Unusual fucoidin-binding properties of chymotrypsinogen and trypsinogen

Biochi,,ttca et Fli,..,l ~ ~i,..d A c t ~

1037 ; 19q0) ..'"": -..,,""

?_27

El~'icr BBAPRO 33552

Unusual fucoidin-binding properties of chymotrypsinogen and trypsinogen Roy Jones Department of Molecular Emb~'ologa'. AFRC Institute oy Animal Ph)'su)logl" and Gem,tws Research. itah.aharr~ Camhrutge (U. &'.~

( I~.~,~cived30,:tobe; 1989)

Key words: Proteinase: Proacrosm; Sulphaled ~lysa.~'charide; Fertiliz. hon: Adhesion mcdc~,.'ule

Previous ~ork (Jones, R. (1987) Cell Biol. Int. Reports II, &it3 and Jones et al. ~i988) De, e!opment 102. T81--792) has shown that sperm proacrosin (the zymogen form of the acrosomal pcoteinase acrosin` EC 3.4.21.10) has the capacity to recognize and bind suiphated polysaccharides and that this property is importam [or the initial stages of fertilization in mammals, To investigate whether this behaviour is specific to Woacrosin` a variety of other proteiemv,es (chymotrypsin o g ~ ~psinngen, thrombin` elastase, plamminogen, pepsin, Sr~eptomyces gr~u,~ proteina~ and V8 pcoteimcse from Stapk)~i~'occm a r e m ) wire inuno~'lized on nitrocellulose and pcobed with I I~-~Ilfucoidin. Only chynmtry~nogen and trypsinogen retained significant amounts of the pcobe, with Ka values of 1.4- 10- 6 M and 3.0- 10- s M, respectively. Proteinase inhi~tors were ineffective as blocking agents suggesting that enzymic activity, is not involved in recognition. However, the tertiary struetm-e of the pcoteins is impoclant, since cleavage of intranmleodar disulphide bomls with 2-mtrcapioethanol reduced binding by 50-60%. Competition experiments with a varlet v o[ mono- and pol?.sacc~arid~ ~ellgest that tide number and disposition of sulphate FToups is critic~.d for interaction with basle residues on the protein. It is e o ~ u d e d that, like proacrosin, chymo~psinogen and trypsinngen are bifnnctional proteins.

Introduction The acrosome of mammalian spermatozoa contains a variety of hydrolytic enzymes the best known of which is proacr~in, the zymogen form of the trypsin-like enzyme acrosin 0evi,:wcd in i~,,~'. I~. Various type.~ of proacr~s,~n have been found in all mammalian spermatozoa investigated ,so far and it. some c a ~ s the protein has been characterized extensively ( e g . boar, 12,3]). Until recently it was thought that the only function of this pro~emase was to help spermatozoa penetrate through the zona pellucida, the extracellular matrix that surrounds mammalian eggs. However, doubts about this role have been raised on ~veral occasions [4,5] and ~n alternative hypothesis has been proposed in which proacrosin serves as an adhesion molecule to cross-link spermatozoa firmly to the zona [6-8]. This hypothesis is based on the ob~rvation that in addition to its na,~ent

Abbreviations: BSA. bovine serum albumin; PBS, phosphate-buffered saline; CRD. carboh~drat¢-rccognizin$domain; PMSF. phcnylmcth:,;sulphonyl fluo:idc. Collespnlldctg~: R. Jones. Department ot" M,.;~cular I(mbr',,(~logy. AFRC Instilutc of Animal Physiology and Genehcs Research, Babraham. Cambridge. Ct'.2 4AT. UK.

