Multiple molecular forms of endothelial cell factor VIII related antigen

361

Biochimica et Biophysica Acta, 667 (1981) 361--369 © Elsevier/North-Holland Biomedical Press

BBA 38611

MULTIPLE MOLECULAR FORMS OF ENDOTHELIAL CELL FACTOR VIII RELATED ANTIGEN RALPH L. NACHMAN *, ERIC A. JAFFE and BARBARA FERRIS

Division of Hematology, Department of Medicine, Cornell University Medical College, 525 East 68th Street, New York, N Y 10021 (U.S.A.) (Received June 10th, 1980) (Revised manuscript received September 16th, 1980)

Key words: Factor VIII related antigen; Molecular heterogeneity; Peptide mapping; (Human endothelial cell)

Summary Heterogeneous factor VIII related antigen isolated from endothelial cen postculture medium was characterized. On crossed immunoelectrophoresis, slow moving less anodal populations of factor VIII related antigen molecules were more prominent in endothelial cell postculture medium than in plasma. The protein synthesized by endothelial cells appeared as two discrete protein bands of different molecular weight in sodium dodecyl sulfate polyacrylamide gels. In contrast the factor VIH related antigen isolated from plasma moved as a single protein band in an identical gel system. The factor VIII related antigen from endothelial cell postculture medium was immunoisolated using monospecific rabbit antibody to normal human plasma factor VIII related antigen, electrophoresed on sodium dodecyl sulfate polyacrylamide gels, radiolabeled with l~sI, trypsinized and subjected to peptide mapping using two
* To w h o m correspondence s h o u l d be addressed. Abbreviation: AGN, factor VIII related antigen; Hepes, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid; PMSF, p h c n y l m e t h a n e s u l f o n y l fluoride.

362

Introduction Factor VIII related antigen is a heterogeneous multimeric protein [1,2] which circulates in plasma. Previous studies have shown that both cultured endothelial cells and megakaryocytes synthesize and contain the protein [3,4]. Factor VIII related antigen is also present in platelets [5] and may differ structurally from plasma factor VIII related antigen [6]. The protein is decreased in the plasma of patients with typical yon Willebrand's disease and is present in normal or increased amounts in the plasma of patients with classical hemophilia. Von Willebrand factor activity which is necessary for normal platelet function and is measured in vivo by the bleeding time and in vitro by ristocetin induced platelet agglutination is the functional attribute of the VIII related antigen system [7] and is also synthesized b y endothelial cells [8,9]. In order to define more precisely the nature of the molecular heterogeneity within the VIII related antigen system we have analyzed the structure of different molecular weight forms released by human endothelial cells in culture. Materials and Methods A n tisera

Rabbit antihuman factor VIII related antigen obtained from Behring Diagnostics (American Hoechst Corp, Somerville, NJ) was used to purify normal plasma factor VIII related antigen which in turn was used as an immunogen. The commercial antiserum was reacted against human factor VIII concentrate (Hemofil, Hyland Division, Costa Mesa, CA) in 1% agarose at room temperature for 72 h. The slides were washed for 96 h at room temperature in phosphatebuffered saline (0.145 M NaC1, 0.01 M phosphate, pH 7.4) and the immunoprecipitin lines cut out with a razor blade. Material from 20 sets of immunodiffusion lines were pooled, lyophilized, and homogenized [4]. Rabbits were immunized as previously described [4]. The resulting antisera were absorbed with 1/10 vol. aluminium hydroxide gel and heated to 56°C for 30 rain to remove residual coagulation proteins. The antisera were absorbed with insolubilized plasma [10] obtained from a patient with severe von Willebrand's disease (approx. 1% VIII related antigen). The antibody precipitated VIII related antigen in normal and hemophilic plasma, inhibited factor VIII coagulant activity and inhibited ristocetin-induced platelet agglutination. The antisera produced one immunoprecipitin arc on immunoelectrophoresis against human plasma and demonstrated a reaction of identity on double immunodiffusion analysis against human plasma and purified plasma VIII related antigen. The antibody did not react on immunodiffusion with purified human fibronectin [11] and did n o t stain cultured human fibroblasts by immunofluorescence. Another independently raised monospecific rabbit antihuman VIII related antigen was kindly provided by Dr. Connie Miller (Cornell University Medical College). This antiserum raised by immunizing rabbits with purified plasma factor VIII related antigen was absorbed with a 3--8% ethanol precipitate from normal human plasma.

