J.

Mol. Riol. (1991) 221, 615-621

Serpin Tertiary Structure Transformation Penelope Stein-f Department of Haematology University of Cambridge Hills Road, Cambridge CB.2 Z&H, U.K.

and Cyrus Chothia Cambridge Centre for Protein Engineering and MRC Laboratory of Molecular Biology Hills Road, Cambridge CB2 2&H, U.K. (Received

26 February

1991; accepted 20 May

1991)

Previous crystallographic analyses have demonstrated that proteolytic cleavage of the serpins can result in a dramatic transformation of their tertiary structure. Some 16 residues on the amino terminal side of the cleavage site are inserted into a large b-sheet to become a central strand, separating the two cleaved residues by about 70 A. We have determined, in outline, the nature of the conformational change responsible for this transformation. After cleavage, a fragment of the protein, consisting of an cc-helix and three strands of /?-sheet, moves away from the rest of the structure to make the space for the new strand. This movement involves a new type of structural change: sheet residues in the small fragment slide along grooves in an a-helix that belongs to the rest of the protein. The general conservation of residues in the regions between the small fragment and the rest of the protein imply that the same mechanism will be found in all serpins that undergo this tertiary structure transformation. Keywords:

serpin; ovalbumin;

a,-antitrypsin;

structure

residues is cleaved if this inhibitor-protease complex is dissociated or on exposure to certain non-target proteases. In most members of the family, cleavage is followed by the insertion of 16 residues on the amino terminal side of the reactive centre into the middle of a large B-sheet so as to form one of its central strands. This change places residues 358 and 359 at opposite ends of the molecule, about 70 A apart (1 A = @l nm), and results in loss of inhibitory activity (Fig. 1). This description of the tertiary structure transformation in the serpins was first proposed by Loebermann et al. (1984) on the basis of the cleaved structure of a,-antitrypsin alone. The recent determination of the structures of the cleaved (Wright et al., 1990) and uncleaved (Stein et al., 1990) forms of ovalbumin, a serpin that does not undergo the transformation, has confirmed the validity of their proposal. Here, we describe the nature of the conformation changes that allow this dramatic and unusual transformation to take place.

1. Introduction Recently it has been demonstrated that the tertiary structures of most members of the serpin family can undergo a dramatic transformation. The serpins are a family of homologous monomeric proteins of some 400 residues with diverse functions, many of which play important physiological roles through their inhibition of the activities of certain serine proteases (for a review, see Huber & Carrell, 1989). They bind to target proteases, through a peptide segment that includes residues 358 and 359$ (the reactive centre loop), forming a stable complex. The peptide bond between the reactive centre

t Present address: Department of Medical Microbiology and Infectious Diseases, 1-41 Medical Sciences Building, University of Alberta, Edmonton, Alberta, T6G 2H7, Canada. 1 Residue numbering and secondary structure assignment is based on sequence alignment with a,-antitrypsin (Huber & Carrell, 1989). 615

0

I991 Academic

Press Limited

P. Stein and C. Chothia

Figure 1. Schematic drawings of the structures of uncleaved ovalbumin (a) and cleaved cc,-antitrypsin (t)) pr~ducetl using the program, RIBBON (Priestle. 1988). strands of p-sheets art=rPpresent,ed bp arrows and a-helices by helical ribbons. The region on the amino terminal side of the cleaved reactive centre bond in a,-antitrypsin (strand &A) and thr homologous intact sequence labelled. h. helix: s, strand.

2.

in ovalbumin

are shaded

black.

a,-Antitrypsin and Ovalbumin as Examples of the Two Tertiary Structures

At present there is no serpin that undergoes the tertiary structure transformation and for which atomic structures of both states are known. The atomic structure of a,-antitrypsin is known only in the cleaved form (Loebermann et al., 1984). The structures of both plakalbumin, a cleaved form of ovalbamin (Wright et al., 1990) and uncleaved ovalbumin (Stein et al., 1990) are known, but this serpin does not undergo the tertiary structure transformation. Heat stability (Carrel1 $ Owen, 1985; Pemberton et aE., 1989), nuclear magnetic resonance (Gettins & Harten, 1988) and circular dichroism studies (Bruch et al., 1988) show that the tertiary structure rearrangement following cleavage is a characteristic feature of serpins with inhibitory properties. a,-antitrypsin, a,-antichymotrypsin. including antithrombin III, Cl-inhibitor and protease nexin 1. Serpins without inhibitory activity can be cleaved at sites that align with the reactive centres of the Among these non-inhibitory serpins, inhibitors. there is evidence for the conformational change in thyroxine-binding globulin and cortisol-binding