proteolytic activity. I~roacrosin ~I:.;;'~ has the capacity to recognize a,)d bind to carbohydrate m o i c t i c s on zona glycoprotcin:, :~.nd ac(~gly¢,,protems [9.iu]. Here we report that this sec.'ndary property tff .~perm proacrosin is also shown by chymotrypsinogen and t .rypsinogen. but not other closely related serine pi,~;eina~s. Thi~ dual capacity adds a new dimension to the functional significance ~,I" -:b,:..,e proteins. Materials ned Methmh Chcm,~,:v. Ah r~,utmc chemicals were of the highest purity available commercially and were purchased from BDtl, Pierce. Sigma or LKB-Pharmacia, Nat"Sl was obtained from Amersham International. Sources of proteinases investigated were as follows: Staphylococcus aureus V-8 proteinase (EC 3.4.21.19; Worthington); trypsinogen (EC 3.4.21 4; bovine: Worthington); chymotrypsinogen A (EC 3.4.21.1; bovine; Bochringer); elastase (EC 3.4.21.11; Porcine; Worthington); thrombin (EC 3.4.21.5; bovine; Sigma); pepsin (EC 3A.23.1; porcine; Miilipor¢). Ilighly purified fucoidin was a gift of Dr. G.M.W. Cooke. University of Cambridge, The polysaccharidc was conjugated with fluore~inamine before iodipation l Ill. Other poly~ccharides, oligosaccharides and monosaccharides were obtained from

0167-483N/90/$03.50 ~' I',SqO Elsevier Science Pubhshcfs B Y (Rion~cdical D,v*sion}

728 .~igma. Total carbohydrate was measured by the phenol suiphuric acid method using fucose as standard [121. Eleciropht,resis and Western Blotting. Proteins were :,cparated on non-reducing 8-1596 polyacrylamide gels containing SDS (SDS-PAGE 1131) and either stained with Coomassie blue R-250 in methanol/acetic acid/ water (40:7.73, v/v) or electrcbiolted onto nitrocellulose paper [14]. Blots were blcz,~;ed with 5% (v/v) BSA for 180 rain, drained and then probed with 13.5-10 ~ cpm/ml of [:2Sl]fucoidin for a further 120 win at room temlxrature. Fucoidin-fluoresceinamine was iodinated by the 'lodogen' procedure [151. Unbound probe was washed away with phosphate-buffered saline (PBS) and bound probe detected by autoradiography with preflashed X-ray film (Fuji). Labelled proteins were quantiffed by densitometry on a Jo)ce Lobel integrating densometer. Total protein was measured ny the method of Bradford 1171 using BSA as a stand:~rd. Solid phase assay for ['2~i]]ucoidin binding. In addition ~o Western blotting, biildtng of I;:Sl]fucoidin to

proteins was assessed quantitively on a solid phase assay system. Equimolar amounts (50-500 pmol) of purified proteins were dot blotted onto squares (1 cm :) of nitrocellulose, blocked with 5% ( w / v ) BSA in PBS for 180 rain and then probed with 0.1 ml of [~:~l]fucoidin t0.5. 106/ml in PBS) for 60 rain at ream temperature. Preliminary experiments showed that t~/: for fucoidin binding was 14 rain. Supports were washed three times (10 min each ~,ith shaking) in PBS and bound radicacti~ty was counted on an LKB gamma counter. For inhibition experiments, immobilized proreins (1013 pmol) were blocked with BSA as desc~'ibed above followed by various polysaccharides and monosaccharides (see Results) for 60 rain before addition of the labelled probe. The final volume was 0,1 ml. After incubation for a further 60 rain, samples were washed and counted. Nonspecific binding was assessed from the amcunt o~" [::Sllfucoidin retained on filters in the at)sence of target proteins. All experiments were carried out in duplicate.

Homology

1 50 VVGGRSAEPGAWPWMVSLQIFMYHHNRRYHTCGGILLNSHWVLTAAHCFK

Acr~in

Ref.

3

(l~)rcinc) Chymolrypsinogcn A

16 IVNbEEAVPGSWPWQVSLq. . . . .

48

DKTGFH;CGGSLINEN¥.VVTAAHC_. . . . .

