Tài liệu Báo cáo khoa học: Mapping of the epitope of a monoclonal antibody protecting plasminogen

Mapping of the epitope of a monoclonal antibody protecting

plasminogen activator inhibitor-1 against inactivating agents

Julie S. Bødker, Troels Wind, Jan K. Jensen, Martin Hansen, Katrine E. Pedersen and Peter A. Andreasen

Laboratory of Cellular Protein Science, Department of Molecular Biology, University of Aarhus, Denmark

Plasminogen activator inhibitor-1 (PAI-1) belongs to the

serpin family of serine proteinase inhibitors. Serpins inhibit

their target proteinases by an ester bond being formed

between the active site serine of the proteinase and the P1

residue of the reactive centre loop (RCL) of the serpin, fol￾lowed by insertion of the RCL into b-sheet A of the serpin.

Concomitantly, there are conformational changes in the

flexible joint region lateral to b-sheet A. We have now, by

site-directed mutagenesis, mapped the epitope for a mono￾clonal antibody, which protects the inhibitory activity of

PAI-1 against inactivation by a variety of agents acting on

b-sheet A and the flexible joint region. Curiously, the epitope

is localized in a-helix Cand the loop connecting a-helix I and

b-strand 5A, on the side of PAI-1 opposite to b-sheet A and

distantly from the flexible joint region. By a combination of

site-directed mutagenesis and antibody protection against an

inactivating organochemical ligand, we were able to identify

a residue involved in conferring the antibody-induced con￾formational change from the epitope to the rest of the

molecule. We have thus provided evidence for communi￾cation between secondary structural elements not previously

known to interact in serpins.

Keywords: cancer; cardiovascular disease; monoclonal

antibody; protease; serpin.

The serpins constitute a protein family of which the best

characterized members, including a1-proteinase inhibitor,

antithrombin III, and plasminogen activator inhibitor-1

(PAI-1), are inhibitors of serine proteinases implicated in

processes such as blood coagulation and turn-over of

extracellular matrix. Of decisive importance for the

inhibitory mechanism of serpins is the surface-exposed,

approximately 20-amino acid long reactive centre loop

(RCL) (see Fig. 1). Biochemical and biophysical evidence

has shown that the reaction between a serpin and its

target proteinase is initiated by formation of a reversible

docking complex in which the P1–P1¢ bond in the RCL

interacts noncovalently with the active site of the

proteinase [1]. In the locking step that follows, the P1–

P1¢ bond is cleaved [2,3] and the P1 residue is coupled to

the active site serine of the proteinase by an ester bond

[4]. The N-terminal part of the RCL then becomes

inserted as strand 4 in b-sheet A (s4A) [5]. Because of the

covalent bond, the proteinase is translocated to the

opposite pole of the serpin [6–8], the active site becoming

distorted, the catalytic machinery inactivated, and the

completion of the catalytic cycle disabled [8–16], resulting

in formation of a stable covalently coupled complex of

1 : 1 stoichiometry (for reviews see [17–19]). The energy

needed for the proteinase distortion comes from stabi￾lization of the serpin in the
relaxed conformation by

insertion of the RCL into b-sheet A, as opposed to the

stressed, relatively unstable active conformation with a

surface-exposed RCL. Under some conditions, proteinase

distortion cannot keep pace with ester bond hydrolysis,

resulting in abortive complex formation, full cleavage of

the P1–P1¢ bond, insertion of the RCL into b-sheet A and

release of an active proteinase (for reviews see [17,20]).

Serpins following this alternative path are said to exhibit

substrate behaviour. Some serpins, including PAI-1 and

antithrombin III spontaneously assume an inactive,

relaxed, so-called latent state in which the intact RCL is

inserted into b-sheet A, after passage through the so￾called gate region between the s3C–s4C loop and the

s3B–hG loop (Fig. 1) [21,22].

RCL insertion is coupled to conformational changes in

the flexible joint region around a-helices D and E. The

flexible joint region of stressed, but not relaxed PAI-1,

binds to the N-terminal 44-amino acid long somatomedin

B domain of the Mr 70 000 glycoprotein vitronectin (VN)

[23,24], which thereby delays the latency transition of

PAI-1 (for a review see [20]). A few organochemical

compounds able to inactivate PAI-1 have been indentified,

including a group of negatively charged amphipathic

compounds like bis-ANS (4,4¢-dianilino-1,1¢-bisnaphthyl￾5,5¢-disulfonic acid) [11,25] and the diketopiperazine

derivative XR5118 ((3Z,6Z)-6-benzylidene-3-(5-((2-dimeth￾ylaminoethyl-thio)-2-thienyl)methylene-2,5-piperazinedione

Correspondence to J. S. Bødker, Department of Molecular Biology,

University of Aarhus, Gustav Wied’s Vej 10C, 8000 C Aarhus,

Denmark. Tel.: + 45 89425079, E-mail: [email protected]

Abbreviations: bis-ANS, 4,4¢-dianilino-1,1¢-bisnaphtyl-5,5¢-disulfonic

acid; h, a-helix; RCL, reactive centre loop; HBS, Hepes buffered

saline; PAI-1, plasminogen activator inhibitor-1; s, b-strand;

S-2444, pyro-Glu-Gly-Arg-p-nitroanilide; uPA, urokinase-type

plasminogen activator; VN, vitronectin; wt, wild-type; XR5118,

((3Z,6Z)-6-benzylidene-3-(5-((2-dimethylaminoethyl-thio)-

2-thienyl)methylene-2,5-piperazinedione hydrochloride).

(Received 3 December 2002, revised 5 February 2003,

accepted 13 February 2003)

Eur. J. Biochem. 270, 1672–1679 (2003) FEBS 2003 doi:10.1046/j.1432-1033.2003.03523.x