Unsaturated fatty acid-activated protein kinase (PKx) from goat testis cytosol

Unsaturated fatty acid-activated protein kinase (PKx) from goat testiscytosol

Koushik Roy a, Atin K. Mandal a, Rita Sikdar a, Subrata Majumdar b,Yoshitaka Ono c, Parimal C. Sen a;*

a Department of Chemistry, Bose Institute, 93/1 A.P.C. Road, Calcutta 700 009, Indiab Department of Microbiology, Bose Institute, 93/1 A.P.C. Road, Calcutta 700 009, India

c Department of Biology, Kobe University, Kobe 657, Japan

Received 10 March 1999; received in revised form 24 June 1999; accepted 23 July 1999

Abstract

The cytosolic fraction of goat cauda epididymis possesses a protein kinase (PKx) activity which is stimulated by a numberof unsaturated fatty acids of which arachidonic acid is the best activator in absence of cAMP or Ca2�. Phosphatidylcholine,phosphatidylethanolamine, phosphatidylserine and diacylglycerol have no effect either alone or in combination. Themembrane fraction does not show any appreciable kinase activity even after detergent treatment. PKx migrates as a singleband of apparent molecular mass of 116 kDa on 10% SDS^PAGE after sequential chromatographic separation on DEAE^cellulose, phenyl^Sepharose, high-Q anion exchange and protamineâgarose affinity column. PKx phosphorylates histoneH1, histone IIIs and protamine sulfate, but not casein. However, the best phosphorylation was obtained with a substratebased on PKC pseudosubstrate sequence (RFARKGSLRQKNV). The kinase phosphorylates two endogenous cytosolicproteins of 60 and 68 kDa. Ser residues are primarily phosphorylated although a low level of phosphorylation is observed onThr residues also. Ca2� and Mn2� inhibit PKx activity in the micromolar range. Staurosporine is found to inhibit the PKxactivity to a significant level at sub-nanomolar concentration. Lyso-phosphatidylcholine and certain detergents at very lowconcentrations (6 0.05%) stimulate enzyme activity to some extent. The immuno-crossreactivity study with antibody againstdifferent PKC isotypes suggests that the protein kinase under study is not related to any known PKC family. Even theantibody against PKN (a related protein kinase reported in rat testis found to be activated by arachidonic acid) does notcross-react with this protein kinase. Hence we believe that the protein kinase (PKx) reported here is different even from thePKN of rat testis. The phosphorylation of endogenous proteins by the protein kinase may be involved in cell regulationincluding fertility regulation and signal transduction. ß 1999 Elsevier Science B.V. All rights reserved.

Keywords: Protein kinase; Phosphorylation; Unsaturated fatty acid; (Goat testis)

0167-4838 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 8 3 8 ( 9 9 ) 0 0 1 7 3 - 9

Abbreviations: PMSF, phenylmethylsulfonyl£uoride; EDTA, ethylenediaminetetraacetic acid; EGTA, ethylene glycol-bis(L-aminoethylether) N,N,NP,NP-tetraacetic acid; LME, L-mercaptoethanol; SDS^PAGE, sodium dodecyl sulfate^polyacrylamide gel electropho-resis ; cAMP, cyclic adenosine monophosphate; cGMP, cyclic guanosine monophosphate; PKA, protein kinase A; PKC, protein kinaseC; PKN, protein kinase N; PKx, protein kinase x

* Corresponding author. Fax: +91-33-350-6790; E-mail : [email protected]

BBAPRO 35987 2-9-99

Biochimica et Biophysica Acta 1434 (1999) 161^169

www.elsevier.com/locate/bba

1. Introduction

The cellular processes that require some form oftransmembrane signalling are diverse and complex,ranging from cellular growth and di¡erentiation tonerve cell communication, learning and memory.Protein phosphorylation is widely recognized as animportant event in transmembrane signal transduc-tion in eukaryotic cells [1]. It is, therefore, not sur-prising that a large number of protein kinases in-volved in protein phosphorylation exist and newkinases are still being discovered.

The signalling systems used by cells often displayextensive heterogeneity and variations exist from onetissue to another. Most tissues seem to have at leasttwo major receptor classes for transducing informa-tion across the cell membranes. One class depends onthe generation of cyclic AMP as a second messenger,while the other class of receptor induces rapid turn-over of inositol phospholipids as well as mobilizationof Ca2�. Stimulation of the latter class normallyleads to release of arachidonate and often increasescyclic GMP. Thus protein kinase C activation, Ca2�

mobilization, arachidonate release and cGMP forma-tion appear to be integrated into a single receptorcascade [2].

