Characterization and partial purification of CD34 ...

Vol. 47, No. 1, January 1999

BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL pages 9-23

C H A R A C T E R I Z A T I O N A N D PARTIAL P U R I F I C A T I O N OF C D 3 4 + P R O G E N I T O R CELL ECT O PHOSPHATIDIC ACID PHOSPHOHYDROLASE Kevin A. Harvey 1, Rafat A. Siddiqui L2, Margaret Reeves l, Thomas Kovala I, Michael Dugan 1, Luke P. Akard 1, and Denis English 1

1Experimental Cell Research Laboratoo1, Methodist Research Institute, Indianapolis, I N USA; 2Department of Biology, Indiana University-Purdue University, Indianapolis, IN USA Corresponding author: Denis English, Methodist Research Institute, 1701 N. Senate Ave., Room MPC, 1417, Indianapolis, IN USA 46202 Tel.: 317-929-2663; e-mail [email protected] Received July 6, 1998. Accepted July 15, 1998. Abbreviations used in this manuscript include: diC8 PA, diotanoyl phosphatidic acid; PAPase, phosphatidic acid phosphohydrolase; SDS, sodium dodecyl sulfate; PBS, phosphate buffered saline; BSA, bovine serum albumin; and PE, phycoerythrin.

Summary Phosphatidic acid and its hydrolysis product, diacylglycerol, play potentially vital roles as extracellular messengers in numerous cellular systems and may play a key role in regulating hematopoiesis. In this study, we describe an ecto-phosphatidic acid phosphohydrolase that potentially regulates cellular responses to phosphatidic acid on bone marrow derived human hematopoietic progenitors. We partially purified hematopoietic progenitor ecto-PAPase using a novel in-gel phosphatase assay and then characterized the enzyme on phenotypically defined subpopulations of hematopoietic CD34 + progenitors isolated by flow cytometry. The most pronounced PAPase activity was confined to uncommitted CD34+/CD38 + hematopoietic progenitors, which lacked the expression of other lineage-associated antigens. We conclude that hematopoietic progenitor cells at various stages of maturation possess a potent ecto-PAPase, an enzyme well positioned to regulate progenitor cell growth and differentiation induced by phosphatidic acid and related lipids. Key words'- phosphatidic acid phosphohydrolase, ecto-PAPase, hematopoietic progenitor cell, CD34 + cells, flow cytometry. Introduction Derived from the activation of phospholipase D and other intracellular enzymes, phosphatidic acid (PA) potentially plays a major role as a messenger of extracellular information (1,2). Studies with neutrophillic leukocytes have implicated exogenous PA as an activator of

9

Copyright 9 1999 IUBMB 1039-9712/99 $12,00 + 0.00


BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL

several inflammatory responses including Ca ++ mobilization (3), superoxide generation (4-6), actin polymerization, and chemotaxis (7).

Phosphatidic acid phosphohydrolase (PAPase)

modulates the role of PA and generates diacyiglycerol, an important intracellular messenger (8). Diacylglycerol is a known activator of superoxide generation in neutrophils which exerts its effects by activating protein kinase C (5,9). Although the exact role of PAPase is not entirely known, it is clear that this enzyme plays a vital role as a modulator in a variety of signal transduction cascades (10). A novel ecto-PAPase has been identified on the outer surface of neutrophillic leukocytes (5,11,12). Ecto-PAPase presumably regulates cellular responses to extracellular PA. Potential sources of extracellular PA include inflammatory exudates, where it may be secreted by stimulated cells,

the outer plasma membrane of stimulated stromal cells and

membrane

fragments released from disrupted, stimulated cells. Various isozymes of PAPase have been purified and cloned.

Kanoh et al. initially

identified an 83kDa PAPase isolated from porcine thymus; however, these investigators subsequently reported that a 35kDa protein and not the 83kDa protein accounted for the PAPase activity within the purified preparation (13,14). potypeptides of 34 and 54kDa

Fleming and Yeaman

identified PAPase in

recovered after separation of cellular extracts (15).

Two

isozymes of N-ethylmaleimide-insensitive PAPase were isolated from rat liver by Waggoner et al. Enzymatic activity was recovered in both an anionic and cationic form of the enzyme, which

possessed apparent molecular weights of 53 and 51kDa, respectively. When treated with Nglycanase F, both of these enzymes were reduced to a single band of 28kDa as determined by electrophoresis (16).

