Chapter 12
Plasma Proteins at Natural and Synthetic Interfaces: A Fluorescence Study 1
1,2
1
1
Proteins at Interfaces Downloaded from pubs.acs.org by UNIV OF MICHIGAN ANN ARBOR on 11/21/18. For personal use only.
E. Dulos , J. Dachary , J. F. Faucon , and J. Dufourcq 1
Centre de Recherche Paul Pascal, Centre National de la Recherche Scientifique, Domaine Universitaire, 33405 Talence Cedex, France Laboratoire d'Hématologie, Université de Bordeaux II, 146, rue Léo Saignat, 33076 Bordeaux Cedex, France 2
Blood clotting proteins bind to charged surfaces mainly by ionic forces often reinforced by hydrophobic contri bution. Cardiotoxin binding to heparin, phospholipids and polymers illustrates both possibilities. The adsorption of the thrombin-antithrombin complex (T-AT) on various anionic anticoagulant polystyrene deri vatives results in a quenching of the T-AT fluorescence. The stability of the resulting complexes depends on the polymer,as shown by desorption by polybrene. From compe tition experiments between heparin and polymers, it is proposed that the T-AT desorption parallels the anticoa gulant activity of the polymers. The binding site of blood clotting proteins on phos pholipids involves both charged and zwitterionic lipids. By fluorescence energy transfer, no selectivity is demons trated for factor Va while its light chain is highly se lective for the negatively charged lipids. In contrast, the complex Va - Xa has some selectivity for the zwitter ionic lipids. The binding of the vitamin K-dependent fac tors does not induce phase separation in lipids. A calcium -independent binding is demonstrated and allows to propose a new model for the interaction. Proteins generally contain aromatic fluorescent residues which are very sensitive to l o c a l events. Their interactions with surfaces can therefore be followed by fluorescence techniques. This study deals with the formation of complexes between blood c l o t t i n g proteins and natural and a r t i f i c i a l surfaces. As these surfaces are generally charged, the behavior of a basic protein, cardiotoxin (CTX), the i n t e r a c t i o n of which i s s t r i c t l y charge-dependent, i s also reported for comparison. Two types of interface have been investigated. F i r s t , phospholipid biiayers which mimic c e l l u l a r membrane and p l a t e l e t factor 3, on which several blood c l o t t i n g factors bind i n order to generate the more e f f i c i e n t cascade of enzymatic reactions ( J _ ) . They are the vitamin K-dependent proteins I I , X and IX which are 0097-6156/87/0343-0180$06.00/0 © 1987 American Chemical Society
12.
Plasma Proteins at Natural and Synthetic Interfaces
DULOS ET AL.
181
supposed to bind to membranes v i a Ca bridges (2) between the Gla residues and the negative charges of the interface and factor V which i s not vitamin K-dependent but interacts with membranes by both elec t r o s t a t i c and hydrophobic forces (3). Second, synthetic polymers designed to be used as parts of i n t r a - or extracorporeal blood c i r c u l a t i o n . These materials must be able to remain i n contact with blood without promoting coagulation. They are polystyrene derivatives (4,5) bearing functional a c i d i c groups of heparin, the natural i n h i b i t o r of coagulation which cata lyses the i n a c t i v a t i o n of Thrombin (T) by Antithrombin (AT). Three polymers with d i f f e r e n t anticoagulant a c t i v i t i e s (6), the so-called "PSS0 Glu", "PAOM" and PSS0 , are compared with respect to t h e i r adsorption properties towards these plasma proteins. In the f i r s t part of t h i s study, we focus on the binding of CTX to the two types of interface. Secondly, we discuss the binding and desorption of the 1:1 T-AT complex from the polymeric interfaces. F i n a l l y , we investigate the charge d i s t r i b u t i o n within the plane of the membrane when blood c l o t t i n g factors II and V are bound. fl
2
If
3
Materials and Methods Materials. Human blood c l o t t i n g factors I I , X and IX were p u r i f i e d from prothrombin concentrates, according to the method of D i Scipio (7)· Human factors V and Va light chain (VaLC) were the generous g i f t of Dr Lindhout, Maastricht University. CTX, (MW 6840) was a g i f t from Pr Rochat and Dr Bougis, Marseille University. Human Thrombin, T, (MW 38,000) and human Antithrombin (AT), (MW 65,000), were obtained from the Centres de Transfusion Sanguine, respectively of Paris and L i l l e . Heparin (mean MW = 15,000) was from Choay (France). Polybrene was from Serva (F.R.G.). The polymers used herein were the generous g i f t of Pr M. Jozefowicz and Dr C. Fougnot, Paris-Nord University (4,6). The polymers consist of a polystyrene backbone bearing, s t a t i s t i c a l l y , a c i d i c groups of heparin. In the "PSS0 derivative, only sulfonate groups are present. Both sulfonate