Studies on factor VIII-related protein

155

Biochimica et Biophysica Acta, 578 (1979) 155--163

@)Elsevier/North-Holland Biomedical Press

BBA 38166

STUDIES ON F A C T O R VIII-RELATED P R O T E I N I. U L T R A S T R U C T U R A L AND E L E C T R O P H O R E T I C H E T E R O G E N E I T Y O F H U M A N F A C T O R VIII-RELATED P R O T E I N

EUGENE A. BECK, LEONE TRANQUI-POUIT, AGNES CHAPEL, BEAT A. PERRET, MIHA FURLAN, GILBERT HUDRY-CLERGEON and MICHEL SUSCILLON Central Hematology Laboratory, Inselspital and University of Berne, School of Medicine, Berne (Switzerland) and Laboratoire d'Hdmatologie, Centre d 'Etudes Nucldaires de Grenoble, Grenoble (France)

(Received August 25th, 1978) Key words: Factor VIII; Antihemophilic factor; Von Willebrand factor

Summary Human factor VIII-related protein precipitates with specific heterologous antibodies directed against purified factor VIII and supports ristocetin-induced aggregation of washed platelets. We purified human factor VIII from cryoprecipitate by subsequent gel filtration on crosslinked large-pore agarose. Factor VIIIrelated protein appeared as a large aggregate following electrophoresis on 3% polyacrylamide gels in the presence of sodium dodecyl sulfate (SDS). The same material was separated into multiple bands (molecular weight in excess of several millions) following electrophoresis on SDS-I% agarose gels. After complete disulfide reduction of factor VIII-related protein and electrophoresis on SDS5% polyacrylamide gels a single subunit chain (Mr ~ 200 000) was revealed. Analysis of this protein, in its non-reduced state, b y negative contrast electron microscopy showed filaments of markedly variable size. The calculated molecular weight of such filaments ranged from a b o u t 0.6 • 106 to 20 • 104. We conclude that size heterogeneity is an essential feature of human factor VIIIrelated protein.

Introduction

Factor VIII has recently been defined as a macromolecular complex carrying several biological functions: coagulant activity (factor VIII:C) corrects the coagulation disorder of patients with hemophilia; factor VIII-related antigen

156

(factor VIIIR:AG) precipitates with specific heterologous antibodies; the socalled von Willebrand factor (factor VIIR:WF) shortens the prolonged bleeding time in patients with typical yon Willebrand's disease [1]. Factor VIIIR :WF is closely related with the capacity of factor VIII to support ristocetin-induced aggregation of washed platelets [2,3 ]. The structure and biological properties of factor VIII:C are largely unknown. However, there is accumulating evidence that factors VIIIR:AG and VIIIR:WF are both part of a partially defined glycoprotein complex [4,5]. Since the relationship between these two properties has not been fully elucidated we prefer, at the present time, the neutral term 'factor VIII-related protein' for description of the normal or abnormal protein moiety of factor VIII. Normal human factor VIII-related protein is composed of repeating subunits which, following disulfide reduction, appear homogeneous on polyacrylamide electrophoresis in the presence of sodium dodecyl sulfate (SDS). On 5% polyacrylamide gels the apparent molecular weight is approximately 200 000 [6]. Conflicting results have been reported concerning the size of non-reduced factor VIII-related protein or proteins which were similarly defined: recently published molecular weight estimates range from less than 0 . 3 . 1 0 6 [7] to approx. 4.5 • 106 [8]. The molecular size of factor VIII-related protein seems to be of importance for its biological functions: the ratio of VIIIR:WF to VIIIR:AG decreases in late eluting fractions from gel filtration columns as compared with material collected close to the void volume [9]. Therefore, it can be speculated that factor VIIIR:WF is preferably associated with high molecular weight fractions. A similar decrease of this ratio has been found following partial proteolytic degradation of factor VIII [9]. We wish to describe our investigations of factor VIII-related protein in a series of interrelated communications. This first article summarises our evidence suggesting a marked size heterogeneity of human factor VIII. It will be followed by reports on quantitative estimates of size distribution, the stability of the factor VIII complex and the functional relevance of the size distribution. In the present study, the size of factor VIII-related protein was estimated by the simultaneous use of different analytical methods, particularly SDS-agarose gel electrophoresis and electron microscopy. Materials and Methods

