THE JOURNAL OF B~KXXCAL CHEMISTRY Q 1990 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 265, No. 19, Issue of July 5, pp. 11056-11061, 1990 Printed in V. S. A.

Activated Protein Kinase C Directly Phosphorylates the CD34 Antigen on Hematopoietic Cells*

(Received for publication, December 18, 1989)

Mary Jo FacklerS, Curt I. Civin@, D. Robert Sutherlandn, Michael A. Bakerll, and W. Stratford May+11 From the iJohns HoDkins Oncolozv Center. Baltimore. Maryland 21231 and the YlOncology Clinic, The Toronto General Hospital, ibronto MiG 2C4, Canada ’

The CD34 antigen is a human leukocyte membrane protein expressed specifically by lymphohematopoietic progenitor cells. We found that CD34 is a phosphopro- tein and therefore examined the regulation of its phos- phorylation. Activation of protein kinase C (PKC) en- hanced CD34 phosphorylation. The PKC activators, 12-0-tetradecanoylphorbol-13-acetate and bryosta- tin- 1, induced rapid, stoichiometric hyperphosphory- lation of CD34 protein in cells, resulting in a 5-fold increase in CD34 phosphorylation. In vitro kinase studies revealed that purified PKC could directly phos- phorylate purified CD34. Only serine phosphorylation was detected in the CD34 molecule. Two-dimensional phosphopeptide mapping experiments indicated that PKC induces the phosphorylation of identical serine residue(s) in vitro and in uiuo (in KG1 cells). These are newly phosphorylated serine residue(s), which are not detectably phosphorylated in CD34 from exponentially growing KG1 cells. These data indicate that the devel- opmental stage-specific molecule, CD34, is a phos- phorylation target for activated PKC. Furthermore, these findings raise the possibility that PKC activation and phosphorylation of the CD34 molecule may play a role in signal transduction during early lymphohema- topoiesis.

CD34l is a membrane-associated glycoprotein whose sur- face expression is confined to lymphohematopoietic (l-4) and some endothelial cells (5, 6). Since CD34’ bone marrow cells contain precursors for all lymphohematopoietic lineages, it is thought, but not demonstrated, that CD34 may play a regu- latory role in progenitor cell function. Recently, structural and biochemical analyses, including partial amino acid se-

* This research was supported by National Institutes of Health Grants CA 44649, CA 47993, CA 32318, CA 06973, and CA 09071 and by the Medical Research Council of Canada. Presented in part at the 18th Annual Meeting of the International Society of Experimental Hematology, Paris, July, 1989. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ Scholar of the Leukemia Society of America. 11 Scholar of the Leukemia Society of America. To whom corre-

spondence and reprint requests should be addressed: The Johns Hopkins Oncology Center, 424 North Bond St., Baltimore, MD 21231. Tel.: 301-955-8696.

1 The abbreviations used are: CD, cluster designation; PKC, protein kinase C; TPA, 12-0-tetradecanoylphorbol-13-acetate; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TPCK, L- l-tosylamido-2-phenylethyl chloromethyl ketone; M-CSF, mono&e/ macrophage colony stimulating factor; EGTA, [ethylenebis (oxylthy- lenenitrilo)]tetraacetic acid.

quencing and determination of the amino acid and carbohy- drate compositions, have revealed that the CD34 antigen is a highly acidic (p1 < 4.0), heavily glycosylated cell surface molecule of 110,000 daltons (6, 7, 48).’ In the course of these studies, our experiments indicated that CD34 is a phospho- protein. Since phosphorylation is often involved as a mecha- nism for functional regulation of surface glycoproteins, such as growth factor receptors, we have further investigated this property of CD34 reasoning that it may provide some clue(s) concerning the regulation and function of CD34 on hemato- poietic cells. The findings reported here indicate that CD34 can be stoichiometrically phosphorylated by protein kinase C (PKC), when the latter is activated by tumor-promoting phor- bol esters or bryostatin-1 (8). Since PKC, a Ca*+/phospho- lipid-dependent serinelthreonine protein kinase, appears to mediate hematopoietic growth and differentiation (9, 10) and can phosphorylate specific membrane-bound growth factor receptors (e.g. epidermal growth factor (ll)), as well as certain leukocyte differentiation antigens (e.g. CD3, CD4, CD8, CD43, CD45 (12, 13)), we sought to determine the mechanism(s) whereby CD34 was phosphorylated and the consequences, if any, of this covalent modification.