and sulfamide aminoacid groups are pre sent i n the other polymers. S p e c i f i c a l l y , glutamic acid and a r g i n y l methyl-ester respectively are present i n the "PSS0 Glu" and "PAOM" derivatives. PSS0 Glu, PAOM, PSS0 , are respectively very active, active and only s l i g h t l y active as anticoagulant materials. They are used as fine p a r t i c l e s with mean diameter lower than 0.1 μπι, giving stable suspensions. Similar l i g h t scattering of a l l polymers at the same concentration indicates that they have almost similar p a r t i c l e size. Natural beef brain phosphatidylserine (PS) was obtained from L i p i d Products (U.K.). Egg l e c i t h i n (PC) was prepared i n the labora tory. Synthetic l i p i d s , namely dimyristoyl- and dipalmitoylglycerophosphocholine (respectively DMPC and DPPC) and dipalmitoylglycerophosphoserine (DPPS), were obtained from Medmark (F.R.G.). The f l u o rescent probes, namely 1-acyl-2-(6-pyrenylbutanoyl)-sn-glycero-3phosphocholine (PBPC) and 1-acyl-2-(6-pyrenylbutanoyl)-sn-glycero-3phosphate (PBPA) were synthesized i n the laboratory. Other analogous probes, with pyrenyldecanoyl chains (respectively PDPC and PDPA) , came from KSV (Finland). Their concentrations were determined using ε ^ = 5 0 , 0 0 0 M" .cm" . The Forster distance R for the Trp-pyrene pair i s 4 nm (9). fl
3
2
2
1
3
7
3
1
Q
182
PROTEINS AT INTERFACES
Methods. I n t r i n s i c fluorescence measurements were carried out on an SLM 8000 spectrofluorometer. " T i t r a t i o n " was performed by adding either phospholipids, heparin or polymers into the reference cuvette and into the measure cuvette which contains protein solutions at fixed concentrations. A l l the corrected spectra were obtained under the following conditions : e x c i t a t i o n wavelength 280 nm, s l i t s 8 mm, temperature 25°C,background subtraction at every stage of the t i t r a t i o n experiments. Moreover, i n the presence of polymers, the f l u o rescence i n t e n s i t i e s were corrected taking into account the absorbance of the samples at the e x c i t a t i o n and at the emission wavelengths, according to the method of Parker and Barnes (10). For energy transfer experiments, fluorescence spectra were obtained by subtracting the fluorescence spectra of the phospholipids, i n the absence or i n the presence of the energy acceptor, to correct for light scattering and for the fluorescence of the probe. Energy transfer e f f i c i e n c i e s (Et) were calculated as previously described (11). The absorbance of samples i n the presence of either heparin or labelled l i p i d s never exceeded 0.1, therefore, the inner f i l t e r effect was n e g l i g i b l e . Fluorescence p o l a r i z a t i o n measurements were performed with an apparatus b u i l t i n the laboratory and connected to a minicomputer ( D i g i t a l LSI I I ) . This was, i n turn, connected to a Vax 11/780, thus allowing complete automatization of the measurements. Diphenylhexatriene (DPH) i n tetrahydrofuran (6 mM) was added to phospholipids i n amounts never exceeding 1 %. Results Interactions of Cardiotoxin with Polyanionic Surfaces Cardiotoxin-lipid complexes. Due to i t s effect on membranes, the b i n ding of CTX to l i p i d interfaces has been extensively studied i n the recent past (8,12). We f i r s t demonstrated that i t can be followed by fluorescence changes of Trp11, the emission being blue-shifted by 15 nm from 350 to 335 nm, and the r e l a t i v e intensity increased by about 200 % when CTX i s bound to PS (13) (Fig. 1). This interaction occurs only with negatively charged species, even i n l i p i d mixtures consisting of charged and zwitterionic l i p i d s . Such behavior implies a s t r i c t s e l e c t i v i t y towards charged groups (13). Seven charged groups have been shown to be involved at the i n terface (13). They have been tentatively attributed to the Lys r e s i dues i n the N-terminal region and at the end of the second loop of the molecule, mainly based on the structure of analogous protein established by X-ray studies (14). By Raman spectroscopy, i t has been demonstrated that the toxin has very similar secondary and t e r t i a r y structures when i n solution or bound to l i p i d s . This i s due to the cross-linking by d i s u l f i d e bridges. Indeed, when these bridges are reduced and methylated, the resultant toxin s t i l l binds, but becomes f l e x i b l e and adopts a different 8 sheet structure at the interface (15). Comparison of different iso-toxins leads to the conclusion that Arg5 residue of CTX IV has a strong s t a b i l i z i n g effect on the protein at the interface (12). Because of the strong s t a b i l i t y of the protein, the effects of extreme pH can be investigated. The CTX-lipid complexes previously formed at pH 7.5 (13) dissociate only at pH values higher than 10,
12.