Finally, Siess and Hofstetter found a similar form of PAPase with a

molecular weight of 34kDa using a novel antibody-bridge to assess distribution of the enzyme in liver extracts separated in SDS polyacrylamide gels (17). The present study was undertaken to identify and characterize ecto-PAPase in CD34 + bone marrow progenitor cells and relate its distribution in terminally differentiated hematopoietic cells. The CD34 antigen is expressed on immature, pluripotent hematopoietic stem cell as well as on more mature progenitor cells which co-express various lineage-associated antigens. Various subpopulations of maturing CD34 + progenitors are defined by their co-expression of lineage-associated antigens such as CDI0 +, CD33 +, and glycophorin A § which are committed to lymphoid, myeloid, and erythroid differentiation, respectively (18-21). CD34 + progenitors that

10



co-express CD38 but not other lineage-associated antigens are relatively immature but are not pluripotent hematopoietic stem cells. In this study, we examined the presence of ecto-PAPase on various subpopulations of phenotypically defined CD34 + progenitors isolated from human bone marrow. To our knowledge, this is the first attempt to identify and characterize this enzyme within hematopoietic progenitors.

Our results identify the presence of a 51-53kDa N-

ethyhnaleimide-insensitive ecto-PAPase on immature CD34 +bone marrow progenitor cells.

Methods Materials Disposable "I" type bone marrow aspiration needles were purchased from Manan Medical Products, Inc. (Northbrook, IL). Monoclonal antibodies were obtained from Becton Dickinson Immunocytometry Systems (San Jose, CA). Ceprate LC column CD34 + cell separation kits were purchased from Cellpro, Inc. (Bothetl, WA). Phosphatidylethanolamine, 1,2 dioctanoylglycerol, and diC8 PA were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL). Diacylglycerol kinase was obtained from Calbiochem (La Jolla, CA). [32p]-labeled ATP was from NEN Life Science Products (Boston, MA). Macrosolute Minicon Concentrators were from Amicon, Inc. (Beverly, MA). Precast tris-glycine polyacrylamide gels were purchased from Owl Scientific, Inc. (Woburn, MA) and silver staining kits were from Bio-Rad Laboratories (Richmond, CA). A non-hydrolyzable PA analogue, phosphonate-1 (5), was a gift of Dr. Theodore Widlanski of the Department of Chemistry, Indiana University, Bloomington, IN. All other reagents and supplies were purchased from Sigma Chemical Co., St. Louis, MO. Enrichment o f bone marrow mononuclear leukocytes Bone marrow samples were collected from consenting healthy adults in accordance with institutional guidelines. Samples of bone marrow were aspirated from the posterior iliac crest by means of a single percutaneous intrusion in conjunction with repositioning the disposable ' T ' type aspiration needle two or three times. Between 50-75 cc of bone marrow was collected into heparinized 60 cc syringes. The marrow was diluted 1:3 in Hanks' balanced salt solution containing phenol red but lacking CaCI2 and MgCI2. Diluted samples were layered over 10 ml cushions of ficoll-Hypaque (density = 1.077) in 50 ml polypropylene conical tubes and centrifuged for 15 minutes at 340 X g. The mononuclear leukocyte layer was recovered from the interface and washed twice using phosphate-buffered saline (PBS) containing I% bovine serum albumin (BSA). The recovered cells were resuspended to 1 x 108 cells/ml in the wash buffer. CD34 +progenitor cell isolation CD34 + progenitor cells were enriched from recovered bone marrow mononuclear leukocytes using Ceprate LC column cell separation kits. Samples were treated and isolated according to the manufacturer's instructions. This system employs biotinylated CD34 antibody and an avidin column. Biotin labeled CD34 + cells adhered to the avidin column, while the majority of the CD34- cells were removed in the flow through portion. The CD34 + cells were dislodged by physical agitation and eluted from the column into PBS containing 1% BSA.