EXPERIMENTAL PROCEDURES

Materials-Murine monoclonal My10 (IgG1 directed against CD34 antigen) and MOPC 21 (IgG1 irrelevant antibody) ascites were ob- tained as described previously (1). TijK3 (IgG3 directed against CD34 antigen) ascites was kindly provided by Dr. B. Uchanska-Ziegler (Institute fur Experimentelle Immunologie, Marburg, Germany) (4). Bryostatin-1 was generously provided by Dr. G. Robert Pettit (Ari- zona State University, Tempe, AZ). 12-O-tetradecanoylphorbol-13- acetate (TPA) and 4-a-phorbol were purchased from Sigma. All other reagents were purchased from commercial sources.

Cell Culture-Human leukemic cell lines were cultured as described (1). KG1 and KGla (an undifferentiated subline of KGl) myeloge- nous leukemia cells (14, 15) were kindly provided by Dr. D. Golde (UCLA). The T-lymphoblastic cell lines RPM1 8402 (16) and MOLT 13 (derived by Dr. J. Minowada, Hayashibara Biochemical Labora- tories, Okayama, Japan) were obtained from the Roswell Park Cell Repository (Buffalo, NY) and Dr. Michael Brenner (Dana Farber Cancer Institute, Boston, MA), respectively.

Metabolic Labeling and Treatment of Cell.-Cells in the exponential phase of growth were washed with phosphate-free RPM1 1640 media (GIBCO) and equilibrated with carrier-free [32P]orthophosphate (Amersham Corp.; 100 &i/ml, 60-75 min, 37 “C). Cells (4 X 10’) were treated for 30 min with PKC activators or growth factors. Following treatment, cells were washed in phosphate-buffered saline at 4 “C. Subsequent manipulations were carried out at 4 ‘C. The cell pellet was lysed in 250 ~1 of detergent disruption buffer (10 mM Tris- HCl, pH 7.5, containing 1% Nonidet P-40, 200 units/ml of Trasylol (Mobay Chemical, New York), 0.005% leupeptin, and 10 mM EDTA), and vortexed vigorously every 5 min, for 30 min. The lysate was centrifuged at 15,000 x g for 10 min in a Beckman microcentrifuge,

’ M. J. Fackler, unpublished observations.

11056

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Protein Kinase C Phosphorylates the CD34 Antigen 11057

and CD34 protein was immunoprecipitated from the resulting super- natant.

Zmmunoprecipitation of CD34-To immunoprecipitate the CD34 antigen, the cellular lysates (the equivalent of 2-4 X 10’ cells) were precleared by adding 100 ~1 of phosphate-buffered saline containing 2% bovine serum albumin, 200 pl of Sansorbin (Calbiochem), and 40 ~1 of Protein A-Sepharose (Sigma). The resulting suspension was mixed by rocking for 30 min and then centrifuged at 15,000 X g for 10 min. The precleared lysate was incubated for 2 h with monoclonal My10 (-+ TDK3) antibody-coated Sepharose (40 ~1; prepared by preincubating Protein A-Sepharose with rabbit anti-mouse IgG (Southern Biotechnology, Birmingham, AL), washing, then incubat- ing with an equal mixture of these two monoclonals). Immune com- plexes were then washed extensively in phosphate-buffered saline, then boiled for 3 min in Laemmli sample buffer (17) before electro- phoresis. In some cases, the CD34 contained in the complexes was desialylated by treatment with neuraminidase (1 unit/ml in 50 ~1 final volume; Genzyme, Boston, MA), as described by the manufac- turer, prior to electrophoresis.

Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS- PAGE)-SDS-PAGE was performed according to the method of Laemmli (17). Following electrophoresis, the poiyacrylamide gel was stained for protein using Coomassie Brilliant Blue and/or the Quick Silver kit (Amersham Corp.). For autoradiography, gels were dried and exposed at -80 “C using Kodak X-Omat AR film.

Determination of the Stoichiometry of CD34 Antigen Phosphoryln- tion-Cells were equilibrated with [32P]orthophosphoric acid for 75 min as described (18). Trichloroacetic acid-soluble material was neu- tralized by extraction in 1,1,2-trichlorotrifluroethane (Aldrich) con- taining 0.5 M trioctylamine (Aldrich), and the non-organic phase (upper phase) was analyzed for ATP concentration by high pressure liquid chromatography as described (19). The specific activity (Ci. mol-‘) of [32P]ATP was calculated by dividing the disintegrations/ min (dpm) contained in the ATP peak by the concentration of ATP determined. The stoichiometry of “P-labeled phosphate incorporated into the CD34 molecule was calculated by dividing the dpm/mol of CD34 antigen auantitativelv immunopreciuitated from 2.7 X lo7 cells (based on the average 3.75 X 10’ cell surface molecules of CD34/KGl cell with a range of -2.5-5.0 x 10’ molecules/cell) (48)* by the specific activity calculated for [32P]ATP (dpm/mol). Since the specific activity of [32P]ATP obtained in these analyses presumably represents phos- phate molecules labeled at the 01,& as well as y position, the absolute specific activity of r-ATP is unknown. For our purpose, we have assumed that the majority of the 32P is r-ATP. Therefore, the calculations will, if anything, actually represent an underestimate of the specific activity of [-y-32P]ATP, since label incorporated in the (Y and @ positions would be unable to enter into the phosphotransfer reaction. Thus it follows that our calculations, if anything, represent an underestimate of the stoichiometry of incorporation of “P into the CD34 molecule. In experiments not shown, we determined that all detectable CD34 antigen is membrane-bound. Results represent the average values obtained from three separate experiments per- formed in triplicate. In a parallel experiment, it was also determined, by sequential rounds of immunoprecipitation with antibody, that all of the CD34 antigen which could be detected was removed by a single round of immunoprecipitation.

