Isolation and characterization of a new Ca2+ ...

Isolation and characterization of a new ~a~+/calmodulin-dependent protein kinase from isoproterenol-stimulated proliferating rat parotid acinar cells KARNAM R. PURUSHOTHAM,' JOSEPH BOLOGNA,AND YOICHI NAKAGAWA Department of Oral Biology, University of Florida, Gainesville, FL 32610, U.S.A . AND

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by ZheJiang University on 06/07/13 For personal use only.

MICHAELG. HUMPHREYS-BEHER Departments of Oral Biology and Pharmacology and Therapeutics, Universityof Florida, Gainesville, FL 32610, U.S.A. Received October 16, 1991

PURUSHOTHAM, K. R., BOLOGNAJ., NAKAGAWA, Y., and HUMPHREYS-BEHER, M. G. 1992. Isolation and characterization of a new ~a~$calmodulin-dependentprotein kinase from isoproterenol-stimulated proliferating rat parotid acinar cells. Biochem. Cell Biol. 70: 250-255. A new ~a~'/calmodulin-dependentserine kinase was isolated from rat parotid gland acinar cells following chronic treatment with the D-agonist isoproterenol. A single-step purification was performed on a calmodulin-agarose affinity column, following solubilization with Triton X-100. Among various substrates tested, bovine galactosyltransferase was the preferred substrate of the kinase, followed by glycogen synthetase > histone > phosphodiesterase > phenylalanine hydroxylase > phosphorylase b > bovine serum albumin. In comparison, a spleen preparation of ~a~+/calmodulindependent kinase did not show galactosyltransferase to be the preferred substrate. Thus, the enzyme would appear to be similar to the human galactosyltransferase-associatedkinase. The kinase activity was saturable with 100 pM c a 2 + and 2 pM calmodulin. The molecular mass determined by nondenaturing and sodium dodecyl sulfate polyacrylamide gel electrophoreses was 75 kDa with a pl of 4.3. The Vm',,was 3500 pmol/(min-mg protein) with a Kmof 1.6 pM for the transferase substrate. Leukotriene C and prostaglandin E, were found to be specific noncompetitive inhibitors of the rat galactosyltransferase-associated kinase. Key words: galactosyltransferase, hypertrophy, cell proliferation, phosphorylation.

PURUSHOTHAM, K. R., BOLOGNA, J., NAKAGAWA,Y., et HUMPHREYS-BEHER, M. G. 1992. Isolation and characterization of a new ~a~+/calmodulin-dependent protein kinase from isoproterenol-stimulatedproliferating rat parotid acinar cells. Biochem. Cell Biol. 70 : 250-255. Une nouvelle strine kinase dtpendante de la ~a~+/calmoduline est isolte des cellules acineuses de la glande parotide de rat aprts un traitement chronique avec un @-agoniste,I'isoprottrtnol. Nous avons effectut une seule ttape de purification sur colonne d'affinitt de calmoduline-agarose apres solubilisation avec le Triton X-100. Parmi les divers substrats essayts, la galactosyltransf&asebovine s'est aver& le substrat prtftrt de la kinase; viennent ensuite la glycogkne synthttase > I'histone > la phosphodiesttrase > la phtnylalanine hydroxylase > la phosphorylase b > l'albumine strique bovine. En revanche, la galactosyltransftrase n'est pas le substrat prtftrt de la kinase dtpendante de la ~a~+/calmoduline extraite de la rate. L'enzyme serait donc semblable a la kinase humaine associk a la galactosyltransftrase. L'activitt kinasique est saturte par le c a 2 + 100 pM et la calmoduline 2 pM. La masse moltculaire dtterminke par l'tlectrophortse sur un gel non dtnaturant et sur un gel de polyacrylarnide avec dodtcylsulfate de sodium est de 75 kDa avec un pI de 4,3. La V,,,',,est de 3500 pmol/(min.mg protkine) avec un Kmde 1,6 pM pour le substrat transftrase. Le leucotritne C et la prostaglandine E, sont des inhibiteurs non compttitifs sptcifiques de la kinase de rat associte a la galactosyltransftrase. Mots clks : galactosyltransferase, hypertrophie, proliftration cellulaire, phosphorylation. [Traduit par la rtdaction]