Protein kinase C (PKC), a Ca2�/phospholipid de-pendent enzyme, has been proposed to play a criticalrole in signal transduction in di¡erent cell systems[3]. Cloning and molecular biology studies have re-vealed the existence of multiple isotypes of PKC withclosely related structures [4]. Di¡erences in tissue dis-tribution, substrate and co-factor speci¢city for dif-ferent isotypes predict selective roles for each iso-form.

PKC isotypes possess a primary structure contain-ing conserved structural motifs with a high degree ofsequence homology. PKC consists of an amino-ter-minal regulatory region and a carboxy-terminal cat-alytic region [5]. The regulatory subunit of PKC con-tains a pseudo-substrate sequence that ful¢ls most ofthe requirements for a PKC phosphorylation site [6].Classical PKCs have two conserved amino acid se-quence domains (C1 and C2) in their regulatory re-gions. The atypical and new PKCs are devoid ofcalcium-binding C2 domain and hence insensitive toCa2�. The carboxy-terminal catalytic regions of allPKC isotypes contain two conserved amino acid se-

quence domains, C3 and C4, responsible for ATPbinding and substrate recognition, respectively [7].The C3 domain is also the site for staurosporine(an inhibitor of PKC) binding [7].

PKC can be activated independent of phospholipidand Ca2� by cis-unsaturated fatty acids [8,9], actingas second messengers, which include arachidonic,oleic, linoleic, linolenic and docosahexaenoic acids.They all are produced from membrane phospholipidsby the action of phospholipase A2-induced unsatu-rated fatty acid liberation, which, in turn, would ac-tivate PKC even after the cytosolic concentration ofCa2� returns to resting level [10]. Potentiation ofcellular responses by cis-unsaturated fatty acids hasalso been reported for diacylglycerol-induced inter-leukin-2 synthesis by human T lymphocytes [11]and for diacylglycerol-induced reduction of K�-channel conductance in Hermissenda photoreceptorcells [12].

Gonads of higher animals are under the control ofthe central nervous system. Luteinizing hormone(LH)-stimulated androgen synthesis in testicular Ley-dig cells is mediated by cAMP and protein kinase A[13]. Reports are available on the prominent role ofCa2� and cAMP in sperm motility and acrosomereaction [14]. Although PKC activity has been foundin di¡erent tissues, low-level activity is found in re-productive organs, e.g., ram and bull spermatozoa,which in turn may be involved in acrosome reaction[15]. Rat testis contains a Ca2�-insensitive long-chainunsaturated fatty acid-stimulated novel protein kin-ase (PKN) [16] that possesses a regulatory subunitstructure di¡erent from PKC, although the catalyticregion shows sequence homology to some extent [17].In this paper we report the identi¢cation, isolationand characterization of a protein kinase (PKx) fromgoat testis cytosol which is activated by unsaturatedfatty acids with arachidonic acid as the best activa-tor. The kinase is di¡erent from PKC and PKN, ascon¢rmed by immunoblotting studies using antibod-ies against PKN and various isoforms of PKC.

2. Materials and methods

2.1. Materials

DEAE^cellulose, phenyl^Sepharose, protamine^

BBAPRO 35987 2-9-99

K. Roy et al. / Biochimica et Biophysica Acta 1434 (1999) 161^169162

agarose, histone H1, histone IIIs, protamine sulfate,fatty acids, phospholipids, detergents, staurosporine,wortmannin, calphostin C, EGTA, PMSF, proteinkinase C substrates ([Ser]25 PKC fragment 19^31,RFARKGSLRQKNV) [6] and VRKRTLRRL [18]and speci¢c PKA substrate (Kemptide, LRRASLG)[19] were obtained from Sigma Chemicals.[Q-32P]ATP (speci¢c activity 5000 Ci/mmol) was ob-tained from Bhabha Atomic Research Centre, Mum-bai, India. Other chemicals of analytical grade werepurchased from Sisco Research Laboratories, India.Nitrocellulose membrane and p81 ¢lter paper werefrom BioRad and Whatman respectively. Monoclo-nal antibodies against human PKCs (K, L, Q, N, O, W,a and V) and PKN were obtained from TransductionLaboratories. Enhanced chemiluminescence kit waspurchased from Amersham.

