Received Date : 22-Jul-2013 Revised Date : 11-Mar-2014
Accepted Article
Accepted Date : 15-Apr-2014 Article type
: Research Article
Combined 1H-NMR and Molecular Dynamics Studies on Conformational Behaviour of a
1
Model Heptapeptide, GRGDSPC
Ashok K. Kulkarni1*, Rajendra P. Ojha2
Department of Physiology, Mediciti Institute of Medical Sciences, Hyderabad - 501401, A.P.,
India 2
Biophysics Unit, Department of Physics, University of Gorakhpur, Gorakhpur - 273001, India
Running Title: Conformation of Heptapeptide by 1H-NMR and Molecular Dynamics
Key Words: Arg-Gly-Asp; β-turn; distance geometry; -turn; 1H-NMR; restrained molecular dynamics.
*Corresponding Author: Department of Physiology, MediCiti Institute of Medical Sciences, Ghanpur, Medchal Mandal, R. R. Dist., Hyderabad – 501401, A. P., India Email: [email protected], [email protected] This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an 'Accepted Article', doi: 10.1111/cbdd.12346 This article is protected by copyright. All rights reserved.
Accepted Article
Phone: +91-9849845967
Abstract Among various strategies, the de novo design and in silico approaches are being used to develop the short peptides, models of modified peptides and mimetics as clinically useful drugs with improved stability and bioavailability. The resulting models will help to isolate the factors behind the folded structure formation, and contribute useful information about de novo peptide design. The combined 1
H-NMR spectroscopic and molecular dynamics methods were used to investigate the
conformational behavior of an Arg-Gly-Asp (RGD) containing peptide, GRGDSPC, the cell binding heptapeptide of extracellular matrix protein, fibronectin. The formation of two fused weak β-turns of type-II (HB, 4→1), and type-II' (HB, 7→4) from simulation studies has been consistent with NMR data. The sustainable ‘S’ shaped molecular structure (which remained unchanged during the entire simulation) and the conformational transitions due to interconversions between multiple turns initiated at Asp4, Ser5 and Cys7 implies that the peptide is flexible in nature. Thus, the model of ‘S’ shaped structure with flexible multiple turns for GRGDSPC peptide may provide the structural rationale for antagonistic properties of this heptapeptide towards the treatment of integrin mediated cellular abnormal behaviors such as thrombosis and metastasis.
Introduction The development of peptides into pharmaceutically acceptable drugs has been reported to be
hindered due to their limited stability and lack of high affinity (1, 2). Various strategies (3) such as de novo design (4), structural characterization and in silico approaches (5) were developed towards producing modified peptides, mimetics and models. The resulting models will help to isolate the factors behind the folded structure formation, and contribute useful information about de novo peptide design. The present study uses NMR and molecular dynamics methods to investigate the conformational behaviour of an Arg-Gly-Asp (RGD)-containing sequence, GRGDSPC. The This article is protected by copyright. All rights reserved.
heptapeptide GRGDSPC has been the topic of immense investigation in lieu of its involvement in cell-adhesion based cellular behaviors such as cell proliferation (6), osteogenic differentiation (7),
Accepted Article
and also in biomedical applications such as gene delivery (8), and tissue engineering of the heart valves (9). The amino acid sequence of Arg-Gly-Asp (RGD) has been reported to be serving as the
primary integrin recognition site in extra-cellular matrix proteins such as fibronectin, vitronectin, laminin, collagen, fibrinogen etc (10-13).
Towards studying the binding specificity of RGD
sequence, it has been shown that the mere positive and negative charges on Arg and Asp, respectively, were not sufficient for its recognition action (14). While proposing a strategy for the design and development of more effective drugs towards controlling platelet aggregation, the crystal structure of RGD containing disintegrin – TRIMESTATIN – has been studied (15). The studies using CD and folding algorithms methods reported that, the substitution of amino acids in RGD-spanning sequences were effective in changing the potency of RGDs as inhibitor of binding ligands (16-19). The synthesis and activity studies (20) on RGD related tetrapeptides RGDS, RGDV and RGDP have reported that the amino acid residues at the C-terminal were very crucial for anti-thrombosis effects in lieu of their total energies. The conformational studies (21) on RGDcontaining tetrapeptides such as H-RGDX-OH (X = Y, W, F, L, V, C, Q and S) reported that more populated β-turn structures were imparted for Gly-Asp (G-D) sequence by more hydrophobic sidechains of X-residues. Towards recognition by integrin receptors, a favorable conformation for solvent-exposed RGD site was reported to be maintained due to contributions from proline residues that were flanked in RGD motif in snake venom disintegrins, Rhodostomin (22) and Dendroaspin (23). However, the structural studies of a neuropeptide precursor protein (produced by snail, Lymnae stagnalis) with an RGD proteolytic site (24) revealed that the minimal structures around RGD region, due to proline residue following RGD site in that protein, do not inhibit platelet aggregation. The peptide with “Ser” residue following RGD site has been reported to be enhancing more reverse turns around RGD region, and was considered to be potently inhibiting the platelet This article is protected by copyright. All rights reserved.
aggregation. Moreover, the incorporation of Pro and Ser residues in RGD-flanking sequences such as GRGDTP, GRGDSP and some cyclic RGD peptides were shown to be decreasing the vasomotor
Accepted Article
response in rat isolated mesenteric arteriole (25). Though the linear peptides containing RGD sequence appeared to be of lower affinity for a bound state with the receptor II/III as compared to those of cyclic ones, additional specificity determinants might be enhanced with greater conformational flexibility for RGD containing linear peptides. The GRGDSPC synthetic peptide, though, has been shown to be non-inhibitory for cell adhesion to the 38-kD fragment of fibronectin (26), some studies reported the role of GRGDSPC sequence as an integrin inhibitor (27), and inhibiting the fibronectin attachment to keratinocytes (28), thymocytes (29), and B-lymphocytes (30). These different versions of interpretations for binding as well as inhibitory roles of the GRGDSPC created interest in further characterizing the heptapeptide. Till date, no structural aspects have been reported for GRGDSPC sequence as such by NMR or by X-ray diffraction methods. Hence, the present work emphasizes the conformational behavior of a model heptapeptide, GRGDSPC, by using 1H-NMR and molecular dynamics methods.
Methods and Materials Sample Preparation The heptapeptide, Gly-Arg-Gly-Asp-Ser-Pro-Cys (GRGDSPC) has been purchased from SigmaGenosys (Cambridge, UK) and used in its original form as it was found to be pure when examined using NMR.
The sample for NMR was prepared in D2O at a peptide concentration of 5 mM (pH,
7.0). After the desired NMR experiments were completed, the sample was lyophilized and redissolved in H2O/D2O (90:10) for the study of exchangeable amide protons.
NMR Experiments All NMR experiments were performed on a Bruker AMX-500 FT NMR Spectrometer. NMR experiments were performed at 25C except for experiments of temperature coefficients of amide
This article is protected by copyright. All rights reserved.
proton resonances. The temperature dependence of the amide proton chemical shift was determined from measurements at 5, 10, 15, 20, 25, 30 and 37 C for the peptide studied here. Phase-sensitive
Accepted Article
COSY (31) and TOCSY (32) experiments were performed to identify spin-systems of protons. Sequence-specific assignments of proton resonances were obtained by NOESY (33) (mixing time = 300, 400 and 500 ms) experiment. All two-dimensional NMR spectra were recorded using the TPPI method with the conventional pulse sequences, with the following parameters: spectral width (6000 Hz), 512 and 256 data points in t1 and t2 dimensions, relaxation delay (1 s). Water signal was suppressed by using presaturation during the recycling delay. The 3JHN-Hα coupling constants were determined by ECOSY (34) experiment.
Distance Restraints The assigned strong, moderate and weak NOE cross-peaks were translated into distances with the lower boundary limits set at 2.0 Å, and the upper boundary limits set at 2.5 Å for strong, 3.0 Å for medium and 4.0 Å for weak peaks (35). 3
J-Dihedral Angle Restraints
The coupling constant values, 3JHN-H and 3JH-H, were converted into dihedral angle information on phi () and chi () angles, respectively, using Karplus equation (36). J = A cos2 + B cos + C
3
where = | 60|. The values A = 6.8, B = 1.3 and C = 1.5 for phi () angles, and A = 9.5, B = 1.6 and C = 1.8 for chi () angles were incorporated.
Energy Calculations Molecular modelling software routines of Insight II and Discover90.0/3.0.0 (Biosym, USA) on Silicon Graphics Iris Power Indigo2 machine have been used for calculations of energy minimization and restrained molecular dynamics. The dielectric constant was fixed at 80.0. The
This article is protected by copyright. All rights reserved.
atom-based non-bonded interactions were calculated by using the consistent valence force field (CVFF) of Biosym with an initial cut-off 9.5 Å. To avoid the perturbations of the forces arising
Accepted Article
from small cut-off, which otherwise mask the effect of non-bonded interactions (37), this cut-off was later increased to 12.0 Å during molecular dynamics simulations. To result an overall neutral system, the atomic charges on the termini on Arg2 guanidine and Asp4 carboxylate groups were scaled by a factor of 0.25 relative to the default values. The distance restraints were employed in energy calculations using a harmonic function based flat-bottomed potential by scaling the force constants for restraints.
In addition to the distance restraints, the 3J-dihedral angle restraints and
hydrogen bonding restraints (NH—O = 1.8 - 2.7 Å, and N—O = 2.5 - 3.7 Å) were employed in restrained molecular dynamics (rMD) simulations for structure calculations.
Refinement Procedure The four steps initial model construction, constrained minimization, restrained molecular dynamics and energy minimization were carried out in refinement procedure.
i) Initial Model Construction While removing the initial strains using a gradient threshold of 0.1 kcal/(mol Å) with the steepest descents and to a maximum derivative of 0.01 kcal/(mol Å) with a conjugate gradient procedure, a linear peptide model was constructed in an extended conformation for Gly-Arg-Gly-Asp-Ser-ProCys sequence. The distance geometry procedure (DG-II) of the NMRchitecht module (Biosym, USA) was performed for the resulting structure by applying the restraints for distance, 3J-dihedral angle and hydrogen bond to generate 100 structures.
ii) Constrained Minimization All these 100 structures from distance geometry procedure were minimized using the steepest descents to a gradient threshold of 0.5 kcal/ (mol Å). These structures were further minimized to a
This article is protected by copyright. All rights reserved.
maximum gradient of 0.001 kcal/ (mol Å) using the conjugate gradients. On account of high energy, 30 structures (from these 100 structures) were discarded, and remaining 70 structures were grouped
Accepted Article
into 7 closely related families using cluster analysis based on internal torsion angles within the peptide backbone as well as hydrogens, heavy atoms and side-chain atom angles. The structures from two families did not yield hydrogen bond pattern to characterize any secondary structural features, hence not considered for further refinement. The structures from the remaining five families exhibited the hydrogen bonds that were characteristic for secondary structural features and were considered for further refinement. Among the 5 selected families, all structures from Family I and II exhibited one type of hydrogen bonding pattern of 41, and the representative structures from these two families were numbered as ‘S-1’ and ‘S-2’, respectively. However, each structure from two other families (Family III and IV) yielded three types of hydrogen bonding pattern, viz, 41, 52 and 74, and the representative structures from these two families were numbered as ‘S-3’ and ‘S-4’. From the remaining family (Family V), 50 % of structures exhibited the hydrogen bond pattern of 41 and 74, while the other 50% of structures yielded hydrogen bond patterns of 52 and 74. The representative conformer of the first 50% structures from this Family V was numbered as ‘S-5’, whereas that of the other 50% of structures was numbered as ‘S-6’. A further refinement was performed for all these six structures (S-1, S-2, S-3, S-4, S-5 and S-6) by using restrained molecular dynamics procedure.
iii) Restrained Molecular Dynamics Procedure By maintaining a constant temperature system and employing the velocity scaling, the restrained molecular dynamics procedure has been performed with restraint functions of distance, 3J-dihedral angle and hydrogen bonds throughout the calculations, which used an integrated time step of 1 fs. The starting structure was initially allowed to equilibrate at 50 K for 5 ps. The temperature of the system was raised to 100 K over a 25 ps interval. The target temperature was raised by 50 degrees every 10 ps until 298 K to equilibrate the system completely. The system was allowed to evolve at This article is protected by copyright. All rights reserved.