Introduction Chronic administration of the 0-adrenergic agonist I S 0 causes increased cell division of the rat parotid gland acinar cells and induces changes in the synthesis of a number of enzymes and other proteins (Ann and Carlson 1985; Humphreys-Beher et al. 1990). Although the exact mechanism of regulation of ISO-mediated cell growth is not clear, it has been shown that IS0 induces an increase in the cell surface form of Gal-Tase (Humphreys-Beher et al. 1990). ABBREVIATIONS: ISO, isoproterenol; Gal-Tase, Dl-4-galactosyltransferase; GTA, galactosyltransferase associated; CDC, cell division control; CHO. Chinese hamster ovary; ALD, aldehyde; PMSF, phenylmethylsulfonyl fluoride; DTT, dithiothreitol; MES, morpholinoethanesulfonic acid; SDS, sodium dodecyl sulfate; PI, isoelectric point; cpm, counts per minute; CaM, calmodulin; TCA, trichloroacetic acid; pNPP, Cnitrophenyl phosphate; TLC, thinlayer chromatography; sp.act., specific activity; BSA, bovine serum albumin; PAGE, polyacrylamide gel electrophoresis. ' ~ u t h o rto whom all correspondence should be addressed. Printed in Canada / Imprime au Canada

Consistent with the appearance of cell surface Gal-Tase has been the observation of ISO-induced transcription of GTA kinase mRNA (Humphreys-Beher et al. 1986, 1989). Subsequent injections of a-lactalbumin (a Gal-Tase specific substrate modifier protein) along with ISO, and the resulting inhibition of growth, suggest a possible receptor-ligand interaction with a surface glycoprotein leading to the signalling for cell proliferation (Humphreys-Beher 1989; Humphreys-Beher et al. 1987, 1990). In subsequent studies on protein phosphorylation in rat acinar cells, we have observed the ability of ~a~+/calmodulin antagonists to also block @-agonist-inducedproliferation. The coadrninistration of trifluoperazineor W-5 with IS0 reduced the parotid gland weight, the activities of cell surface Gal-Tase, and [3~]thymidine incorporation compared with administration of IS0 alone (Purushotham ei a/. 1991). A number of protein kinases with the ability to regulate eukaryotic cell DNA synthesis and division in response to external stimuli have been isolated (Hunter and Cooper 1986). Within these is a class of proteins, known as CDC


NOTES

proteins, with the enzymatic capacity t o transfer the Y-phosphoryl group from A T P t o the amino acid serine o n substrate cellular proteins. These proteins, when overexpressed, interfere with the normal cell cycle phenotype and the late events associated with normal mitosis. A human G T A kinase cDNA has been isolated (Humphreys-Beher et al. 1986, 1989; Bunnell et al. 1990c), which has about 50% homology t o the human C D C 2 and yeast C D C 28 protein kinases. The protein sequence of the G T A kinase has revealed < 30% identity t o all other known protein kinases. However, the kinase has a unique 74 amino acid N-terminal region that includes the amino acids and structural motif necessary for a calmodulin-binding site (Bunnell et al. 1 9 9 0 ~ ) .T h e overexpression of the G T A kinase cDNA in C H O cells causes decreases in cell surface Gal-Tase activity and may therefore directly o r indirectly control plasma membrane localization of the transferase. T h e 3 ' untranslated region, having a number of A U rich elements, is also responsible for changes in m R N A stability (Bunnell et al. 1990b). T h e present study reports the isolation of the rat G T A kinase protein induced in the ISO-treated parotid gland a n d its biochemical characterization.

Materials and methods Materials Radioisotope [Y-~~P]ATP (specific activity, 7000 Ci/mmol; 1 Ci = 37 GBq) was purchased from New England Nuclear (Boston, Mass.). Sprague-Dawley rats were purchased from the University of Florida breeding colony. Leukotriene C and prostaglandin E, were obtained as a gift from Dr. Nasser Chegini, University of Florida. Calmodulin-agarose-ALD was from GibcoBRL. Bovine Gal-Tase, phosphorylase b, and other substrates were purchased from Sigma Co. (St. Louis, Mo.). Ultrapure chemicals for electrophoresis were obtained from Bio-Rad (Richmond, Calif.). All other chemicals were reagent-grade quality. Animals used and I S 0 administration Male Sprague-Dawley rats (170-200 g) were maintained in the laboratory with food and water ad libitum for 1 week prior to experimentation. Groups of animals received twice daily i.p. injections of either 0.5 mL saline (control) or 0.5 mL of 10 mg ISO/mL for 3 days. On day 4, the animals were killed by exsanguination. Calmodulin affinity chromatography The parotid glands and spleen were isolated following identification by gross morphology, and homogenized in 2 mL of 10 mM Tris (pH 8.0) containing 0.1% Triton X-100. The kinases from these two tissues were isolated by the protocol specifically developed for use with the BRL calmodulin affinity matrix (Purushotham and Humphreys-Beher 1991). In brief, the cell lysates were solubilized overnight at 4OC in a buffer containing 1% w/v Triton X-100, 250 mM NaCI, 20 mM Tris (pH 7.4), and 4 pM PMSF. It was observed that Triton X-100 has no influence on the activity of either tissue kinase. The suspensions were centrifuged at 20 000 x g for 45 min. In the absence of detergent, 80% of the parotid kinase was localized to the soluble fraction following centrifugation; the other 20% was found in the membrane pellet. The supernatant was adjusted to 2 mM c a 2 + with CaCI, and applied to a calmodulin-agarose-ALD column which had been preequilibrated with calcium-containing buffer (20 mM Tris (pH 7.4), 1 mM MgCl,, 1 mM DTT, 2 mM CaCI,, and 0.1% (W/VTriton X-100) at a flow rate of 10 mL/h. The column was then washed with 7-10 bed volumes of ca2+-containingbuffer and subsequently eluted with the above buffer containing 10 mM EGTA. The eluates were concentrated in an Amicon filtration unit using a PM30 filter and dialyzed overnight at 4°C to remove salts.