58~

18

49 NSGYHFCGGSLIN._$SQW__VVSAAHCYK

52~

19

I 4O VVCL~TRAAQ§EFPFR_.VRLSM . . . . . . . . . . GC_.G_~LYAQDIV~!AA~¢¥S

50;

20

1 48 VVGGTEAQRNSW.._PSQI~LQ.__YR--$G$$WAHTCGGTLIRQNWVMTAAHeVO

46 (g

21

561 605 VVGGCVAHPHSWPWQV~t.RTRF. . . . . . GRHFCG_~.T~ISPC~V~TAAHCLt

33~

22

144 184 ~FGLSETEP..G6SFLYYAPFbGILGLAYPSISASGATPVFDNL. . . . . . . . .

12~

23

367 41 lVE§QDAEVGLSPM . . . . QVRLFRKSPOELLCGASLISDRMV~TAAH~L

42~

24

146 194 -G§FTAEQITKYSGEGbLA|~KFSP~EQNKHIaEVVKPATRSNNAETOVN

12'~

25

282 331 VG6..~PIqGGPI~FGALPPEQ_ADTDFGSSSSSVDG__6DTTISARVRDDIKAVL

I'/~

26

(bovine)

7 !V£.GYTCGANTVPYQVSL. . . . . . .

lryp~*n~en (l~)wnc)

Trs'psin (

Streptomyces gr,xeut~ )

[':lastase (l'~rcin~.,)

(human,* i)cp.~im~gcn A (~rcinc; Prothromhln (~)vin¢) V8 Prot¢ina~:

.Vtaphvle~ aur~ ) (

,wcu~

Bindin ( SIr(~LI,,I~'~ trom.~ p~o'p~atu~ )

Fill. I. Sequence homolol~y between N-ternmnal region of boar sperm acr(~tn, .~a urchil, bindin and several pro~einases. Positions idenw:al to acrosin are underlined. Computer alignment allowed for no mismatches. Numbers above the leUelrs refer to the l~.)sition of the anruno acid from the N-terminal end of the protein and the ix-rc~ntag¢ homology is relative to acrosin ll~.avy chain.

229 Results and

Discussion

In previous work v,e demonstrated that [~-'-~l]fucoidin binds to proacrosin from boar, bull, ram and human spermatozoa on Western blots and that recognition is selective as the probe doe~ not bind to a wide range of other proteins present in crude sperm extracts [9,17]. Proacrosin belongs to the family of serine proteinases as one of its active forms, 8-acrosin, has similar substrate specificity to pancreatic typsin [1]. In addition, a comparison to the N-terminal sequence of boar sperm proacrosin heavy chain with aligned residues selected from other serine pvoteinases shows considerable homology with chymotrypsinogen, irypsinogen, elastase, plasminogen and prothrombin (Fig. 1). Therefore, to investigate whether binding of [12,,l]fucoidin was common to all serine proteinases or whether it was a specific property of proacrosin, a variety of proteins were t~,ted for there ability to behave as ligands for the polysaccharide. The mechanism of the interaction has also been investigated.

i

ii

iii

iv

v

vi

vii

viii

0

Y

IP

Binding of [ I"~i]fucoidin to proteinases When West-,m blots containing six mammalian and two bacterial proteinases were probed with [l'~llfucoidin, strong retention of the polysacchanae was observed only over chymotrypsinogen and trypsinogen (Fig. 2). A weak signal was present on a minor component in the sample containing elastase but plasminogen, pepsinogen, prothrombin, Streptomyces griseus trypsin and Staphylococcus aureus V8 proteinase all faibd to retain significant amounts of [12Sl]fucoidin. As found previously with proacrosin 19], uptake of [12Sl]fucoidin was not affected by the presence of 2 mM p-aminobenzamidine (a potent inhibitor of proacrosin activation and ~crosin activity) or even a cocktail of inhibitors (1 mM PMSF plus 10 ~tg/ml aprotinin plus 10 ttg/mi leupeptin). On the contrary, ul~take was always enhanced by 5-15%. Thus, enzymic activity is unlikely to be involved in interaction of the protein wi:h the polysaccharide probe. The above results were confirmed by a quantitative comparison of equimolar amounts of each protein on a solid phase assay s~stem. Resu!t~ showed that chymotrypsinogen and trypsinogen were the only pro',eins tested to bind [tZ~l]lucoidin (Fig. 3). Saturation kinetics revealed that 1 ~g of chymotrypsinogen bound a maximum of 23 ng Of fucoidin and i ~tg of trypsi~logen bound 7 ng of fucoidin (Fig. 4). Scatchard plot analyses gave K d values of 1A. 10- ~ M for chymotrypsinogen and 3.0.10- s M for trypsinogen (Fig. 4). The curvilinear functions suggests the presence of several binding sites on each protein with different affinities for fucoidin.