In Vitro Labeling of Purified CD34 Antigen-CD34 antigen was purified from KGla cell lysate to a single band on SDS-PAGE gel, as assessed by silver stain (results not shown) (48).’ Protein kinase C was purified from rat brain essentially as described (20). Purified CD34 antigen (0.1 rg) was added to 50 ~1 of a standard reaction mixture containing 10 pM ATP, 1 mM MgClp, 1 mM MnC12, 15 mM NaCl, 1 mM dithiothreitol, 1 &i of [Y-~‘P]ATP f 10 nM TPA, 40 &ml nhosnhatidvlserine (Avanti. Pelham. AL), and 100 uM CaCL, .-. - _ _ . in 10 mM Tris-HCl, pH 7.4. Various components of the reaction mixture were omitted as controls. To initiate the reaction, purified protein kinase C was added as described (18), and the incubation continued for 25 min at room temperature. To stop the reaction, samples were boiled in Laemmli sample buffer.

Two-dimensional Separation of CD34 Phosphopeptides-Peptide mapping was performed essentially as described (18) with the follow- ing modifications. Briefly, bands containing the purified 32P-labeled CD34 antigen were excised from the dried SDS-polyacrylamide gel, rehydrated, and digested in NH,HCO, buffer containing 50 pg of TPCK-treated trypsin for 16 h at 37 “C. An additional 50 rg of TPCK-treated trypsin was added, the digestion continued for 3 h, and the suspension was centrifuged. The supernatant was collected, lyophilized, and washed several times. The resulting digests were

dissolved in 5 ~1 of electrophoresis buffer (pyridine/acetic acid/ acetone/water (1:2:8:40)) and applied to thin layer plates. Electro- phoresis was carried out in the first dimension at 1000 V for 30 min with cooling. After drying, the plate was chromatographed in the second dimension in n-butyl alcohol/pyridine/acetic acid/water (45:30:9:36) until the leading edge of the solvent had reached l-2 cm from the top (-45 min).

Phosphoamino Acid Analysis-A portion of the resulting tryptic digests from 32P-labeled CD34 antigen was incubated in 500 ~1 of 6 N HCl at 105 “C for 60 min as described (21). The standards were identified with ninhydrin spray (Sigma). “P-Labeled phosphoamino acids were visualized by autoradiography.

RESULTS

Activators of Protein Kinase C Mediate Hyperphosphoryla- tion of the CD34 Antigen in Cells-The CD34’ cell lines, KG1 and KGla myelogenous leukemia cells, and RPM1 8402 and MOLT 13 T-lymphoblastic leukemia cells were equilibrated in [32P]orthophosphate, and the CD34 antigen was immuno- precipitated from cellular lysates by using specific antibodies. Low level incorporation of 32P-labeled phosphate into the llO- kDa CD34 molecule was detected in exponentially growing CD34+ cells (Fig. lA, lone I). To confirm that the immuno- precipitated, 32P-labeled IlO-kDa protein identified was the CD34 molecule, we treated the resulting immune complex with neuraminidase to release terminal sialic acid residues before SDS-PAGE. The apparent molecular weight of the CD34 molecule shifted up from M, = 110,000 to M, = 150,000 after this treatment, a characteristic previously described for the asialo-CD34 molecule (Fig. 1B) (6, 7).*

Since PKC activators such as the phorbol ester, TPA, and the macrocyclic lactone, bryostatin-1, have been reported to stimulate hematopoietic progenitor cells in culture (22-24), we wished to determine whether the CD34 antigen could be phosphorylated by activated PKC. As a first approach, after addition of a biologically active concentration of TPA (23, 24), hyperphosphorylation of the CD34 antigen was detected in all CD34+ cell lines tested (Fig. lA, lanes 2). TPA and bryostatin-1 (as well as 4-P-phorbol 12,13didecanoate, in experiments not shown) induced similar hyperphosphoryla- tion of the CD34 antigen in KG1 cells (Fig. 2). In contrast, the biologically inactive 4-cu-phorbol or the solvent dimethyl sulfoxide, failed to increase CD34 phosphorylation (Fig. 2). Hyperphosphorylation was rapid and could be detected within 1 min following TPA addition (results not shown). The irrel- evant isotype-matched antibody, MOPC 21, did not immu- noprecipitate this phosphoprotein (Fig. L4). These results suggest that activated PKC may directly phosphorylate CD34.

