Tải bản đầy đủ - 0 (trang)


Tải bản đầy đủ - 0trang

Table 1 Structural Statistics and Residual Violations of the 30 Conformers Used

to Represent the Solution Structure of the Inhibited Catalytic Domain of




DIANA target function (A˚ )

FANTOM energy (kcal/mol)

Lennard-Jones energy (kcal/mol)

Distance constraint violations (A˚)




Exp. angle constraint violations (j)




Rmsd residues 83 – 250 (A˚)

backbone (Ca,N,C’,O)

all heavy atoms


10.01 F 0.76


À191.0 F 52.8

À605.4 F 48.4

35.2 F 1.2

0.48 F 0.06

0.06 F 0.01

61.6 F 1.4

0.39 F 0.03

0.08 F 0.01

93.4 F 11.1

7.5 F 0.9

0.90 F 0.08

112.1 F 19.9

12.6 F 5.0

1.2 F 0.3

0.48 F 0.06

0.94 F 0.06

0.55 F 0.06

0.97 F 0.05

Source: Ref. 7.

one antiparallel strand and the topology À1x, +2x, +2, À1, using

the Richardson nomenclature [50]. The h-sheet lies on two helices (helix

A and B); a third helix (helix C) is near the C-terminus. The molecule

has two zincs: a catalytic zinc is located at the bottom of a cleft, and

a structural zinc above the h-sheet. The overall fold of sfSTR may

be described as follows. The N-terminus is located near the N-terminal

end of helix C. The protein backbone forms a poorly defined irregular

strand for the first 13 residues before entering strand I of the h-sheet,

then descending through helix A. Helix A acts as a backbone to the

protein, spanning its full length. The pronounced amphipaticity of this

helix provides hydrophobic residues for internal packing to helix B and

to the h-sheet, and the hydrophilic residues are exposed to the solvent.

After helix A the protein backbone turns to form strand II of the h-sheet,

which lies parallel to and outside strand I. This strand rises steeply, giving

the h-sheet a distinctly twisted appearance. It is connected by a short

loop to strand III, which is parallel to and inside of strand I. A long loop

connects strands III and IV, crossing over strand V and placing strand IV

along the ligand-binding cleft and antiparallel to strand V. Another small

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

Figure 9 Backbone (Ca, C’, N) trace from residues 83 to 250 of 30 conformers

of inhibited sfSTR. Residues 251 to 255 are disordered and are not included. All

the heavy atoms of the inhibitor are shown. The family of structures are viewed

along the long axis of the catalytic helix B. The inhibitor (I) binds to the protein in

a well-defined cleft and runs antiparallel to the outer strand of the h-sheet with

the ring of P1V homophenylalanine (hP) buried in a bottomless S1V subsite and the

P2V arginine (R) is exposed to the solvent.

loop connects strand IV to V, which runs parallel to strand III. The

structural zinc is ligated by three His, one each from strands IV and V

(His-166 and -179, respectively), and the third (His-151) from the long

loop connecting strands III and IV. The fourth ligand of this zinc appears

to be Asp-153. After strand V the backbone loops to form helix B. The

two His residues of helix B, His-201 and -205, ligate the catalytic zinc. A

short turn then enters an extended strand containing His-211, a third

ligand of the catalytic zinc. From His-211 to Leu-218 several short range

NOEs, in particular between the side chains of Ser-212 and Ala-217,

ChH3 of Ala-217 to the NH protons of Leu-218 and Met-219, and the

backbone atoms of Thr-215 to Ala-217, describes the presence of two

tight turns. An invariant residue, Met-219, which is residue three in one

of these turns, is positioned below the three His residues that ligate the

catalytic zinc and shows NOEs to all three. Except for helix C, the

remainder of the protein is irregular, but well-defined. Helix C runs

perpendicular to helix A; the segments C-terminal to these helices are

near each other.