Effect. of p,otem folding on uptake of [t:Ji]fucoidin It is well known that many properties are crucially dependent on their secondary or tertiary structures. To

Fig. 2. SDS-PAGF. of protcinase- in) (.'~a~mas,dcblue slam of (l) trypsinogcn; (ii) chymott3/psinogun;(iii) plasmmogei,;(iv) Staphv/oc ( ~ auregY V8 prot,:a~; (v) elasta,,,¢; (vl) pep.sin: (vii) lhrt)mhin; (~,iii) Streptomyce,"lCrtse,,¢ proteinase: 5 tO pg protein/track. (h) Autoradlograph of Western blot laken from :'- parallel gel to (a) probed withIt2, Ilfucoidi,,

examine whether intramolecular folding due to disulphide bonds is important fo" ~he ability of chymotrypsinogen and trypsinogen to hind [~2~I lfucoidin, proteins were reduced with 5% (v/v) mercaptoethanol for 5 min at 100°C before SDS-PAGF and Western blotting. Autoradiographs were quantified by densitometry. Results showed that, relative to non-reduced controls, this treatment causes a decreased in uptake of labelled probe by 63% [or chymotrypsinogen and 52% for trypsinogen. Values ol 61% and 55%, respc-ctively, were obtained using the solid phase assay system. Since chymotrypsinogen contains two internal disulphide bol~ds and trypsinogen six, considerable anfolding is likely to occur following reduction. These results suggest, therefore, that tertiary protein structure is important for fucoidin binding.

Carbohydrate binding properties of chymoto'psinogen and trypsinogen Carbohydrate-recogniz.ing domains (CRDs) have been d e l e t e d in a diverse range of vertebrate and inverlebrate proteins, e.g., asialoglycoprotein receptors,

230

~48Fi

i /

Y I c3 0

~. u

!

7ai-

10 ¢

s~,~/

,,~ ~

~ ..... 0

0

~00

200 7' MOLES

300

4o Bz I0-~1M

O0

¢ 3 "-

7_0

35

7[0

4C

PROTF-IN

Fig. 3. ~ ! i d p h a ~ hindin 8 assay of proteinases w~lh [~Sl]fucoidin probe. Eouimolar amounls ol :he proteinascs listed in Fig. 2a 0-vii) w,:,: immoniii?cd onto ai~-occl:ulos¢ paper and probed with 0.5- l06 c p m / m l of I~'~l]fucoidin (1360(0')cpm//~g). Significant binding wcs ohtain,=d only to ch~motiypsinogcn (~ , --~) and ~cyp>inog.en (11 -II); ~11 ot~:r proteinascs ,este,d (F_J E3: see Fig. 2a) were close to hackgrc ,rod.