Purification of factor VIII Factor VIII was prepared from lipid-poor plasma by cryoprecipitation and subsequent filtration on Sepharose CL-2B (Pharmacia Fine Chemicals AB, Uppsala, Sweden} as previously described [9]. For elution 0.01 M Tris/0.01 M citrate/0.13 M NaC1 buffer (pH 7.4) was used. Fractions which were simultaneously analysed by electrophoresis and electron microscopy were eluted close to the void volume and had an absorbance (A2a0--A320) less than 0.08; they corresponded to the fractions with the highest VIII:C, VIIIR:WF and VIIIR:AG content. Assuming that a frozen plasma pool from healthy adult blood donors contained 1 unit per ml we found 2 to 4 VIIIR:WF (ristocetin cofactor) units and a b o u t 2 units of VIIIR:AG per ml of undiluted purified

157 factor VIII. Purified factor VIII was neither frozen nor lyophilised. Upon storage at 4°C coagulant activity was lost within a few days (T1/2 ~ 1 day). VIIIR:WF and VIIIR:AG remained, however, constant up to several weeks.

Measurement of factor VIII-related activities VIII:C was estimated by a semi-automated one-stage procedure [10]. VIIIR: WF activity was determined by measuring the rate of aggregation of washed, formalin-treated human platelets in the presence of ristocetin [ 11]. Factor VIIIR:AG was measured by one-dimensional electrophoresis on agarose gels containing specific antiserum raised in rabbits with purified human factor VIII [12]. Disulfide reduction of purified factor VIII Disulfide reduction of factor VIII, resulting in the appearance of homogeneous monomeric subunit, was performed by heating factor VIII at 95°C for 10 min, following addition of 2-mercaptoethanol (Merck, Darmstadt, F.R.G., 1% final concentration) in 1% SDS and 0.8 M urea. SDS-polyacry lamide gel electrophoresis Polyacrylamide gel electrophoresis in the presence of SDS (BDH Chemicals, Poole, England) was performed either on 3% gels (containing 3% by weight bisacrylamide : acrylamide) or on 5% gels (containing 5% by weight bisacrylamide : acrylamide) as previously described [13,14]. Acrylamide (Serva, Feinbiochemica, Heidelberg, F.R.G.) and methylene bisacrylamide (Eastman Organic Chemical Co., Rochester, NY, U.S.A.) were disolved in 6 M urea/0.1 M Tris-HC1/0.2% SDS, pH 7.4, and polymerised in 8 cm glass tubes with an inner diameter of 7 ram. Aliquots of 0.2 ml samples in elution buffer were applied following addition of 20 pJ of 10% SDS in 8 M urea and the marker dye (bromophenol blue). The electrophoresis chamber contained 0.2% SDS in 0.1 M Tris-HC1 buffer, pH 7.4. Electrophoresis at 7 mA/gel was usually terminated when the marker dye had reached the bottom of the gels. SDS-1% agarose gel electrophoresis Agarose (Serva, Feinbiochemica, Heidelberg, F.R.G.) was dissolved in 6 M urea/0.075 M sodium 5,5~liethylbarbiturate/0.01 M ethylenediaminotetraacetate (EDTA)/0.2% SDS buffer (pH 8.6), boiled for 15 min and poured into 12 cm glass tubes with an inner diameter of 7 ram. The gels were cooled at room temperature for 30 min and then brought to 4°C for at least 12 h prior to use. Electrophoresis was carried out at room temperature in a buffer containing 0.075 M sodium 5,5
158

Immunological identification of factor VIII-related protein after SDS-1% agarose electrophoresis Factor VIII-related protein was first electrophoresed on SDS-I% agarose electrophoresis as described above. Gels were then split longitudinally. One half was stained with Coomassie blue while the other was subjected to second electrophoresis at a right-angle as described by Converse and Papermaster [ 16]. Instead of Lubrol XP, we used Triton X-100 (1.5%, v/v) to remove SDS. The second layer contained rabbit antiserum against human factor VIII.

Two-dimensional

immunoelectrophoresis

was

essentially performed

as

described by Zimmerman et al. [17].

