Conformational studies on forms of plasminogen activator inhibitor, PAI-1

Conformational studies on forms of plasminogen activator inhibitor, PAI-1

ORAL COMMUNICATIONS l-3: Plasminogen Activator Inhibitors 47 THE STRUCTURAL BASIS OF LATENCY H. M. Tucker, J. Mottonen. Department of Biochemistry, ...

150KB Sizes 0 Downloads 22 Views

ORAL COMMUNICATIONS l-3: Plasminogen Activator Inhibitors

47 THE STRUCTURAL BASIS OF LATENCY H. M. Tucker, J. Mottonen. Department of Biochemistry,

IN PAl-1 E. J. Goldsmith and R.D. Gerard UT Southwestern Medical Center,

Dallas, Texas. The active to latent transition of PAI- is accompanied by the movement of reactive center residues from their position in an arched loop on the surface of the protein to insertion of these residues as an antiparallel P-strand into sheet A. Using the known structure of latent PAl-1 as a rational guide, mutations were designed to prevent or slow the insertion of the reactive center residues into the A sheet. Three types of mutations have been constructed. In the first, we have attempted to block the insertion of p-strand 4A into the A sheet by altering the sequence of amino acid residues in the strand such that the substitutions are less favorable to the formation of P-sheet. Such mutations include the substitution of glutamic acid or proline residues for those normally found within strand 4A at positions P14-P4. Second, we have introduced specific changes into a region of the molecule that we call the “gate” in order to prevent the active to latent

48

ANALYSIS OF PLASMINOGEN ACTIVATOR INHIBITOR TYPE 1 (PAI-1) BY LIGHT SPECTROSCOPY: ‘Fa M ‘Aleshkov S, ‘Strandhere L, ‘Karolin J ‘Johansson L. B-A, -9 and&T Department of ‘Medical Biochemistry and Biophysics, and *Physical Chemistry, University of Urn&, 901 87 Urn&, Sweden. PAI- is a Serpin family that has its reactive centre on a so called strained loop. The inherited ability of this loop to move partly into a neighhouring p-sheet may be important for both the activity and specificity of the serpins. PAI- lacks cyst&e residues which makes it possible to use mutagenesis to introduce unique cysteines for Structural and attachment of extrinsic fluorescence probes. intramolecular changes can than be investigated by time-resolved fluores’cence spectroscopy and electronic energy transfer methods. To study the mobility of the strained loop in PAI- during interaction with PAS, vitronectin and heparin as well as during the conversion from the active to the latent form, we have replaced Ser 344 ( P3) and Met 347 (Pl’) separately with cysteines. After expression in E. coli and purification to homogeneity the mutant proteins were labelled with the fluorescent probe BODIPY ( Molecular Probes). The labelling did not significantly decrease the activity of PAI-1.

49

CONFORMATIONAL

STUDIES

ON FORMS

OF

PLASMINOGEN ACTIVATOR INHIBITOR, PAI1E Sancho, *PJ Declerck, 3NC Price. 3SM Kelly, 4pI &, 4D ChaDma, and 1NA Booth, Universities of

‘Aberdeen, *Leuven, 3Stirling and 4London. Plasminogen activator inhibitor 1, the major inhibitor of t-PA and u-PA, is a member of the serpin superfamily. PAI-I can exist in three forms, an active inhibitory form, a non-inhibitory, cleavable, substrate form and a latent form. We present data on the conformation of active, latent, substrate and cleaved PAI-1. All forms were expressed in E cd The active form was produd in a high level expression system and purified at pH 5.6 and in 1M salt, conditions under which it is active and stable. PAI- 1, cleaved at PI-Pl’, was prepared from a mutant (A335P), which acts predominantly as a substrate for t-PA. Structural differences were assessed in terms of

17

transition. Third, using the model of antithrombin Ill as a guide, we have introduced a pair of cysteine residues into PAl-1 where none previously existed. The mutations have been engineered into our new bacterial expression vector in which PAl-1 expression is driven from the T7 gene 10 promoter, which in turn depends on the IPTG-inducible expression of T7 RNA polymerase. A polyhistidine tag permits affinity purification of the expressed protein on a nickel-agarose column, and further purification of PAl-1 is accomplished on a heparin-Sepharose column. This system permits the rapid expression and purification of milligram quantities of PAI- required for analysis of the “latency-resistant” mutants. The results of stability experiments on these variants support three conclusions: 1) strand 4A is partially inserted into sheet A and this insertion is required for inhibitory activity of PAI-1; 2) substitution mutations in the gate region and also in the reactive center can stabilize the active form of PAl-1 such that the transition to the latent form is significantly slower; 3) the two cysteine residues that were introduced can form a disulfide bond which will stabilize PAl-1 in the active state.

The time resolved fluorescence anisotropy was determined for active and latent BODIPY labelled PAI- as well as for complexes with tPA and uPA. The anisotropy data was analyzed in terms of local mobility and restriction of BODIPY in the reactive centre loop of PAImolecule. The orientational restrictions for BODIPY in the P3 and Pl’ positions were significantly different. Moreover, for both positions there are less restrictions in the latent forms as compared to the active form. This indicates that the local environment of the P3 and Pl’ residues in the loop are more dense in the active form than in the latent form. After complex formation with tPA or uPA the orientational distribution of the BODIPY in P3 position was more restricted whereas the orientational restrictions of BODIPY in Pl’ position decreased. These results suggest that complex formation may result in strain loop cleavage without dissociation of the complex. We have also studied the effect of vitronectin and heparin on the fluorescence anisotropy of BODIPY in the P3 and Pl’ position. The data shows that addition of heparin or vitronectin results in less orientational restriction of BODIPY, suggesting that PAI- undergoes a conformational changes upon the interaction with vitronectin and heparin.

thermostability, changes in secondary structure (far UV-CD) and environment of aromatic side chains (near UV-CD, fluorescence spectroscopy) and IT-IR spectra.

Active and latent PAI- showed small but distinct differences, consistent with decreased a-helix from 26 to 22% and increased P-sheet from 23 to 34% as active converts to latent. Latent PAI- was more thermostable than active PAI-1, which sugggests its similarity to cleaved serpins. However, cleaved PAI- was much more resistant to unfolding in the presence of guanidinium chloride (GdnHCl), 50% denaturation occurring in 4.5 M Gdn-HCl, while both active and latent PAI- were 50% denatured in 2M Gdn-HCl. Further, differences in fluorescence emission maxima, latent 339 nm, active and substrate 336 nm, cleaved 334nm, underlined the contrast between latent and cleaved PAI-1. Conformational changes occurring on cleavage are more profound than those on transition between active and latent PAI-1.