2.2. Puri¢cation of PKx

Goat testes collected from a local slaughterhouseimmediately after killing of the animals were broughtto the laboratory on ice. All the subsequent proce-dures were carried out at 4³C. The cauda regionswere separated and minced with bu¡er A (50 mMTris^HCl (pH 7.5) containing 0.25 M sucrose, 1 mMEDTA, 1 mM EGTA, 1 mM PMSF, 1 Wg/ml leupep-tin, 0.01% NaN3 and 2 mM LME). The sample washomogenized for 5 min in a motor driven homoge-nizer and centrifuged at 12 000Ug for 15 min. Thesupernatant was collected and spun at 100 000Ug for1 h in a Hitachi ultracentrifuge (Model 55P-72). Post100 000Ug supernatant (cytosol) was used as thesource of the protein kinase. The clear cytosol wasincubated with 40 WM cAMP for 30 min at 4³C andloaded on to a DEAE^cellulose column (60 ml) equil-ibrated with bu¡er B (20 mM Tris^HCl (pH 7.5)containing 1 mM EDTA, 1 mM EGTA, 1 Wg/mlleupeptin, 0.01% NaN3 and 1 mM LME). The col-umn was washed with ¢ve column volumes of bu¡erB containing 100 mM NaCl. The kinase enrichedfractions were eluted with the above bu¡er contain-ing 200 mM NaCl. Fractions rich in PKx activitywere pooled and brought to a ¢nal concentrationof 1 M NaCl. It was loaded onto a phenyl^Sepharosehydrophobic column (10 ml) equilibrated with 1 MNaCl in bu¡er B. The column was washed sequen-

tially with 30 ml each of bu¡er B containing 1 M and0.5 M NaCl. The PKx-rich fraction was eluted withbu¡er B and was then subjected to further puri¢ca-tion on a high-Q anion-exchange column (5 ml) andeluted with 0^0.25 M continuous NaCl gradient.PKx-enriched fractions (0.15^0.18 M NaCl eluent)were pooled and applied to a protamineâgarose af-¢nity column (3 ml). The column was pre-equili-brated with bu¡er B and after application of thesample, washed sequentially with bu¡er B and bu¡erB containing 0.2 M NaCl, three column volumeseach. A continuous 0.2^2 M NaCl gradient was ap-plied and fractions of 1 ml each were collected in 20tubes. The kinase was eluted in fractions 14^16. Thispuri¢ed protein kinase was used for all the subse-quent studies.

2.3. Protein estimation and SDS^PAGE

Protein was estimated following the method ofBradford [20]. SDS^PAGE (10%) was performed ac-cording to the procedure of Laemmli [21] and silverstaining was done following the method described byMorrissey [22].

2.4. Protein kinase assay

The assay medium in a volume of 50 Wl contained20 mM Tris^HCl (pH 7.5), 5 mM MgCl2, 100 WMPKC substrate ([Ser]25 PKC fragment 19^31), 10 WM[Q-32P]ATP, without and with 50 WM arachidonicacid (eicosatetraenoic acid, C20:4). The reaction wasstarted with the addition of 100 ng of enzyme andincubated for 10 min at 30³C. It was terminated byspotting on to Whatman p81 paper strips, immersedimmediately into 75 mM phosphoric acid andwashed with the same solution for 2 h with fourchanges in every 30 min. Incorporation of [32P]-phosphate into substrate was assessed by scintillationcounting of the dried papers. PKx activity was calcu-lated from the di¡erence in radioactive counts be-tween the absence and presence of arachidonic acid.

E¡ects of di¡erent modi¢ers, fatty acids, deter-gents, protein kinase inhibitors, such as calphostinC, staurosporine and wortmannin were studied withpuri¢ed protein kinase as described in appropriate¢gure legends or in Section 3.

BBAPRO 35987 2-9-99

K. Roy et al. / Biochimica et Biophysica Acta 1434 (1999) 161^169 163

2.5. Endogenous protein phosphorylation

Endogenous protein phosphorylation was carriedout under the same standard conditions of enzymeassay as described above with 1 Wg of puri¢ed PKxand 20 Wg of cytosolic protein (heat denatured) inabsence of any exogenous substrate. The reactionwas terminated with gel electrophoresis sample bu¡ercontaining 1% SDS and subsequently separated by10% SDS^PAGE [21]. The autoradiogram of thedried gel was taken on Kodak X-ray ¢lm.

2.6. Immunoblotting

This technique was applied to detect if PKx hasany sequential homology with PKN or PKC. Sam-ples of PKx, transferred from gel to nitrocellulosepaper, were treated with antibodies against PKNand di¡erent PKC isotypes, and immunoreactivebands were visualized by the enhanced chemilumi-nescence method.