Since PKC activators can mimic certain effects of hema- topoietic growth factors (23, 25, 26), we assessed whether hematopoietic growth factors could mediate CD34 phos- phorylation. KG1 cells are known to bind and metabolize IL- 3 and GM-CSF (27), and we would expect these growth factors to mediate a biological response in these cells. Neither IL-3, GM-CSF, G-CSF, M-CSF, nor IL-la, added at saturating concentrations for 30 min at 37 “C, altered the phosphoryla- tion status of the CD34 molecule in KG1 cells (data not shown). The lack of effect of these growth factors on CD34 phosphorylation is difficult to interpret since we cannot rule out the possibility that this is a KGl-specific phenomenon. However, neither TPA nor the hematopoietic growth factors changed the level of CD34 expression on the KG1 cell surface, as assessed by flow cytometry and Western blotting (results not shown). Also, no change in CD34 expression was observed when cells were cultured with 100 nM TPA for up to 3 days (results not shown). These results suggest that while PKC might directly phosphorylate the CD34 antigen, hyperphos- phorylation of CD34 by PKC does not regulate its level of

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11058 Protein Kinase C Phosphorylates the CD34 Antigen RPM1

A KGlo KGI 6402 MOLT 13 My10 MOPC My10 MOPC MylO MOPC MylO MOPC I 2 3 I 2 3 I 2 3 I23

ZOOKO-

II6 - 92

-a -Y -

66 -

(3,

-+- -+ + -++ -++ TPA TPA TPA TPA

6 Neuraminidase -+

200kO-

116- 92-

t 5 E

++ TPA

FIG. 1. CD34 antigen is a phosphoprotein. The CD34+ KGla and KG1 myelogenous leukemia and RPM1 8402 and MOLT 13 T- lymphoblastic leukemia cell lines were equilibrated with [32P]ortho- phosphate and incubated ?lOO nM TPA for 30 min. Cells were washed and lysed, and CD34 antigen was immunoprecipitated before SDS- PAGE. Gels were dried, and autoradiography was performed. In A: for each panel, lanes I and 2 represent untreated and TPA-treated cells, respectively, which were lysed, and then CD34 antigen was immunoprecipitated with My10 monoclonal CD34 antibody. In con- trol lane 3, the cellular lysate was immunoprecipitated with the irrelevant, isotype-matched monoclonal antibody, MOPC 21. In B: immune complexes containing3*P-labeled CD34 antigen isolated form KG1 cells were treated with neuraminidase before SDS-PAGE. The apparent molecular weight of the CD34 antigen increased with neur- aminidase treatment (from M, = 110,000 to A4, = 150,000), as is characteristic of the CD34 antigen (7,48).* The mobility of molecular weight standards is indicated at the left.

surface expression, at least in KG1 cells. PKC Directly Phosphorylates the CD34 Antigen in Vitro-

To determine whether PKC directly phosphorylates CD34, purified CD34 protein was mixed with purified rat brain PKC in a cell-free phosphorylation system (Fig. 3, lane 1) (28). The results indicate that the CD34 antigen can be phosphorylated directly by the activated PKC (Fig. 3, lane 1). CD34 phos- phorylation did not occur in the absence of calcium, phospha- tidylserine, and TPA, indicating that CD34 phosphorylation was a Ca*+/phospholipid-dependent reaction, consistent with the known requirements for PKC activation (Fig. 3, lane 4). In the absence of PKC, no CD34 phosphorylation was ob- served, indicating that CD34 does not contain intrinsic auto- kinase activity under these assay conditions (Fig. 3, lane 2). These results indicate that the CD34 antigen can be phos- phorylated directly by activated PKC. As can be seen in Fig. 3, autophosphorylation of PKC (Mr = 80,000) was detected whether or not CD34 antigen was present (lane 1 and lane 3). We also noted that purified CD34, when added at higher concentrations (up to 4 pg), could enhance the autophosphor-

200 kD- x f

116- 92-

0

66-

FIG. 2. Protein kinase C (PKC) activators induce phos- phorylation of CD34 antigen in KG1 cells. In an experiment similar to Fig. 1, cells were treated with TPA (200 nM), bryostatin-1 (200 nM), 4-cu-phorbol (200 nM), or dimethyl sulfoxide (0.2-2.0s; DMSO). Phosphorylation of the llO,OOO-dalton CD34 antigen was amplified after addition of TPA and bryostatin. The mobility of molecular mass standards is indicated at the left.

co34: + + - + PKC: + - + +

TPA,Ca*,PS: + + + - EGTA. - - - +

I234

ZOOKD- .