363

Immunoisolation of factor VIII related antigen from endothelial cell postculture medium Human endothelial cells were derived from umbilical cords and cultured using methods and materials previously described [12]. The endothelial cells were cultured in medium 199 (Flow Laboratories, Inc., Rockville, MD) contalning 20% normal rabbit serum (Pel-Freeze Farms, Inc., Rogers, AR). The media also contained 100 U/ml penicillin, 100 ~g/ml streptomycin, 2 mM L-glutamine (all Flow Laboratories) and 15 mM Hepes buffer, pH 7.4, Sigma). Six days after passage, when the cells had become confluent, the culture medium was removed and the ceils washed 3-times with buffer (140 mM NaC1, 4 mM KC1, 10 mM Hepes, pH 7.5, 10 mM glucose). The cells were then incubated for 30 min at 37°C with Medium 199 containing 10% tryptose phosphate broth (Flow), 50 units/ml Trasylol, (FBS Pharmaceuticals, New York, NY), and 1 pM pepstatin (Protein Foundation, Osaka, Japan). This medium was removed and the cells were then cultured for 18 h in Medium 199 containing 10% tryptose phosphate broth and 50 units/ml Trasylol. The post~ulture medium was removed and the following additional inhibitors added to yield the listed final concentrations: 0.4 mM phenylmethanesulfonyl fluoride (PMSF), 5 mM N~thylmaleimide (Sigma), 10 mM EDTA, 1/~M pepstatin, and 0.02% sodium azide. In some experiments, the N-ethylrnaleimide was omitted. The post~ulture medium was centrifuged at 40 000 × g for 15 min at 20°C and the factor VIII related antigen isolated by the m e t h o d of Kessler [13]. The postculture medium (75 ml) was first pre-absorbed with 200 pl packed, prewashed Pansorbin (fixed Staphylococcus aureus Cowan I strain, Calbiochem-Behring Corp, San Diego, CA) for 1 h at 20°C. The bacteria were removed by centrifugation; 100 #l anti-factor VIII related antigen was added, and the post-culture medium incubated for 1 h at 20°C. Pansorbin (100 #1 packed bacteria) was then added and the mixture incubated for 1 h at 20°C. The bacteria were washed 8-times with NET-Triton [13] (until A2a0 of the wash reached background) and then eluted by boiling for 3 min in 2% SDS~ M urea containing the protease inhibitors. The bacteria were pelleted and the supernatant removed and frozen. Plasma factor VIII related antigen was immunoisolated using the same technique. The immunoisolation of all-labeled endothelial cell factor VIII: AGN was cartied out using [aH ] leucine-labeled (NH,)2SO4-precipitated endothelial ceil postculture medium prepared as previously described [ 11]. The immunoisolate was electrophoresed on 3% polyacrylamide~).5% agarose gels. The gels were sliced and counted as previously described [11]. Crossed immunoelectrophoresis was performed according to the method of Zimmerman et al. [2]. The.second dimension agarose contained 0.1% antifactor VIII related antigen. Two-dimensional immunoelectrophoresis was performed by the method of Converse and Papermaster [14]. In the first dimension, the sample was electrophoresed unreduced in SDS-3% polyacrylamide-0.5% agarose gels. The gel was sliced longitudinally and a section of the gel then electrophoresed through a layer of agarose containing 1.5% Lubrol PX into a second layer of agarose containing 0.15% antifactor VIII related antigen. SDS-polyacrylamide gel electrophoresis was performed using 3% polyacryl-