DULOS ET AL.
Plasma Proteins at Natural and Synthetic Interfaces
183
when the net charge of the protein i s severely decreased or, conversely, at pH values lower than 4, when the l i p i d s are protonated. The binding of constitutive peptides of CTX with blocked COO-terminal, namely the peptides 6-12 and 5-12, the sequence of which i s Arg5-LeuIle-Pro-Pro-Phe-Trp-Lys12, demonstrates that Arg5 plays an important role i n maintaining the peptide at the interface up to pH 11 (8). Since charged groups are involved at the interface, i t i s possible to dissociate the complex by increasing the ionic strength. The release of the protein, inferred from the recovery of the fluorescence of protein free i n solution and centrifugation experiments (13), occurs at concentrations of NaCl higher than 1 M (12) (Table I ) . In contrast, the small constitutive peptides investigated have a lower s t a b i l i t y at interface since lower amounts of s a l t , or calcium i n the mM range, are able to dissociate the lipid-peptide complexes ( 8 ) . Although the major effect i s c l e a r l y n e u t r a l i z a t i o n of charges involved at the interface, one has to keep i n mind that the burying of Trp11, detected by fluorescence, can be interpreted as a passage of t h i s group from a highly polar water environment to a less polar one where solvent relaxation cannot occur. Therefore, hydrophobic forces also are involved and we proposed that the f i r s t loop ( r e s i dues 6-11) may serve as hydrophobic anchor (12). The protein must penetrate through the polar group region of the bilayer and compress the l i p i d molecules. This has been inferred more d i r e c t l y from a monolayer study by Bougis et a l . who demonstrated that at high l a t e r a l pressures, compatible with those i n a b i l a y e r , the protein occupies at the interface an area of about 500 Â (16). 2
Cardiotoxin-heparin complexes. The strong binding of CTX to anionic l i p i d s , as indicated above, suggests that the protein would interact with other strongly a c i d i c surfaces or polyanions. Therefore, the fluorescence of CTX i n the presence of heparin, one of the more r e l e vant polyanionic compounds involved i n coagulation, was studied. As shown i n F i g . 1, one can follow the increase of fluorescence intens i t y of the protein which p a r a l l e l s the addition of heparin. The t o t a l increase i s r e l a t i v e l y small (about + 45 % ) , and only a s l i g h t blue s h i f t of the emission wavelength occurs (about 2 nm). S i m i l a r l y , i t has been found that the antithrombin binding to heparin results i n a 25 % increase of the protein fluorescence (17-19). These features immediately indicate that a complex i s formed between CTX and hepar i n , giving r i s e to an increase of the quantum y i e l d without serious burying of Trp. The stoichiometry of these complexes can be estimated as about 3 to 5 CTX bound per heparin molecule and the d i s s o c i a t i o n constant as about 0.5 χ 10~ M" . Addition of C a i n the mM range which allows a quick recovery of the fluorescence of the toxin free in solution ( F i g . 2), i s interpreted as a d i s s o c i a t i o n of the com plexes and thereby demonstrates the ionic pairing e f f e c t . 6
1
+ +
Cardiotoxin-polymer complexes. F i n a l l y , the interaction of CTX was also studied with a polymeric surface, PSS0 Glu, the composition and properties of which resemble those of heparin. As shown i n F i g . 1, a result quite different from those for heparin or l i p i d binding i s observed. When the surface i s maximally covered, at a polymer to protein r a t i o of 12 mg of polymer per ymole of CTX, the quantum y i e l d of Trp11 i s decreased by about 50 %, while the emission wavelength remains almost unchanged at about 345 nm. This can be interpreted 2
PROTEINS AT INTERFACES
184
Table
I . Binding characteristics of cardiotoxin to different polyanionic interfaces PS