3. Results

Protein kinase x was puri¢ed 1600-fold to homo-geneity after sequential separation by DEAE^cellu-lose, phenyl^Sepharose, high-Q anion exchange andprotamineâgarose a¤nity chromatography. Thesummary of puri¢cation is shown in Table 1. Incu-bation of sample with 40 Wm cAMP prior to DEAEanion exchange chromatography eliminated a con-taminating cAMP-dependent protein kinase (PKA).PKx bound to DEAE^cellulose and phenyl^Sephar-ose hydrophobic matrix was eluted with 0.2 M and

0 M NaCl in bu¡er B, respectively. High-Q fractionsshowed PKx enriched activity at about 0.15^0.18 MNaCl gradient. The ¢nal step of puri¢cation by prot-amineâgarose a¤nity column chromatography isshown in Fig. 1. 1.6 M NaCl eluent showed a singleband of PKx (molecular mass approx. 116 kDa) on10% SDS^PAGE after silver staining of the gel (Fig.2).

E¡ects of various modi¢ers of di¡erent kinases areshown in Table 2. It is evident that arachidonic acidis the best activator of PKx. When the assay wascarried out at di¡erent concentrations of arachidonicacid, the optimum concentration was found to beabout 50 WM with an A0:5 of 20 WM (Fig. 3). Inaddition to arachidonic acid, other unsaturated fatty

Table 1Summary of puri¢cation of PKx from goat testis cytosol

Puri¢cation steps Total protein(mg)

Total activity(nmol/min)

Speci¢c activity(nmol/min per mg)

Yield(%)

Puri¢cation(fold)

Supernatant 395 7.90 0.02 100 ^DEAE^cellulose 87 6.96 0.08 88 4Phenyl^Sepharose 5 4.30 0.86 54 43High-Q 1.5 3.24 2.16 41 108Protamineâgarose 0.027 0.864 32.00 11 1600

Cytosolic protein (395 mg) was loaded on to a DEAE^cellulose column and stepwise puri¢cation was done as described in Section 2.The kinase activity was assayed in the absence and presence of 50 WM arachidonic acid as described in Section 2 using [Ser]25

PKC(19^31) peptide as substrate.

Fig. 1. Protamineâgarose a¤nity chromatography of proteinkinase x. Pooled fractions eluted from phenyl^Sepharose col-umn enriched with PKx activity was loaded on to protamineâgarose a¤nity column (3 ml). Protein was eluted with a con-tinuous gradient of NaCl. The column fractions (1 ml each)were collected and assayed for protein kinase activity with[Ser]25 PKC(19^31) peptide as substrate as described in Section2. Protein kinase activity was monitored in the presence (b) orabsence (a) of arachidonic acid. Dotted line, absorbance at 280nm; dashed line, NaCl concentration.

BBAPRO 35987 2-9-99


acids were also found to be e¡ective to di¡erent ex-tents (Fig. 4). Saturated fatty acids had no appreci-able e¡ect. Various detergents stimulated PKx at lowconcentrations, while high concentrations were detri-mental to the enzyme (Table 3). Biological surfactantlyso-phosphatidylcholine could enhance the enzymeactivity to some extent (Table 2).

Staurosporine inhibited PKx activity completelyeven at low concentration with an IC50 value of 0.3

nM (Fig. 5), but wortmannin showed no e¡ect (datanot shown).

Like all other kinases, PKx also showed a speci¢cphosphorylation site motif. Phosphorylation of dif-ferent substrates in the absence and presence ofarachidonic acid are shown in Table 4. In the ab-sence of arachidonate, protamine sulfate could be agood substrate. Histone H1 and histone IIIs werephosphorylated moderately in presence of arachido-nate. Casein and Kemptide were not e¡ective at allas substrates. PKC substrate based on classical PKCpseudosubstrate sequence, [Ser]25 PKC fragment 19^

Fig. 2. SDS^PAGE at di¡erent stages of puri¢cation of PKx.Proteins were resolved on 10% polyacrylamide gel according tothe method of Laemmli [21]. Lanes: 1, standard molecularmass protein; 2, crude cytosolic protein (30 Wg); 3, DEAE^cel-lulose fraction (20 Wg); 4, phenyl^Sepharose fraction (5 Wg) 5,High-Q fraction (3 Wg); 6, protamineâgarose fraction (1 Wg).The gel was stained with silver nitrate as described by Morris-sey [22].