Electron microscopy Factor VIII was applied onto ionised carbon membranes at ambient temperature. Negative contrast was achieved with 2% phosphotungstic acid (potassium salt) at pH 7.0. Evaporation of salt-containing solutions was avoided prior to

A

B C 543

2

1

D

{IgM) n 5"4'3'

2

~

M r x 10-6

Fig. 1. E l e c t r o p h o r e t i c a n a l y s i s o f f a c t o r V I I I - r e l a t e d p r o t e i n . P u r i f i e d f a c t o r V I I I w a s e l e c t r o p h o r e s e d on S D S - I % agaxose. O n e h a l f o f an e l e c t r o p h o r e t i c gel w a s s t a i n c d w i t h C o o m a s s i e b l u e , p h o t o g r a p h e d (C) a n d s c a n n e d ( A ) . ,The s e c o n d h a l f w a s u s e d f o r i m m u n o e l e c t r o p h o r e s i s (B). G e l D r e p r e s e n t s t h e e l e c t r o p h o r e t i c p a t t e r n o f I g M o U g o m e r s . T h e m o l e c u l a r w e i g h t s , i n d i c a t e d b e l o w , axe b a s e d o n m o l e c u l a r w e i g h t e s t i m a t e s o f IgM.

159 fixation by keeping the loaded grids in a moist chamber. The procedure corresponds to that developed for studies on fibrin formation [18]. Results

The following characteristic properties of human factor VIII were found at the time when samples were analysed b y SDS-1% agarose gel electrophoresis or electron microscopy: following comple.te disulfide reduction, the protein yielded a single band with a molecular weight of a b o u t 200 000 on SDS-5% polyacrylamide gel electrophoresis; non-reduced material barely penetrated SDS-3% polyacrylamide gels, even following prolonged electrophoresis: the ratio of V I I I R : W F / V I I I R : A G remained > 1 . 0 in fractions eluting close to the

A

B a

b

c

d

F i g . 2. E l e c t r o n m i c r o s c o p y o f p u r i f i e d f a c t o r V I I I . M o l e c u l a r w e i g h t s o f f o u r t y p i c a l f i l a m e n t s (B) a t 1 . 5 × o r i g i n a l m a g n i f i c a t i o n ( A ) w e r e c a l c u l a t e d as f o l l o w s : a ~ 0 . 6 • 1 0 6 ; b ~ 0 . 8 • 1 0 6 ; c ~ 1 . 5 • 1 0 6 ; d 5.8 • 106.

160 void volume; two-dimensional immunoelectrophoresis suggested that slowmigrating factor VIII was predominant. In contrast to SDS-polyacrylamide electrophoresis, multiple bands appeared when purified factor VIII was electrophoresed on SDS-I% agarose gels (Fig. 1). Comparison with migration of appropriate molecular weight markers, e.g. crosslinked IgM, suggested that the major bands had a molecular weight in excess of several millions (Fig. 1). Each band clearly reacted with specific heterologous antibody as shown b y immunoelectrophoresis (Fig. 1). Semiquantitative evaluation of antigen suggested proportionality between the staining intensity of protein bands on SDS-1% agarose gels (first dimension) and the distance of the precipitin line from the origin. Furthermore, we observed several peaks in the precipitin line corresponding exactly to the faster moving bands on the first dimension electrophoretic gel. Coomassie blue staining of the first dimension electrophoretic gels did not reveal any protein which did not react with antibody, at least in fractions eluting close to the void volume. Electron microscopy of purified factor VIII, characterized as described above, revealed multiple molecular forms. Representative examples are shown in Figs. 2 and 3. The most consistent structures seen in our preparations (6 fractions from 4 different agarose columns) were filaments of variable diameter

F i g . 3 . E l e c t r o n m i c r o s c o p y o f p u r i f i e d f a c t o r V I I I . E s t i m a t i o n o f m o l e c u l a r w e i g h t o f s o m e large aggreg a t e s gave t h e f o l l o w i n g v a l u e s : a ~ 1 0 - 1 0 6 ; b ~ 1 6 - 1 0 6 ; c ~ 2 0 • 1 0 6 .