PSS0 Glu and the same sequence was found for polymer effects on hepa rin-thrombin and on heparin-antithrombin complexes (data not shown). This sequence can be related to that of the anticoagulant effects of polymers, namely, PSS0 < PAOM< PSS0 Glu. 3
2
3
2
PROTEINS AT INTERFACES
188
The experimental results of T-AT adsorption on PSS0 and PSS0 Glu in the presence and i n the absence of heparin were further examined by looking at the normalized curves shown i n F i g . 5. Making again the hypothesis of simple e q u i l i b r i a , the amounts of protein bound can be compared using these curves. For the same quantity of polymer added (dashed v e r t i c a l line i n F i g . 5), i t can be seen that : i ) In the absence of heparin, the f r a c t i o n of protein adsorbed on PSS0 Glu (point a) i s higher than on PSS0 (point b). i i ) The f r a c t i o n of prot e i n adsorbed on PSS0 i s the same (point b) i n the absence and i n the presence of heparin, i i i ) In the presence of heparin, the f r a c t i o n of protein adsorbed on PSS0 i s higher than on PSS0 Glu. In other words, i n the absence of heparin, the T-AT adsorption occurs more readily on PSS0 Glu than on PSS0 . Heparin reverses this order so that, i n the presence of heparin, the T-AT adsorption occurs more readily on PSS0 than on PSSO Glu. This result may be expressed as follows : i n the presence of Heparin, the T-AT complex can be r e l e a sed more e a s i l y from the PSS0 Glu than from the PSS0 surface. This observation could be of some significance for the functioning of polymers as anticoagulant materials, i f the quantitative difference between their anticoagulant a c t i v i t i e s i s taken into account. 3
2
2
3
3
3
2
2
3
3
2
3
Search for Topological Changes i n the D i s t r i b u t i o n of Charged Groups at the Phospholipid Interface, Induced by Blood C l o t t i n g Factors Due to t h e i r requirement for negatively charged groups at the phospholipid interface, one can expect that, upon binding, blood c l o t t i n g proteins change the l a t e r a l d i s t r i b u t i o n of l i p i d s i n the membrane at their binding s i t e . In order to document such an e f f e c t , the composit i o n of the binding sites of blood c l o t t i n g factors was estimated i n two ways : i ) Through changes i n the thermotropic properties of l i pids, i i ) Using fluorescence energy transfer data under isothermal conditions. Thermotropic behavior of phospholipids i n the presence of blood c l o t t i n g factors. The changes i n the t r a n s i t i o n temperature (Tm) of phospholipid mixtures induced by the presence of the proteins were monitored by fluorescence p o l a r i z a t i o n of the hydrophobic probe, DPH, inserted i n the bilayers. These changes are generally interpreted as indicative of phase separation, as already demonstrated for the e f fect of calcium, at concentrations higher than 10 mM i n PS-rich b i layers (23), and i n the case of some proteins (24). For CTX, i t has already been shown that, on pure negatively charged l i p i d s , phase separation i s induced between areas of pure l i pid and aggregated CTX-lipid complexes whose thermal transitions d i sappear (24). Moreover, i n binary PC-PS mixtures, CTX always strongly s h i f t s the melting temperature Tm, towards that of the pure PC component (25). The presence of the c l o t t i n g factor II (prothrombin) at a l i p i d to protein molar r a t i o of 20, induces a s h i f t i n Tm of pure l i p i d s such as PS, DMPC, DPPC, from 8 to 12°C for PS, from 21 to 23.5°C for DMPC and from 38 to 43°C for DPPC,as seen on F i g . 6a. These results are indicative of the binding of factor II with the pure lipids.With DMPC and DPPC, they further suggest that, at the very least, an aggregation of the vesicles does occur, the t r a n s i t i o n p r o f i l e being
12.
189
Plasma Proteins at Natural and Synthetic Interfaces
DULOS ET AL.