Table 2E¡ects of various modi¢ers on kinase activity

Modi¢er Activity (% of control)

cAMP 107 þ 16cGMP 97 þ 15Ca2� (0.1 mM) 90 þ 18Calmodulin (0.1 mg/ml) 96 þ 21Ca2� (0.1 mM) + calmodulin (0.1 mg/ml) 92 þ 12Phosphatidylserine (10 Wg/ml) 95 þ 22Diacylglycerol (1 Wg/ml) 98 þ 17Phosphatidylserine (10 Wg/ml) + diacylglycerol (1 Wg/ml) 99 þ 22Ca2� (0.1 mM) + phosphatidylserine (10 Wg/ml) + Diacylglycerol (1 Wg/ml) 91 þ 15Phosphatidylcholine (10 Wg/ml) 87 þ 19Phosphatidylethanolamine (10 Wg/ml) 93 þ 14Lyso-phosphatidylcholine (20 Wg/ml) 163 þ 34Arachidonic acid (50 WM) 480 þ 56

PKx activity with puri¢ed protein (100 ng) was assayed using [Ser]25 PKC(19^31) peptide as substrate in presence of 5 mM MgCl2and various modi¢ers as described in the text. Data are expressed as percentage activity relative to that obtained in presence of MgCl2alone which was taken as 100. Each value represents the mean þ S.E.M. of three separate determinations.

Fig. 3. Stimulation of puri¢ed PKx by di¡erent concentrationsof arachidonic acid (C20:4). Assay was performed as describedin Section 2 using [Ser]25 PKC(19^31) peptide as substrate. Thedata shown are mean values of three separate experiments.

BBAPRO 35987 2-9-99


31 (RFARKGSLRQKNV), was heavily phosphoryl-ated by the kinase. Another PKC substrate, contain-ing Thr as phosphate acceptor (VRKRTLRRL), wasalso phosphorylated to some extent. In the presenceof arachidonate, a Vmax of 44 nmol/min per mg pro-

tein and a Km value of 70 WM for phosphorylation ofPKC substrate [Ser]25 PKC(19^31) by PKx were ob-tained from a Hanes^Woolf plot (data not shown).

Protein kinases exhibit their intracellular activitiesthrough phosphorylation of membrane or cytosolicproteins. Our kinase did not phosphorylate anymembrane protein signi¢cantly. However, two cyto-solic proteins of molecular masses 60 and 68 kDashowed pronounced phosphorylation (Fig. 6).

No immunoreactive band was observed at the re-gion of 116 kDa with antibodies against PKN or anyisotype of PKC (data not shown).

4. Discussion

In the present study, we have isolated and charac-terized a protein kinase from goat testis cytosol thatshows some similarities in co-factor and substratespeci¢cities with that of rat testis PKN [16], althoughstudies with PKN antibody suggests dissimilarities inamino acid alignment in their catalytic regions. Theantibody against PKN from rat testis did not cross-react with the PKx of goat testis. However, when apositive control of goat brain cytosol was triedagainst the above antibody, a cross-reactive bandwas obtained, again suggesting that PKx is di¡erentfrom PKN. PKx is not responsive to calcium andphospholipids (Table 2), indicating the absence of

Fig. 4. E¡ects of di¡erent free fatty acids and fatty acyl coen-zyme A on puri¢ed PKx activity. Control value represents kin-ase activity in the presence of 5 mM MgCl2 alone and the as-say conditions were as described in Section 2. Concentrationsof fatty acids and fatty acyl-CoA are given in parentheses. a,control ; b, arachidonic acid (50 WM); c, eicosapentaenoic acid(50 WM); d, oleic acid (50 WM); e, linolenic acid (50 WM); f, li-noleic acid (50 WM); g, palmitic acid (50 WM); h, stearic acid(50 WM); i, arachidoyl CoA (5 WM); j, linoleoyl CoA (5 WM).These data are given as mean values of three separate experi-ments.

Fig. 5. Inhibition of arachidonic acid-activated protein kinase(puri¢ed) by staurosporine. Assay conditions were as describedin the text. PKx activity was calculated from the di¡erence of32P incorporation into [Ser]25 PKC(19^31) peptide substrate inthe presence (50 WM) and absence of arachidonic acid. Thedata represent the mean of three separate observations.

Table 3E¡ect of di¡erent detergents on PKx activity

Detergent Concentration(%)

Activity(% of control)

Deoxycholate 0.05 148 þ 330.50 18 þ 5

Sodium dodecyl sulfate 0.005 137 þ 340.05 6 þ 2

Triton X-100 0.005 163 þ 450.05 61 þ 19

Octyl glucoside (C12 E8) 0.05 178 þ 400.50 44 þ 12

PKx activity (100 ng) was assayed in the presence of di¡erentconcentrations of detergents. Phosphate incorporation into[Ser]25 PKC(19^31) peptide in presence of 5 mM MgCl2 alonewas taken as 100 and the activities shown in the presence of de-tergents are relative to that activity. Each value represents themean þ S.E.M. of three separate determinations.