300 K for 100 ps followed by 2000 ps MD run at 300 K during which the structures were sampled every 0.5 ps. The structures obtained from the trajectories were used for subsequent geometric and
Accepted Article
energetic analysis. The conformational behaviour of structures obtained from molecular dynamics simulations were analysed for (a) key backbone dihedral angles (38), (b) the inter-proton distances, and (c) the calculated Cα-Cα (i, i+3) (39, 40) distances.
iv) Final Minimization The sampled lowest-energy-contained structure from the dynamics trajectories was minimized using the steepest descent method to a gradient threshold of 0.5 kcal/(mol Å) and further minimized to a maximum gradient of 0.001 kcal/(mol Å) using the conjugate gradients method. The geometric statistics similar to those for structures from dynamics trajectories were also calculated for the minimized structures.
Results Amide Proton Chemical Shifts and Temperature Dependence The 1D 1H-NMR spectrum of exchangeable amide protons of heptapeptide, GRGDSPC, is shown in Figure 1. The Arg2 NH was the one-proton signal (at about 7.2 ppm) at temperatures 5C – 37C with broadening and reduction in intensity of signal. NOE Connectivities The values of chemical shift for the 1H resonances, temperature coefficients of the amide protons and coupling constants of NH-CαH peaks are given in Table 1. The fingerprint region of the NOESY experiment (Figure 2) provided reliable sequential information and various NOE connectivities are summarized in the NOE connectivity diagram (Figure 3).
This article is protected by copyright. All rights reserved.
Structures for Analysis The ensemble of conformations of NMR structures obtained from a mixed protocol of distance
Accepted Article
geometry and simulated annealing is represented in Figure 4. All the six structures (S-1 through to S-6) generated from constrained minimization and cluster analysis (Section 2.6.ii.) were considered for further refinement using restrained molecular dynamics procedure. Consequently, the six structures obtained from dynamics trajectories and minimizations designated as Structure [S-1], [S-2] and [S-3] in Figure 5, and Structure [S-4], [S-5] and [S-6] in Figure 6 yielded a mixture of different type of turns. Inter-proton distances The NOE dependent distance restraints applied in the simulation, the comparison of relative magnitudes of NOE signals, and inter-proton distances generated by simulation of 2000 ps rMD are shown in Table 2 for intra-residue cross-peaks, and in Table 3 for inter-residue cross-peaks, and the distance restraints violations are summarized in Table 4. DISCUSSION Amide Proton Chemical Shifts and Temperature Dependence As compared to the peak of Gly3NH proton, the peak of Arg2 NH proton exhibited a significantly increased broadening, varying intensities and gradually vanishing as temperature raised, suggesting that Arg2NH proton was exchanging faster and, thus, not involved in hydrogen bonding, whereas Gly3 amide proton was associated in weak hydrogen bonding. The relative gradual broadening and reduction in intensities at higher temperatures for NH protons of Asp4 and Ser5, suggesting that amide protons of Asp4 and Ser5 were involved in weak hydrogen bonding. However, the temperature coefficient value of 3.42 (during 25 - 37º C) for the amide proton of Ser5 in one of ensembles was suggestive of involvement of Ser5 NH proton in strong hydrogen bond in that particular ensemble. The intensity of peak of Cys7 NH proton remained unchanged significantly as the temperature raised; suggesting that amide proton of Cys7 was involved in strong hydrogen bonding. The observations that the presence of more than one peak for the residues Asp4, Ser5 and This article is protected by copyright. All rights reserved.
Cys7 in temperature experiments, and that the temperature coefficient values (Table 1) being in more than one form, suggest that Asp4, Ser5 and Cys7 residues were involved in various different
Accepted Article
conformations. The splitting-up into two peaks by amide protons of Asp4 from 10C onwards suggests that, the amide protons of Asp4 were involved in two conformations after this temperature. Similarly, the presence of three different peaks of amide protons of Ser5 and Cys7 indicates that each of these two residues is involved in three conformations. NOE Connectivities The cross-peaks of (i) Arg2NH-Gly3NH, (ii) Gly3NH-Asp4NH, (iii) Arg2NH-Asp4NH, and (iv) Asp4NH-Ser5NH were indicative of the presence of a -turn. The observations of (i) the moderate to strong peak of Gly3NH-Asp4NH [dNN (3, 4)], and (ii) the weak or absence of Arg2HAsp4NH [dN (2, 4)] peaks were indicative of the presence of type I/II -turns. To distinguish between these two types, one can examine the NOE connectivities between the backbone protons of residues 2 and 3 of the turn. We have observed the cross peak of ArgH-Gly3NH [dN (2, 3)], which was an indication for the presence of type II -turn. This is also corroborated by two observations. Firstly, the distance corresponding to a weak peak of Arg2H-Gly3NH [dN (2,3)] should be more for type II than type I, and we obtained the larger distance corresponding to [dN (2,3)]. Secondly, the 3JNH-H coupling constant value (Table 1) of the second residue of the turn was 6.72 Hz, which was very close to those of standard values for type-II -turn. Therefore, the type-II -turn is predicted between the residues Gly1 and Asp4 (HB, 4→1).
The weak cross peak of Ser5NH-Cys7NH, the very weak cross peak of Asp4NH-Cys7NH and
a cross peak of Asp4H-Cys7NH were also observed, which suggest the presence of another turn between the residues Asp4 and Cys7. Because of the presence of Pro6 at the third position of this turn, it is impossible to obtain the cross-peak corresponding to: (i) [dN (2,3)], (ii) [dN (1,3)], (iii) [dNN (2,3)], (iv) [dNN (3,4)], (v) [dNN (1,3)], and (vi) [dN (2,3)]. We have not observed the cross This article is protected by copyright. All rights reserved.
peaks corresponding to (i) [dN (3, 4)], (ii) [dN (2, 4)], and (iii) [dN (1, 4)]. The weak cross peaks of Pro6H-Cys7NH and Pro6H-Ser5NH were observed to be overlapping. The 3JNH-H coupling
Accepted Article
constant value of Ser5 is 7.15 Hz, while that of Pro6 was absent. It is, therefore, inferred that the turn between Asp4 and Cys7 (HB, 74) is a II’ type of -turn. Structures for Analysis It is known from the structures in Figure 5 and Figure 6 that, the peptide shows tendency to form multiple conformations such as -turn (HB, 42), inverse -turn (HB, 31) in addition to -turn, type-III’ (HB, 41) and -turn type-II’ (HB, 74).
This suggests that the peptide exhibits
inherently weaker -turn types due to conformational transitions.
Inter-proton distance analysis A short-range distance of Arg2HN-Arg2Hα was violated through lower-bound limits in three structures from both dynamics trajectories as well as from minimizations. Similarly, a sequential medium-range distance of Arg2Hα-Gly3HN was violated through both lower and upper bound limits in almost all structures from dynamics trajectories as well as minimizations. Except these two distances, almost all inter-proton distances obtained from sampled structures and minimized structures were in fairly good agreement with the experimentally obtained distances, thus indicating the stable structures. The distances between C(Gly1) and C(Asp4), and between C(Asp4) and C(Cys7) in these structures from both the dynamics trajectories and minimizations were less than 7.0 Å, which also corroborate the presence of folded -turns between Gly1 and Asp4, and between Asp4 and Cys7. Flexibility of conformations in GRGDSPC sequence It is clear that, the -turn (HB, 31) was common in structures [S-1] and [S-2] from both the dynamic trajectories and minimization. The -turn (HB, 42) was exhibited by structure [S-3]
This article is protected by copyright. All rights reserved.
from dynamic trajectories. The hydrogen bond, 41 was shown to be common in structure [S-3] from trajectories and minimization but no definitive β-turn. The hydrogen bond, 74, exhibited by
Accepted Article
structures [S-1] and [S-3] from trajectories, and by structure [S-1] from minimization, yielded βturn type-III only in structure [S-3] from dynamics trajectory. The hydrogen bonds, 51 and 4→2 were common in structures [S-4], [S-5] from both the trajectories and minimization. However, in the structure [S-6] from trajectory the HB, 4→2 yielded inverse -turn, and in structure [S-6] from minimization the HB, 4→1 yielded β-turn type-II. The distribution of , angles and the pattern of hydrogen bond formations, therefore, indicate that residues Asp4, Ser5 and Cys7 were playing key roles in adopting the flexible conformations within the sequence of GRGDSPC. Consequently, the conformations like -turns, -turns, inverse -turns and random coil were observed during simulation. It has been shown that the GRGDSP peptide exhibits the duplicating ability to binding
activity of fibronectin with multiple -turns of type III-III or III-I (HB, 41 and 52) conformations (41). We have shown here that the GRGDSPC peptide prefers two fused -turns between Asp4 and Gly1 (HB, 41), and Cys7 and Asp4 (HB, 74). The formation of the -turn type-II between the residues Gly1 and Asp4 (HB, 41) from simulation is consistent with that from NMR data. But, the formation of -turn type-II’ between the residues Cys7 and Asp4 (HB, 74) in one of the distance geometry structures is also consistent with that from NMR data, suggests that the -turn type-II’ between the residues Cys7 and Asp4 (HB, 74) is inherently weaker, flexible and exists in the lesser population with tendency for multiple conformational transitions. However, our simulation studies suggest that in addition to these -turns of types said-above, the GRGDSPC peptide may also adopt a -turn of type-III (HB, 74), reverse -turn of type-III (HB, 47), turns (HB, 31 and 42), inverse -turn (HB, 42) and random coil.
This article is protected by copyright. All rights reserved.