25 1

Approximately 100 pL of the concentrated sample was subjected to a Bio-Rad desalting column. To improve the performance of the column on subsequent use, the packing material of the column was eluted with 8-10 bed volumes of regenerating buffer (0.1 M MES, 2 M urea, 0.02% NaN,, 1 mM EGTA, and 1.1 mM CaCl,, pH 7.0), followed by equilibration with the above ca2'-containing buffer.

Electrophoretic analysis of the GTA kinase The kinase activity present in the sample following elution from the calmodulin affinity column was then analyzed by 10% SDSpolyacrylamide gels, using a modification of the Laemmli discontinuous buffering system described by Pugsley and Schnaitman (1979). The gels were fixed and subsequently stained with Coomassie brilliant blue R-250 (Fairbanks et al. 1971) or silver stain using a Bio-Rad Silver Stain Kit as per the manufacturer's instructions. Analysis of the purified kinase on nondenaturing polyacrylamide gels was performed by the omission of 0.1% SDS from the Tris-glycine buffer and 8% acrylamide mixture during polymerization. The sample was additionally dissolved in sample buffer without SDS. The kinase band(s) were visualized by Coomassie brilliant blue staining. Isoelectric focusing was performed on an LKB1804-101 Ampholine PAG (polyacrylamide gel) plate system (pH 3.5-9.5). Twenty microlitres of dialyzed sample (1 pg) was applied to the plate. Samples were electrophoresed at 1500 V for 90 min. The gel was fixed and stained using Coomassie brilliant blue R-250 as per the manufacturer's instructions. Following staining, the isoelectric point of the rat GTA kinase was determined by plotting the pI marker standards (Sigma IEF-Mi pI = 3.6-9.3) versus their migration in centimetres. Assay of calmodulin kinase activity The kinase activity was assayed by the phosphorylation of purified protein as described by Yamauchi et al. (1989). The reaction mixture (50 pL) contained 50 pM [Y-~~P]ATP (2 x 10' 6 x 10' cpm), 8 mM magnesium acetate, 0.2 mM CaCl,. 0.1 mM EGTA, 50 mM Hepes buffer (pH KO), 2 pM CaM (Sigma), and 0.5 pg GTA kinase protein purified by CaM affinity chromatography from the ISO-treated parotid gland or spleen kinase. The reaction also contained 0.5 pM soluble bovine Gal-Tase purified by a-lactalbumin-Sepharose chromatography as the acceptor substrate (Humphreys-Beher et al. 1984). Control reactions were carried out in the absence of CaM or c a 2 + . The reactions were preincubated at 30°C for 2 min prior to the addition of enzyme. Incubation was continued at 30°C for 90 s and subsequently terminated by the addition of 10 pL of 0.4 M EDTA. The incorporation of 3 2 label ~ into protein was measured by 10% TCA precipitation of the reaction on glass fiber filters followed by liquid scintillation counting. Characterization of Gal-Tase-associated kinase specificity The specificity of Gal-Tase as a substrate for the GTA kinase or spleen ca2+/ca~-dependentkinase activity was determined by including various concentrations of the transferase or other known ca2+/ca~-dependentkinase substrates in the assay described above. The apparent K,,, was calculated by the double-reciprocal plot. In similar reactions, 25 pmol of the arachidonic acid metabolites was included to ascertain their inhibitory nature. In vitro dephosphorylation and Gal-Tase assay Bovine Gal-Tase (0.5 U) was treated with either 0.12 U potato acid phosphatase for 15 min at 37°C or 0.12 U potato acid phosphatase plus 100 mM Cnitrophenyl phosphate in 10 mM sodium acetate buffer (pH 4.5) prior to Gal-Tase enzyme assay, as described by Bunnell et al. (1990a). Gal-Tase activity was measured as previously described with the pH for transferase activity adjusted to pH 6.3 (Humphreys-Beher et al. 1987). Typically the reaction consisted of 0.1 M MES (pH 6.3), 25 mM MnCl,, 0.5% Triton X-100, 1 mM UDP-['~C]G~I (2 mCi/mM),

252

BIOCHEM. CELL BIOL. VOL.