11



Fluorescent-activated cell sorting of CD34 +progenitors Fluorescent-activated cell sorting was perfomaed using a Becton Dickinson FACStar PLUs flow cytometer equipped with a water-cooled argon laser emitting at a wavelength of 488 nm as previously described (20). Prior to sorting, enriched CD34 + progenitor cells obtained from the Ceprate LC column were concentrated by centrifugation into 1 ml of PBS containing 1% BSA. The CD34 + progenitor cells were labeled with PE-conjugated CD34 (8G12) antibody by incubation with 20 lal of antibody for fifteen minutes at 4~ in the dark. Subsequently, the cells were washed once and resuspended in PBS containing 1% BSA. A relatively low forward scatter threshold was established tO detect cellular targets, while ensuring that debris and electronic noise would not be ascertained as legitimate events. In every sort performed, the cell population of interest was gated based on light scatter and fluorescence. Cells were sorted at a rate of 1,000-1,500 events/second. Two different sorting approaches were used for these experiments. In most cases, all CD34 + cells were sorted directly into one 12 x 75 mm glass tube. When sorting subpopulations of CD34 + progenitors, various other antibodies were used to define progenitor cell subsets as determined by their phenotype as previously described (20,21). Erythroid committed progenitor cells were labeled with fluorescein-conjugated antibodies to the glycophorin A antigen. Myeloid CD34 ~ progenitors were labeled with fluorescein-conjugated anti-CD33 and lymphoid committed progenitors were obtained by labeling with fluorescein-conjugated anti-CD10. In some experiments, neutrophils, lymphocytes, and erythrocytes were sorted and labeled with phycoerythrin or fluorescein-conjugated antibodies to C D l l b , CD3, and glycophorin A, respectively. When sorting CD34+ subpopulations, 5,000 gated events were collected per tube using a 10raM HEPES buffered saline sorting solution (pH 7.2). For every progenitor cell sort, a test sort was performed to ensure that < 98.5% homogenous population was harvested.

Synthesis q/[r

phosphatidic acid

Into each of eight 12 x 75 mm glass test tubes, 10 l-tl of 1,2 dioctanoyl glycerol (1 mM stock in CHCI3) and 200 ~,tl of phosphatidylethanolamine/cardiolipin (1 mM each in CHCI3: MeOH, 1:1) were combined and dried under nitrogen gas. The lipids were subsequently sonicated into buffer containing 100 mM NaCI, 50 mM Tris base (pH 6.6), 10 mM MgC12, 1 mM 2-mercaptoethanol, and 0.7% Triton X-100. Following the addition of 10 ~tl of lmg/ml diacylglycerol kinase and a five minute pre-incubation at 37~ 500 Ci of [32p]-ATP (specific activity = 6,000 Ci/mmol) was placed into each tube and incubated at 37~ for two hours. Subsequently, the reaction was tenaainated with the addition of 2ml of CHCI3/MeOH/HCI (1:2:0.03). Phases were resolved by the addition of 500 #l of CHC13 and 500 ~1 of H20 and the CHCI3 layers were collected and washed twice with H20 to remove contaminating [32p]-labeled ATP. The CHC13 was dried under nitrogen gas and the lipids were resuspended in spotting solution (CHCI3/MeOH, 90:10). In order to isolate the radiolabelled PA, lipid suspensions were spotted onto a thin-layer chromatography (TLC) plate and separated using a CHCI3/MeOH/20% methylamine (60:35:10) solvent system. The dried TLC plate was then autoradiographed for visualization of [32 P]-labeled phosphatidic acid. The [32.P]-PA was isolated by scraping the targeted lipid from the plate and eluting the radiolabelled lipid from the silica using methanol. A 0.2 ~tm filter disc was attached to a 20cc syringe for the elution process, thus enabling the [32p]_ PA to be collected into a glass tube while removing the silica gel particles. The methanol eluant was dried under nitrogen gas and resuspended into 5ml of CHC13/MeOH/HC1 (1:2:0.03) by

12



vortexing and sonication. [32p]-PA was back extracted using an equal volume of CHC13 and H20 to ensure that any contaminates had been removed. The CHCI3 layers were dried under nitrogen gas, and stored at -20~ until further use.