Change in Rigidity and Activity as an Antagonist for Multiple Ligand Binding
Accepted Article
The proposed conformation is playing an inhibitory role in the interaction between glycoprotein GPIIbIIIa and fibrinogen. The side-chains of Ser5 and Cys7 in our present sequence GRGDSPC were not imparting a net hydrophobic movement to the molecule. They may enhance the binding by unshielding the Arg and Asp side-chain interactions with GPIIbIIIa from competition due to solvent. This suggests that its flexible conformational behaviour may play an important role in the activity of the peptide. Such a view is consistent with the comparative studies on structural characterization of RGD-cell-adhesion recognition sites of Streptavidin that despite the large conformational space occupied the surface loops (42) the high affinity for appropriate receptors has been shown to be displayed by only a few sampled structures from dynamics trajectories. Also a corroborative view is consistent with the comparative studies that up on incorporating Pro and Ser residues in RGD-flanking sequences such as GRGDTP, GRGDSP and some cyclic RGD peptides lead to a decreasing vasomotor response in rat isolated mesenteric arteriole (25). The literature abounds with the data that some of RGD containing peptides have been shown
to be flexible on account of non-established hairpin loop conformations (43-45), and the potency of peptides containing RGD motif in integrin binding has been shown to be in the order of GRGDNP > GRGDSP > GRGDTP = cyclo-RGD (46).
The structural studies on present sequence of
GRGDSPC show that there is a conformational change among two fused -turns between Asp4 and Gly1 (HB, 41), and Cys7 and Asp4 (HB, 74), and the peptide may also adopt a -turn of type-III (HB, 74), reverse -turn of type-III (HB, 47), -turns (HB, 31 and 42), inverse -turn (HB, 42).
Conclusion The present results imply that the GRGDSPC peptide exhibits a two-folded state of -turns, type-II (41) and type-II’ (74) between the residues Gly1 and Asp4, and Asp4 and Cys7, respectively.
This article is protected by copyright. All rights reserved.
Due to the key roles played by the residues Asp4, Ser5 and Cys7, this two-folded -turn state is flexible and tends to adopt into additional conformations like, -turn type-III (74), reverse -turn
Accepted Article
type-III (47), -turns (31 and 42) and inverse -turn (42). Moreover, the sustainable ‘S’ shaped structure with two-folded turns remained unchanged during the entire simulation. So, the conformational transitions with multiple turns within the sequence enhance less rigidity and lower activity in integrin-ligand interactions. Thus, the model of ‘S’ shaped structure with flexible multiple turns for GRGDSPC peptide may provide the structural rationale for antagonistic properties of this heptapeptide.
Acknowledgements The authors thank to high field 500 MHz FT-NMR National Facility, Tata Institute of Fundamental Research, Mumbai, India for providing us with experimentation facility.
References 1.
Humphrey M.J., Ringrose P.S. (1986) Peptides and related drugs: A review of their absorption, metabolism, and excretion. Drug Metab Rev;17:283–310.
2.
Plattner J.J., Norbeck D.W. (1990) Obstacles to drug development from peptide leads. In Clark, C.R., Moos, W.H., editors. Drug Discovery Technologies. Horwood, Chichester, England, p. 92-126.
3.
Adessi C., Soto C. (2002) Converting a peptide into a drug: strategies to improve stability and bioavailability. Curr Med Chem;9:963678.
4.
Cait E.M., Derek N.W. (2004) Engineered and designed peptide-based fibrous biomaterials. Curr Opin Sold State Mater Sci;8:141149.
5.
Mariana R.A., Francisco J.B., Luis S. (1996) De novo design and structural analysis of a model beta-hairpin peptide system. Nature Struct Biol;3:604612.
This article is protected by copyright. All rights reserved.
6. Jiang X.S., Chai C., Zhang Y., Zhuo R.X., Mao H.Q., Leong K.W. (2006) Surfaceimmobilization of adhesion peptides on substrate for ex vivo expansion of cryopreserved
Accepted Article
umbilical cord blood CD34+ cells. Biomaterials;27:2723-2732.
7. Zouani O.F., Chollet C., Guillotin B., Durrieu M.C. (2010) Differentiation of pre-osteoblast cells on poly(ethylene terephthalate) grafted with RGD and/or BMPs mimetic peptides. Biomaterials;31:8245-8253.
8. Haitao P., Qixin Z., Xiaodong G. (2007) A novel synthetic peptide vector system for optimal gene delivery to bone marrow stromal cells. J Pept Sci;13:154-163.
9. Shi J., Dong N., Sun Z. (2009 ) Immobilization of decellularized valve scaffolds with Arg-GlyAsp-containing peptide to promote myofibroblast adhesion. J Huazhong Univ Sci Technolog Med Sci;29:503-507.
10. Hayman E.G., Pierschbacher M.D., Ohgren Y., Rouslahti E. (1983) Serum spreading factor (vitronectin) is present at the cell surface and in tissues. Proc Natl Acad Sci USA;80:40034007.
11. Carlson R., Eugvall E., Freeman A. (1981) Rouslahti E. Serum spreading factor (vitronectin) is present at the cell surface and its tissues. Proc Natl Acad Sci USA;78:24032406.
12. Kleinman H.K., Klebe R.J., Martin G.R. (1981) Laminin and fibronectin in cell adhesion: enhanced adhesion of cells from regenerating liver to laminin. J Cell Biol;88:473485.
13. Assa-Munt N., Jia X., Lakhonen P., Rouslahti E. (2001) Solution structure and integrin binding activities of an RGD peptide with two isomers. Biochemistry;40:23732378.
14. Pierschbacher M.D., Rouslahti E. (1984) Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. Nature (London);309:491497.
15. Fujii Y., Okuda D., Fujimoto Z., Horii K., Morita T., Mizuno H. (2003) Crystal structure of trimestatin, a disintegrin containing a cell adhesion recognition motif RGD. J Mol Biol;332:111522.
This article is protected by copyright. All rights reserved.
16. Johnson W.C.Jr., Pagano T.G., Basson C.T., Madri J.A., Gooley P., Armitage I.M. (1993) Biologically active Arg-Gly-Asp oligopeptides assume a type II β-turn in solution.
Accepted Article
Biochemistry;32:268273.
17. Harris N.L., Rattray K.R., Tye C.E., Underhill T.M., Somerman M.J., D'Errico J.A., Chambers A.F., Hunter G.K., Goldberg H.A. (2000) Functional analysis of bone sialoprotein: Identification of the hydroxyapatite-nucleating and cell binding domains by recombinant peptide expression and site-directed mutagenesis. Bone;27:795802.
18. Suzuki K., Grinneld A.D., Kidokiro Y. (2002) Hypertonicity induced transmitter released at Drosophilla neuromuscular junction is partly mediated by integrins and cAMP/protein kinase. J Physiol;538:103109.
19. Smith J.B., Theakston R.D., Coelho A.L., Barja-Fidalgo C., Calvete J.J., Marcinkiewicz C. (2002) Characterization of a monomeric disintegrins, Ocellatussin, present in the venom of the Nigerial Carpet viper, Echiocellatus. FEBS Lett;512:111115.
20. Zhao M., Peng S.Q., Dilinuer S., Wang Y.Y., Zhou Q.L., Zhao W.Z. (1997) Studies on the synthesis and activities of RGD related peptides. Yao Xue Xue Bao;32:271277.
21. Park H.S., Kim C., Kang Y.K. (2002) Preferred conformations of RGDX tetrapeptides to inhibit the binding of fibrinogen to platelets. Biopolymers;63:298313.
22. Chang C.P., Chang J.C., Chang H.H., Tsai W.J., Lo S.J. (2001) Positional importance of Pro 53 adjacent to the Arg 49-Gly 50-Asp 51 sequence of rhodostomin in binding to integrin, alpha IIb beta 3. Biochem J;357:5764.
23. Lu X., Sun Y., Shang D., Wattam B., Egglezou S., Hughes T., Hyde E., Scully M., Kakkar V. (2001) Evaluation of the role of proline residues flanking the RGD motif of dendroaspin, an inhibitor of platelet aggregation and cell adhesion. Biochem J;355:633638.
24. Zachariah C., Cameron A., Lindberg I., Kao K.J., Beinfeld M.C., Edison A.S. (2001) Structural studies
of
a
neuropeptide
precursor
protein
Biochemistry;40:87908799. This article is protected by copyright. All rights reserved.
with
an
RGD
proteolytic
site.
25. Bekker E.N., Balt J.C., Pfaffendor M., Spann J.A., Van banell E. (2001) Vasomotor effects of arg-gly-asp (RGD) peptides are limited and not related to endothelium-derived hyperpolarizing
Accepted Article
factor-mediated relaxation in rat mesenteric arteries. Clin Exp Pharmacol Physiol;28:873876.
26. Ferreira O.C Jr., Garcia-Pardo A., Bianco C. (1990) Specific binding of the human monocytic cell line U937 to the alternatively spliced connecting segment (IIICS) of fibronectin. J Exp Med;171:351-356.
27. Barletta E., Mugnai G., Ruggieri S. (1993) Complex gangliosides modulate the integrinmediated adhesion in a rat hepatoma cell line. Biochem Biophys Res Commun;192:214-222.
28. Toda K., Tuan T.L., Brown P.J, Grinnell F. (1987) Fibronectin receptors of human keratinocytes and their expression during cell culture. J Cell Biol;105:3097-3104.
29. Torimoto Y., Kinebuchi M., Matsuura A., Kikuchi K., Uede T. (1990) A monoclonal antibody (8H3) that binds to rat T lineage cells and augments in vitro proliferative responses. J Exp Med;172:1315-1323.
30. Sánchez-Aparicio P., Dominguez-Jiménez C., Garcia-Pardo A. (1994) Activation of the alpha 4 beta 1 integrin through the beta 1 subunit induces recognition of the RGDS sequence in fibronectin. J Cell Biol;126:271-279.
31. Aue W.P., Bartholdi E., Ernst R.R. (1976) Two-dimensional spectroscopy: Application to nuclear magnetic resonance. J Chem Phys;64:22292245.
32. Brawnshweiler L., Ernst R.R. (1983) Coherence transfer by isotropic mixing: application to proton correlation spectroscopy. J Magn Reson;53:521528.
33. Kumar A., Ernst R.R., Wuthrich K. (1980) A two-dimensional nuclear Overhauser enhancement (2D NOE) experiment for the elucidation of complete proton-proton crossrelaxation networks in biological macromolecules. Biochem Biophys Res Commun;95:16.
34. Griensinger C., Sorensen O.W., Ernst R.R. (1985)Two-dimensional correlation of connected NMR transitions. J Am Chem Soc;107:63946396.
35. Wuthrich K. (1986) NMR of Proteins and Nucleic acids. John Wiley., New York, USA. This article is protected by copyright. All rights reserved.
36. Karplus M. (1959) Contact electron-spin coupling of nuclear magnetic moments. J Phys Chem;30:1115.
Accepted Article
37. Constantine K.L., Friedricks M.S., Stouch T.R. (1996) Extensive molecular dynamics simulation of a β-hairpin forming peptide. Biopolymers;39:591614.
38. Ramachandran G.N., Sasisekharan V. (1968) Conformation of polypeptides and proteins. Adv Protein Chem;23:283438.
39. Lewis P.N., Momany F.A., Scheraga H.A. (1973) Chain reversals in proteins. Biochem Biophys Acta;303:211229.
40. Efimov A.V. (1993) Standard structures in proteins. Prog Biophys Mol Biol;60:201239. 41. Reed J., Hull W.E., von der Lieth C.W., Kübler D., Suhai S., Kinzel V. (1988) Secondary structure of the Arg-Gly-Asp recognition site in proteins involved in cell-surface adhesion: Evidence for the occurrence of nested beta-bends in the model hexapeptide GRGDSP. Eur J Biochem;178:141154.