70, 1992


TABLE1. Incorporation of [ Y - ~ ~ P I A Tinto P protein substrates catalyzed by purified 75-kDa protein kinase

FIG. 1. Isolation and polyacrylamide gel analysis of calmodulin-agarose binding proteins following isoproterenol treatment. (A) Parotid gland cell lysates from control (lane 1) and ISO-treated (lane 2) rats were solubilized in Triton X-100 for 24 h at 4°C. After centrifugation at 20 000 rpm for 45 min, 35 pg of protein per sample supernatant was separated by 10% SDSpolyacrylamide gel electrophoresisand stained with Coomassie blue R-250. (B and C) The above supernatant fraction, after adjusting to 2 mM CaC12, was subjected to calmodulin-agarose-ALD affinity chromatography. Samples were eluted with the EGTAcontaining buffer. The eluates from control (lane 1) and ISOtreated (lane 2) samples were concentrated and 1 pg of purified protein was run on 10% SDS-PAGE and stained with either Coomassie blue (B) or silver stain (C). The lanes furthest to the left of B and C are prestained molecular mass standards (Bio-Rad): phosphorylase b, 110 kDa; bovine serum albumin, 84 kDa; ovalbumin, 47 kDa; carbonic anhydrase, 33 kDa; soybean trypsin inhibitor, 24 kDa; lysozyme, 16 kDa. 10 mM ovalbumin as acceptor, and untreated or dephosphorylated Gal-Tase. All reactions contained 100 mM pNPP, so as to normalize for the phosphatase substrate in the reaction during the enzyme assay. Dephosphorylation of Gal-Tase with bacterial alkaline phosphatase (1 U) or calf intestinal phosphatase (1 U) was performed at 37°C for 15 min in 10 mM Tris buffer of pHs 8 and 7.4, respectively, prior to adjusting the pH to 6.3 with MES buffer for the Gal-Tase enzyme assay. In another set of assays, Gal-Tase was incubated with the purified rat GTA kinase prior to the addition of substrate UDP-Gal and ovalbumin. Activity of the phosphatases was monitored by the hydrolysis of thepNPP which is a chromogenic substrate for these enzymes. Phosphoamino acid analysis Bovine Gal-Tase was incubated with [ Y - ~ ~ P I A T under P conditions described above for active phosphorylation by the GTA kinase. The transferase was subsequently purified by a-lactalbumin chromatography and the protein hydrolyzed at 100°C for 2 h in 50 pL of 6 N HCI. The protein was spotted onto TLC plates and electrophoresed at 1000 V for 2 h in acetic acid - formic acid H 2 0 (78:25:89 by volume; pH 1.9), followed by ascending chromatography in isobutyric acid - 0.5 N NH,OH (5:3 v/v) for 6 h at room temperature (Ottaviano and Gerace 1985). Nonradioactive phosphoamino acid standards (Sigma) were visualized by ninhydrin spraying. Experimental samples were exposed to X-ray film for 48 h at - 80°C using enhancing screens and Kodak XARS film.

Results Purification of the GTA kinase The protein profile of the 1% Triton X-100 solubilized rat parotid gland homogenates showed at least six distinct bands detected with Coomassie blue stain in both control and ISO-treated crude supernatants. However, when the samples were applied to the CaM-agarose-ALD column and eluted with the buffer containing 10 mM EGTA, only one band with a molecular mass of 75 kDa in the ISO-treated

Condition

Kinase specific activity

No c a 2 + , bovine Gal-Tase 0.1 mM c a 2 + ,bovine Gal-Tase 0.1 mM c a 2 + , histone 0.1 mM c a 2 + ,bovine serum albumin No CaM, 0.1 mM c a 2 + , bovine Gal-Tase

1800k 300 3200 + 400 1800+ 150 300 + 50 500 + 100

NOTE: GTA kinase (0.5 pg) and CaM (2 pM) were added to initiate phosphorylation in the reaction mixtures as described in Materials and methods. After incubation at 30°C for 90 s, the reactions were terminated by addition of LO pL of 0.4 M EDTA and the radiolabel associated with the samples was measured by precipitation on glass fiber filters and scintillationcounting. All values are expressed as means f SD of three experimental determinations, corrected for background with no substrate present. The specific activity of the GTA kinase is expressed as pmol ATP hydrolyzed/(min .mg purified enzyme).