364

amide-0.5% agarose gels by the method of Weinstein [15]. For reduction, proteins were incubated with 14 mM dithiothreitol. Plasma factor VIII related antigen was purified from normal plasma as previously described [5]. Plasma fibronectin was purified as previously described [11 ]. 12sI Labeling and peptide mapping Radioiodination of fixed and stained proteins within SDS gel slices followed by tryptic digestion and peptide mapping was performed by the method described by Elder et al. [16]. Cellulose TLC plates [EM Laboratories, Elmsford, NY] (20-X 20 cm) were used. Quantitative yon Willebrand factor assay This was performed using paraformaldehyde fixed washed normal platelets as described by Reisner et al. [17]. The assay is based upon the fact that ristocetin-induced agglutination of formaldehyde fixed platelets is dependent on the presence of VIII: yon Willebrand factor. The ability of endothelial cell post~ulture medium and immunoabsorbed post~culture medium to support ristocetininduced fixed platelet agglutination was compared to a normal plasma control. For immunoabsorption studies the antisem were coupled to CNBr-activated Sepharose beads [10] and the post~ulture medium incubated with 0.1 vol. packed antibody coated beads for 1 h at room temperature with shaking. Antihuman albumin (Behring Diagnostics, Summit, NJ) was used as the control immunoabsorbent. Results

Immunoisolation o f endothelial cell factor VIII related antigen The factor VIII related antigen in the concentrated endothelial cell postculture medium was easily detected by crossed immunoelectrophoresis (Fig. 1). The slower moving, less anodal factor VIII related antigen forms were more prominent in the endothelial post-culture medium than in normal plasma. The distribution of endothelial cell factor VIII related antigen molecules suggested the presence of two populations, a slow moving cathodal group and a faster

Fig. 1. Crossed i m m u n o e l e c t r o p h o r e s i s of plasma and endothelial cell factor VIII related antigen. Normal h u m a n p!a_ama (left) and endothelial cell post-culture m e d i u m (right) were analyzed by crossed i m m u n o elect~ophoresis in gels containing anti-VIII related antigen. The sample origin is i ndi c a t e d b y the vertical i n k marks m a r k i n g the middle of each sample well. The anode is to t he right in the first dimension and to the top in the second dlmens~on.

365 EC Plasma ClG

]Zm v.,

[

Bond

Fig. 2. ( L e f t p a n e l ) . S D S - p o l y a c r y l a m i d e gel e l e c t z o p h o r e s i s of f a c t o r V I I I r e l a t e d a n t i g e n l m m u n o l s o l a t e d f r o m e n d o t h e l i a l p o s t - c u l t u r e m e d i u m , pun:fled p l a s m a f a c t o r V I I I r e l a t e d a n t i g e n , a n d p u r i f i e d h u m a n f l b r o n e c ~ n r e f e r r e d t o as CIG or c o l d i n s o l u b l e g l o b u l i n . Up-reduced. V I I I , f a c t o r V I I I r e l a t e d A G N ; EC, e n d o t h e l i a l cell. Fig. 3. (Middle panel). S D S - p o l y a c r y l a m i d e gel e l e c t r o p h o r e s i s of f a c t o r V I I I r e l a t e d a n t i g e n i m m u n o i s o f a t e d f r o m e n d o t h e l i a l cell p o s t - c u l t u r e m e d i u m . A n a l i q u o t of the s a m p l e s e e n i n Fig. 2 w a s r e d u c e d w i t h d i t h i o t h r e i t o l a n d t h e n a n a l y z e d . A r r o w i n d i c a t e s t h e f a c t o r V I I I r e l a t e d a n t i g e n of a p p r o x i m a t e M r 220 000. Fig. 4. ( R i g h t p a n e l ) . T w o - d l m e n s i o n a l i m m u n o e l e c t r o p h o r e s i s of f a c t o r V I I I r e l a t e d a n t i g e n i m m u n o l s o f a t e d f r o m e n d o t h e l i a l cell p o s t - c u l t u r e m e d i u m . F i r s t d i m e n s i o n : the f a c t o r V I I I r e l a t e d a n t i g e n i m m u n o i s o l a t e was e l e c t r o p h o r e s e d u n r e d u c e d i n a n S D S - p o l y a c r y l a m i d e gel ( h o r i z o n t a l l y ) ; t h e t o p of gel is t o w a r d the l e f t . The gel was t h e n s l i c e d i n h a l f l e n g t h w i s e a n d o n e slice s t a i n e d w i t h C o o m a s s i e blue. T h e o t h e r slice was s u b j e c t e d t o i m m u n o e l e c t ~ o p h o r e s i s i n a s e c o n d d i m e n s i o n ( v e r t i c a l l y ) i n t o agarose conraining 0.15% anti-factor VIII related antigen.