42. Le Trong I., McDevitt T.C., Nelson K.E., Stayton P.S., Stenkamp R.E. (2003) Structural characterization and comparison of RGD cell-adhesion recognition sites engineered into streptavidin. Acta Crystallogr D Biol Crystallogr;59:828834.
43. Dennis M.S., Henzel W.J., Pitti R.M., Lipari M.T., Napier M.A., Deisher T.A., Bunting S., Lazarus R.A. (1990) Platelet glycoprotein IIb-IIIa protein antagonists from snake venoms: evidence
for
a
family
of
platelet-aggregation
inhibitors.
Proc
Natl
Acad
Sci
USA;87:24712475.
44. Cooke R.M., Carter B.G., Martin D.M., Murray-Rust P., Weir M.P. (1991) Nuclear magnetic resonance studies of the snake toxin echistatin:1H resonance assignments and secondary structure. Eur J Biochem;202:323328.
45. Dalvit C., Widmer H., Bowermann G., Breckenridge R., Metlernich R. (1991) 1HNMR studies of echistatin in solution: sequential resonance assignments and secondary structure. Eur J Biochem;202:315321. This article is protected by copyright. All rights reserved.
46. Umesh A., Thompson M.A., Chini E.N., Yip K.P., Sham J.S. (2006) Integrin ligands mobilize Ca2+ from ryanodine receptor-gated stores and lysosome-related acidic organelles in pulmonary
Accepted Article
arterial smooth muscle cells. J Biol Chem;45:3431234323.
Figure Legends Figure 1: Variable temperature 500-MHz 1H-NMR spectra of the exchangeable amide protons for GRGDSPC in H2O/D2O (90:10), pH 7. Temperatures covering the range of 5C - 37C are indicated.
Figure 2: Cross-relaxation peaks for amide protons of GRGDSPC in H2O at 25C.
A two-
dimensional cross-relaxation experiment (NOESY) was performed with a mixing time of 500 ms. The vertical axis (F1 domain) gives the chemical shifts of protons participating with NH (F2 domain; horizontal axis) in cross-relaxation (HN-H connectivities).
Figure 3: Schematic diagram of sequential, medium, and long distance NOE connectivities found in the heptapeptide GRGDSPC. The thickness of the bars reflects the strength of the NOE connectivities grouped into strong, medium, and weak.
Figure 4: Superimposition of backbone structures obtained from NMR data by using a mixed protocol of distance geometry and simulated annealing. Five groups of structures representing for five families from distance geometry procedure are shown. Each structure exhibited two turns, one with 41 hydrogen bonded conformer (the first turn, the upper one) and the other with 74 hydrogen bonded conformer (the second turn, the lower one). The structures from two families with no characteristic main-chain hydrogen bonds are not shown.
This article is protected by copyright. All rights reserved.
Figure 5: Molecular graphics renditions of -turns, inverse -turns, -turns of type II’ and III of GRGDSPC obtained from a 2000 ps rMD run. The structures designated by [S-1], [S-2] and [S-3]
Accepted Article
were obtained from dynamics trajectories (upper row) and minimizations (lower row). The carbons are labelled with the amino acid residue numbers in the sequence GRGDSPC. Hydrogen bonds, characteristics for secondary structures, are labelled with proton donors and proton acceptors.
Figure 6: Molecular graphics renditions of -turns and -turn of type II of GRGDSPC obtained from a 2000 ps rMD run. The structures designated by [S-4], [S-5] and [S-6] from dynamics trajectories (upper row) and minimizations (lower row). The -carbons are labelled with the amino acid residue numbers in the sequence GRGDSPC. Hydrogen bonds, characteristics for secondary structures, are labelled with proton donors and proton acceptors.
Table 1: Chemical shifts (ppm), Temperature Coefficients of NH Protons (ppm K-1 x 10–3) and Coupling - Constants (Hz) of GRGDSPC in H2O at 300K Protons
Gly1
Arg2
Gly3
Asp4
Ser5
Pro6
Cys7
HN
-
8.67
8.61
8.34
8.26
-
8.19
H
3.88
4.34
3.93
4.79
4.48
4.46
4.57
H
-
1.85
-
2.85
3.85
2.29
3.25
1.78
-
2.79
3.82
2.03
2.96
1.65
-
-
-
2.0
-
1.65
-
-
-
2.0
-
3.20
-
-
-
3.79
-
H
H
-
-
This article is protected by copyright. All rights reserved.
Accepted Article
HN
/
-
-
3.20
-
-
-
3.73
-
7.18
-
-
-
-
-
7.18
-
-
-
-
-
7.43
8.75
(ppb/K)
3
JNH--H
3
JH--H
6.19 (i)
8.47 (i)
9.06 (ii)(a)
8.66 (ii)(a)
10.5 (ii)(b)
9.56 (i) -
3.42 (ii)(b)
14.66 (ii)(d) 6.11 (ii)(e)
8.5 (ii)(c)
10.06(iii)(d)
6.75 (iii)(b)
6.94 (iii)(e)
-
6.72
-
9.80
7.15
-
6.19
-
8.3
-
5.2
5.5
10.0
6.5
Temperature coefficients taken at (a) 10-25C, (b) 25-37C, (c) 15-25C, (d) 5-20C, and (e) 20-37C for ensembles (i), (ii) and (iii).
This article is protected by copyright. All rights reserved.
Table 2: Comparisons between Intra-residue NOE Cross-Peak Intensities and Distances Obtained from
Accepted Article
Dynamics Trajectories and Minimization of Molecular Dynamics run of 2000 ps From Dynamic
Cross-Peaks
Trajectories I*
#
Arg2HN-
From Final Minimization
M
Arg2H
L
U
B
B
*
*
2. 00
1
2
3
4
5
6
1
2
3
4
5
6
3.
1.7
1.
1.
1.
1.
2.
1.
1.6
1.
1.2
1.
2.
00
3
77
29
26
23
57
71
9
30
2
21
4 6
Gly3HN-
S
Gly3XH
2.
2.
2.6
2.
2.
2.
2.
2.
2.
2.6
2.
2.5
2.
2.
00
50
1
69
62
55
84
52
62
1
58
8
57
5 9
Arg2HN-
M
Arg2XH
2.
3.
3.0
3.
2.
3.
3.
2.
3.
3.1
2.
3.0
3.
2.
00
00
6
06
46
02
03
64
06
0
40
3
04
6 2
Arg2HN-
M
Arg2XH
2.
3.
3.0
2.
3.
3.
2.
3.
2.
3.0
3.
3.0
2.
2.
00
00
0
98
02
02
53
00
85
1
02
2
59
7 0
Arg2XH-
S
Arg2XH
2.
2.
2.5
2.
2.
2.
2.
2.
2.
2.5
2.
2.5
2.
2.
00
50
5
62
55
55
55
55
55
5
57
7
55
7 9
Arg2XH-
M
Arg2XH
2.
3.
3.0
3.
3.
2.
3.
3.
3.
3.0
3.
3.0
3.
3.
00
00
1
03
01
98
01
01
01
1
04
4
01
2 1
Arg2XHArg2XH
W
2.
4.
4.0
3.
4.
4.
4.
4.
4.
4.0
3.
3.3
4.
2.
00
00
3
33
03
04
03
02
03
2
43
0
03
9 4
This article is protected by copyright. All rights reserved.
*I = Intensity (i.e. S for Strong, M for Medium and W for Weak); * LB = Lower Bound; *UB = Upper Bound.
Accepted Article
# Atoms designated ‘X’ (e.g. XH) are pseudo-atoms representing the average location of the unresolved constituent atoms (XH1 and XH2 in this example). For peaks involving ambiguous prochiral hydrogens, the atom pair resulting in the shortest average distance was used for the comparison.
Table 3: Comparisons between Inter-residue NOE Cross-Peak Intensities and Distances Obtained from
Dynamics Trajectories and Minimization of Molecular Dynamics run of 2000 ps
From Dynamic Trajectories
Cross-Peaks #
Gly1XH-
I*
S
Arg2HN Gly3HN-
W
Arg2XH Gly3HN-
W
Arg2XH Gly3XH-
M
Asp4HN Arg2HN-
M
Gly3HN Arg2HN-
W
Ser5HN Gly3HNAsp4HN
M
LB
UB
*
*
2.
From Final Minimization
1
2
3
4
5
6
1
2
3
4
5
6
2.
2.5
2.
2.
2.
2.
2.5
2.
2.5
2.
2.
2.
2.
00
50
2
52
51
55
54
2
51
3
51
53
53 83
2.
4.
1.7
1.
1.
1.
1.
1.3
1.
1.7
1.
1.
1.
00
00
3
66
79
84
83
6
79
3
78
83
83 87
2.
4.
3.1
3.
2.
3.
2.
3.1
3.
3.1
2.
2.
2.
00
00
5
15
96
04
71
1
11
6
94
70
82 91
2.
3.
3.0
2.
3.
3.
3.
2.9
3.
3.0
3.
3.
3.
00
00
2
95
02
02
03
2
00
0
01
02
02 17
2.
3.
3.1
3.
3.
3.
3.
3.0
3.
3.1
3.
3.
3.
00
00
4
14
19
13
12
1
15
1
19
12
12 49
2.
4.
4.0
4.
4.
4.
4.
4.0
3.
4.0
4.
3.
3.
00
00
1
07
01
02
03
1
64
2
03
69
51 81
2.
3.
1.7
1.
2.
3.
3.
2.9
1.
1.8
2.
3.
3.
00
00
0
84
86
02
01
8
80
7
28
04
01 08
This article is protected by copyright. All rights reserved.
2.
3.
3.
4.
4.
2.
Asp4HN-
S
Accepted Article
Ser5HN Asp4HN-
W
Cys7HN Ser5HN-
W
Cys7HN Arg2H-
S
Gly3HN Asp4XH-
M
Cys7HN Ser5XH-
M
Cys7HN Pro6HCys7HN
M
2.
2.
2.1
2.
2.
1.
1.
1.6
2.
2.5
1.
1.
1.
00
50
4
21
11
49
48
9
00
1
84
92
99 40
2.
4.
3.0
2.
3.
4.
4.
1.8
3.
3.7
1.
4.
3.
00
00
3
24
80
01
02
3
74
2
70
01
55 72
2.
4.
2.8
3.
2.
3.
3.
2.4
2.
3.9
2.
4.
2.
00
00
8
36
91
78
67
6
98
0
80
00
60 01
2.
2.
2.9
3.
1.
1.
1.
3.4
2.
3.0
1.
1.
1.
00
50
7
12
29
37
41
0
97
5
28
42
45 91
2.
3.
3.0
1.
3.
1.
1.
3.0
3.
1.9
3.
1.
1.
00
00
7
91
06
43
71
4
04
6
02
75
20 26
2.
3.
3.0
3.
3.
3.
3.
3.0
3.
3.0
3.
3.
3.
00
00
9
09
09
07
05
5
09
6
09
02
10 68
2.
3.
2.8
3.
2.