161 and length. Many larger filaments appear to be composed of smaller fibrillar structures. We attempted to calculate the molecular size of individual filaments by assuming that subsegments corresponded to cylindric structures; the mass of dry protein was based on a specific volume of 0.72. Thus, a molecular weight ranging from 0.6 • 106 to 6 • 106 were obtained for the four elements which are shown in more detail on Fig. 2. Larger aggregates, similar to those illustrated in Fig. 3, had a calculated molecular weight of between 10 • 106 and 20 • 106. Discussion Despite numerous recent studies on the relationship between factor VIII structure and function it is n o t yet clear how the biological properties VIII:C and VIIIR:WF are associated with the polypeptide chains of factor VIII. Methods, described in this paper, demonstrate the size heterogeneity of the proteins which, by agarose gel filtration, are eluted together with VIII:C and VIIIR:WF; they do, however, not allow the measurement of biological activities associated with the individual size species. Our results indicate that the protein bands, separated by SDS-agarose electrophoresis, carry antigenic sites (VIIIR:AG) reacting with specific heterologous antibodies. In view of the reasonable purity of our factor VIII preparations we assume t h a t most of the material shown by electron microscopy also corresponds to factor VIII-related protein. Two features of factor VIII-related protein emerge from our present study, one being the enormous size of protein aggregates, the other a prominent size heterogeneity. Since the introduction of crosslinked large-pore (2%} agarose for separation of factor VIII from contaminating proteins in cryoprecipitate (9) we assume that the factor VIII, eluted close to the void volume, may have a molecular weight in excess of 10 • 106, because the exclusion limit of this agarose for proteins corresponds to a molecular weight of about 40 • 1 0 6. Comparison of factor VIII-related protein with chemically crosslinked IgM, following electrophoresis on SDS-I% agarose gels, suggests t h a t the smallest species of our gel filtration fraction, which exhibits the highest biologic activities, had an apparent molecular weight in excess of 5 • 106. We are, however, aware t h a t this figure is only a rough estimate, since molecular weight determinations on SDS-gels depend on m a n y factors, such as intrinsic charge and configuration of the protein as well as the pore-size of the gel. Fass et al. [19] reported that purified porcine factor VIII, according to its electrophoretic mobility on SDS-2% agarose, may have a molecular weight of up to 21 • 106. We chose electron microscopy as a further tool for analysis of factor VIIIrelated protein, because the time needed for preparation of samples was minimal and addition of denaturing agents, such as SDS or urea, was avoided. The estimates of the molecular size of factor VIII aggregates fit with the high molecular weight values extrapolated from SDS-agarose gels. In fact, the largest clearly isolated particles seen in our factor VIII preparations have a calculated molecular weight of about 20 • 106. We wish to stress that these values are only approximations. Errors of individual measurements are possibly caused by irregularities of filaments, such as ramifications and/or superpositions of adja-