more anodal group. The factor VIII related antigen in the endothelial cell postculture medium was immunoisolated using monospecific antiserum and killed Staphylococcus aureus Cowan Strain I as described by Kessler [13]. The immunoisolates were analyzed without reduction by an SDS-polyacrylamide gel system (3% polyacrylamide4).5% agarose) that clearly differentiated factor VIII: AGN from fibronectin since both are synthesized by endothelial cells and comigrate following reduction in conventional 5% polyacrylamide gel systems. The Factor VIII related antigen present in endothelial postculture medium migrated as two high molecular weight bands in the unreduced state (Fig. 2). Band 1 barely entered the gel and moved to the same position as a purified plasma factor VIII related antigen marker; band 2 moved into the gel with an apparent molecular weight of 450 000 and migrated to the same position as a purified fibronectin marker. Identical patterns were obtained when the postculture medium was processed in the absence of the sulfhydryl blocker, N-ethylmaleimide. The low molecular weight bands in the endothelial cell factor VIII related antigen immunoisolate moved to the same region of the gel as 7~globulin markers and thus represent the 7~lobulin used to immunoisolate the factor VIII related antigen. When the immunoisolate was reduced with dithiothreitol and e!ectrophoresed, only one factor VIII related antigen band

366

was seen and it migrated with an apparent molecular weigt of 220 000 (Fig. 3). To demonstrate that bands I and 2 both migrated after reduction with a molecular weight of approx. 220 000, bands 1 and 2 were eluted separately from unstained unreduced gels, reduced with dithiothreitol and separately electrophoresed. Under these conditions both bands 1 and 2 moved to the identical position which corresponds to a molecular weight of 220 000. In order to determine the relationship of bands 1 and 2 to the factor VIII related antigen system, two
Von Willebrand factor activity of endothelial cell post-culture medium In order to determine whether bands 1 and 2 were functionally related to von Willebrand factor activity, the post-culture medium was assayed for the ability to support ristocetin-induced platelet agglutination before and after the immunoisolation procedure. As we have previously demonstrated, preculture medium contained no yon Willebrand factor activity [8] while the post-culture medium contained activity equal to 32% of that found in pooled normal human plasma. The yon Willebrand factor activity in the post~ulture medium was completely removed with anti-factor VIII related antigen. Control absorption with antihuman serum albumin had no effect.

Synthesis of endothelial cell factor VIII related antigen Factor VIII related antigen synthesized by endothelial cells was immunoisolated from fractions of post-culture media derived from endothelial cells cultured in the presence of [3H]leucine. The 3H-labeled factor VIII related antigen was isolated using the kiUed Staphylococcus aureas system and characterized by SDS-polyacrylamide electrophoresis (Fig. 5). Radioactivity was detected in 240 220 ZOO 180

2C

Slice nomber Fig. 5. S D S p o l y a c r y l a m i d e gel electzophoresis of factor V I I I related antigen s y n t h e s i z e d b y c u l t u r e d e n d o t h e l i a l cells. E n d o t h e l i a l cells w e r e l a b e l e d w i t h [ 3 H ] l e u c i n e a n d F a c t o r V I I I related antigen i m m u n o i s o l a t e d f r o m t h e p o e t - c u l t u r e m e d i u m u s i n g a n t i - f a c t o r V I I I : A G N a n d StaphylococcuS a u r e a s C o w a n S t r a i n I as d e s c r i b e d i n M e t h o d s . T h e gels w e r e r u n u n r e d u c e d , sliced, and c o u n t e d . In c o m p a n i o n gels, i s o l a t e d p l a s m a factor f a c t o r V I I I r e l a t e d a n t i g e n as a m a ~ k e r m o v e d t o the same p o s i t i o n as the high m o l e e u l a z w e i g h t p e a k a t the t o p o f the gel.