Elements

of secondary

st’ructure

referred to in the text are

et al.. 1988). but not in ovalglobulin (Pemberton bumin or angiotensinogen (Bruch et al.. 1988: Gettins, 1989; Stein et al.! 1989). The crystal structures of plakalbumin (Wright et al., 1990) and intact ovalbumin (Stein et al.. 1990) confirm that the native conformation is retained after cleavage. In ovalbumin, P-sheet A lacks the fourth strand (s4A) and consists of only five strands, compared with six strands in cleaved a,-antitrypsin (Fig. 1). The peptide segment corresponding to strand s4A is exposed and forms an a-helix. Thus, both intact ovalbumin and plakalbumin provide models for the inhibitory serpins in their native, uncleaved forms. They imply that the intact reactive centre lies on an exposed peptide loop and that the typical rearrangement on cleavage involves insertion of the residues on the amino terminal side of the cleavage site in the centre of sheet A to form an additional et al. strand (s4A), as predicted by Loebermann (1984). This is consistent with the known suscept$$tv to proteolytic cleavage of residues of strand s4A ‘in the intact structure (Kress et al.. 1979: Potempa et al., 1986). In the analysis described here we have used the atomic co-ordinates of intact, ovalbumin (Stein et

Serpin

al., 1990) and et d., 1984): Brookhaven (Bernst’ein et respectively.

cleaved a1 -antitrypsin (Loebermann which are both available from the Protein Structure Data Bank al., 1977), entries 1OVA and 5AP1,

3.

The Effect of Sequence Differences on the Structures of Ovalbumin and a,-Antitrypsin

Ovalbumin and cr,-antitrypsin differ in sequence; at homologous positions only 30% of the residues are identical, and insertions and deletions are found in certain loop regions. Sequence differences in homologous proteins are accommodated by changes in structure. Loops and peripheral elements of secondary structure may change their fold, while t,he major elements of secondary structure retain t#heir fold, but may change the exact details of their conformation and usually shift in relation to each other by small amounts (Chothia & Lesk, 1986). Therefore, in our comparison of a,-antitrypsin and ovalbumin. we must distinguish structural differences that are clearly a consequence of the tertiary structure t,ransformation from those that arise from just the sequence differences between the two proteins. Examination of the two proteins shows that significant difference in fold is confined to a number of loops. Residues 27 to 80, 93 to 147, 148 to 174, 179to192, 195to246,247to257,260to277,281 to 343 and 361 t#o 391 have similar local structures in t,he two proteins. If these regions in ovalbumin and cr,-antitrypsin are individually superposed, the rootmean-square (r.m.s.t) difference in the position of equivalent’ main-chain atoms (N, C”, C and 0) varies between 0.7 and 1.2 A, with the exception of the region 93 to 147 for which the r.m.s. difference is 1.8 8. The main reason for this high value is that have structures within this region secondary different relative positions in the two proteins, as discussed below. These results show that the two proteins contain some 330 residues with the same local conformation, about 85% of each molecule. Residues 343 to 358 have quite different positions, not’ because of evolution but because of the tertiary struct’ure transformation. Thus, the large sequence differences between ovalbumin and a,-antitrypsin produce changes in folding of only a few surface loops. Sequence differences in regions that retain their fold will produce small changes in structure. For two protein structures with 30% residue identity and low experimental errors, we would expect the or*erall r.m.s. difference in position of the main-chain regions that do not change their fold to be approximately 1.5 19 (Chothia & Lesk, 1986). Large fragments would be expected to have differences somewhat smaller than this overall figure. These figures place limits on the significance for the t)ertiary structure transformation of small differences bet)ween the two structures. 7 Abbreviat’ion used: r.m.s..

617

Tertiary Structure Transformation

root-mean-square.