116-w *

FIG. 3. PKC directly phosphorylates purified CD34 antigen on a cell-free system. Highly purified rat brain protein kinase C (M, = 80,000; the mobility of PKC is indicated by the smaller arrolu) was incubated with purified CD34 antigen (M? = 110,000; the mobility of CD34 is indicated by the larger arrow; CD34 antigen was purified to a single band in SDS-PAGE/silver-stained gels) f activators and cofactors of PKC, as indicated. Proteins were resolved by SDS-PAGE, gels were dried, and autoradiography was performed. Protein kinase C activity is dependent on calcium and phosphatidylserine (PS) and is enhanced by the addition of TPA. The mobility of molecular mass standards is indicated at the left.

ylation of PKC; however, the significance of this finding is unknown (data not shown).

CD34 Antigen Is Phosphorylated on Serine Residues-Since PKC is a Ser/Thr protein kinase, we determined which amino acid(s) of the CD34 protein were phosphorylated in response to TPA. Phosphoamino acid analysis was performed on acid- hydrolyzed, 32P-labeled immunoprecipitated CD34 antigen. Serine was the only amino acid found to be phosphorylated in CD34, isolated from untreated (Fig. 4, left lane) or TPA- stimulated (middle lane) cells or from in uitro phosphorylated CD34 (right lane). These findings are compatible with hyper- phosphorylation of CD34 being mediated by activated PKC.

CD34 Antigen Is Stoichiometrically Phosphorylated-If the hyperphosphorylation of CD34 in cells is a physiologically relevant effect, the CD34 phosphorylation would be stoichio- metric. In contrast, if the observed incorporation of phosphate represented only a slight (i.e. nonstoichiometric) increase in the total phosphate incorporated per mol of CD34, then it could be reasonably argued that such a covalent modification

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Protein Kinase C Phosphorylates the CD34 Antigen 11059

KGI Cell

PO;. Q *

Cell- Free

a - PO;

P-TYR. -P-THR

(P-TYR

- + + TPA TPA

FIG. 4. Phosphoamino acid analysis of CD34 antigen. Phos- phoamino acid analysis was performed on 32P-labeled CD34 antigen, immunoprecipitated from whole cells (treated + TPA, left and middle panel), or phosphorylated in vitro (right panel), as described under “Experimental Procedures. ” 32P-labeled phosphoserine was detected, after exposure at -80 “C for 1 week. Samples were spotted at the bottom. The mobility of free phosphate (PO:-), phosphotyrosine (P- TYR), phosphothreonine (P-THR) and phosphoserine (P-SER) amino acid standards is indicated at the far left and the fur right. Incompletely hydrolyzed 3ZP-labeled peptides migrated between the point of application of the sample and the phosphotyrosine standard.

TABLE I Effect of TPA on intracellular ATP levels

Results represent the average of three separate experiments. From these values, the stoichiometry of phosphorylation of CD34 antigen in control (untreated) and TPA-stimulated KG1 cells was 0.20 and 0.99 mol of 32P incorporated/mol of CD34 antigen, respectively.

Treatment ATP

Control TPA

pmol~10-6celk 2062 f 67 1970 2 60

Ci. mol-’ 1322 + 10 1343 + 9

may have no physiological consequence. Accordingly, the stoichiometry of incorporation of “P into the CD34 molecule was determined in KG1 cells. We estimate that following TPA stimulation, at least 1 mol of [32P]ATP is incorporated for every mole of CD34 (Table I). Since only -20% of the CD34 molecules were found to contain 32P in the untreated cells, this indicates that the addition of TPA induced a substantial increase in total phosphorylation of the CD34 antigen (i.e. -&fold), which resulted in stoichiometric phos- phorylation.

PKC Mediates Phosphorylation of Distinct Sites on the CD34 Antigen-In order to determine whether activated PKC phosphorylates the same or unique serine site(s), peptide mapping studies were performed. Comparison of the site(s) of phosphorylation of CD34 was performed by two-dimensional separation of tryptic fragments of immunoprecipitated 32P- labeled CD34 antigen, which was purified from TPA-treated or untreated KG1 cells or from in vitro studies. Two new CD34 phosphopeptides were identified when KG1 cells were stimulated with TPA (peptides 1 and 2, Fig. 5, middle panel) which were not detected in untreated cells (top panel). The same two new phosphopeptides were present when purified CD34 antigen was phosphorylated directly by activated PKC in a cell-free system (Fig. 5, bottom panel). Furthermore, several small CD34 phosphoserine-containing peptides were identified in untreated KG1 cells (Fig. 5, top panel, arrow- heads). The amplitude of phosphorylation of these smaller phosphopeptides was not altered by incubation of the cells