367 Band 1

Bond 2

CIG

I

|

Fig. 6. A u t o r a d i o g r a p h s o f two-dirnens~onal t r y p t i c 125I p e p t i d e m a p s of: i m m u n o i s o l a t e d f a c t o r VIII r e l a t e d a n t i g e n f r o m e n d o t h e l i a l ceil p o s t - c u l t u r e m e d i u m ( b a n d s 1 a n d 2), a n d p u r i f i e d h u m a n f l b r o n e c tin r e f e r r e d to as CIG o r c o l d i n s o l u b l e g l o b u l i n . B a n d s 1 a n d 2 r e f e r t o the b a n d s d e p i c t e d i n Fig. 2. The t r y p t i c p e p t i d e s were a p p l i e d t o t h e l o w e r r i g h t h a n d c o m e r a n d s u b j e c t e d t o h i g h voltage e l e c t r o p h 0 r e s i s in t h e h o r i z o n t a l d i r e c t i o n w i t h t h e a n o d e t o the right. T h e s e c o n d d i m e n s i o n utilizing t h i n - l a y e r c h r o m a t o g r a p h y was c a r r i e d o u t w i t h t h e origin a t t h e b o t t o m .

the region of bands 1 and 2. Following reduction, the radioactivity moved to the 220 000 molecular weight region of the gel.

Two-dimensional tryptic 12SIpeptide maps o f endothelial post-culture medium To determine the structural relationships between the proteins, the fixed and stained proteins in the SDS gels were removed, radioiodinated, and subjected to tryptic digestion. The autoradiographs of the two dimensional peptide maps are shown in Fig. 6. Endothelial cell factor VIII related antigen bands 1 and 2 were essentially identical. All of the peptide spots present in band I were present in band 2 although the relative intensities of some of the spots varied. The factors responsible for the variability in the intensity of some peptide spots in different runs were not evident. A peptide map of equal mixtures of band 1 and band 2 peptides was essentially identical to the peptide map of either band alone. The two~limensional peptide map of fibronectin differed markedly from the factor VIII related antigen proteins (Fig. 6). Discussion Studies by Shearn et al. have previously demonstrated electrophoretic differences between endothelial cell and plasma factor VIII related antigen [ 18 ]. Our studies suggest that human endotheliel cells synthesize two immunologically related molecular families or forms of factor VIII related antigen which differ in their apparent molecular weights. Structurally both forms of factor VIII related antigen are closely related as evidence by the fact that the two
368 noabsorption method identical to that used for isolating the endothelial cell factor VIII related antigen. It would thus appear that the endothelial cell VIII related antigen system shows a greater degree of molecular heterogeneity than the plasma VIII related antigen system. This possibility was also suggested by the molecular distribution of endothelial cell VIII related antigen on crossed immunoelectrophoresis (Fig. 1). Currently studies are underway to determine whether there are any significant structural differences between the endothelial cell and plasma VIII related antigen proteins (Nachman, R~L. and Jaffe, E.A., unpublished data). The factor VIII related antigen system in normal plasma is heterogeneous as evidenced by sucrose density centrifugation [ 1 9 ~ 0 ] , large pore polyacrylamide electrophoresis [ 1 ~ 1 ] , agarose gel chromatography [2], and crossed immunoelectrophoresis [2]. The molecular basis for this heterogeneity is not fully understood. The molecules appear to represent polymeric combination of subunits [22]. There is some controversy regarding the stability of the polymeric forms. Zimmerman et al. [2] indicated that the separate factor VIII:AGN forms were independent molecular entities which did not recombine. On the other hand, Ekert and Chavin [23] presented studies suggesting that the factor VIII related antigen molecules exist in polymeric forms which spontaneously dissociate and exist in equilibrium with a pool of partially dissociated subunits. Several studies suggest that the multimers of the factor VIII related antigen system are held together by labile disulfide bonds with ready disulfide interchange between the subunits [24]. Recently Ruggeri and Zimmerman [25] using a radiolabeled affinity purified antibody as a detection probe have demonstrated up to ten multimeric molecular subsets in plasma factor VIII related antigen. It will be of great interest to apply a similar technique to the study of endothelial cell factor VIII related antigen. Transfusion experiments using radiolabeled factor VIII related antigen in human volunteers have raised the possibility that there may be a continuous conversion of higher molecular forms into lower molecular weight forms in vivo [26,27 ]. Similar observations have been made in a patient with yon Willebrand's disease [28]. The absence of in vivo circulation and conversion may explain the altered ratio of polymer subsets and the greater degree of molecular heterogeneity in endothelial cell factor VIII related antigen. Acknowledgements This work was supported by a Specialized Center of Research in Thrombosis Grant from the National Institutes of Health (NIH HL 18828) and by support from the Arnold Krakower Foundation. One of us (E.A.J.) is a recipient of a Research Career Development Award (5K04 HL 00237) ~nd a Career Scientist Award from the Irma T. Hirschl Trust. References 1 Van Mourik, J.A., Bouma0 B.N., La B ~ y i r e , W.T., de Graaf, S. a nd Mochta~, I.A. (1974) T h r o m b . Res. 4, 155--165 2 Zimmerraan, T.S., Roberts, J. and Edgington, T.S. (1975) Proc. Natl. Acad. Sci. U.S.A. 72, 5121--5125