apoprot¢in of pulmonary surfaoant and sea urchin lectin [27]. These CRDs have beer shown to reside in a series of 18 i~.variant amino acids arranged in a structural motif. Two types of CRD can be distinguisheu, C-typ.~ (Ca 2+ and disulphide dependent) and S-type (:hiol dependent). However, neither of these CRDs were detected in chymotrypsinogen or trypsinogea ~vhen the appropriate motif search was carried out. Interestingly, we also failed to detect CRDs in the primary seqaence of bindin [261 from sea urchin spermatozoa {$trongrlorentrotus purpuratu: ). Bindin is a protein found within the acrosomal granule th,qt is responsible for the ".'ttachment of acrosome-reacted spermato~o., to tl,~ v-tilline layer of sea urchin eggs [28 I. It is classified as a iectin as it has haemagglutinafing properties [29]. Kecently, it has been sequenced [25] and its carbohydrate-binding properties have been elucidated [30,31]. It would seem that there is preference for sulphated polysaccharide str,.ctures and that the composifon of the polysaccharide backbone is less important than thc position and arrangement of sulphate ester groups. To investigate the mechanism of interaction of [Á2sl][ucoidin with chymotrypslnogen and tD'psinogen, a variety of monosaccharidcs and polysaccharides were tested for their ability to inhibit competitively the uptake of the fucoidin probe in the solid phase assay. Results are summarized in Table l. No inhibition was observed with any of the monosaccharides listed even at high concentrations (0.4 M). However, unlabelled fucoidin, dextran sulphate (500 kDa and 5 kDa) and mannan ~,3,~ kDa) were all effective to varying degr~s.

72 "7,

121 7_ D 3t~ 0

40[ 8,

I 0 " 9 ~

o

c/

0

lag ~DDED Fig. 4. Binding of I;~Slllucoidm to (a) chymotrypsmogen am:l (b) trypsinogen. 5 ~ag of ta.'get protein was immobilized on mtrocellulose and incubai,~i wi,h various amounts of [Izsl]fu~idin in a total volume of O.l ml in PBS (pH 7.2). Specific binding was determined from Io:al cpm bound minus background. Inset shows a Scatchard plot of the data ob'cie: 0 f:om the binding assay.

Non-sulphated dextran ~500 kDa) on the other hand, did not block, Collectively, these results suggest that interaction between [l:silfucoidin and chymotrypsin+ gen/trypsinogen is mcdiated primarily by sulpha,e groups on the polysaccharid¢. In addition, die densby and arrangement of sulphate groups are important, since glucose 6-sulGha'e..glucosamine 2.3-disulphat¢, hyaiuronic acid and chondroitin sulphates A and C arc ineffective inhibitors (.f fucnidin binding. These conclusions are similar to t[os¢ drawn by DeAng©lis and Glabc [30] for the interaction between bindin and sulphated fucans. Based on chemical mu~agencsis studies DcAngelis and Glabe [311 have also suggested that basic amino acids on bindin are the :,rincipal residues mediating interaction with sulphate groups. If so, it is to be

231

Fig. S, 'Rihboa' modeb ¢/tire ~ ~ ef (a) d k ~ and (b) ~ . la each Imir the model on the ~ t ~ side has been rotated th~m~ 90" from left to right. Lysinenaddu~ are coiotwedred, aqgininesyellowa~l histidiaes light blue. The dart blue line is the pol~eptide backboneand the pink Rliom fepfeseat intenud ~-sheets. Residuescomprising the active site (histidin¢ asparagine, senne) are shown with projectingaromatic rings and the C-terminal region as an e-helix.

expected that such residues would lie mainly on the surface of the molecule. The tertiary structure of bindin is not known but Fig, 5 shows 'ribbon' models of chymotrypsinogen and t rypsinogen obtained from Evans and Sutherland computer programmes. Chymotryosinogen contains 14 lysines, two arginines and three histidines per mol©cule. By rotating the models around their vertics! axis it is po~ible t.o judge the relative TABLE I ICjo (concentraticm /or .SO~ tnhmmon) o/ various m¢,~o. , (it. aJ J mdysacchari~te~ for bmdm~ of / I 2si lfucl~tdtn te ch.rmot~psmoge- m , frypstaogen Where no inhihitk,n was obtained the values shown are the hight:st concentration tested.