Electrophoresis

I

FIG. 5. Comparative peptide mapping of phosphorylated CD34 antigen. Peptides from TPCK-treated “P-labeled immuno- precipitated CD34 antigen, from whole cells (treated ? TPA; top and middle panel) or from in vitro phosphorylated CD34 antigen (lower panel), were separated by electrophoresis and then by ascending chromatography as indicated. Peptides labeled I, 2, and possibly 3 (cross-hatched in the extreme lower panel) represent PKC serine phosphorylation site(s). Phosphopeptides marked with arrowhead (open symbols in the extreme lower panel) may represent non-PKC phosphorylation site(s).

with TPA (middle panel, arrowheads), indicating that they may not be sites of PKC phosphorylation.

DISCUSSION

The CD34 antigen is a developmentally restricted lymphoh- ematopoietic progenitor cell surface molecule (l-4). While little is known about the regulation of hematopoietic cell growth/differentiation, evidence suggests that this may in- volve the biochemical transduction of cellular signals, me- diated at least in part by protein kinase C. For example, activators of PKC can induce differentiation of human leu- kemic HL60 and KG1 cells (29, 30). Furthermore, the PKC activators TPA and bryostatin-1 stimulate growth of human bone marrow progenitor cells (22, 23, 31), while PKC inhibi- tors (e.g. staurosporine or H-7) block proliferation of both these normal cells (9,10) and HL60 leukemia cells (32). Since activated PKC can regulate hematopoietic progenitor cells, and since CD34 is a progenitor cell-specific molecule, we were interested in whether CD34 is phosphorylated and if it is a physiologic substrate for PKC.

Our findings raise the possibility that CD34 antigen phos- phorylation may be important for signal transduction in hu- man lymphohematopoietic progenitor cells. Although earlier studies failed to demonstrate that the CD34 structure was phosphorylated in uiuo (7), the more recent availability of higher affinity antibodies led us to retest this possibility. We found that the CD34 antigen is a phosphoprotein and repre- sents a membrane target for activated PKC. Based on two- dimensional phosphopeptide maps, we found that hyperphos- phorylation of CD34 is directly mediated by PKC (Fig. 5). Activated PKC selectively phosphorylates at least one unique

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11060 Protein Kinme C Phosphorylates the CD34 Antigen

site in the CD34 molecule which is not identified in growing cells (e.g. in the steady state). Furthermore, PKC activation is stoichiometric and results in a -5-fold increase in 32P incorporation into CD34, and, thus, phosphorylation of CD34 may represent a physiologically significant post-translational modification.

Analysis of the human CD34 cDNA sequence reveals that the protein appears to be a type I transmembrane molecule.3 The predicted internal portion of the protein appears to contain basic amino acid residues adjacent to Ser residues which represent at least two potential target sites for PKC phosphorylation (Arg x Ser and Ser x Arg) (33, 34; serine protein kinase target motifs reviewed by Traugh (35)). Fur- thermore, the serine from the latter of the two PKC sites may also serve as the distal phosphoserine in a consensus motif potentially recognized by glycogen synthase kinase-3 (Fa) [Ser(X)., Ser(P)] (36). In addition, there are two other con- sensus motifs that correspond to potential target sites for a Ca*‘/calmodulin-dependent kinase and/or protease-activated kinase I (Arg X X Ser) (37,38). Interestingly, in exponentially growing KG1 cells there are additional phosphopeptides de- tected by two-dimensional phosphopeptide mapping which are not hyperphosphorylated following activation of PKC (Fig. 5). Thus, it is possible that these may be additional target sites in the CD34 molecule for one or more serine protein kinases other than PKC. This possibility is currently under investigation.

The functional significance of phosphorylation of the CD34 molecule is as yet unknown, but it might be comparable to the role that PKC plays in the activation of certain other molecules. PKC is known to phosphorylate leukocyte-associ- ated differentiation antigens (12). PKC-mediated phosphoryl- ation of certain of these surface proteins induces or is asso- ciated with surface receptor modulation, either down-regula- tion (e.g. CD4, CDB-r, CD71 (transferrin receptor), and the M-CSF receptor (c-fms)) or up-regulation (e.g. fibronectin receptor (39)). However, CD34 phosphorylation is apparently not associated with altered levels of surface expression. On the other hand, functional activation of surface proteins as a consequence of PKC-mediated phosphorylation may occur in the absence of surface modulation. This is apparently the case for CD8, which is linked to signal transduction via the CD2 and the CD3-Ti pathways (40-43). Moreover, the sialophorin/ leukosialin glycoprotein (CD43), another lymphohemato- poietic cell-specific molecule, shares many biochemical/struc- tural properties with the CD34 antigen (44,45), and can even be phosphorylated by activators of PKC (12). While the physiological effect of CD43 phosphorylation is not known, it has been suggested that CD43 may be important for monocyte activation and possibly cellular adhesion (46,47).