2.
2.
3.0
2.
2.4
2.
3.
2.
00
00
8
02
87
79
90
2
84
3
83
02
84 23
*I = Intensity (i.e. S for Strong, M for Medium and W for Weak); * LB = Lower Bound; *UB = Upper Bound. # Atoms designated ‘X’ (e.g. XH) are pseudo-atoms representing the average location of the unresolved constituent atoms (XH1 and XH2 in this example). For peaks involving ambiguous prochiral hydrogens, the atom pair resulting in the shortest average distance was used for the comparison.
This article is protected by copyright. All rights reserved.
2.
1.
3.
1.
3.
3.
3.
Table 4: NOE based Distance Restraint Violations in Structures Obtained from Restrained Molecular Dynamics of 2000 ps Run
Accepted Article
From Dynamic Trajectories Distances
From Final Minimization
1
2
3
4
5
6
1
2
3
4
5
6
7
7
7
7
7
7
7
7
7
7
7
7
Lower Bound
-
-
1
1
1
-
-
-
1
1
1
-
Upper Bound
-
-
-
-
1
-
-
-
-
-
-
-
-
-
14.2
14.2
14.2
-
-
-
14.2
14.2
14.2
-
8%
8%
8%
8%
8%
8%
-
-
14.2
Short range Violations
(>0.3Å)
%
Violations Lower Bound
Upper Bound
-
-
-
-
-
-
-
-
-
8%
Long range
14
14
14
14
14
14
14
14
14
14
14
14
Lower Bound
-
2
1
3
2
1
-
-
1
1
2
-
Upper Bound
-
-
-
-
-
1
1
1
-
=
-
3
-
14.2
7.14
21.4
14.2
7.1
-
-
7.14
7.14
-
-
8%
%
2%
8%
4%
%
%
-
-
-
-
7.1
7.1
7.1
-
-
14.2
21.4
4%
4%
4%
8%
2%
Violations
(>0.3Å)
%
Violations Lower Bound
Upper Bound
-
This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article
Accepted Date : 15-Apr-2014 Article type
: Research Article
Combined 1H-NMR and Molecular Dynamics Studies on Conformational Behaviour of a
1
Model Heptapeptide, GRGDSPC
Ashok K. Kulkarni1*, Rajendra P. Ojha2
Department of Physiology, Mediciti Institute of Medical Sciences, Hyderabad - 501401, A.P.,
India 2
Biophysics Unit, Department of Physics, University of Gorakhpur, Gorakhpur - 273001, India
Running Title: Conformation of Heptapeptide by 1H-NMR and Molecular Dynamics
Key Words: Arg-Gly-Asp; β-turn; distance geometry; -turn; 1H-NMR; restrained molecular dynamics.
*Corresponding Author: Department of Physiology, MediCiti Institute of Medical Sciences, Ghanpur, Medchal Mandal, R. R. Dist., Hyderabad – 501401, A. P., India Email: [email protected], [email protected] This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an 'Accepted Article', doi: 10.1111/cbdd.12346 This article is protected by copyright. All rights reserved.
Accepted Article
Phone: +91-9849845967
Abstract Among various strategies, the de novo design and in silico approaches are being used to develop the short peptides, models of modified peptides and mimetics as clinically useful drugs with improved stability and bioavailability. The resulting models will help to isolate the factors behind the folded structure formation, and contribute useful information about de novo peptide design. The combined 1
H-NMR spectroscopic and molecular dynamics methods were used to investigate the
conformational behavior of an Arg-Gly-Asp (RGD) containing peptide, GRGDSPC, the cell binding heptapeptide of extracellular matrix protein, fibronectin. The formation of two fused weak β-turns of type-II (HB, 4→1), and type-II' (HB, 7→4) from simulation studies has been consistent with NMR data. The sustainable ‘S’ shaped molecular structure (which remained unchanged during the entire simulation) and the conformational transitions due to interconversions between multiple turns initiated at Asp4, Ser5 and Cys7 implies that the peptide is flexible in nature. Thus, the model of ‘S’ shaped structure with flexible multiple turns for GRGDSPC peptide may provide the structural rationale for antagonistic properties of this heptapeptide towards the treatment of integrin mediated cellular abnormal behaviors such as thrombosis and metastasis.
Introduction The development of peptides into pharmaceutically acceptable drugs has been reported to be
hindered due to their limited stability and lack of high affinity (1, 2). Various strategies (3) such as de novo design (4), structural characterization and in silico approaches (5) were developed towards producing modified peptides, mimetics and models. The resulting models will help to isolate the factors behind the folded structure formation, and contribute useful information about de novo peptide design. The present study uses NMR and molecular dynamics methods to investigate the conformational behaviour of an Arg-Gly-Asp (RGD)-containing sequence, GRGDSPC. The This article is protected by copyright. All rights reserved.
heptapeptide GRGDSPC has been the topic of immense investigation in lieu of its involvement in cell-adhesion based cellular behaviors such as cell proliferation (6), osteogenic differentiation (7),
Accepted Article
and also in biomedical applications such as gene delivery (8), and tissue engineering of the heart valves (9). The amino acid sequence of Arg-Gly-Asp (RGD) has been reported to be serving as the
primary integrin recognition site in extra-cellular matrix proteins such as fibronectin, vitronectin, laminin, collagen, fibrinogen etc (10-13).
Towards studying the binding specificity of RGD
sequence, it has been shown that the mere positive and negative charges on Arg and Asp, respectively, were not sufficient for its recognition action (14). While proposing a strategy for the design and development of more effective drugs towards controlling platelet aggregation, the crystal structure of RGD containing disintegrin – TRIMESTATIN – has been studied (15). The studies using CD and folding algorithms methods reported that, the substitution of amino acids in RGD-spanning sequences were effective in changing the potency of RGDs as inhibitor of binding ligands (16-19). The synthesis and activity studies (20) on RGD related tetrapeptides RGDS, RGDV and RGDP have reported that the amino acid residues at the C-terminal were very crucial for anti-thrombosis effects in lieu of their total energies. The conformational studies (21) on RGDcontaining tetrapeptides such as H-RGDX-OH (X = Y, W, F, L, V, C, Q and S) reported that more populated β-turn structures were imparted for Gly-Asp (G-D) sequence by more hydrophobic sidechains of X-residues. Towards recognition by integrin receptors, a favorable conformation for solvent-exposed RGD site was reported to be maintained due to contributions from proline residues that were flanked in RGD motif in snake venom disintegrins, Rhodostomin (22) and Dendroaspin (23). However, the structural studies of a neuropeptide precursor protein (produced by snail, Lymnae stagnalis) with an RGD proteolytic site (24) revealed that the minimal structures around RGD region, due to proline residue following RGD site in that protein, do not inhibit platelet aggregation. The peptide with “Ser” residue following RGD site has been reported to be enhancing more reverse turns around RGD region, and was considered to be potently inhibiting the platelet This article is protected by copyright. All rights reserved.
aggregation. Moreover, the incorporation of Pro and Ser residues in RGD-flanking sequences such as GRGDTP, GRGDSP and some cyclic RGD peptides were shown to be decreasing the vasomotor
Accepted Article
response in rat isolated mesenteric arteriole (25). Though the linear peptides containing RGD sequence appeared to be of lower affinity for a bound state with the receptor II/III as compared to those of cyclic ones, additional specificity determinants might be enhanced with greater conformational flexibility for RGD containing linear peptides. The GRGDSPC synthetic peptide, though, has been shown to be non-inhibitory for cell adhesion to the 38-kD fragment of fibronectin (26), some studies reported the role of GRGDSPC sequence as an integrin inhibitor (27), and inhibiting the fibronectin attachment to keratinocytes (28), thymocytes (29), and B-lymphocytes (30). These different versions of interpretations for binding as well as inhibitory roles of the GRGDSPC created interest in further characterizing the heptapeptide. Till date, no structural aspects have been reported for GRGDSPC sequence as such by NMR or by X-ray diffraction methods. Hence, the present work emphasizes the conformational behavior of a model heptapeptide, GRGDSPC, by using 1H-NMR and molecular dynamics methods.
Methods and Materials Sample Preparation The heptapeptide, Gly-Arg-Gly-Asp-Ser-Pro-Cys (GRGDSPC) has been purchased from SigmaGenosys (Cambridge, UK) and used in its original form as it was found to be pure when examined using NMR.
The sample for NMR was prepared in D2O at a peptide concentration of 5 mM (pH,
7.0). After the desired NMR experiments were completed, the sample was lyophilized and redissolved in H2O/D2O (90:10) for the study of exchangeable amide protons.
NMR Experiments All NMR experiments were performed on a Bruker AMX-500 FT NMR Spectrometer. NMR experiments were performed at 25C except for experiments of temperature coefficients of amide
This article is protected by copyright. All rights reserved.
proton resonances. The temperature dependence of the amide proton chemical shift was determined from measurements at 5, 10, 15, 20, 25, 30 and 37 C for the peptide studied here. Phase-sensitive
Accepted Article
COSY (31) and TOCSY (32) experiments were performed to identify spin-systems of protons. Sequence-specific assignments of proton resonances were obtained by NOESY (33) (mixing time = 300, 400 and 500 ms) experiment. All two-dimensional NMR spectra were recorded using the TPPI method with the conventional pulse sequences, with the following parameters: spectral width (6000 Hz), 512 and 256 data points in t1 and t2 dimensions, relaxation delay (1 s). Water signal was suppressed by using presaturation during the recycling delay. The 3JHN-Hα coupling constants were determined by ECOSY (34) experiment.
Distance Restraints The assigned strong, moderate and weak NOE cross-peaks were translated into distances with the lower boundary limits set at 2.0 Å, and the upper boundary limits set at 2.5 Å for strong, 3.0 Å for medium and 4.0 Å for weak peaks (35). 3
J-Dihedral Angle Restraints
The coupling constant values, 3JHN-H and 3JH-H, were converted into dihedral angle information on phi () and chi () angles, respectively, using Karplus equation (36). J = A cos2 + B cos + C
3
where = | 60|. The values A = 6.8, B = 1.3 and C = 1.5 for phi () angles, and A = 9.5, B = 1.6 and C = 1.8 for chi () angles were incorporated.
Energy Calculations Molecular modelling software routines of Insight II and Discover90.0/3.0.0 (Biosym, USA) on Silicon Graphics Iris Power Indigo2 machine have been used for calculations of energy minimization and restrained molecular dynamics. The dielectric constant was fixed at 80.0. The
This article is protected by copyright. All rights reserved.
atom-based non-bonded interactions were calculated by using the consistent valence force field (CVFF) of Biosym with an initial cut-off 9.5 Å. To avoid the perturbations of the forces arising
Accepted Article
from small cut-off, which otherwise mask the effect of non-bonded interactions (37), this cut-off was later increased to 12.0 Å during molecular dynamics simulations. To result an overall neutral system, the atomic charges on the termini on Arg2 guanidine and Asp4 carboxylate groups were scaled by a factor of 0.25 relative to the default values. The distance restraints were employed in energy calculations using a harmonic function based flat-bottomed potential by scaling the force constants for restraints.