100
Moles of POLYBRENE/Mole of
Τ_ΔΤ
Figure 4. Effects of polybrene on the fluorescence of T-AT i n i t i a l l y adsorbed onto polymers :• PSS0 Glu, Δ PAOM, · PSS0 . ([T-AT] » 10" M, 40 mg of polymer per u mole of T-AT.) 2
3
6
m
9
if
POLYMER/μΜοΙβ
of
T-AT
Figure 5. Quenching of the r e l a t i v e fluorescence intensity of T-AT versus the quantity of polymer added. In the absence of he parin, adsorption of T-AT onto PSS0 Glu, ο PSS0 . In the pre sence of heparin, adsorption of T-AT onto :• PSS0 Glu, · PSS0 . ([T-AT] = 0.5 χ 10~ M.) 2
3
2
6
3
PROTEINS AT INTERFACES
190
sharpened and resembling that of large v e s i c l e s . This was recently confirmed by Lentζ et al.(26)who showed by freeze fracture electron microscopy that fragment I which i s the N-terminal part of prothrom bin induces a fusion of DMPC v e s i c l e s , while negatively charged v e s i cles remain i n t a c t . With binary mixtures such as DPPC-PS (50-50), the s h i f t of Tm was from 32 to 35°C i n the presence of 5 mM calcium. Blood c l o t t i n g factors I I , X and IX induced a further small s h i f t . With the DMPCDPPS binary mixture, calcium s h i f t s the Tm from 37.5°C down to 33.5°C, t h i s temperature being again decreased by 1 to 2°C i n the presence of blood c l o t t i n g factors (Fig. 6b). These effects are quite weak when compared to those induced by CTX on the same l i p i d mixtures : an increase of 8°C for DPPC-PS and a decrease of 9°C for DMPC-DPPS (25). For t h i s protein, which only binds to PS, the Tm observed i s that of a PC-enriched phase. There appears to be no reason to conclude, therefore, that phase separation occurs as a consequence of blood c l o t t i n g factor binding, i n contrast to the conclusion of Mayer and Nelsestuen (27,28) from similar experiments. The lack of phase separation i n similar systems was recently confirmed by Lentζ (26) describing the whole phase d i a gram of a PC-PG mixture. The binding s i t e for blood c l o t t i n g factors can be enriched i n negative l i p i d s but, i f domains of d i f f e r e n t composition compared to the bulk are formed, they must be very small i n size and/or their composition must be very close to that of the i n i t i a l mixture. Energy transfer experiments with vitamin K-dependent factors. Equimolar mixtures of PC-PS were labelled with either PBPC or PBPA, at the same concentration, namely 3.8 %. The transfer e f f i c i e n c y values (Et) are shown i n Table I I . Since the two probes have i d e n t i c a l spectros copic features (9), the comparison of Et values for the membranes having similar amounts of label, allows a direct comparison of the l i p i d environment of the factors. The higher Et value indicates a lower s t a t i s t i c a l distance and/or a greater number of l a b e l l e d phos pholipids i n the neighborhood of the protein. Results show that the energy transfer process can occur with both types of l i p i d probe for factors I I and IX, indicating that both types of l i p i d are involved i n the binding s i t e s of these f a c tors. Energy transfer e f f i c i e n c y values were lower f o r PBPC- than f o r PBPA-labelled membranes i n the case of factor I I . This indicates a PS s e l e c t i v i t y for factor I I . On the other hand, for factor IX, the e f f i c i e n c i e s are very similar (0.35 and 0.45 for PBPC and PBPA respec t i v e l y ) , so that both types of l i p i d seem to be present i n similar proportions i n the factor IX binding s i t e . Experiments were also done with PC-PS (80-20) mixtures. Compari son of the Et values shows that the binding s i t e s for both factors have a higher PS content i n the PC-PS (50-50) membrane compared to that i n the PC-PS (80-20) membrane. This indicates that the factors do not have a single, well-defined binding s i t e , independent of the i n i t i a l membrane composition. Another interesting result from our work i s that the transfer process was also observed, with both probes, i n the absence of calcium. This again leads to the conclusion that the two types of l i p i d are present i n the Ca -independent binding s i t e of the factors. Since these experiments are direct studies of binding, the asso++
12.
DULOS ET AL.
191 Plasma Proteins at Natural and Synthetic Interfaces
Figure 6. a) Phase transitions of PS, DMPC, DPPC i n the presence of factor II :0 PS alone,» PS, factor I I , =50 ;^ DMPC alone, •DMPC, factor I I , R £ = 2 0 ; ο DPPC alone, · DPPC, factor II,R£=50. b) Phase t r a n s i t i o n of DMPC-DPPS (1:1) : + alone ; or with : ο CTX ; C a 5mM; Δ C a 5mM, factor IX,R£=29; · C a 5mM, factor X, R£ = 20. + +
+ +
+ +
192
PROTEINS AT INTERFACES