BBAPRO 35987 2-9-99


C2- and C1-like consensus domains of PKC iso-forms. Immunological experiments using antibodiesof PKC-K, L, Q, N, O, W, a and V ruled out the possi-bility of its structural homology with classical, atyp-ical or new PKC. However, most of them reactedwith PKC from goat brain.

In the puri¢cation process of PKx, incubation ofcrude sample with 40 WM cAMP dissociates a con-taminating PKA into catalytic and regulatory sub-

units. The catalytic sub-unit does not bind to anionexchange material [23] and is successfully eliminatedafter DEAEânion exchange column chromatogra-phy. PKx binds to phenyl^Sepharose hydrophobicresin (Table 1) which suggests the existence of ex-posed hydrophobic sites on the active molecule. Ex-ploiting the idea from the observation that PKx canuse protamine sulfate as its substrate, protamineâgarose a¤nity chromatography is used in the ¢nalstep of puri¢cation (Fig. 1). The puri¢ed fraction isfound to contain a single protein of molecular mass116 kDa as evident from SDS^PAGE (Fig. 2).

For further proof, the puri¢ed protein kinase wasrun on 10% polyacrylamide native gel. The bandcorresponding to 116 kDa was cut across the geland the protein was eluted from the gel. When analiquot was run on 10% SDS^PAGE, a single proteinband of molecular mass 116 kDa was obtained. Thepart used for protein kinase assay showed appreci-able arachidonic acid-stimulated PKx activity using[Ser]25 PKC substrate. The experiments conclusivelyprove that the protein of approx. 116 kDa is relatedto PKx (data not shown).

Arachidonic acid is found abundantly in the sn-2position of phospholipids and is liberated in unesteri-¢ed form due to selective cleavage by phospholipaseA2 [24]. This free fatty acid is believed to play im-portant roles (as second messenger) in signal trans-duction processes as a precursor of a variety ofbiological products such as prostaglandins and leu-kotrienes [25]. Cis-unsaturated fatty acids enhancethe diacylglycerol dependent activation of PKC andallow PKC to exhibit nearly full activity in the pres-

Table 4Substrate speci¢city of PKx

Substrate PKx activity (nmol/min per mg)

without arachidonic acid with arachidonic acid

Histone H1 (0.2 mg/ml) 8 þ 2 15 þ 4Histone IIIs (0.2 mg/ml) 9 þ 4 20 þ 6Protamine sulphate (0.2 mg/ml) 18 þ 4 23 þ 5Casein (0.2 mg/ml) 3 þ 1 4 þ 2[Ser]25 PKC (19^31) ^ RFARKGSLRQKNV (100 WM) 11 þ 3 43 þ 10PKC substrate ^ VRKRTLRRL (100 WM) 7 þ 2 15 þ 3PKA substrate (Kemptide) ^ LRRASLG (100 WM) 2 þ 1 3 þ 1

PKx activity with puri¢ed protein (100 ng) was assayed to determine its phosphotransferase activity to di¡erent protein kinase sub-strates in the absence and presence of 50 WM arachidonic acid. The assay conditions are as described in the text and each value repre-sents the mean þ S.E.M. of three separate determinations.

Fig. 6. A typical autoradiogram showing the phosphorylationof endogenous substrate proteins by puri¢ed PKx. Heat-dena-tured (90³C, 5 min) cytosolic proteins (20 Wg) of goat testiswere phosphorylated in the absence (lane 1) and presence (lane2) of 50 WM arachidonic acid and resolved on 10% SDS^PAGE. Phosphorylation was carried out as described in thetext. Molecular mass markers are indicated in the margin.

BBAPRO 35987 2-9-99


ence of sub-micromolar Ca2� concentration [26].Studies in several laboratories have found that inthe absence of phosphatidylserine, unsaturated fattyacids may activate PKC to various degrees, moste¡ectively activating the Q-isoform [27,28]. The sig-ni¢cant level of stimulation of PKx activity by lowconcentration of arachidonate suggests the kinase tobe a physiological target of arachidonic acid andother related fatty acids in vivo (Figs. 3 and 4). Arecent report suggested the presence of a phospho-lipid-dependent and fatty acid-activated proteinkinase from human platelets which is di¡erent fromPKC [29].

A number of unsaturated fatty acids and fatty acylCoA are found to be e¡ective but not to a compar-able level of arachidonic acid. Failure of saturatedfatty acids to stimulate the kinase (Fig. 4) may bedue to the absence of double bonds in their carbonchains, that possibly play vital roles in the interactionwith the kinase.