Thus, like the antigens described above, the CD34 molecule may transduce signal(s) which ultimately lead to cellular activation events including proliferation, differentiation, and/ or cellular adhesion, even in the absence of surface modula- tion. While highly speculative, in the case of cellular adhesion, it is possible that phosphorylation of CD34 may induce con- formational changes which modify its function.

Since CD34 is uniquely distributed on lymphohemato- poietic progenitor cells and is phosphorylated by PKC, the CD34 antigen is a candidate molecule for participation in signal transduction in lymphohematopoietic progenitor cells of all lineages.

Acknowledgments-We are grateful to Dr. Thomas Zimmerman (Burroughs-Wellcome Co., Research Triangle Park, NC) for gener-

3 D. B. Seed, personal communication.

ously performing the intracellular measurements for ATP and deter- mining the specific activity of [32P]ATP. We wish to thank Dr. Brian Seed (Massachusetts General Hospital, Boston, MA) for providing the human CD34 cDNA sequence prior to publication, and Dr. M. B. Kastan (Johns Hopkins Oncology Center) for helpful scientific dis- cussions.

REFERENCES

1. Civin, C. I., Strauss, L. C., Brovall, C., Fackler, M. J., Schwartz, J. F., and Shaper, J. H. (1984) J. Zmmunol. 133, 157-165

2. Katz, F. E., Tindle, R., Sutherland, D. R., and Greaves, M. F. (1985) Leuk. Res. 9, 191-198

3. Andrews, R. G., Singer, J. W., and Bernstein, I. D. (1986) Blood 67,842-845

4. Civin, C. I., Trischmann, T. M., Fackler, M. J., Bernstein, I. D., Buehring, H. J., Campos, L., Greaves, M. F., Kamoun, M., Kratz, D. R., Lansdorp, P. M., Look, A. T., Seed, B., Suther- land, D. R., Tindle, R. W., and Uchanska-Ziegler, B. (1989) in Leukocyte Typing IV (Knapp, W., Diirken, B., Gilks, W. R., Rieber, E. P., Stein, H., Schmidt, R. E., von den Borne, A. E. G., eds) pp. 818-825, Oxford University Press, Oxford, 1989

5. Beschorner, W. E., Civin, C. I., and Strauss, L. C. (1985) Am. J. Pathol. 119, l-4

6. Watt, S. M., Karhi, K., Gatter, K., Furley, A. J. W., Katz, F. E., Healy, L. E., Altass, L. J., Bradley, N. J., Sutherland, D. R., Levinsky, R., and Greaves, M. F. (1987) Leukemia 1,417-426

7. Sutherland, D. R., Watt, S. M., Dowden, G., Karhi, K., Baker, M. A., Greaves, M. F., and Smart, J. E. (1988) Leukemia 2, 793-803

8. Nishizuka, Y. (1988) Nature 334, 661-665 9. Katayama, N., Nishikawa, M., Minami, N., and Shirakawa, S.

(1989) Blood 73,123-130 10. Jenis, D. M., Johnson, C. S., and Furmanski, P. (1989) Znt. J.

Cell Cloning 7, 190-202 11. Livneh, E., Dull, T. J., Berent, E., Prywes, R., Ullrich, A., and

Schlessinger, J. (1988) Mol. Cell. Biol. 8, 2302-2308 12. Chatila, T. A., and Geha, R. S. (1988) J. Zmmunol. 140, 4308-

4314 13. Hunter, T. (1989) Cell 58, 1013-1016 14. Koeffler, H. P., and Golde, D. W. (1980) Blood 56,344-350 15. Koeffler, H. P., Yen, J., and Lowe, L. (1981) Cancer 48, 1958-

1965 16. Huang, C. C., Hou, Y., Woods, L. K., Moore, G. E., and Mino-

wada, J. (1974) J. Natl. Cancer Inst. 63,655-660 17. Laemmli, U. K. (1970) Nature 227,680-685 18. May, W. S., Sahyoun, N., Jacobs, S., Wolf, M., and Cuatrecasas,

P. (1985) J. Bial. Chem. 260,9419-9426 19. Zimmerman, T. P., Ridout, J. L., Wolberg, G., Duncan, G. S.,

and Ellion, G. B. (1977) J. Biol..Ckem. 2&1,6757-6766 20. Wolf. M.. Sahvourn. N.. Le Vine. H.. III. and Cuatrecasas. P.