369 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Jaffe, E.A., Hoye~, L.W. and Nachman, R~L. (1973) J. Clin. Invest. 52, 2 7 5 5 - - 2 7 6 4 Nachman, R.L., Levine, R. and Jaffe, E.A. (1977) J. Clin. Invest. 60, 914--921 Naehman, R.L. and Jaffe, E.A. (1975) J. Exp. Med. 141, 1101--1113 Nachman, R.L., Jaffe, E.A. and Ferris, B. (1980) Bioehem. Biophys. Res. Commun. 92, 1208--1214 Weiss, H.J., Hoyer, L.W., Rickles, F.R., Varma, A. and Rogers, J. (1973) J. CUn. Invest, 52, 2708--2716 Jaffe, E.A., Hoyer, L.W. and Nachman, R.L. (1974) Proc. Natl. Acad. Sei. U.S.A. 71, 1906--1909 Jaffe, E.A. and Nachman, R.L. (1975) J. Clin. Invest, 56, 698--702 Cuatrecasas, P. (1970) J. Biol. Chem. 245, 3059---3065 Jaffe, E.A. and Moshex, D, (1975) J. Exp. Med. 147, 1779--1791 Jalfe, E.A., Naehman, R.L., Becket, C.G. and Miniek, C.R. (1973) J. Clin. Invest. 52, 2475--2756 Kessler, S.W. (1975) J. I m m u n o l . 115, 1617--1624 Converse, C.A. and Papermaster, D.S. (1975) Science 189,469---472 Weinstein, M.J., Deykin, D. and Davie, E.W. (1976) Br. J, Haematol. 33, 343--355 Elder, J.H., Pickett, R.A., Hampton, J. and Lernex, R.A. (1977) J. Biol. Chem. 252, 6510--6515 Reilmero H.M., Katz, H.J., Goldin, L.R., Barrow, E.S. and G~aham, J.B. (1978) Br. J. Haematol. 40, 339--350 Shearn, S.A.M., Peake, I.R., Giddings, J.C., Humphrey, J. and Bloom, A.L. (1977) Thzomb. Res. I i , 43--56 Weiss, H.J. and Kochwa, S. (1970) Br. J. Haematol. 18, 89--95 Zimmel~man, T.S,, Abilgaard, C.F. and Edgington, T.S. (1976) Clin. Res. 2 4 , 4 4 4 A Meyer, O., Obert, B., Pietu, G., Lavergne, J.M. and Zimmexman, T.S. (1980) J. Lab. Clin. Med. 95, 590--602 Van Mourik, J.A. and Bolhuis, P.A. (1978) Thromb. Res. 13, 15--24 Ekert, H. and Chavin, S.I. (1977) Br. J. Haematol. 36, 271--279 Counts, R.B,, Pasken, S.L. and Elgee, S.K. (1978) J. Clin. Invest. 62, 702--709 Ruggeri, Z.M. and Zimmerman, T.S. (1980) J. Clin. Invest. 65, 1318--1325 Over, J., Sixma, J.J., Doueet De Bruine, H.M., Trieschnlgg, A.M.C., Vlooswijk, R.A.A., Besger-Viuer, N.H. and Bouma, B.N. (1978) J. CKn. Invest. 62, 223--234 Over, J., Bouma, B.N., Sixma, J.J,, Bolhuis, P.A. and Vlooswijk, R.A.A. (1980) J. Lab. Clin. Med. 95, 323--334 Sultan, Y., Simeon, J. and Caen, J.P. (1976) J. Lab. Clin, Med. 87, 185--197