4. Ovalbumin and a,-Antitrypsin contain Two Large Fragments with the Same Fold but Different Relative Positions (a) Re.sidues in the two fragments Visual inspection of ovalbumin and a,-antitrypsin shows that each protein contains two large fragments that have the same fold but different relative positions (Fig. 2). This impression is confirmed by fits of the atomic co-ordinates of the two molecules. The larger of the two fragments (fragment 1) consists of t’he six regions listed in Table 1. It contains 229 residues comprising helices A, B, C, G, H and I, sheets B and C and strands s5A and s6A of sheet A. The r.m.s. difference in positions of equivalent main-chain atoms of these regions in the two proteins is 1.61 8. The fit is improved. with an r.m.s. difference of 1.1 A, by exclusion of helices G and H and 14 further residues in loops and turns. The positions of the G and H helices: relative to the rest of fragment 1, differ by about 3 A between the two proteins. Only four of the 19 residues that form these two helices are the same in cc,-antitrypsin as they are in ovalbumin. Mutations that change the shape or volume of residues buried in the interface between G-H and the rest of the protein include (going from a,-antitrypsin to ovalbumin) Leu267 to Ile, Ile272 to Leu, Phe275 to Trp. Ser237 to Met, Va1239 to Ile, Phe253 to Leu and Va1371 to Leu. The smaller of the two fragments (fragment 2) consists of four regions listed in Table 1, It contains 63 residues comprising helix F and strands slA, s2A and s3A of sheet A. The r.m.s. difference in the fit for this fragment between the two proteins is 1.25 8.

(b) Relative position

of the two .fragments

The relative positions of the two fragments in ovalbumin and a,-antitrypsin can be determined by first superposing the two structures using a leastsquares’fit of the main-chain atoms of fragment 1 and then calculating the additional translation and rotation required to superpose fragment 2. This calculation shows that fragment 2 in a,-antitrypsin, compared to ovalbumin, has shifted by a translation of 3.9 A and a rotation of 6” away from fragment 1 (Fig. 2).

(c) Flexible joints

between the two fragments

Helices D and E do not form part of either fragment. These helices pack against fragment 1: helix D against helices A, B and C, and helix E against helix B. The two helices are covalently linked to fragment 2, at the earbonyl end in the case of helix D and at both ends in the case of helix E (Fig. 2). These helices move to accommodate the shift of fragment 2 relative to fragment 1 in a similar, but not the same, direction. Helix D shifts by 4 a and helix E by 3 A relative to fragment 1. In addition there are smaller conformational changes

618

P. Stein and C. C’hothia

Figure 2. Ribbon representations of the structures of uncleaved legend to Fig. 1) showing the different relative positions of the 2 (unshaded); flexible joints that move to accommodate relative a,-antitrypsin and the homologous intact sequence in ovalbumin

ovalbumin (a) and cleaved a,-antitrypsin (b) (see the fragments: small fragment (hatched): large fragment shift of the 2 fragments (stippled): strand ~4.4 in (black).

Table 1 Fits of intact ovalbumin to cleaved antitrypsin different Residue spans used for superposition

Fragment Fragment

1

Fragment

1 “core”

Fragment

2

Fragment 1 and fragment 2 Fragment 1 “core” and fragment 2 t Main-chain

atoms

after superposition

u.sing

fragments Number of residues in fragment

r.m.s. deviation for main-chain atoms? (A)

27-80 195-246 247-257 260-277 S-343 361-391 27-80 195-210 215-224 226-232 237-245 247-256 282-323 327-343 361~-377 380-391 109-124 141-147 148-174 179-191 see above

229

1.61

194

1.09

63

I.25

292

2.17

see above

167

1.86

used for superposition

are N, C”. C and 0.

Serpin

S3A

S5A

Tertiary

s2A

(a)

Figure

3. (‘ross-sections

(b)

through space-filling models of uncleaved ovalhumin

illustrating the change in packing at the interface strand s4A (white): strands s2A, s3A. s5A (grey);

(a) and cleaved r,-antitrypsin (b) between the 2 fragments on moving from uncleaved to cleaved forms: helix B and other residues packing against sheet A (black).

in the peptides linking these helices to the fragments. However, sequence differences between the two proteins will contribute to these differences in position. so we can not determine the exact extent to which the changes are a response to the tertiary structure transformation. Residues 192 to 194 link the end of strand s3A in fragment 2 (residue 191) to fragment 1 (residue 195). This peptide segment, through small changes in forms a hinge about which the conformation, relative movement of the two fragments takes place.

