent protein kinase regulates pseudohyphal di¡erentiation in Saccharomyces cerevisiae, Mol. Cell. Biol. 19 (1999) 4874^4887. [5] F.T. Sabie, G.M. Gadd, E¡ect of nucleosides and nucleotides and the relationship between cellular adenosine 3P:5P-cyclic monophosphate cyclic AMP and germ tube formation in Candida albicans, Mycopathologia 119 (1992) 147^ 156. [6] M. Niimi, Dibutyryl cyclicAMP-enhanced germ tube forma-

BBAMCR 14819 5-2-02

A. Zelada et al. / Biochimica et Biophysica Acta 1542 (2002) 73^81

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

tion in exponentially growing Candida albicans cells, Fungal Genet. Biol. 20 (1996) 79^83. A. Zelada, R. Castilla, S. Passeron, M.L. Cantore, Reassessment of the e¡ect of glucagon and nucleotides on germ tube formation in Candida albicans, Cell. Mol. Biol. 42 (1996) 567^576. R. Castilla, S. Passeron, M.L. Cantore, N-Acetyl-D-glucosamine induces germination in Candida albicans through a mechanism sensitive to inhibitors of cyclic AMP-dependent protein kinase, Cell. Signal. 10 (1998) 713^719. A. Sonneborn, D.P. Bockmu«hl, M. Gerards, K. Kurpanek, D. Sanglard, J.F. Ernst, Protein kinase A encoded by TPK2 regulates dimorphism of Candida albicans, Mol. Microbiol. 35 (2000) 386^396. A. Zelada, S. Passeron, S. Lopez Gomes, S.M.L. Cantore, Isolation and characterization of cAMP-dependent protein kinase from Candida albicans. Puri¢cation of the regulatory and catalytic subunits, Eur. J. Biochem. 252 (1998) 245^252. R. Rangel-Aldao, O. Mendelsohn Rosen, Dissociation and reassociation of the phosphorylated and dephosphorylated forms of adenosine 3P:5-monophosphate-dependent protein kinase from bovine cardiac muscle, J. Biol. Chem. 11 (1976) 3375^3380. B.A. Hemmings, A. Aitken, P. Cohen, M. Rymond, F. Hofmann, Phosphorylation of the type-II regulatory subunit of cyclic-AMP-dependent protein kinase by glycogen synthase kinase 3 and glycogen synthase kinase 5, Eur. J. Biochem. 127 (1982) 473^481. R.K. Braun, P.R. Vulliet, D.A. Carbonaro-Hall, F.L. Hall, Phosphorylation of RII subunit and attenuation of cAMPdependent protein kinase activity by proline-directed protein kinase, Arch. Biochem. Biophys. 289 (1991) 187^191. F.M. Ausubel, R. Brent, E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, K. Struhl (Eds.), Current Protocols in Molecular Biology, Wiley and Sons, New York, 1993. L.C. Etchebehere, M.X.P. Van Bemmelen, C. Anjard, F. Traincard, K. Assemat, C. Reymond, M. Ve¨ron, The catalytic subunit of Dictyostelium cAMP-dependent protein kinase. Role of the N-terminal domain and of the C-terminal residues in catalytic activity and stability, Eur. J. Biochem. 248 (1997) 820^826. S.O. DÖskeland, D. Ògreid, Ammonium sulfate precipitation assay for the study of cyclic nucleotide binding to proteins, Methods Enzymol. 159 (1988) 147^150. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of the bacteriophage T4, Nature 227 (1970) 680^685. J.R. Roskoski, Assay of protein kinase, Methods Enzymol. 99 (1983) 3^6.

81

[19] G.M. Rubin, The nucleotide sequence of Saccharomyces cerevisiae 5.8 ribosomal ribonucleic acid, J. Biol. Chem. 248 (1973) 3860^3875. [20] M. Kuwahara, K. Fushimi, Y. Terada, L. Bai, F. Marumo, S. Sasaki, cAMP-dependent phosphorylation stimulates water permeability of aquaporin-collecting duct water channel protein expressed in Xenopus oocytes, J. Biol. Chem. 270 (1995) 10384^10387. [21] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248^ 354. [22] S.S. Taylor, J. Bubis, J. Toner-Webb, L.D. Saraswat, E.A. First, J.A. Buechler, D.R. Knighton, D.R. Sowadski, cAMPdependent protein kinase: prototype for a family of enzymes, FASEB J. 2 (1988) 2677^2685. [23] T. Toda, S. Cameron, P. Sass, M. Zoller, J.D. Scott, B. McMullen, M. Hurwitz, Cloning and characterization of BCY1, a locus encoding a regulatory subunit of the cAMP dependent protein kinase in yeast, Mol. Cell. Biol. 7 (1987) 1371^1377. [24] S. Whitehouse, J.R. Feramisco, J.E. Casnellie, E.G. Krebs, D.A. Walsh, Studies on the kinetic mechanism of the catalytic subunit of the cAMP-dependent protein kinase, J. Biol. Chem. 258 (1983) 3693^3701. [25] R. Rangel-Aldao, O.M. Rosen, E¡ect of cAMP and ATP on the reassociation of phosphorylated and nonphosphorylated subunits of the cAMP-dependent protein kinase from bovine cardiac muscle, J. Biol. Chem. 252 (1977) 7140^7145. [26] J. Granot, A.S. Midvan, K. Hiyama, H. Kondo, E.T. Kaiser, Magnetic resonance studies of the e¡ect of the regulatory subunit on metal and substrate binding to the catalytic subunit of bovine heart protein kinase, J. Biol. Chem. 255 (1980) 4569^4573. [27] J. Kuret, K.E. Johson, C. Nicolette, M.J. Zoller, Mutagenesis of the regulatory subunit of yeast cAMP-dependent protein kinase, J. Biol. Chem. 263 (1988) 9149^9154. [28] Y. Su, W.R. Dostmann, F.W. Herberg, K. Durick, N.H. Xuong, L. Ten Eyck, S.S. Taylor, K.I. Varughese, Regulatory subunit of protein kinase A: structure of deletion mutant with cAMP binding domains, Science 269 (1995) 807^ 813. [29] L. Jun-shen, S.S. Taylor, Dissecting cAMP binding domain A in the RI alpha subunit of cAMP-dependent protein kinase. Distinct subsites for recognition of cAMP and the catalytic subunit, J. Biol. Chem. 273 (1998) 26739^26746. [30] A. Bairoch, P. Bucher, K. Hofmann, The PROSITE database, its status in 1997, Nucleic Acids Res. 24 (1997) 217^ 221.

BBAMCR 14819 5-2-02