sample was observed using 1.0 pg of the purified protein (Fig. 1). Further purity was established by silver staining of the SDS gel and by isoelectric focusing. Under this sensitive detection system, again only a single protein band was identified at 75 kDa. Electrophoresis on a nondenaturing 8% polyacrylamide gel did not reveal multimeric forms of the kinase (data not presented). Only one form was observed for the kinase when the sample buffer contained or was devoid of 0.1 % 2-mercaptoethanol. A single isoelectric point protein of pI = 4.3 for the kinase was also determined (not shown). The recovery of the protein resulted in a 2500-fold purification (or approximately 7.5% of the initial kinase activity of the parotid acinar cell lysate) based on kinase activity from the crude supernatant (sp.act. = 1.28 pmol ATP/(min .mg protein)) and the activity of eluate bound to the column (sp.act. = 3200 pmol ATP/(min-mg protein)). The molecular weight of the rat parotid kinase was similar to that detected in CHO cells by Bunnell et al. (1990~)using antibody to a peptide fragment from the human GTA kinase. Enzymatic properties of the GTA kinase Incubation of the purified parotid gland GTA kinase with [Y-~~PIAT resulted P in a ca2+/cah4-dependent phosphorylation of substrate Gal-Tase, which proceeded linearly during the incubation from 30 to 120 s. Under the conditions of increasing concentrations of cofactors, the activity of the GTA kinase was saturable with 100 pM c a 2 + and a 2 pM concentration of CaM. These concentrations were found to also elicit optimal enzyme activity. The phosphorylation capacity of the kinase was reduced by 44% with the omission of c a 2 + and 84% by the omission of CaM (Table 1). Kinase activation required the presence of ca2+, CaM, and ATP as substrate, thus indicating that the parotid GTA kinase belongs to the CaM-dependent kinase family. There was no detection of kinase autophosphorylation in the presence or absence of c a 2 + , as determined by SDSpolyacrylamide gel electrophoretic separation of the reaction mixture and subsequent autoradiography (data not shown). This is consistent with the observations of Bunnell et al. (1990~)using histone H1 as substrate with immunopurified human GTA kinase. The 75-kDa kinase showed c a 2 +- and CaM-independent kinase activity towards the substrate Gal-Tase (Table 1). The substrate specificity of parotid GTA kinase was examined by carrying out phosphorylation reactions with


FIG. 2. Substrate preference for the purified GTA kinase and spleen Ca2+/ca~-dependent kinase. Purified rat GTA kinase was incubated under the conditions of kinase activity described in Materials and methods (A) or spleen kinase (B). Substrate bovine Gal-Tase and glycogen synthetase were assayed at concentrations ranging from 0 to 10 pM. All other substrates ranged from 0 to 50 pM. The x-axis coordinates represent concentrations of the substrates: 1.0 = 2.5 or 12.5 pM; 2.0 = 5.0 or 25 pM; 3.0 = 7.5 or 37.5 pM; 4.0 = 10 or 50 pM. The substrates are bovine Gal-Tase (w), glycogen synthetase (o), histone (A), phosphorylase b (o),phosphodiesterase ( x ) , phenylalanine (a), and BSA (A).

various substrate proteins using 250 ng purified protein from rat parotid gland in the presence and absence of c a 2 + and CaM. The kinase catalyzed phosphorylation of bovine Gal-Tase to the greatest level, followed by glycogen synthetase, histone, phosphorylase b, phosphodiesterase, phenylalanine hydroxylase, and BSA (Fig. 2A). Similar phosphorylation substrates have been reported for rat spleen and liver CaM kinase by Sato et al. (1990). As a comparison, the spleen kinase was isolated under identical conditions to the GTA kinase and the substrate specificity profile is presented in Fig. 2B. The ability to phosphorylate Gal-Tase was substantially lower for the spleen kinase relative to that observed for the rat parotid kinase. The substrate preference was glycogen synthetase, followed by phenylalanine hydroxylase, phosphodiesterase, BSA, histone, Gal-Tase, and phosphorylase b. The parotid GTA kinase is unique in that the Gal-Tase was the most preferred substrate in terms of the phosphorylation reaction, using commercially available substrates for this comparison. ~ ATP into bovine GalThe incorporation of 3 2 from Tase was saturated at 5 min of incubation at 30°C. Using a-lactalbumin-purified soluble bovine Gal-Tase, the maximal incorporation of radiolabel was 1.21 mol 3 2 ~ / m oGall Tase. This was based on a molecular mass of 48 kDa for the transferase. Kinetic analysis of the GTA kinase The apparent Km value for the Gal-Tase substrate with parotid GTA kinase was found to be 1.6 pM and the Vm, was 3500 pmol/(min-mg protein). In general, arachidonic acid metabolites inhibited activity of purified GTA kinase in a concentration-dependent manner. In the presence of leukotriene C and prostaglandin E2, the V,, was decreased (2000 and 1700 pmol/(min. mg protein), respectively) without altering the K, value. Thus, the GTA kinase inhibition by these metabolites was found to be noncompetitive with respect to either ATP or Gal-Tase as substrate. The reversibility of the inhibitory effects was demonstrated by taking advantage of BSA to bind to fatty acids. Addition of BSA blocked the inhibition of kinase activity by the metabolites of arachidonic acid. Role of Gal- Tase phosphorylation Since we examined the ability of the isolated kinase to phosphorylate bovine Gal-Tase in vitro, it was of interest