(1984) &o&m. B&hys. Res. Com&un.‘122, 1268-1275 21. Cooper, J. A., Sefton, B. M., and Hunter, T. (1983) Methods

Enzymol. 99,387-402 22. Griffin, J. D., Beveridge, R. P., and Schlossman, S. F. (1983)

Leuk. Res. 7,43-49 23. May, W. S., Sharkis, S. J., Esa, A. H., Gebbia, V., Kraft, A. S.,

Pettit, G. R., and Sensenbrenner, L. L. (1987) Proc. Natl. Acad. Sci. U. S. A. 64,8483-8487

24. Fields, A. P., Pincus, A. M., Kraft, A. S., and May, W. S. (1989) J. Biol. Chem. 264,21896-21901

25. Whetton, A. D., Valiance, S. J., Monk, P. N., Cragoe, E. J., Dexter, M. T., and Heyworth, C. M. (1988) B&hem. J. 266, 585-592

26. Stuart, R. K., and Hamilton, J. A. (1980) Science 208,402-404 27. Gesner, T. G., Mufson, R. A., Norton, C. R., Turner, K. J., Yang,

Y.-C., and Clark, S. C. (1988) J. Cell. Physiol. 136, 493-499 28. May, W. S., and Tyler, G. (1987) J. Biol. Chem. 262, 16710-

16718 29. Ferrero, D., Pessano, S., Pagliardi, G. L., and Rovera, G. (1983)

Blood 61.171-179 30. Kiss, Z., Deli, E., Shoji, M., Koeffler, P., Pettit, G. R., Vogler, W.

R., and Kuo, J. F. (1987) Cancer Res. 47, 1302-1307 31. Abrahm, J. L., and Smiley, R. (1981) Blood 68,1119-1126 32. Nishikawa, M., Uemura, Y., Hidaka, H., and Shirakawa, S. (1986)

Life Sci. 39,1101-1107 33. House, C., Wettenhall, R. E. H., and Kemp, B. E. (1987) J. Biol.

&em. 262.772-777

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Protein Kinase C Phosphorylutes the CD34 Antigen 11061

34. Woodgett, J. R., Gould, K. L., and Hunter, T. (1986) Eur. J. 42. Rudd, C. E., Trevillyan, J. M., Dasgupta, J. D., Wong, L. L., and Biochem. 161, 177-184 Schlossman, S. F. (1988) Proc. Natl. Acad. Sci. U. 5’. A. 85,

35. Traugh, J. A. (1989) in Aduances in Regulation of Cell Grozuth 5190-5194 (Mend, J. J., Cambier, J. C., and Weiss, A., eds) Vol. 1, pp. 43. Perlmutter, R. M., Marth, J. D., Ziegler, S. F., Garvin, A. M., 173-202, Raven Press, New York Pawar, S., Cooke, M. P., and Abraham, K. M. (1988) Biochim.

36. Fiol, C. J., Mahrenholz, A. M., Wang, Y., Roeske, R. W., and Biophys. Acta. 948, 245-262 Roach, P. J. (1987) J. Biol. Chem. 262,14042-14048 44. Remold-O’Donnell, E., Kenney, D. M., Parkman, R., Cairns, L.,

37. Leis, J., Johnson, S., Collins, L. S., and Traugh, J. A. (1984) J. Savage, B., and Rosen, F. S. (1984) J. Exp. Med. 159, 1705- Biol. Chem. 259,7726-7732 1723

38. Tuazon, P. T., and Traugh, J. A. (1984) J. Biol. Chem. 259,541- 45. Carlsson, S. R., and Fukuda, M. (1986) J. Biol. Chem. 261, 546 12779-12786

39. Symington, B. E., Symington, F. W., and Rohrschneider, L. R. 46. Nong, Y. H., Remold-O’Donnell, E., LeBien, T. W., and Remold, (1989) J. Biol. Chem. 264, 13258-13266 H. G. (1989) J. Exp. Med. 170,259-267

40. Blue, M-L., Hafler, D. A., Craig, K. A., Levine, H., and Schloss- 47. Silverman, L. B., Wong, R. C. K., Remold-O’Donnell, E., Vercelli, man, S. F. (1987) J. Immunol. 139,3949-3954 D., Sancho, J., Terhorst, C., Rosen, F., Geha, R., and Chatila,

41. Veillette, A., Bookman, M. A., Horak, E. M., and Bolen, J. B. T. (1989) J. Zmmunol. 142,4194-4200 (1988) Cell 55, 301-308 48. Fackler, M. J., and Civin, C. I. (1988) Exp. Hematol. 16, 336a

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

M J Fackler, C I Civin, D R Sutherland, M A Baker and W S Mayhematopoietic cells.

Activated protein kinase C directly phosphorylates the CD34 antigen on

1990, 265:11056-11061.J. Biol. Chem. 

  http://www.jbc.org/content/265/19/11056Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/265/19/11056.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from