c i a t i o n parameters can be obtained from the transfer experiments. In the presence of calcium, the values of Kd are about 10 M" for each factor. These values agree with those of some authors (29,30), but are lower by one or two orders of magnitude than those determined by others (31,32). In the absence of calcium, the Kd values we obtained, which are the f i r s t reported i n the l i t e r a t u r e , are of the same order of magnitude, suggesting that t h i s ion has no great influence on the binding process. 1
Energy transfer experiments with factor V, factors Va and VaLC. Similar experiments have been performed with another factor involved i n prothrombin a c t i v a t i o n but which i s not vitamin K-dependent, namely factor V. The experiments were done with PC-PS (50-50) mixture l a b e l led with 5 % PDPC or PDPA, at a single phospholipid-to-protein molar r a t i o , R^ = 530. Fluorescence quenching values are reported i n Table III. They are of the same magnitude with both probes, for factor V, factor Va and for factor VaLC i n the absence of calcium. We conclude, therefore, that no s i g n i f i c a n t s e l e c t i v i t y for a p a r t i c u l a r class of l i p i d occurs. On the contrary, i n the presence of Ca , some s p e c i f i c i t y for the negatively charged PS occurs with VaLC, since the fluorescence quenching i s greater with PBPA- than with PBPC-labelled membranes. When the complexes V-Xa and Va-Xa were the energy donors, the f l u o rescence quenching was greater with PBPC than with PBPA. This i n d i c a tes some s e l e c t i v i t y for PC, i n contrast to each individual component where no s e l e c t i v i t y was detected. F i n a l l y , when the influence of calcium was studied, the results indicated that the fluorescence intensity corresponding to the f r a c t i o n of bound protein decreases as calcium concentration increases (Fig. 7a), r e f l e c t i n g the decrease of the a f f i n i t y of VaLC for the membrane (3). However, transfer e f f i c i e n c i e s are constant up to 8 mM calcium, suggesting that calcium does not change the composition of the binding s i t e (Fig. 7b). At 15 mM, the Et value i s greater for PBPA and decreases for PBPC. This suggests that the binding s i t e i s then enriched i n PS molecules. This result runs p a r a l l e l to the h i gher degree of phase separation induced by high concentration of Ca In conclusion, even though e l e c t r o s t a t i c forces between factors Va and VaLC with membranes are now well documented, the present r e sults demonstrate that hydrophobic contacts between these proteins and the hydrocarbon region of the phospholipids are probably also important . Conclusion The o v e r a l l features developed i n this study a l l imply that the strong binding of the proteins studied requires ion p a i r i n g with negatively charged groups located at interface. Such effects are c l e a r ly dominant i n the case of cardiotoxin and are probably s u f f i c i e n t to s t a b i l i z e CTX-heparin complexes documented for the f i r s t time i n t h i s report. The binding of CTX and VaLC to l i p i d interface seems also to be relevant to such a type of interaction . However, i n t h i s case, the occurrence of hydrophobic "burying" i s c l e a r l y documented. In the case of CTX, t h i s i s well proved and hydrophobic effect provides a better s t a b i l i t y of the toxin at the l i p i d interface. This leads to the suggestion that, although the use of heparin as an agonist against
+ +
+ +
3 mM
V Va VaLC V-Xa Va-Xa Without C a Va
With C a
PC-PS 50:50
-
-
PC-PS 80-20
3.84 3.04
3.84
-
o
r
I
58 46 21
X
FDA FD 62 49 0.20 0.64
0.21
Et o
r
I
Fluorescence of [Ca++] = 5 mM
a
65 56 47
77 59
*FD
FDA
0.14 0.28
0.23
Et
factor II EDTA I
62 40 34 38
86 59
FDA
0.35 0.45 0.38
0.32
Et
100
95.5 87.5 90 100 100
PC-PS
83.5
- 16.5
PC - PS + 5 % PDPA Δ % *DA 86 - 10 82.5 - 5.7 56 - 37.7 90 - 10 90 - 10
9
F
D
83.5
85.5 83.25 62 80 86
0.33
0.07
10 4.8 31 20 14 - 16 5
-
Et
0.25
PC - PS + 5 % PDPC Δ %
47
72 67
87 65.5
I
FDA or l
Fluorescence of factor IX EDTA LCa++J = 5 mM
Table I I I . Fluorescence i n t e n s i t i e s of the proteins i n the presence of unlabelled PC-PS (50:50) membrane (Ip), and of l a b e l l e d membrane ( 1 ^ ) · Δ i s the percentage of fluorescence decrease i n r e l a t i o n to Ip. Concentration of the proteins : 22.5 10~ M.
-
3.84
PBPA %
PBPC %
Lipids
Label
e x t r a
Table I I . F i n a l fluorescence i n t e n s i t i e s i n the absence (Ιρ$) and i n the presence of acceptor (IpDA.) P ° l t e d f o r i n f i n i t e phospholipid concentration, and energy transfer e f f i c i e n c y values (Et). 0.165 uM, factor i x For PC-PS (80:20) membrane 0.205 μΜ. factor II For PC-PS (50:50) membrane 0.080 uM, factor ixl 0.088 uM. factor II
PROTEINS AT INTERFACES
194
*
I
Ob
or
'DAb
20
[>"]
15
10
*M
Figure 7. a) Fluorescence i n t e n s i t i e s corresponding to the frac tions of VaLC bound to the membrane, with (I DAb) and without (I Db) acceptor, as a function of the C a concentration. ( [VaLC] = 22.5 χ 10"* M.) b) Energy transfer e f f i c i e n c i e s as a function of C a concentration. + +
9
+ +
12.