162 cent fibrils. Considerable size variations of factor VIII have been suggested by several earlier studies on the ultrastructure of factor VIII. Thus, filaments of variable size were found after shadowing with carbon-platinum [8] or by negative staining [20]. In a recent publication, Tan and Andersen described a molecular size range similar to ours [21]. However, the predominant shape which they ascribed to human factor VIII, both in its hydrated and dehydrated state, was rod-like or spherical [21]. Marder et al. also described a predominance of 'prolate ellipsoid forms' of factor VIII, besides filaments or variable length and thickness [20]. SDS agarose gel electrophoresis of human factor VIII-related protein shows a regular pattern of multiple bands, similar to that reported for porcine factor VIII [19]. As already mentioned the smaller species of the peak factor VIII fractions from agarose column had an apparent molecular weight in excess of several millions on electrophoresis gels. In contrast to this, we found a significant proportion of smaller aggregates, in the same preparation, by electron microscopy. We were unable to detect any effect of phosphotungstic acid on the electrophoretic behavior of factor VIII. Salt concentration was also considered as a factor responsible for dissociation of larger aggregates; however, dialysis against volatile buffers did not change the size-distribution suggested by electron microscopy (unpublished results). We conclude that factor VIII-related protein, as defined in our study, may form (or consist of) aggregates of very large size. Size heterogeneity was found b y t w o radically different analytical tools, i.e. SDS-agarose electrophoresis and electron microscopy. The regular pattern of protein bands, found on electrophoretic gets, prompted a more quantitative approach which will be described in the accompanying paper [22]. Acknowledgements This work was supported by grants from the Stanley Thomas Johnson Foundation, the Central Laboratory of the Swiss Red Cross National Transfusion Service and grant No. 3.515-0.75 from the Swiss National Science Foundation. The skilful technical assistance of Miss. A. Glanzmann is gratefully acknowledged. References 1 T a s k F o r c e o f t h e I n t e r n a t i o n a l C o m m i t t e e o n T h r o m b o s i s a n d Haemostasis: Nomenclature o f F a c t o r VIII r e l a t e d activities. Paris, 1 9 7 5 ( p r e l i m i n a r y report) 2 Howard, M.A., Sawers, R.J. and Firkin, B.G. (1972) Blood 41,687--690 3 Weiss, H . J . , H o y e r , L.W., R i c k l e s , F . R . , V a r m a , A. a n d R o g e r s , J. ( 1 9 7 3 ) J. Clin. I n v e s t . 5 2 , 2 7 0 8 - 2716 4 M c K e e , P.A., A n d e r s e n , J . C . a n d S w i t z e r , M.E. ( 1 9 7 5 ) A n n . N . Y . A c a d . Sci, 2 4 0 , 8 - - 3 3 5 G r a l n i c k , H . R . a n d CoUer, B.S. ( 1 9 7 5 ) B l o o d 4 6 , 4 1 7 - - 4 3 0 6 S h a p i r o , G . A . , A n d e r s e n , J.C.~ Pizzo, S.V. a n d M c K e e , P.A. ( 1 9 7 3 ) J. Clin. Invest. 5 2 , 2 1 9 8 - - 2 2 1 0 7 N e w m a n , J., H a r r i s , R.B. a n d J o h n s o n , A . J . ( 1 9 7 6 ) N a t u r e 2 6 3 , 6 1 2 - - - 6 1 3 8 H e r s h g o l d , E . J . ( 1 9 7 5 ) A n n . N.Y. A c a d . Sci. 2 4 0 , 7 1 - - 7 3 9 F u r l a n , M. a n d B e c k , E . A . ( 1 9 7 7 ) T h r o m b o s . R e s . 1 0 , 1 5 3 - - 1 5 8 1 0 B e c k , E . A . ( 1 9 7 5 ) in H a e m o p h i l i a ( U l u t i n , O . N . a n d P e a k e , I . R . , eds.), p p . 1 8 7 - - 1 8 9 E x c e r p t a Medica, Amsterdam

163 11 12 13 14 15 16 17 18 19 20 21 22

Cooper, H.A., Reisner, F.F., Hall, M. and Wagner, R.H. (1975) J. Clin. Invest. 56,751--760 Meyer, D., Dreyfus, M.D. and Larrieu, M.J. (1973) Path. Biol. 21 (Suppl.), 66--71 Shapiro, A.L., Vinuela, E. and Maizel, J.V. (1967) Biochem. Biophys. Res. Commun. 28,815--820 Furlan, M., Jakab, T. and Beck, E.A. (1977) Thrombos, Res. 10, 431--443 Furman, M. and Beck, E.A. (1975) Thrombos. Res. 7 , 8 2 7 - - 8 3 8 Converse, C.A. and Papermaster, D.S, (1975) Science 189,469--472 Zimmerman, T.S., Roberts, J. and Edgington, T.S. (1975) Proc. Natl. Acad. Sei. U.S., 72, .5121--5125 Pouit, L., Marcille, G., Suscillon, M. and Hollard, D. (1972) Thrombos. Diathes. Haemorrh. (Stuttg.) 27,559--572 Fass, D.N., Knutson, G.J. and Bowie, E.J.W. (1978) J. Lab. Clin. Med. 9 1 , 3 0 7 - - 3 2 0 Marder, V.J., Tranqui-Pouit, L., Hudry-Clergeon, G., Atichartakarn, V. and Suscillon, M. (1977) Thrombos. Haemostas. (Stuttg.) 38, 88 Tan, H.K. and Andersen, J.C. (1978) Science 198,932--934 Perret, B.A., Furlan, M. and Beck, E.A. (1979) Biochim. Biophys. Acta 578,164--174