FIG. 3. Phosphoamino acid analysis of GTA kinase substrate bovine Gal-Tase. Thin-layer chromatography was performed on bovine Gal-Tase (20 000 cpm) following purification by crlactalbumin affinity chromatography. Phosphoamino acid standards are as follows: P-Tyr, phosphotyrosine; P-Thr, phosphothreonine; P-Ser, phosphoserine. Analysis was performed as described in Materials and methods. C, chromatography; E, electrophoresis.

to determine the possible role phosphorylation may play on enzyme activity. In the presence of potato acid phosphatase, calf intestinal phosphatase, or bacterial alkaline phosphatase, Gal-Tase activity was not reduced compared with native Gal-Tase. Gal-Tase activity averaged 0.33 nmol Gal incorporated into substrate acceptor per hour. Further incubation of a-lactalbumin-purified Gal-Tase with the GTA kinase and subsequent increased phosphorylation of the transferase did not, again, alter activity towards the substrate glycoprotein (0.31 nmol/h versus 0.32 nmol Gal incorporated/h into ovalbumin for bovine Gal-Tase preincubated in the GTA kinase reaction and no preincubation, respectively). As a further control, we preincubated UDP-Gal with


254

BIOCHEM. CELL BIOL. VOL. 70.

the acid phosphatase before the addition of bovine Gal-Tase. This procedure did not, however, decrease transferase activity either, as would be the case if the nucleotide sugar was being hydrolyzed independently of transferase utilization. Phosphoamino acid analysis An analysis of 32~-containing amino acids from bovine Gal-Tase following repurification by a-lactalburnin chromatography, hydrolysis by boiling in 6 N HC1, and twodimensional chromatography (Ottaviano and Gerace 1985) was performed. Essentially all radiaoctivity cornigrated with phosphoserine (Fig. 3). This observation is consistent with the previous results of Strous et al. (1987).