DULOS ET AL.
Plasma Proteins at Natural and Synthetic Interfaces
195
the toxic action of CTX has been proposed (33), i t probably would not be very e f f i c i e n t when dealing with membrane e f f e c t s . On the other hand, we propose that the polymeric "heparin-like" materials here i n vestigated, would be much more e f f i c i e n t , due to the higher s t a b i l i t y of CTX bound to these i n t e r f a c e s . The binding of CTX to l i p i d mixtures strongly modifies the thermotropic behavior of the l i p i d s and induces phase separation i n the gel phase, as a result of ion p a i r i n g between the Lys residues of the peptide and the negative charges of the l i p i d s (25,34). Such an ion pairing may be involved i n the binding of the vitamin K-dependent blood c l o t t i n g factors, p a r t i c u l a r l y , i n the absence of C a . However, the perturbation of the thermal t r a n s i t i o n of the l i p i d s i s weak. One can then propose that the main effect of C a would not be to form bridges between the proteins and the polar head groups of the l i p i d s . More probably, as recently documented from lanthanide luminescence (35), C a binding induces a conformational change which shields the Gla negative charges and allows a better approach to the interface by Lys and hydrophobic residues. The lack of s t r i c t s e l e c t i v i t y of the vitamin K-dependent factors for charged l i p i d s i s strongly supported by FET results presented here above and leads to the conclusion that neutral phospholipids are involved i n the binding s i t e . The composit i o n of the binding s i t e depends on the i n i t i a l composition of the mixture, suggesting that the c l o t t i n g factors are not able to modify the d i s t r i b u t i o n of the l i p i d s i n the plane of the membrane, at constant temperature, i n order to provide a more favorable binding s i t e . Comparing now these features to that observed on anticoagulant interfaces, i t appears that the binding of proteins i s probably stronger and less r e v e r s i b l e . This i s again well demonstrated by the behav i o r of CTX. The stronger binding to polymers i s not only due to the d i f f e r e n t nature of ion p a i r i n g with s u l f a t e or sulfonate groups, but also to an important hydrophobic contribution, as suggested by desorption experiments. When comparing the various polymers which only d i f f e r i n t h e i r charged groups, anticoagulant a c t i v i t y seems to be correlated to the easier release of T-AT from the interface. This conclusion makes more a t t r a c t i v e the idea that these polymers act as catalysts i n the i n a c t i v a t i o n at thrombin by antithrombin, as already proposed (36). Thus, active polymers would be endowed with heparinl i k e properties. However, i t must be pointed out that the experiment a l conditions used herein are far from physiologic where polymers are i n contact with a complex mixture of proteins and c e l l s carried by the blood stream. + +
+ +
+ +
Acknowledgment s We are g r a t e f u l to Pr H. Rochat and v e r s i t y , for providing cardiotoxin, Fougnot, Paris-Nord University, for compounds, to Pr H.C. Hemker and Dr s i t y , for providing factor V and to the pyrene-labelled phospholipids. This research was supported by "Polymères Hémocompatibles". One of an I.N.S.E.R.M. grant n° 835003.