In addition to the distance restraints, the 3J-dihedral angle restraints and
hydrogen bonding restraints (NH—O = 1.8 - 2.7 Å, and N—O = 2.5 - 3.7 Å) were employed in restrained molecular dynamics (rMD) simulations for structure calculations.
Refinement Procedure The four steps initial model construction, constrained minimization, restrained molecular dynamics and energy minimization were carried out in refinement procedure.
i) Initial Model Construction While removing the initial strains using a gradient threshold of 0.1 kcal/(mol Å) with the steepest descents and to a maximum derivative of 0.01 kcal/(mol Å) with a conjugate gradient procedure, a linear peptide model was constructed in an extended conformation for Gly-Arg-Gly-Asp-Ser-ProCys sequence. The distance geometry procedure (DG-II) of the NMRchitecht module (Biosym, USA) was performed for the resulting structure by applying the restraints for distance, 3J-dihedral angle and hydrogen bond to generate 100 structures.
ii) Constrained Minimization All these 100 structures from distance geometry procedure were minimized using the steepest descents to a gradient threshold of 0.5 kcal/ (mol Å). These structures were further minimized to a
This article is protected by copyright. All rights reserved.
maximum gradient of 0.001 kcal/ (mol Å) using the conjugate gradients. On account of high energy, 30 structures (from these 100 structures) were discarded, and remaining 70 structures were grouped
Accepted Article
into 7 closely related families using cluster analysis based on internal torsion angles within the peptide backbone as well as hydrogens, heavy atoms and side-chain atom angles. The structures from two families did not yield hydrogen bond pattern to characterize any secondary structural features, hence not considered for further refinement. The structures from the remaining five families exhibited the hydrogen bonds that were characteristic for secondary structural features and were considered for further refinement. Among the 5 selected families, all structures from Family I and II exhibited one type of hydrogen bonding pattern of 41, and the representative structures from these two families were numbered as ‘S-1’ and ‘S-2’, respectively. However, each structure from two other families (Family III and IV) yielded three types of hydrogen bonding pattern, viz, 41, 52 and 74, and the representative structures from these two families were numbered as ‘S-3’ and ‘S-4’. From the remaining family (Family V), 50 % of structures exhibited the hydrogen bond pattern of 41 and 74, while the other 50% of structures yielded hydrogen bond patterns of 52 and 74. The representative conformer of the first 50% structures from this Family V was numbered as ‘S-5’, whereas that of the other 50% of structures was numbered as ‘S-6’. A further refinement was performed for all these six structures (S-1, S-2, S-3, S-4, S-5 and S-6) by using restrained molecular dynamics procedure.
iii) Restrained Molecular Dynamics Procedure By maintaining a constant temperature system and employing the velocity scaling, the restrained molecular dynamics procedure has been performed with restraint functions of distance, 3J-dihedral angle and hydrogen bonds throughout the calculations, which used an integrated time step of 1 fs. The starting structure was initially allowed to equilibrate at 50 K for 5 ps. The temperature of the system was raised to 100 K over a 25 ps interval. The target temperature was raised by 50 degrees every 10 ps until 298 K to equilibrate the system completely. The system was allowed to evolve at This article is protected by copyright. All rights reserved.
300 K for 100 ps followed by 2000 ps MD run at 300 K during which the structures were sampled every 0.5 ps. The structures obtained from the trajectories were used for subsequent geometric and
Accepted Article
energetic analysis. The conformational behaviour of structures obtained from molecular dynamics simulations were analysed for (a) key backbone dihedral angles (38), (b) the inter-proton distances, and (c) the calculated Cα-Cα (i, i+3) (39, 40) distances.
iv) Final Minimization The sampled lowest-energy-contained structure from the dynamics trajectories was minimized using the steepest descent method to a gradient threshold of 0.5 kcal/(mol Å) and further minimized to a maximum gradient of 0.001 kcal/(mol Å) using the conjugate gradients method. The geometric statistics similar to those for structures from dynamics trajectories were also calculated for the minimized structures.
Results Amide Proton Chemical Shifts and Temperature Dependence The 1D 1H-NMR spectrum of exchangeable amide protons of heptapeptide, GRGDSPC, is shown in Figure 1. The Arg2 NH was the one-proton signal (at about 7.2 ppm) at temperatures 5C – 37C with broadening and reduction in intensity of signal. NOE Connectivities The values of chemical shift for the 1H resonances, temperature coefficients of the amide protons and coupling constants of NH-CαH peaks are given in Table 1. The fingerprint region of the NOESY experiment (Figure 2) provided reliable sequential information and various NOE connectivities are summarized in the NOE connectivity diagram (Figure 3).
This article is protected by copyright. All rights reserved.
Structures for Analysis The ensemble of conformations of NMR structures obtained from a mixed protocol of distance
Accepted Article
geometry and simulated annealing is represented in Figure 4. All the six structures (S-1 through to S-6) generated from constrained minimization and cluster analysis (Section 2.6.ii.) were considered for further refinement using restrained molecular dynamics procedure. Consequently, the six structures obtained from dynamics trajectories and minimizations designated as Structure [S-1], [S-2] and [S-3] in Figure 5, and Structure [S-4], [S-5] and [S-6] in Figure 6 yielded a mixture of different type of turns. Inter-proton distances The NOE dependent distance restraints applied in the simulation, the comparison of relative magnitudes of NOE signals, and inter-proton distances generated by simulation of 2000 ps rMD are shown in Table 2 for intra-residue cross-peaks, and in Table 3 for inter-residue cross-peaks, and the distance restraints violations are summarized in Table 4. DISCUSSION Amide Proton Chemical Shifts and Temperature Dependence As compared to the peak of Gly3NH proton, the peak of Arg2 NH proton exhibited a significantly increased broadening, varying intensities and gradually vanishing as temperature raised, suggesting that Arg2NH proton was exchanging faster and, thus, not involved in hydrogen bonding, whereas Gly3 amide proton was associated in weak hydrogen bonding. The relative gradual broadening and reduction in intensities at higher temperatures for NH protons of Asp4 and Ser5, suggesting that amide protons of Asp4 and Ser5 were involved in weak hydrogen bonding. However, the temperature coefficient value of 3.42 (during 25 - 37º C) for the amide proton of Ser5 in one of ensembles was suggestive of involvement of Ser5 NH proton in strong hydrogen bond in that particular ensemble. The intensity of peak of Cys7 NH proton remained unchanged significantly as the temperature raised; suggesting that amide proton of Cys7 was involved in strong hydrogen bonding. The observations that the presence of more than one peak for the residues Asp4, Ser5 and This article is protected by copyright. All rights reserved.
Cys7 in temperature experiments, and that the temperature coefficient values (Table 1) being in more than one form, suggest that Asp4, Ser5 and Cys7 residues were involved in various different
Accepted Article
conformations. The splitting-up into two peaks by amide protons of Asp4 from 10C onwards suggests that, the amide protons of Asp4 were involved in two conformations after this temperature. Similarly, the presence of three different peaks of amide protons of Ser5 and Cys7 indicates that each of these two residues is involved in three conformations. NOE Connectivities The cross-peaks of (i) Arg2NH-Gly3NH, (ii) Gly3NH-Asp4NH, (iii) Arg2NH-Asp4NH, and (iv) Asp4NH-Ser5NH were indicative of the presence of a -turn. The observations of (i) the moderate to strong peak of Gly3NH-Asp4NH [dNN (3, 4)], and (ii) the weak or absence of Arg2HAsp4NH [dN (2, 4)] peaks were indicative of the presence of type I/II -turns. To distinguish between these two types, one can examine the NOE connectivities between the backbone protons of residues 2 and 3 of the turn. We have observed the cross peak of ArgH-Gly3NH [dN (2, 3)], which was an indication for the presence of type II -turn. This is also corroborated by two observations. Firstly, the distance corresponding to a weak peak of Arg2H-Gly3NH [dN (2,3)] should be more for type II than type I, and we obtained the larger distance corresponding to [dN (2,3)]. Secondly, the 3JNH-H coupling constant value (Table 1) of the second residue of the turn was 6.72 Hz, which was very close to those of standard values for type-II -turn. Therefore, the type-II -turn is predicted between the residues Gly1 and Asp4 (HB, 4→1).
The weak cross peak of Ser5NH-Cys7NH, the very weak cross peak of Asp4NH-Cys7NH and
a cross peak of Asp4H-Cys7NH were also observed, which suggest the presence of another turn between the residues Asp4 and Cys7. Because of the presence of Pro6 at the third position of this turn, it is impossible to obtain the cross-peak corresponding to: (i) [dN (2,3)], (ii) [dN (1,3)], (iii) [dNN (2,3)], (iv) [dNN (3,4)], (v) [dNN (1,3)], and (vi) [dN (2,3)]. We have not observed the cross This article is protected by copyright. All rights reserved.
peaks corresponding to (i) [dN (3, 4)], (ii) [dN (2, 4)], and (iii) [dN (1, 4)]. The weak cross peaks of Pro6H-Cys7NH and Pro6H-Ser5NH were observed to be overlapping. The 3JNH-H coupling
Accepted Article
constant value of Ser5 is 7.15 Hz, while that of Pro6 was absent. It is, therefore, inferred that the turn between Asp4 and Cys7 (HB, 74) is a II’ type of -turn. Structures for Analysis It is known from the structures in Figure 5 and Figure 6 that, the peptide shows tendency to form multiple conformations such as -turn (HB, 42), inverse -turn (HB, 31) in addition to -turn, type-III’ (HB, 41) and -turn type-II’ (HB, 74).
This suggests that the peptide exhibits
inherently weaker -turn types due to conformational transitions.
Inter-proton distance analysis A short-range distance of Arg2HN-Arg2Hα was violated through lower-bound limits in three structures from both dynamics trajectories as well as from minimizations. Similarly, a sequential medium-range distance of Arg2Hα-Gly3HN was violated through both lower and upper bound limits in almost all structures from dynamics trajectories as well as minimizations. Except these two distances, almost all inter-proton distances obtained from sampled structures and minimized structures were in fairly good agreement with the experimentally obtained distances, thus indicating the stable structures. The distances between C(Gly1) and C(Asp4), and between C(Asp4) and C(Cys7) in these structures from both the dynamics trajectories and minimizations were less than 7.0 Å, which also corroborate the presence of folded -turns between Gly1 and Asp4, and between Asp4 and Cys7. Flexibility of conformations in GRGDSPC sequence It is clear that, the -turn (HB, 31) was common in structures [S-1] and [S-2] from both the dynamic trajectories and minimization. The -turn (HB, 42) was exhibited by structure [S-3]
This article is protected by copyright. All rights reserved.
from dynamic trajectories. The hydrogen bond, 41 was shown to be common in structure [S-3] from trajectories and minimization but no definitive β-turn. The hydrogen bond, 74, exhibited by
Accepted Article
structures [S-1] and [S-3] from trajectories, and by structure [S-1] from minimization, yielded βturn type-III only in structure [S-3] from dynamics trajectory. The hydrogen bonds, 51 and 4→2 were common in structures [S-4], [S-5] from both the trajectories and minimization. However, in the structure [S-6] from trajectory the HB, 4→2 yielded inverse -turn, and in structure [S-6] from minimization the HB, 4→1 yielded β-turn type-II. The distribution of , angles and the pattern of hydrogen bond formations, therefore, indicate that residues Asp4, Ser5 and Cys7 were playing key roles in adopting the flexible conformations within the sequence of GRGDSPC. Consequently, the conformations like -turns, -turns, inverse -turns and random coil were observed during simulation. It has been shown that the GRGDSP peptide exhibits the duplicating ability to binding
activity of fibronectin with multiple -turns of type III-III or III-I (HB, 41 and 52) conformations (41). We have shown here that the GRGDSPC peptide prefers two fused -turns between Asp4 and Gly1 (HB, 41), and Cys7 and Asp4 (HB, 74). The formation of the -turn type-II between the residues Gly1 and Asp4 (HB, 41) from simulation is consistent with that from NMR data. But, the formation of -turn type-II’ between the residues Cys7 and Asp4 (HB, 74) in one of the distance geometry structures is also consistent with that from NMR data, suggests that the -turn type-II’ between the residues Cys7 and Asp4 (HB, 74) is inherently weaker, flexible and exists in the lesser population with tendency for multiple conformational transitions. However, our simulation studies suggest that in addition to these -turns of types said-above, the GRGDSPC peptide may also adopt a -turn of type-III (HB, 74), reverse -turn of type-III (HB, 47), turns (HB, 31 and 42), inverse -turn (HB, 42) and random coil.