1992

most notably the requirement for both c a 2 + and CaM for optical activity (Sato et al. 1990). However, it is unique in that it shows a preference for Gal-Tase as substrate and exhibits certain biochemical characteristics such as monomeric activity, which are different for other CaM kinases reported. The kinase itself does not appear to undergo autophosphorylation of Ser or Thr, or c a 2 + / CaM-independent activity. The multifunctional CaM kinase is ubiquitous in its tissue distribution and is reported to regulate numerous physiological functions such as glycogen metabolism, neurotransmitter exocytosis, catecholamine biosynthesis, and cytoskeletal interactions (Colbran and Soderling 1990). Our present and previous data (Purushotham et al. 1991) suggest that the kinase is involved Discussion in regulating targeting rather than the enzymatic activity of In this report, we have presented evidence for the isolaGal-Tase. In conclusion, results presented here detail the tion and initial characterization of a new c a 2 + / c a ~ - isolation and characterization of a new c a 2 + / c a ~ binding serine kinase that is induced following chronic I S 0 dependent serine kinase from isoproterenol-stimulated treatment in the parotid gland of rats. The molecular mass parotid rat acinar cells. The nature and extent of involveof the purified kinase is consistent with the mass reported ment of GTA kinase in localization of the Gal-Tase versus for GTA protein observed in CHO cells, although this may regulation of transferase activity in vivo is presently being be anomalous based upon the mobility of the human recomfurther investigated. binant protein in SDS gels (Bunnell et al. 1990~).Evidence Acknowledgements for the possible involvement of a CaM-binding protein in regulating acinar cell proliferation was observed by reduced The authors would like to thank Mr. James Fischer for parotid gland hypertrophy and acinar cell hyperplasia in the technical assistance and Ms. Marilyn Lietz for preparation presence of CaM antagonists trifluoperazine and W-5, when of our manuscript. We thank Dr. N. Chegini for providing given to animals along with chronic I S 0 treatment lipid substrate inhibitors used in these studies. This work (Purushotham et al. 1991). These inhibitors also prevented was supported by National Institute of Dental Research the cell surface appearance of Gal-Tase. DNA sequence grants DE 08778 and DE 00293 to M.H.B. analysis of the human GTA kinase has shown a high degree of homology to human and yeast CDC kinases (Bunnell Ann, D.K., and Carlson, D.M. 1985.The structure and organization of a proline-rich protein gene of a multigene family. J. Biol. et al. 1990b). This kinase further contains overlapping Chem. 260: 15 863 - 15 872. regions of known CaM-binding sites. Variations in the Bunnell, B.A., Adams, D.E., and Kidd, V. J. 1990a. Transient cellular level of the GTA kinase affect not only the cell cycle, expression of a p58 protein kinase cDNA enhances mammalian but also levels of cell surface Gal-Tase (Bunnell et al. 1990a, glycosyltransferase activity. Biochem. Biophys. Res. Commun. 1990b, 1990~).Contrarily, we were unable to demonstrate 121: 196-203. a role for phosphorylation in the regulation of enzyme Bunnell, B.A., Fillmore, H., Gregory, P., and Kidd, V.J. 1990b. activity. A dominant negative mutation in two proteins created by ectopic The de novo expression of the GTA kinase that is initiated expression of an AU-rich 3 ' untranslated region. Somatic Cell by increased cAMP may occur in much the same fashion Mol. Genet. 16: 151-162. as the proline-rich proteins, since injection of the Bunnell, B.A., Heath, L.S., Adams, D.E., et al. 1990c. Increased expression of a 58 kDa protein kinase leads to changes in the P-adrenergic receptor antagonist propranolol blocks the CHO cell cycle. Proc. Natl. Acad. Sci. U.S.A. 87: 7467-7471. appearance of surface Gal-Tase and cAMP accumulation Colbran, R.J., and Soderling, T.R. 1990. Calcium/camoldulin (Ann and Carlson 1985). The subsequent event, namely dependent protein kinase 11. In Current topics in cellular regulatargeting of Gal-Tase from Golgi to plasma membrane, may tion. Vol. 31. Edited by B.L. Horecker, E.R. Stadtman, in part be mediated by intracellular mobilization of c a 2 + P.B. Chock, and A. Levitzki. Academic Press. New York. and interaction with CaM and GTA kinase. Strous et al. pp. 181-221. (1987) have isolated Gal-Tase from the Golgi and extraFairbanks, G., Steck, T.L., and Wallach, D.F.H. 1971. Electrocellular media following pulse-chase labeling with phoretic analysis of the major polypeptides of the human [32~]orthophosphate from HepG2 cells. They noted that erythrocyte membrane. Biochemistry, 10: 2606-2617. the level of phosphorylation of the transferase decreases in Humphreys-Beher, M.G. 1989. Restoration of a-lactalbumin inhibited rat parotid salivary gland hypertrophy and hyperplasia transit through the cell. This could, in part, be accounted by agents specific for membrane glycoprotein N-acetylfor by the loss of a membrane-anchoring peptide. Using a glucosamine. Arch. Oral Biol. 34: 81 1-819. dephosphorylated soluble form of bovine Gal-Tase, we Humphreys-Beher, M.G., Immell, M., Jentoft, N., et al. 1984. observed a stoichiometry of phosphorylation of 1.2 mol to Isolation and characterization of UDP-ga1actose:N-acetyl1 mol transferase. This variability in molar ratio of glucosamine 4P-galactosyltransferase activity induced in rat phosphorylation may be accounted for by the fact that parotid glands treated with isoproterenol. J. Biol. Chem. 259: soluble Gal-Tase is composed of 48- and 42-kDa proteolytic 5799-5802. s ecies (Magee et al. 1974). Therefore, the molar ratio of Humphreys-Beher, M.G., Bunnell, B., Van Tuinen, P., et al. 1986. 3P' to transferase may actually be higher. Molecular cloning and chromosomal localization of human The GTA kinase reported here shares several character4P-galactosyltransferase. Proc. Natl. Acad. Sci. U.S.A. 83: 8918-8923. Correction: PNAS 86: 8747, 1989. istics belonging to the broad class of c a 2 + / c a kinases, ~

255


NOTES

Humphreys-Beher, M.G., Schneyer, C.A., Kidd, V.J., and Marchase, R.B. 1987. Isoproterenol-mediated parotid gland hypertrophy is inhibited by effectors of 40-galactosyltransferase. J. Biol. Chem. 262: 11 706 - 11 713. Humphreys-Beher, M.G., Bunnell, B., Van Tuinen, P., et al. 1989. Correction: Molecular cloning and chromosomal localization of human 40-galactosyltransferase.Proc. Natl. Acad. Sci. U.S.A. 86: 8747. Humphreys-Beher, M.G., Zelles, T., Maeda, N., et al. 1990. Cell surface galactosyltransferase acts as a general modulator of rat acinar cell proliferation. Mol. Cell. Biochem. 95: 1-1 1. Hunter, T., and Cooper, J.A. 1986. Viral oncogenes and tyrosine phosphorylation. In The enzymes. Vol. 17. Edited by P.D. Boyer and E.G. Krebs. Academic Press, Orlando. pp. 191-246. Magee, S.C., Mawal, R., and Ebner, K.E. 1974. Multiple forms of galactosyltransferase from bovine milk. Biochemistry, 23: 99-107. Ottaviano, Y., and Gerace, L. 1985. Phosphorylation of the nuclear lamins during interphase and mitosis. J. Biol. Chem. 260: 624-632. Pugsley, A.D., and Schnaitman, C.A. 1979. Factors affecting the electrophoretic mobility of the major outer membrane proteins