Dr P. Bougis, Marseille-Nord Unito Pr M. Jozefowicz and Dr C. providing polymeric heparin-like Th. Lindhout, Maastricht UniverJ . Lalanne for the synthesis of a C.N.R.S. grant, GRECO 130048 us, J . Dachary, i s recipient of
196
PROTEINS AT INTERFACES
Literature Cited 1. Zwaal, R.F.A. Biochim. Biophys. Acta 1978, 515, 163-205. 2. Nelsestuen,G.L. Fed. Pro.,Fed. Am. Soc. Exp. Biol. 1978,37,2621-25. 3. Van De Waart, P. Ph.D. Thesis, Maastricht University, Holland, 1984. 4. Fougnot, C. ; Jozefonvicz, J . ; Samama, M. ; Bara, L. Annals of Biochem. Eng. 1979, 7, 429-39. 5. Boisson, C. ; Gulino, D. ; Jozefonvicz, J . ; Fischer, A.M. ; Tapon-Brétaudière, J . Thrombosis Research 1984, 34, 269-76. 6. Fougnot, C.;Dupillier, M.P.; Jozefowicz,M. Biomaterials 1983,4, 101-4. 7. Di Scipio, G. ; Hermodson, A. ; Yates, G. ; Davie, E.W. Bioche mistry 1977, 16, 698-706. 8. Bougis, P. ; Tessier, M. ; Van Rietschotten, J . ; Rochat, H. ; Faucon, J.F. ; Dufourcq, J . Mol. Cell. Biochem.1983, 55, 49-64. 9. Freire, E. ; Markello, T. ; Rigell, C. ; Holloway, P.W. Bioche mistry 1983, 22, 1675-80. 10. Parker, C.A.; Barnes, W.J.,in Fluorescence Spectroscopy;Pesce,A.J.; Rosén,C.G.;Pasby,T.L., Ed.; M. Dekker : New York, 1971, pp. 166-68. 11. Prigent-Dachary, J . ; Faucon, J.F. ; Boisseau, M.R. ; Dufourcq, J. Eur. J . Biochem. 1986, 155, 133-40. 12. Dufourcq,J.;Faucon,J.F.;Bernard,Ε.;Pézolet,M.;Tessier,M.;Bougis, P.;Delori,P.;Van Rietschotten,J.;Rochat,H. Toxicon 1982,20,165-74. 13. Dufourcq, J . ; Faucon, J.F. Biochemistry 1978, 17, 1170-76. 14. Tsernoglou,D.;Petzko,G.A. Proc. Nat. Acad. Sci. USA 1977,74,971. 15. Pézolet, M. ; Duchesneau, L. ; Bougis, P. ; Faucon, J.F. ; Dufourcq, J. Biochim. Biophys. Acta 1982, 704, 515-23. 16. Bougis, P. ; Rochat, H. ; Pieroni, G. ; Verger, R. Biochemistry 1980, 20, 4915-20. 17. Dachary, J . ; Dulos, E. ; Faucon, J.F. ; Boisseau, M.R. ; Dufourcq, J. Colloid and Surfaces 1984, 10, 91-99. 18. Einarsson,R.;Anderson,L.O. Biochim. Biophys. Acta 1977,490, 104-11. 19. Olson,S.T. ; Shore, J.D. J . Biol. Chem. 1981, 256, 11065-72. 20. Wong,R.F.;Windwer,S.R.;Feinman,R.D. Biochemistry 1983,22,3994-99. 21. Fougnot,C.;Jozefowicz,M.;Rosenberg, R.D. Biomaterials 1983,4,294-98. 22. Boisson,C.;Gulino,D.;Jozefonvicz,J.Thrombosis Research in press. 23. Hui,S.W.;Boni,L.T.; Stewart,Τ.P.;Isac,T. Biochemistry 1983,22,3511-16. 24. Dufourcq, J. ; Faucon, J.F. ; Maget-Dana, F. ; Piléni,Μ.Ρ.;Hélène, C. Biochim. Biophys. Acta 1981, 649, 67-75. 25. Faucon, J.F. ; Bernard, E. ; Dufourcq, J . ; Pézolet, M. ; Bougis, P. Biochimie 1981, 63, 857-61. 26. Lentz, B.R. ; Alford, D.R. ; Jones, M.E. ; Dombrose, F.A. Bio chemistry 1985, 24, 6997-7005. 27. Mayer, L.D. ; Nelsestuen, G.L. Biochemistry 1981, 20, 2457-63. 28. Mayer, L.D. ; Nelsestuen, G.L. Biochim. Biophys. Acta 1983,734,48-53. 29. Lecompte, M.F. ; Miller, I.R. ; Elion, J . ; Bennarous, R. Bio chemistry 1980, 19, 3434-39. 30. Mersten,K.;Cupers,R.;Van Winjgaarden,A. Biochem J. 1984,223,599-605. 31. Nelsestuen,G.L.;Kisiel,W.;Di Scipio,G. Biochemistry 1978,17,2134-8. 32. Dombrose, F.A. ; Gitel, S.N. ; Zouvalich K. ; Jackson, C.M. J . Biol. Chem. 1979, 254, 5027-40. 33. Lin,S.Y.S.;Hsia,S.;Lee,C.Y. J . Chinese Biochem.Soc. 1973,2,38-44. 34. Faucon, J.F. ; Dufourcq, J . ; Bernard, E. ; Duchesneau, L. ; Pézolet, M. Biochemistry 1983, 22, 2179-85. 35. Rhee,M.J.;Horrocks,W.D.;Kosow,D.P. Biochemistry 1982,21,4524-28. 36. Fougnot,C.;Jozefowicz,M.;Rosenberg,R.D. Biomaterials 1984,5, 94-99. RECEIVED February 18, 1987