(d) Comparison

619

Structure Transformation

of a-antitrypsin

and plakalbumin

Engh et al. (1990) carried out a molecular dynamics simulation of the tertiary structure transformat’ion in u,-antitrypsin. In the course of this work they made some comparisons of the structures of cleaved a,-antitrypsin and plakalbumin (cleaved ovalbumin which does not undergo the structure transformation). They found that strands slA, s2A and s3A of sheet A and helix F shift relative to the remainder of the structure to accommodate the new strand. They also discuss the position of the hinge for insertion of the strand and note differences in the positions of helices D and E.

5. The Change in Packing of the Small Fragment against Helix B and Accommodation of the New Strand in Sheet A Fragment 2 is covalently linked to fragment 1 by helix D and at the end of strand ~3-4. ln addition, fragment 2 is linked to both ends of helix E which packs against fragment 1. The changes in these regions that accommodate the relative shift of the two fragments were described above. The other area of contact involves strands s2A and s3A of sheet A in fragment 2. These strands pack against residues 244 and 384 and helix B in fragment 1 and, in the uncleaved form, strand s3A hydrogen bonds to strand s5A which is also in fragment 1. Changes in contacts in these regions are crucial to the tertiary structure transformation. The relative shift of the fragments on going from the uncleaved to the cleaved forms separates the adjacent strands s3A and s5A. changes the packing of part of sheet A against helix B and creates a space into which the new central strand (s4A) can pack (Fig. 3).

(a) The packing

of the two fragments

in ovalbumin

The packing of sheet A of fragment 2 against helix B of fragment 1 in ovalbumin is illustrated in

P. Stein and C. Chothia

Figure 3(a). The helix axis is tilted relative to the strand direction and at the interface there is “ridge and groove” packing of side-chains. This packing is unlike that usually found at helix-sheet interfaces (Janin & Chothia, 1980). /?-sheet hydrogen bonds between strands s3A and s5A are formed for only part of the strand length. They do not occur between residues 190 to 192 of strand s3A and residues 339 to 341 of strand s5A, making a gap in the A-sheet near the hinge region. (This gap accommodates part of the inserted strand when the tertiary structure transformation takes place. explaining why larger changes in packing occur at the opposite end of the sheet, as described below.)

of the Asn 186 side-chain allows it to hydrogen bond to the carbonyl oxygen atom of residue 56 in both the cleaved and uncleaved forms (ND2-0 distance> of -3.0 a and NDS-CO a,ngle of - 115” in both struct.ures). There are also small differences in t,he conformation of other residues. However, the determination of the exact role of both the small and large differences in side-chain conformations will require high resolution structures of cleaved and uncleaved forms of the same serpin. The insertion of residues 343 to 358 into the sheet results in a large reduction in their acctessiblr surfact, 1971). In ovaibumin these area (he 8~ Richards, residues bury 55 ‘?d of their surface; in cr,-antitrypsin they bury 8504.

(b) The packing of the two fragments in a,-antitrypsin and accommodation of the inserted strand

(c) Neyuanc:e conservation of residues involved in paclci,ng at the interface between the tw?o,frcqmrnts

In cleaved a,-antitrypsin, strand s3A retains interactions with helix B, although it is translated relative to its position in ovalbumin to make space for the new strand s4A (Fig. 3). M-Carbon positions of residues 190 and 192 of strand s3A are shifted by only 2 to 3 A relative to their positions in ovalbumin. This does not involve any repacking of their side-chains and residue 384 (Fig. 3), just’ small adjust’ments in conformation and contacts. Larger shifts are found further down the strand, where it forms b-sheet hydrogen bonds with strand s5A in the uncleaved structure. u-Carbon positions of residues 184, 186 and 188 move by about 5 A in a direction perpendicular to the strand lengt’h. (‘omparison of the residue packing in ovalbumin and cc,-antit’rypsin (Fig. 3(a) and (b)) shows that t,he side-chains of the sheet, residues “slide” in the groove of helix H int#o which they are packed. Three helix ridges are formed by residues 53 and 56; by 57. 60 and 63, and by 61 and 64 (Fig. 3(a)). On going to the cleaved form, residue 186 moves between 53, 56 and 57, 60; 63 to occupy the space of residue 116: residue 184 moves between 60. 63 and 61, 64 to occupy the space of 118 (Fig. 3). St,rand s2A is translated with &and s3A and makes no significant contacts with fragment 1 in the cleaved &ucture. Tn both ovalbumin and a,-antitrypsin the interface between the two fragments is close packed. This was found by calculating the volumes occupied by residues at the int,erface between t’he two fragments using the Voronoi polyhedron procedure (Richards, 1974). In ovalbumin and a,-antitrypsin the volumes occupied by the residues are close to those they occupy in crystals of t’heir amino acid ((‘hothia. 1984). (lertain side-chains at or near the interface between the two fragments show large differences in caonformation. Residues wit,h x1 differences of - 120” are Leu/Tyrl IO. Leu/Phell2, Leu I 18 and AsnlX6; Met63 differs by - 180”. These conformational differences probably arise from the tertiary st,ruct#ure transformation rather than from sequence differences between the two proteins. The rotation