Fucoidin Dextran su',ph:'.h: {500 kDa) Destr~n salp~at, (5 kDa) [:)¢xlr;,n ~500 kDa) Man..,",-. Choraln>itin sulphate A Chom:lr0dtin sulphate C Hyaluronic ~'id D( + )-Xylos¢ D( ÷ )-Fucos¢ 13( + ~-Manno~¢ D( + )..Galactose 13( + )-Glucose o - G l m a n e 6-sulphate o-Gluco~minc 2,3-sulphat© Lactose

Chymotrypsinc,gen

Trypsinogen

0.22 p M 0 08/AM ? ~2 .aM > 160/~M :]90 ~M > 400 ~M > 400 tAM > 2 mii/ml > 400 mM > 400 m M > 4NO m M > 400 m M > 400 m M > 400 m M > 400 m M > 400 m M

0.98 p M 0.10 ;~M > 20 p M > 160 p M > 1400 p M > 4(}0 I~M > 400 tt M > 2 mg/ml > 400 mM > 400 m M > 400 m M > 4o0 m M > 400 m M > 400 m M > 400 m M > 400 m M

position (surface or internal) of the above.named residues. In both chymotrypsinogen and trypsinogea all the lysincs, arg/nines and histidines are located either on surface loops or on fl-shee~s that approach the exterior face of the molecule. Thus, the position of these basic residues is compatiuIc with the hypothesis that they are primarily responsible for interacting with specific sulphate groups on polysaccha~des or synthetic compounds. Essentially similar results have been obtained in modelling experiments with boar sperm proacrosin (Coadwell. W.J. and. ;ones, R., unpubhshcd ob,~rvati,~r.~) i.3 which there is a pre.oond,'rance of arginines on the surface (boar proacrosin co,,'ains 15 lysines. 25 arglnines and seven histidines per molecule). Binding of fucoidin, tl~erefore, probably involves a large number of low affinity sites. in conclusion, therefore, it has been shown qualitatively and quantitatively that fucoidin binds selectvely :c :h:,'motr~sinogen and trypsinc'gcn. The interaction is oi moderate affinity and i; inhibited by certain of odnhated .nolysacch:.ride, but not by conslituent monosaccharides or non-suiphared polymers. The results suggest that the polysaccharide backbone does not contribute directly to the interaction with fucoidin except insofar as it provides the correct spacial orientation of sulphate ester groups that are critical for binding. Relayed to this is the three-dimensional structure of the target protein. Moderate perturbation of tertiary form by reduction of intramolecular disulphide bor~ds causes a significant decrease in fucoidin binding. Presumably this is due to re-arrangement of surface orientated basic

232

residues such as lysines or arginines tha;. carry a positive charge for interaction with sulphate groups on the polysacchadde. Lastly, it is interesting 1o note that fucoidin bir, ds selectively to several other well-known adhesion proteins. These include yon Willebrand factor, laminin and thrombospondin [321. In fact dextran sulphate (M, = 500000) is a more potent inhibitor of yon Willebrand factor binding than fucoidin [32] Thvs, sea urchin bindin, mammalian sperm proacrosin, c h y m e t o p sinogen and trypsinogen represent an emerging class of bifunctional proteins that have evoh'ed independently the capacity to interact with sulphated polysaccharides. It is conceivable that this property may even usurp the

original function of the protein, e.g., sperm proacrosin [9]. These obervations may have important consequen~.es lot the action of active trypsin and chymotrypsin on glycoprotein substrates. Acknowledgements ! am indebted to Mr. J. Coadwell for carrying out the computer graphics and Mrs. Lirlda Norton for preparation of the manuscript. I also thank Dr. Dennis Beale for helpful discussions.