of Escherichia coli in polyacrylamide gels. Biochim. Biophys. Acta, 581: 163-178. Purushotham, K.R., and Humphreys-Beher, M.G. 1991. Use of calmodulin agarose ALD in the purification of calcium dependent calmodulin kinase. Focus, 13: 63-65. Purushotham, K.R., Zelles, T., and Humphreys-Beher, M.G. 1991. Role of protein phosphorylation and inositol phospholipid turnover in rat parotid gland proliferation. Mol. Cell. Biochem. 102: 19-33. Sato, H., Yamauchi, T., and Fujisawa, H. 1990. Purification and characterization of calmodulin dependent protein kinase I1 from rat spleen: a new type of calmodulin dependent protein kinase 11. J. Biochem. (Tokyo), 197: 802-809. Strous, G.J., Van Kerkhof, P., Fullon, R.J., and Schwartz, A.L. 1987. Golgi galactosyltransferase contains serine-linked phosphate. Eur. J. Biochem. 169: 307-31 1. Yamauchi, T., Ohsako, S., and Deguchi, T. 1989. Expression and characterization of calmodulin dependent protein kinase I1 from cloned cDNAs in Chinese hamster ovary cells. J. Biol. Chem. 264: 19 108 - 19 116.

Accumulation of starch in Chlamydomonas reinhardtii flagellar mutants BRADFORD S. HAMILTON Institute of Medical Science and Department of Medicine, Sunnybrook Health Science Centre University of Toronto, Toronto, Ont., Canada M4N 3M5

K ~ z u oNAKAMURA Department of Biological Sciences, University of Lethbridge, Lethbridge, Alta., Canada TIK 3M4 AND

DANIELA. K.

RONCARI'

Institute of Medical Science and Department of Medicine, Sunnybrook Health Science Centre University of Toronto, Toronto, Ont., Canada M4N 3M5 Received July 23, 1991 HAMILTON, B. S., NAKAMLTRA, K., and RONCARI,D. A. K. 1992. Accumulation of starch in Chlamydomonas reinhardtii flagellar mutants. Biochem. Cell Biol. 70: 255-258. Paralyzed flagellar mutants pf-1, pf-2, pf-7, and pf-18 of the green alga Chlamydomonas reinhardtii (Dangeard) were shown to store a significantly greater amount of starch than the motile wild type 137c + . The increase in starch storage was significant relative to protein, chlorophyll, and cell number. Analysis of average cell size revealed that the paralyzed mutants were larger than the wild type. This increase in storage molecule accumulation supports an inverse relationship between chemical energy storage and energy utilization for biomechanical/motile cellular functions. Chlamydomonasreinhardtii provides a useful model for studies of the role of cytoskeletal activity in the energy relationship and balance of organisms. Key words: Chlamydomonas, cytoskeleton, paralyzed flagella, starch, bioenergetics. K., et RONCARI, D. A. K. 1992. Accumulation of starch in Chlamydomonasreinhardtii HAMILTON,B. S., NAKAMURA, flagellar mutants. Biochem. Cell Biol. 70 : 255-258. Les mutants flagellaires paralysCs pf-1, pf-2, pf-7 et pf-18 de l'algue verte, Chlamydomonasreinhardtii (Dangeard) emmagasinent une quantitt beaucoup plus grande d'amidon que le type sauvage mobile 137c + . Cette augmentation d'emmagasinage de l'amidon est relike de facon importante aux protCines, B la chlorophylle et au nombre de cellules. L'analyse des dimensions moyennes des cellules rCvtle que celles des mutants paralysCs sont plus grandes que celles de type sauvage. Cet accroissement de l'accumulation des mol&ules de storage supporte une relation inverse entre l'emmagasinage de 1'Cnergie chimique et l'utilisation CnergCtique pour les fonctions cefiulaires biomCcaniques/motiles. Chlam~domonasreinhardtii fournit un modtle utile Dour les ttudes du rSle de I'activitC cvtosauelettiaue dans la relation - et lYCq-uilibreCnergCtiques des organismes. Mots elks : Chlamydomonas, cytosquelette, flagelles paralysts, amidon, biotnergttique. [Traduit par la rCdaction] I

Author to whom all correspondence should be sent at the following address: Sunnybrook Health Science Centre, 2075 Bayview Ave., North York, Ont., Canada M4N 3M5. Printed in Canada / Imprime au Canada