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

Figure 10 Ribbon diagrams of a single conformer of inhibited sfSTR from

residues 83 to 250. (A) The complex is viewed from above the h-sheet. The

positions of the two zincs are indicated as large balls. The strands of the hsheet (I – V) and helices (A – C) are indicated. The heavy atoms of the inhibitor

and residues of the protein that ligate zinc are shown. The inhibitor runs

antiparallel to strand IV. The structural zinc lies above the h-sheet and is

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

B. Conformation of the Inhibitor

The inhibitor binds to the protein in a well-defined cleft (Figs. 9 and 10) and

in an extended fashion, running antiparallel to strand IV of the sheet, as

indicated by the strong CaH-CaH NOE between the P2V residue and Val163 (Fig. 8) and parallel to the nonregular loop region encompassing Pro221 to Tyr-223. One of the most striking features of the structure is that the

S1V subsite appears to pass through the entire structure. Indeed the

aromatic ring of the homophenylalanine group is clearly observed from

below the S1V pocket [7]. The S1V subsite is lined with hydrophobic residues

including Leu-164, Leu-197, His-201, Val-198, Leu-218, Tyr-220, Leu-222

and Tyr-223. The residues Leu-197, Val-198 and His-201 are from the

catalytic helix, whereas Tyr-220 and -223 and Leu-218 and -222 are from

the loop following this helix. Contacts between the protein and inhibitor

are summarized in Figure 11. Despite the P1V group appearing in contact

with a number of residues, the ring of this residue can clearly undergo ring

flips, as indicated by the degeneracy of the H3,5 and H2,6 resonances

(Fig. 7) thus indicating that this ring is not especially restricted. Similarly

not all residues of the S1V are restricted in motion. For example, both

methyls of Leu-197 show intraresidue NOEs to the CaH proton of Leu-197

suggesting that motion around the torsion angles m1, m2 is present. We

note that this residue shows strong NOEs to the homophenylalanine ring

of the inhibitor (Fig. 8) indicating that it is in contact with the inhibitor.

Analysis of spectra with other inhibitors with extensions to the homophenylalanine showed this residue became restricted in motion, and thus

subtle changes to residue mobility is inhibitor dependent.

The family of conformers were analyzed for hydrogen bonds, where

acceptor-donor (N-H. . .O) distance was set to an upper limit of 2.4 A˚ and

ligated by His-166 from strand IV, His-179 from strand V, and His-151 and

Asp-153 both from a 14 residue loop. The catalytic zinc is ligated by His-201

and – 205 from helix B and His-211. (B) The complex is viewed from below S1V

subsite. The heavy atoms of the inhibitor and the residues that are in

intermolecular contact (Leu-164, Leu-197, Val-198, His-201, Leu-218, Tyr-220,

Leu-222, Tyr-223) with the P1V homophenylalanine are shown. To reduce

crowding in the figure not all these residues are labelled. (*) marks Leu-218

and His-201. Val-198 is below Leu-197. Leu-164 is at the N-terminal end of

the h-strand that appears above Leu-197 in this figure. The ribbon diagrams

were produced by MOLSCRIPT [62].

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

Figure 11 Potential hydrogen bond partners to the backbone atoms of the

inhibitor and the residues of the S1V subsite that are in intermolecular contact with

the P1V homophenylalanine.

the angle to 35j. The analysis suggests that the NH of the P3V hydrogen

bonds to the carbonyl of Asn-162; the carbonyl of P1V hydrogen bonds to

the NH of the Leu-164 (which is slowly exchanging with deuterium); and

the amine of P1V hydrogen bonds with the carbonyl of Ala-165. The

structures described here do not show hydrogen bonds between the NH

and the carbonyl of the P2V arginine to the protein, which is in contrast to

reported crystal structures which show a hydrogen bond to the NH of Tyr223 [10]. Although Pro-221 and Tyr-223 are near atoms of the inhibitor, for

example, the NH of Tyr-223 shows weak NOEs to the ring of P3V, their

distances in the structure models are not in agreement with these residues

participating in hydrogen bonds. The NH of Tyr-223 does not show slow

exchange with 2H2O and analysis of 2-D saturation transfer difference


H,15N HSQC spectra suggested that the exchange rate of the NH of Tyr223 was one to two orders slower than a free amide proton further

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

Tài liệu bạn tìm kiếm đã sẵn sàng tải về


Tải bản đầy đủ ngay(0 tr)