Detergents and fatty acids aggregate to formmonomeric micelles in aqueous solution at low con-centration due to their amphiphilic structures. Acti-vation of PKx by a low concentration of detergent(Table 3) may be due to direct interaction with thesemicelles. A high concentration of detergent may dis-rupt secondary structure of the enzyme leading toinactivation. It is pertinent to mention in this con-nection that not any of the detergent used here hasany e¡ect on the activation of PKx by arachidonicacid.

The inhibition by staurosporine (Fig. 5) indicatesthat a consensus sequence of ATP binding domain ofprotein kinase [7] may exist on PKx. The insensitivityto wortmannin, a phosphatidylinositol 3-kinase in-hibitor [30], ruled out the possibility of any homol-ogy of the kinase with that of PI 3-kinase. No e¡ecton our kinase was found even with PKC-speci¢c in-hibitor, calphostin C (data not shown). The ¢ndingagain conclusively suggests that PKx is di¡erent fromPKC.

Most of the protein kinases reported so far arefound to phosphorylate speci¢c amino acid resi-due(s). When we used histone IIIs as substrate andthe phosphorylated product was analysed, it wasfound that phosphorylation took place only on Serresidues (data not shown). However, when a Thr-containing peptide (VRKRTLRRL) was used as sub-

strate, weak phosphorylation was also observed (Ta-ble 4). It is therefore predicted that the kinase prob-ably phosphorylates both Ser and Thr residues butSer acts as a preferred phosphate-acceptor moleculewhen both are present.

The Km value 70 WM for [Ser]25 PKC(19^31) sub-strate is much higher than the Km determined forPKC derived from rat brain [6] and human neutro-phil [31]. This ¢nding suggests that the peptide is nota speci¢c substrate for PKx.

Phosphorylation of endogenous cytosolic proteins(Fig. 6) suggests the biological importance of thekinase, since most biological systems that are con-trolled by protein phosphorylation are believed tobe under the complex interaction of second messen-gers and protein kinases. Several laboratories havereported that arachidonic acid and to a lesser extent,related fatty acids, are able to activate several signaltransduction mechanisms such as activation of Ca2�

entry through N-methyl-D-aspartate receptor-ori-ented Ca2� channels [32], inhibition of Ras-GTPaseactivating protein [33] and activation of a GTP bind-ing protein in the neutrophil plasma membrane [34].Thus it has been suggested that arachidonic acid mayact as a `second messenger' [35,36]. In most casessecond messengers stimulate the activities of speci¢cprotein kinases, which phosphorylate and hencemodulate the activity of a large number of enzymes,membrane ion channels and structural proteins[37,38]. More recently, arachidonic acid-stimulatedprotein kinase (PKN) has been shown to bind tothe activated RhoA GTPase [39], and involvestress-induced translocation to nucleus from cytosol[40] and interaction with cytoskeleton protein [41].Hence, free unsaturated fatty acid-mediated activa-tion of PKx may have a crucial role in the regulationof reproductive systems of mammals.

Acknowledgements

This work is supported in part by grants fromCouncil of Scienti¢c and Industrial Research, Gov-ernment of India (37/0845/94-EMR-II and 37/0942/97-EMR-II). We are thankful to Prof. P.K. Ray,Director, Bose Institute, for his keen interest in thiswork and Prof. A. Ghosh of our department forproviding us with eicosapentaenoic acid.

BBAPRO 35987 2-9-99


References

[1] P. Cohen, Trends Biochem. Sci. 17 (1992) 409^413.[2] Y. Nishizuka, Trends Biochem. Sci. 8 (1983) 13^16.[3] Y. Nishizuka, Annu. Rev. Biochem. 58 (1989) 31^44.[4] H. Hug, T.F. Saree, Biochem. J. 291 (1993) 329^343.[5] M.H. Lee, R.M. Bell, J. Biol. Chem. 261 (1986) 14867^

14870.[6] C. House, B.E. Kemp, Science 238 (1987) 1726^1728.[7] A. Azzi, D. Boscoboinik, C. Hensey, Eur. J. Biochem. 208

(1992) 547^557.[8] K. Murakami, A. Routtenberg, FEBS Lett. 192 (1985) 189^

193.[9] K. Murakami, S.Y. Chan, A. Routtenberg, J. Biol. Chem.