The residues involved in the packing of fragments 1 and 2 show strong, but not absolute, conservation in all those serpins (see section 2, above) which are known to undergo the conformational change. In sheet A, residues 184, 186 and 188 are st,rongly conserved, consistent with the packing constraints described in section 5(b), above. ITsually. residue 184 is Leu. 186 is Asn and 188 is Ile. In angiotensinogen. which retains its native nonformation after cleavage, position 184 is occupied by an atypical residue (Phe). Residues 345. and 347 and 349 of the inserted strand s4A near the hinge typically have small sidrchains (Ala. Gly, Ser or Thr) that’ are readily acconmoclated in the gap between 190 to 192 and 33X to 340. Large side-chains in t)hese positions must contribute to the observed failure of ovalburnin (where 345 is Arg) and angiotensinogen (where 345 is Arg and 349 is Glu). to undergo the conformational change (Stein Pt al.. 1989; Wright Pt al.. 1990). The side-chains of residues 351 and 353. at ttrt, opposite end of this strand. are typically of a similar size to those of the equivalent residues (186 and 188. respectively) of strand s3A that they replace,. Residue 351 is usually Thr or Ser. while residue 353 is typically Leu. Val or Ilr. Similarly. the general nature. if not t.he exac.1 identity of residues 53, 56. 57 and 64 of helix I< and residues 244 and 384 is strongly conserved in t)hosc serpins which undergo the transformation. Residue 53 is usually Ser or Ala: -56 Glp. Ala or Ser: 57 and 64 Tie, Val or Leu; 244 Tyr, Phe or Leu and 384 Phc. 6. Conclusion We have determined in outline the mechanism of the tertiary structure transformat~ion that t’akes place when a typical serpin is cleaved by comparing the atomic structures of uncleaved ovalbumin and cleaved a,-antitrypsin. The serpin st#ructure can br considered as two fragment,s that, behave to a good approximation as rigid bodies, the smaller fragment comprising three strands of sheet A and helix F. and the larger fragment comprising most of the rest of’

Serpin Tertiary Structure Transformation the molecule. Flexible joints between the fragments include helices D and E and a short segment at the end of strand s3A. They accommodate the relative shift of the two fragments that takes place on moving from uncleaved to cleaved forms. The relative movement of the two fragments makes a space in the centre of sheet A into which the new carboxy terminus formed by the cleavage is inserted to form an additional strand in this sheet. The movement changes the packing contacts at the interface between the fragments. Residues in strand s3A of sheet A slide along the grooves of helix B. To our knowledge, this is the first example of this type of dynamic behaviour. Changes in the packing of helices occurs at subunit interfaces in the allosteric transitions of haemoglobin and phosphorylase, but in these cases it involves residues jumping between different grooves (for a review, see Perutz, 1989). As expected, the residues involved in packing at the interface between the fragments are conserved in those serpins that undergo the tertiary structure transformation. Detailed understanding of the conformational transformation will require high resolution structures of the same serpin in both uncleaved and cleaved forms. However, the general structural and sequence conservation within the family strongly implies that the overall mechanism will be that described here. We thank Dr A. M. Lesk for computer programs used in this work. Professor R. W. Carrel1 and Dr A. G. W. Leslie for useful discussion and the Medical Research Council and the Lister Institute of Preventive Medicine for their support.