This article is protected by copyright. All rights reserved.
Change in Rigidity and Activity as an Antagonist for Multiple Ligand Binding
Accepted Article
The proposed conformation is playing an inhibitory role in the interaction between glycoprotein GPIIbIIIa and fibrinogen. The side-chains of Ser5 and Cys7 in our present sequence GRGDSPC were not imparting a net hydrophobic movement to the molecule. They may enhance the binding by unshielding the Arg and Asp side-chain interactions with GPIIbIIIa from competition due to solvent. This suggests that its flexible conformational behaviour may play an important role in the activity of the peptide. Such a view is consistent with the comparative studies on structural characterization of RGD-cell-adhesion recognition sites of Streptavidin that despite the large conformational space occupied the surface loops (42) the high affinity for appropriate receptors has been shown to be displayed by only a few sampled structures from dynamics trajectories. Also a corroborative view is consistent with the comparative studies that up on incorporating Pro and Ser residues in RGD-flanking sequences such as GRGDTP, GRGDSP and some cyclic RGD peptides lead to a decreasing vasomotor response in rat isolated mesenteric arteriole (25). The literature abounds with the data that some of RGD containing peptides have been shown
to be flexible on account of non-established hairpin loop conformations (43-45), and the potency of peptides containing RGD motif in integrin binding has been shown to be in the order of GRGDNP > GRGDSP > GRGDTP = cyclo-RGD (46).
The structural studies on present sequence of
GRGDSPC show that there is a conformational change among two fused -turns between Asp4 and Gly1 (HB, 41), and Cys7 and Asp4 (HB, 74), and the peptide may also adopt a -turn of type-III (HB, 74), reverse -turn of type-III (HB, 47), -turns (HB, 31 and 42), inverse -turn (HB, 42).
Conclusion The present results imply that the GRGDSPC peptide exhibits a two-folded state of -turns, type-II (41) and type-II’ (74) between the residues Gly1 and Asp4, and Asp4 and Cys7, respectively.
This article is protected by copyright. All rights reserved.
Due to the key roles played by the residues Asp4, Ser5 and Cys7, this two-folded -turn state is flexible and tends to adopt into additional conformations like, -turn type-III (74), reverse -turn
Accepted Article
type-III (47), -turns (31 and 42) and inverse -turn (42). Moreover, the sustainable ‘S’ shaped structure with two-folded turns remained unchanged during the entire simulation. So, the conformational transitions with multiple turns within the sequence enhance less rigidity and lower activity in integrin-ligand interactions. Thus, the model of ‘S’ shaped structure with flexible multiple turns for GRGDSPC peptide may provide the structural rationale for antagonistic properties of this heptapeptide.
Acknowledgements The authors thank to high field 500 MHz FT-NMR National Facility, Tata Institute of Fundamental Research, Mumbai, India for providing us with experimentation facility.
References 1.
Humphrey M.J., Ringrose P.S. (1986) Peptides and related drugs: A review of their absorption, metabolism, and excretion. Drug Metab Rev;17:283–310.
2.
Plattner J.J., Norbeck D.W. (1990) Obstacles to drug development from peptide leads. In Clark, C.R., Moos, W.H., editors. Drug Discovery Technologies. Horwood, Chichester, England, p. 92-126.
3.
Adessi C., Soto C. (2002) Converting a peptide into a drug: strategies to improve stability and bioavailability. Curr Med Chem;9:963678.
4.
Cait E.M., Derek N.W. (2004) Engineered and designed peptide-based fibrous biomaterials. Curr Opin Sold State Mater Sci;8:141149.
5.
Mariana R.A., Francisco J.B., Luis S. (1996) De novo design and structural analysis of a model beta-hairpin peptide system. Nature Struct Biol;3:604612.
This article is protected by copyright. All rights reserved.
6. Jiang X.S., Chai C., Zhang Y., Zhuo R.X., Mao H.Q., Leong K.W. (2006) Surfaceimmobilization of adhesion peptides on substrate for ex vivo expansion of cryopreserved
Accepted Article
umbilical cord blood CD34+ cells. Biomaterials;27:2723-2732.
7. Zouani O.F., Chollet C., Guillotin B., Durrieu M.C. (2010) Differentiation of pre-osteoblast cells on poly(ethylene terephthalate) grafted with RGD and/or BMPs mimetic peptides. Biomaterials;31:8245-8253.
8. Haitao P., Qixin Z., Xiaodong G. (2007) A novel synthetic peptide vector system for optimal gene delivery to bone marrow stromal cells. J Pept Sci;13:154-163.
9. Shi J., Dong N., Sun Z. (2009 ) Immobilization of decellularized valve scaffolds with Arg-GlyAsp-containing peptide to promote myofibroblast adhesion. J Huazhong Univ Sci Technolog Med Sci;29:503-507.
10. Hayman E.G., Pierschbacher M.D., Ohgren Y., Rouslahti E. (1983) Serum spreading factor (vitronectin) is present at the cell surface and in tissues. Proc Natl Acad Sci USA;80:40034007.
11. Carlson R., Eugvall E., Freeman A. (1981) Rouslahti E. Serum spreading factor (vitronectin) is present at the cell surface and its tissues. Proc Natl Acad Sci USA;78:24032406.
12. Kleinman H.K., Klebe R.J., Martin G.R. (1981) Laminin and fibronectin in cell adhesion: enhanced adhesion of cells from regenerating liver to laminin. J Cell Biol;88:473485.
13. Assa-Munt N., Jia X., Lakhonen P., Rouslahti E. (2001) Solution structure and integrin binding activities of an RGD peptide with two isomers. Biochemistry;40:23732378.
14. Pierschbacher M.D., Rouslahti E. (1984) Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. Nature (London);309:491497.
15. Fujii Y., Okuda D., Fujimoto Z., Horii K., Morita T., Mizuno H. (2003) Crystal structure of trimestatin, a disintegrin containing a cell adhesion recognition motif RGD. J Mol Biol;332:111522.
This article is protected by copyright. All rights reserved.
16. Johnson W.C.Jr., Pagano T.G., Basson C.T., Madri J.A., Gooley P., Armitage I.M. (1993) Biologically active Arg-Gly-Asp oligopeptides assume a type II β-turn in solution.
Accepted Article
Biochemistry;32:268273.
17. Harris N.L., Rattray K.R., Tye C.E., Underhill T.M., Somerman M.J., D'Errico J.A., Chambers A.F., Hunter G.K., Goldberg H.A. (2000) Functional analysis of bone sialoprotein: Identification of the hydroxyapatite-nucleating and cell binding domains by recombinant peptide expression and site-directed mutagenesis. Bone;27:795802.
18. Suzuki K., Grinneld A.D., Kidokiro Y. (2002) Hypertonicity induced transmitter released at Drosophilla neuromuscular junction is partly mediated by integrins and cAMP/protein kinase. J Physiol;538:103109.
19. Smith J.B., Theakston R.D., Coelho A.L., Barja-Fidalgo C., Calvete J.J., Marcinkiewicz C. (2002) Characterization of a monomeric disintegrins, Ocellatussin, present in the venom of the Nigerial Carpet viper, Echiocellatus. FEBS Lett;512:111115.
20. Zhao M., Peng S.Q., Dilinuer S., Wang Y.Y., Zhou Q.L., Zhao W.Z. (1997) Studies on the synthesis and activities of RGD related peptides. Yao Xue Xue Bao;32:271277.
21. Park H.S., Kim C., Kang Y.K. (2002) Preferred conformations of RGDX tetrapeptides to inhibit the binding of fibrinogen to platelets. Biopolymers;63:298313.
22. Chang C.P., Chang J.C., Chang H.H., Tsai W.J., Lo S.J. (2001) Positional importance of Pro 53 adjacent to the Arg 49-Gly 50-Asp 51 sequence of rhodostomin in binding to integrin, alpha IIb beta 3. Biochem J;357:5764.
23. Lu X., Sun Y., Shang D., Wattam B., Egglezou S., Hughes T., Hyde E., Scully M., Kakkar V. (2001) Evaluation of the role of proline residues flanking the RGD motif of dendroaspin, an inhibitor of platelet aggregation and cell adhesion. Biochem J;355:633638.
24. Zachariah C., Cameron A., Lindberg I., Kao K.J., Beinfeld M.C., Edison A.S. (2001) Structural studies
of
a
neuropeptide
precursor
protein
Biochemistry;40:87908799. This article is protected by copyright. All rights reserved.
with
an
RGD
proteolytic
site.
25. Bekker E.N., Balt J.C., Pfaffendor M., Spann J.A., Van banell E. (2001) Vasomotor effects of arg-gly-asp (RGD) peptides are limited and not related to endothelium-derived hyperpolarizing
Accepted Article
factor-mediated relaxation in rat mesenteric arteries. Clin Exp Pharmacol Physiol;28:873876.
26. Ferreira O.C Jr., Garcia-Pardo A., Bianco C. (1990) Specific binding of the human monocytic cell line U937 to the alternatively spliced connecting segment (IIICS) of fibronectin. J Exp Med;171:351-356.
27. Barletta E., Mugnai G., Ruggieri S. (1993) Complex gangliosides modulate the integrinmediated adhesion in a rat hepatoma cell line. Biochem Biophys Res Commun;192:214-222.
28. Toda K., Tuan T.L., Brown P.J, Grinnell F. (1987) Fibronectin receptors of human keratinocytes and their expression during cell culture. J Cell Biol;105:3097-3104.
29. Torimoto Y., Kinebuchi M., Matsuura A., Kikuchi K., Uede T. (1990) A monoclonal antibody (8H3) that binds to rat T lineage cells and augments in vitro proliferative responses. J Exp Med;172:1315-1323.
30. Sánchez-Aparicio P., Dominguez-Jiménez C., Garcia-Pardo A. (1994) Activation of the alpha 4 beta 1 integrin through the beta 1 subunit induces recognition of the RGDS sequence in fibronectin. J Cell Biol;126:271-279.