Rderences I Hedfick, J.L..Urch. U.A. and Hardy. D.M. (198~,)in Enzymes in AgricuRural Biotech~.ology (Shoemaker. S., Sonnet. P. and Whitaker. J., eds.), pp. 55-73. ACS Books, Washington. IX', 2 Polakoskl, K.L and Parnsh, R.F. (1977) J. Biol. Chem 252. 1861'~-1894. 3 Fock-Nuzel. R., Lottspeich, F.. Henschen. A.. Muller-L~terl. W. and Fritz. H. 0980) Hopper-qeylers Z. Phyiiol. Chem. 361, 1823-1828. 4 Bedford. J.M. and Cross, N.L. 0976) J. Repr(',d Fcrtil 54. 385-392.

5 Sating, P.M, (198:) Proc. Natl. Acad Sci. U S A 78. 6231-6235. 6 Fness, A.E.. "t'~pfer-Peterscn,E., Nguyen. H a r d Schill. W-B. ~1987) Histochendstry 86, 297-303. 7 Jones, R. (1987) Cell Biol. Int. Rep. II. 633. 8 Jones. R. and Brown. C.R. (1987) Expl. Cell Re~ 17). 503-.~)8. 9 Jones. R.. Brown, C.R. and Lancaster. R.L. (1988) Dcvel,~ment 102. 781-792. I0 ]'6pfer-Petersen.E. and Henschen, A. (1988) Biol. Chon. HoppeSeyler 369, 69-76. II Glabe, C.(~., Hart~ P.K and Ro.~n. S.D, (1963) Anal. Biochem. 130 287-294. 12 Dubois. M., Giqes, K.A., Ha~Iton, J.K., Rebers, P.A. and Smith, 3. (19661 Anal. Chem. 7,S,350-356. 13 Laemmli, U.K. (1970) Nature (London) 227. 680-665. 14 Towbin, H., Staehehn. T.T. and Gordon, J. (1979) Proc. Nail. Acad. Sci. U S A 76. 4350-4354. 15 Markwell. P.J. and Fox. C.F. (19781 ~iochemist~ 17. 4807-4817. 16 Bradford, M.M. (1976) Analyt. Biochem. 72. 248-254. 17 Jones, R. (1989) H u m a n Reprod. 4. No. 5. 550-557. 18 Brown. JR. and Hartley. B.S. (1966) Biochem. J. 101, 214- 228. 19 Mikes, O.. Holeysossky, V., Tomasek0 V and Sorm, F. (1966) Biochem Biophys. Res. Commun. 24. 346-352. 20 Olafson, R.W.. Jurasek, L.. Carpenter, M.R. and Smillie. L.B. (1975) Biochemistry 14. 1168-I 177. 21 Sh,)tton.D.M. and Hanley, B.S. (1973) Biochem. J. 131,643-675, 22 M21;nowski, D.P.. Sadler. J.E and Favie. E W . (]984) [tiochemistry 23. 4243-4250.2~. 23 M,)ravck, L. and Kostka. V. (1974) FEBS Left. 43, 207-211. 24 MacGillivray, R.T.A ~nd Da:~e, E.W {1984) Biochem. 23. 1626 - ! 63,4. 25 Ambler, R.P, (1980) Phil. Trans. R. Soc. London B 289. 321-33. 2'~ Gao. B,, Klein. L E . , Britten. R.J. and Davidson, E.H. (I986) Proc. Natl. Acad. Sci. USA 83, 8634-8636. 27 Drickamer. K. (1988) J. Biol. Chem. 263, 9557-9560. 28 Vacquier. V.D. (1986) Trends Biochem Sci. 11.77-81. 29 Glabe, C G.. Grabel, L.B., Vacquier. V.D. and Rossen. SD. (1982) J. Co)) Hiol. 94. 123-128. 30 DcAng¢lis, P.L and Glabe, C.G. (1987) J. Biol. C h e m 262, 13946-13952. 31 DeAngelis, P.L. and Glabe. C . G (1988) Biochemistry 27, 8189-8194. 32 H,~ylaerts.M . (')wen. W,G. and Cullen. D. (19~,t)J. Biol. Chem. 259. 5670-567 ").