References Bernstein. F. C., Koetzle, T. F.. Williams, G. J. B., Meyer, E. F.. Brice, M. D., Rodgers, J. R., Kennard, O., Shimanouchi, T. & Tasumi. M. (1977). The protein data bank: a computer-based archival file for macromolecular structures. J. Mol. Biol. 112, 535-542. Bruch, M.. Weiss, V. & Engel, J. (1988). Plasma serine proteinase inhibitors (serpins) exhibit major conformational changes and a large increase in conformational stability on cleavage at their reactive sites, J. Biol. Chem. 263. 16626-16630. Carrel], R. W. & Owen, M. C. (1985). Plakalbumin, a,-antitrypsin, antithrombin and the mechanism of inflammatory thrombosis. Nature (London), 317> 736732. Chothia. (‘. (1984). Principles that determine the strurture of proteins. rlnnu. Rev. Biochem. 53. 537-572. Chothia, c’. & Lesk, A. (1986). The relation between the divergence of sequence and structure in proteins. EMBO J. 5, 823~-826. Engh, R. A., Wright. H. T. C Huber, R. (1996). Modeling

621

the intact form of the a,-proteinase inhibitor. Prot. Engng, 3, 469-477. Gettins, P. (1989). Absence of large scale conformational change upon limited proteolysis of ova,lbumin, the prototypic serpin. J. Biol. Chem. 264, 3781-3785. Gettins, P. & Harten, B. (1988). Properties of thrombin and elastase modified human antithrombin III. Biochemistry, 27, 3634-3639. Huber, R. & Carrell, R. W. (1989). Implications of the 3-dimensional structure of a,-antitrypsin for the structure and function of serpins. Biochemistry, 28, 8951-8965. Janin, J. & Chothia, C. (1980). Packing of a-helices onto p-pleated sheets and the anatomy of a//l proteins. J. Mol. Biol. 143, 95-128. Kress, L. F., Kurecki, T., Chan, S. K. & Laskowski, M. (1979). Characterisation of the inactive fragment resulting from limited proteolysis of human a,-proteinase inhibitor by Crotalus adamanteus proteinase II. J. Biol. Chem. 254, 5317--5320. Lee, B. & Richards, F. M. (1971). The interpretation of protein structures: estimation of static accessibility. J. Mol. Biol. 55, 379-400. Loebermann, H., Tokuoka, R., Deisenhofer, J. & Huber, R. (1984). Human a,-proteinase inhibitor. Crystal structure analysis of 2 crystal modifications, molecular model and preliminary analysis of the implications for function. J. Mol. Biol. 177. 531-556. Pemberton, P. A., Stein, P. E., Pepys, M. B.. Potter, J. M. & Car&l, R. W. (1988). Hormone binding globulins undergo serpin conformational change in inflammation. Nature (London), 336, 257-258. Pemberton, P. A., Harrison, R. A., Lachman P. J. & Carrell, R. W. (1989). The structural basis for neutrophi1 inactivation of Cl-inhibitor. Biochem. J. 258, 1933198. Perutz, M. F. (1989). Mechanisms of cooperativity and allosteric regulation in proteins. &art. Rc~. Biophys. 22, 139-236. Potempa, J., Watorek. W. & Travis. J. (1986). The inactivation of human plasma a,-proteinase inhibitor by proteinases from Staphylococcus aureus. J. Biol. Chem. 261, 14330614334. Priestle, J. P. (1988). RIBBON: a stereo cartoon drawing program for proteins. J. Appl. C’rystallogr. 21, 572-576. Richards, F. M. (1974). The interpretation of protein structures: total volume, group volume distributions and packing density. J. Mol. Biol. 82. l--14. Stein, P. E., Tewksbury, D. A. & Carrel], R. W. (1989). Ovalbumin and angiotensinogen lack serpin S-R conformational change. Biochem. J. 262, 103-107. Stein, P. E., Leslie, A. G. W., Finch, J. T., Turnell, W. G., McLaughlin, P. J. & Carrel]. R. W. (1990). Crystal structure of ovalbumin as a model for the reactive centre of serpins. Nature (London), 347, 99-102. Wright, H. T., Qian, H. X. & Huber, R. (1996). Crystal structure of plakalbumin, a proteolytically nicked form of ovalbumin. Its relationship to the structure of cleaved a-1-proteinase inhibitor. J. Mol. Biol. 213, 513-528.

Edited by W. Hendrickson

Serpin tertiary structure transformation.

Previous crystallographic analyses have demonstrated that proteolytic cleavage of the serpins can result in a dramatic transformation of their tertiar...
950KB Sizes 0 Downloads 0 Views