261 (1986) 15424^15429.[10] Y. Nishizuka, Science 258 (1992) 607^614.[11] M. Szamel, B. Rehermann, B. Krebs, R. Kurrle, K. Resch,

J. Immunol. 143 (1989) 2806^2813.[12] D.S. Lester, C. Collin, R. Etcheberrigaray, D.L. Alkon, Bio-

chem. Biophys. Res. Commun. 179 (1991) 1522^1528.[13] A.K. Christensen, Leydig cells, in: R.O. Greep, E.B. Ast-

wood (Eds.), Handbook of Physiology, sect. 7, vol. 5, Amer-ican Physiological Society, Washington, DC, 1975, p. 57.

[14] D.L. Garbars, G.S. Kopf, Adv. Cyclic Nucleotide Res. 13(1980) 251^305.

[15] H. Breitbart, J. Lax, R. Rotem, Z. Naor, Biochem. J. 281(1992) 473^476.

[16] M. Kitagawa, H. Mukai, H. Shibata, Y. Ono, Biochem. J.310 (1995) 657^664.

[17] H. Mukai, Y. Ono, Biochem. Biophys. Res. Commun. 199(1994) 897^904.

[18] C. House, R.E.H. Wettenhall, B.E. Kemp, J. Biol. Chem.262 (1987) 772^777.

[19] B.E. Kemp, D.J. Graves, E. Benjamini, E.G. Krebs, J. Biol.Chem. 252 (1977) 4888^4894.

[20] M.M. Bradford, Anal. Biochem. 72 (1976) 248^254.[21] U.K. Laemmli, Nature 227 (1970) 680^685.[22] J.H. Morrissey, Anal. Biochem. 117 (1981) 307^310.[23] T.H. Lubben, J.A. Traugh, J. Biol. Chem. 258 (1983) 13992^

13997.

[24] H. Van den Bosch, Biochim. Biophys. Acta 604 (1980) 191^246.

[25] D. Piomelli, P. Greengard, Trends. Pharmacol. Sci. 11 (1990)367^373.

[26] T. Shinomura, Y. Asaoka, M. Oka, K. Yoshida, Y. Nishi-zuka, Proc. Natl. Acad. Sci. USA 88 (1991) 5149^5153.

[27] K. Murakami, A. Routtenberg, FEBS Lett. 192 (1985) 189^193.

[28] M.S. Shearman, Z. Naor, K. Sekiguchi, A. Kishimoto, Y.Nishizuka, FEBS Lett. 243 (1989) 177^182.

[29] W.A. Khan, G.C. Blobe, A.L. Richards, Y.A. Hannun,J. Biol. Chem. 269 (1994) 9729^9735.

[30] A. Arcaro, M.P. Wymann, Biochem. J. 296 (1993) 297^301.[31] S. Majumdar, L.H. Kane, M.W. Rossi, B.D. Volpp, W.M.

Nauseef, H.M. Korchak, Biochim. Biophys. Acta 1176(1993) 276^286.

[32] B. Miller, M. Sarantis, S.P. Traynelis, D.D. Attwell, Nature355 (1992) 722^725.

[33] M.H. Tsai, C.L. Yu, F.S. Wei, D.W. Stacey, Science 243(1988) 522^526.

[34] S.B. Abramson, J. Leszczynska-Piziak, G. Weissmann,J. Immunol. 147 (1991) 231^236.

[35] Z. Naor, Mol. Cell. Endocrinol. 80 (1991) C181^C186.[36] P.M. Jones, S.J. Persaud, J. Endocrinol. 137 (1993) 7^14.[37] E.J. Nestler, P. Greengard, Protein Phosphorylation in the

Nervous System, John Wiley and Sons, New York, 1984.[38] D.E. Harrison, S.J. Ashcroft, M.R. Christie, J.M. Lord, Ex-

perientia 40 (1984) 1075^1084.[39] M. Amano, H. Mukai, Y. Ono, K. Chihara, T. Matsui, Y.

Hamajima, K. Okawa, A. Iwamatsu, K. Kaibuchi, Science271 (1996) 648^650.

[40] H. Mukai, M. Miyahara, H. Sunakawa, H. Shibata, M.Toshimori, M. Kitagawa, M. Shimakawa, H. Takanaga,Y. Ono, Proc. Natl. Acad. Sci. USA 93 (1996) 10195^10199.

[41] H. Mukai, M. Toshimori, H. Shibata, H. Takanaga, M.Kitagawa, M. Miyahara, M. Shimakawa, Y. Ono, J. Biol.Chem. 272 (1997) 4740^4746.

BBAPRO 35987 2-9-99


Documents

Unsaturated fatty acid-activated protein kinase (PKx) from goat testis cytosol