Ecto-phosphatidic acid phosphohydrolase assay Sorted intact CD34 + progenitor cells were assayed for ecto-PAPase activity by incubation with [32p]-labeled phosphatidic acid using modifications to the method previously used with neutrophils (12). Hydrolysis of the [32p]-labeled phosphatidic acid was quantitated by the release of [32p] from the lipid substrate into the aqueous solution. The reaction was typically carried out in a volume of 100 ~tl, which consisted of 80 gl of ecto-PAPase assay buffer (1 mM EGTA in 10 mM HEPES buffered saline, pH 7.0), 10 ~1 of [32P]-labeled phosphatidic acid containing 0.01 mg/ml of carrier phosphatidic acid, and 10/al containing 2.5 X 10 4 intact cells (unless otherwise noted). Reactions were carried out at 37~ for various lengths of time as described within the figure legends. The reactions were stopped by the addition of 500 ~tl of CHC13/MeOH/HCI (1:2:0.03). An equal volume of CHCI3 and diH20 were added in order to separate the labeled PA from the hydrolyzed [3Zp]. Samples were vortexed followed by centrifugation at 450 X g for five minutes to separate the organic and aqueous phases. The aqueous phase was aspirated from each sample and its radiation assayed in a scintillation counter. Modification of this assay was necessary when determining ecto-PAPase activity in sorted subpopulations of CD34 + progenitors due to the limited number of cells available. In these experiments, each assay tube contained 5,000 cells in approximately 85 ~tl of assay buffer. In order to increase sensitivity, 50 ~tl of 130 mM sodium citrate was added to reduce the pH to 6.6, thereby enhancing enzyme activity (12). The same amount of substrate was utilized as describes above. Further modification was needed for the in-gel PAPase assay, which was undertaken to determine the molecular weight and purity of the enzyme in resolved extracts of CD34 + cell subpopulations as described below. In these assays, gel slices were physically disrupted into 150 ~tl of H20 by pulverization and sonication (12). For each gel slice, 125 ~tl of 130 mM sodium citrate was added in order to reduce the pH to 6.0. The reactions were initiated with the addition of carrier-free [32p]-labeled PA and were incubated at 37~ for four hours. The extraction and quantification processes were performed as described in the text above.

Partial purification of ecto-PAPase Ecto-PAPase was liberated by incubating hematopoietic cell subpopulations with low concentrations of a non-hydrolyzable analog of phosphatidic acid as described previously in our study with neutrophillic leukocytes (12). Briefly, isolated CD34 + progenitor cells were treated with 25 gg/ml of a non-hydrolyzable phosphatidic acid analogue for thirty minutes at 37~ The treated progenitors were centrifuged at 450 x g for ten minutes. Following centrifugation, the supernatant was combined and concentrated approximately twenty-five fold using a Macrosolute Minicon concentrator (MW cutoff = 15,000). The recovered protein concentrate was diluted 1:1 in 2X SDS sample buffer and separated on two precast 10 cm 12% Tris-glycine polyacrylamide gels. One gel was stained for protein visualization using a BioRad silver staining kit. The proteins within the second gel were denatured followed by renaturation in order to generate enzymatically active proteins (22). The gel was submerged in 50 mM tris-HC1 (pH 8.0) containing 20% 2-propanol for one hour followed by an additional hour in 50 mM tris-HC1 containing 5 mM 2-mercaptoethanol. Following this equilibration period, the gel was placed into a 6 M guanidine-HCl solution, which was changed once over a one hour period. Finally, the

13

Vol. 47, No. 1, January-1999


gel was placed at 4~ in renaturation buffer consisting of 50 mM Tris-HC1 (pH 8.0) containing 5 mM 2-mercaptoethanol and 0.04% Tween 40. The gel was immersed in this buffer for no less than sixteen hours with at least four buffer changes. After renaturation, gels were sliced and sections pulverized and used for assaying PAPase activity. Results are expressed as counts/minute above background released from each gel slice recovered. Results

Characteristics o f ecto-PAPase activity in CD34 +progenitor cells

Various numbers of CD34 + progenitor cells isolated by flow cytometry were assessed for PA hydrolysis by incubation with PA for various lengths of time up to two hours. Figure 1 demonstrates the linear relationship of ecto-PAPase activity over time. Additionally, figure 1 portrays the sensitivity of the ecto-PAPase assay by showing detectable [3Zp] release effected by a restricted number of flow-isolated progenitor cells. Due to the relatively low number of sorted CD34 + progenitor cells that can be obtained from a given bone marrow collection (