31. Aue W.P., Bartholdi E., Ernst R.R. (1976) Two-dimensional spectroscopy: Application to nuclear magnetic resonance. J Chem Phys;64:22292245.
32. Brawnshweiler L., Ernst R.R. (1983) Coherence transfer by isotropic mixing: application to proton correlation spectroscopy. J Magn Reson;53:521528.
33. Kumar A., Ernst R.R., Wuthrich K. (1980) A two-dimensional nuclear Overhauser enhancement (2D NOE) experiment for the elucidation of complete proton-proton crossrelaxation networks in biological macromolecules. Biochem Biophys Res Commun;95:16.
34. Griensinger C., Sorensen O.W., Ernst R.R. (1985)Two-dimensional correlation of connected NMR transitions. J Am Chem Soc;107:63946396.
35. Wuthrich K. (1986) NMR of Proteins and Nucleic acids. John Wiley., New York, USA. This article is protected by copyright. All rights reserved.
36. Karplus M. (1959) Contact electron-spin coupling of nuclear magnetic moments. J Phys Chem;30:1115.
Accepted Article
37. Constantine K.L., Friedricks M.S., Stouch T.R. (1996) Extensive molecular dynamics simulation of a β-hairpin forming peptide. Biopolymers;39:591614.
38. Ramachandran G.N., Sasisekharan V. (1968) Conformation of polypeptides and proteins. Adv Protein Chem;23:283438.
39. Lewis P.N., Momany F.A., Scheraga H.A. (1973) Chain reversals in proteins. Biochem Biophys Acta;303:211229.
40. Efimov A.V. (1993) Standard structures in proteins. Prog Biophys Mol Biol;60:201239. 41. Reed J., Hull W.E., von der Lieth C.W., Kübler D., Suhai S., Kinzel V. (1988) Secondary structure of the Arg-Gly-Asp recognition site in proteins involved in cell-surface adhesion: Evidence for the occurrence of nested beta-bends in the model hexapeptide GRGDSP. Eur J Biochem;178:141154.
42. Le Trong I., McDevitt T.C., Nelson K.E., Stayton P.S., Stenkamp R.E. (2003) Structural characterization and comparison of RGD cell-adhesion recognition sites engineered into streptavidin. Acta Crystallogr D Biol Crystallogr;59:828834.
43. Dennis M.S., Henzel W.J., Pitti R.M., Lipari M.T., Napier M.A., Deisher T.A., Bunting S., Lazarus R.A. (1990) Platelet glycoprotein IIb-IIIa protein antagonists from snake venoms: evidence
for
a
family
of
platelet-aggregation
inhibitors.
Proc
Natl
Acad
Sci
USA;87:24712475.
44. Cooke R.M., Carter B.G., Martin D.M., Murray-Rust P., Weir M.P. (1991) Nuclear magnetic resonance studies of the snake toxin echistatin:1H resonance assignments and secondary structure. Eur J Biochem;202:323328.
45. Dalvit C., Widmer H., Bowermann G., Breckenridge R., Metlernich R. (1991) 1HNMR studies of echistatin in solution: sequential resonance assignments and secondary structure. Eur J Biochem;202:315321. This article is protected by copyright. All rights reserved.
46. Umesh A., Thompson M.A., Chini E.N., Yip K.P., Sham J.S. (2006) Integrin ligands mobilize Ca2+ from ryanodine receptor-gated stores and lysosome-related acidic organelles in pulmonary
Accepted Article
arterial smooth muscle cells. J Biol Chem;45:3431234323.
Figure Legends Figure 1: Variable temperature 500-MHz 1H-NMR spectra of the exchangeable amide protons for GRGDSPC in H2O/D2O (90:10), pH 7. Temperatures covering the range of 5C - 37C are indicated.
Figure 2: Cross-relaxation peaks for amide protons of GRGDSPC in H2O at 25C.
A two-
dimensional cross-relaxation experiment (NOESY) was performed with a mixing time of 500 ms. The vertical axis (F1 domain) gives the chemical shifts of protons participating with NH (F2 domain; horizontal axis) in cross-relaxation (HN-H connectivities).
Figure 3: Schematic diagram of sequential, medium, and long distance NOE connectivities found in the heptapeptide GRGDSPC. The thickness of the bars reflects the strength of the NOE connectivities grouped into strong, medium, and weak.
Figure 4: Superimposition of backbone structures obtained from NMR data by using a mixed protocol of distance geometry and simulated annealing. Five groups of structures representing for five families from distance geometry procedure are shown. Each structure exhibited two turns, one with 41 hydrogen bonded conformer (the first turn, the upper one) and the other with 74 hydrogen bonded conformer (the second turn, the lower one). The structures from two families with no characteristic main-chain hydrogen bonds are not shown.
This article is protected by copyright. All rights reserved.
Figure 5: Molecular graphics renditions of -turns, inverse -turns, -turns of type II’ and III of GRGDSPC obtained from a 2000 ps rMD run. The structures designated by [S-1], [S-2] and [S-3]
Accepted Article
were obtained from dynamics trajectories (upper row) and minimizations (lower row). The carbons are labelled with the amino acid residue numbers in the sequence GRGDSPC. Hydrogen bonds, characteristics for secondary structures, are labelled with proton donors and proton acceptors.
Figure 6: Molecular graphics renditions of -turns and -turn of type II of GRGDSPC obtained from a 2000 ps rMD run. The structures designated by [S-4], [S-5] and [S-6] from dynamics trajectories (upper row) and minimizations (lower row). The -carbons are labelled with the amino acid residue numbers in the sequence GRGDSPC. Hydrogen bonds, characteristics for secondary structures, are labelled with proton donors and proton acceptors.
Table 1: Chemical shifts (ppm), Temperature Coefficients of NH Protons (ppm K-1 x 10–3) and Coupling - Constants (Hz) of GRGDSPC in H2O at 300K Protons
Gly1
Arg2
Gly3
Asp4
Ser5
Pro6
Cys7
HN
-
8.67
8.61
8.34
8.26
-
8.19
H
3.88
4.34
3.93
4.79
4.48
4.46
4.57
H
-
1.85
-
2.85
3.85
2.29
3.25
1.78
-
2.79
3.82
2.03
2.96
1.65
-
-
-
2.0
-
1.65
-
-
-
2.0
-
3.20
-
-
-
3.79
-
H
H
-
-
This article is protected by copyright. All rights reserved.
Accepted Article
HN
/
-
-
3.20
-
-
-
3.73
-
7.18
-
-
-
-
-
7.18
-
-
-
-
-
7.43
8.75
(ppb/K)
3
JNH--H
3
JH--H
6.19 (i)
8.47 (i)
9.06 (ii)(a)
8.66 (ii)(a)
10.5 (ii)(b)
9.56 (i) -
3.42 (ii)(b)
14.66 (ii)(d) 6.11 (ii)(e)
8.5 (ii)(c)
10.06(iii)(d)
6.75 (iii)(b)
6.94 (iii)(e)
-
6.72
-
9.80
7.15
-
6.19
-
8.3
-
5.2
5.5
10.0
6.5
Temperature coefficients taken at (a) 10-25C, (b) 25-37C, (c) 15-25C, (d) 5-20C, and (e) 20-37C for ensembles (i), (ii) and (iii).
This article is protected by copyright. All rights reserved.
Table 2: Comparisons between Intra-residue NOE Cross-Peak Intensities and Distances Obtained from
Accepted Article
Dynamics Trajectories and Minimization of Molecular Dynamics run of 2000 ps From Dynamic
Cross-Peaks
Trajectories I*
#
Arg2HN-
From Final Minimization
M
Arg2H
L
U
B
B
*
*
2. 00
1
2
3
4
5
6
1
2
3
4
5
6
3.
1.7
1.
1.
1.
1.
2.
1.
1.6
1.
1.2
1.
2.
00
3
77
29
26
23
57
71
9
30
2
21
4 6
Gly3HN-
S
Gly3XH
2.
2.
2.6
2.
2.
2.
2.
2.
2.
2.6
2.
2.5
2.
2.
00
50
1
69
62
55
84
52
62
1
58
8
57
5 9
Arg2HN-
M
Arg2XH
2.
3.
3.0
3.
2.
3.
3.
2.
3.
3.1
2.
3.0
3.
2.
00
00
6
06
46
02
03
64
06
0
40
3
04
6 2
Arg2HN-
M
Arg2XH
2.
3.
3.0
2.
3.
3.
2.
3.
2.
3.0
3.
3.0
2.
2.
00
00
0
98
02
02
53
00
85
1
02
2
59
7 0
Arg2XH-
S
Arg2XH
2.
2.
2.5
2.
2.
2.
2.
2.
2.
2.5
2.
2.5
2.
2.
00
50
5
62
55
55
55
55
55
5
57
7
55
7 9
Arg2XH-
M
Arg2XH
2.
3.
3.0
3.
3.
2.
3.
3.
3.
3.0
3.
3.0
3.
3.
00
00
1
03
01
98
01
01
01
1
04
4
01
2 1
Arg2XHArg2XH
W
2.
4.
4.0
3.
4.
4.
4.
4.
4.
4.0
3.
3.3
4.
2.
00
00
3
33
03
04
03
02
03
2
43
0
03
9 4
This article is protected by copyright. All rights reserved.
*I = Intensity (i.e. S for Strong, M for Medium and W for Weak); * LB = Lower Bound; *UB = Upper Bound.
Accepted Article
# Atoms designated ‘X’ (e.g. XH) are pseudo-atoms representing the average location of the unresolved constituent atoms (XH1 and XH2 in this example). For peaks involving ambiguous prochiral hydrogens, the atom pair resulting in the shortest average distance was used for the comparison.
Table 3: Comparisons between Inter-residue NOE Cross-Peak Intensities and Distances Obtained from
Dynamics Trajectories and Minimization of Molecular Dynamics run of 2000 ps
From Dynamic Trajectories
Cross-Peaks #
Gly1XH-
I*
S
Arg2HN Gly3HN-
W
Arg2XH Gly3HN-
W
Arg2XH Gly3XH-
M
Asp4HN Arg2HN-
M
Gly3HN Arg2HN-
W
Ser5HN Gly3HNAsp4HN
M
LB
UB
*
*
2.
From Final Minimization
1
2
3
4
5
6
1
2
3
4
5
6
2.
2.5
2.
2.
2.
2.
2.5
2.
2.5
2.
2.
2.
2.
00
50
2
52
51
55
54
2
51
3
51
53
53 83
2.
4.
1.7
1.
1.
1.
1.
1.3
1.
1.7
1.
1.
1.
00
00
3
66
79
84
83
6
79
3
78
83
83 87
2.
4.
3.1
3.
2.
3.
2.
3.1
3.
3.1
2.
2.
2.
00
00
5
15
96
04
71
1
11
6
94
70
82 91
2.
3.
3.0
2.
3.
3.
3.
2.9
3.
3.0
3.
3.
3.
00
00
2
95
02
02
03
2
00
0
01
02
02 17
2.
3.
3.1
3.
3.
3.
3.
3.0
3.
3.1
3.
3.
3.
00
00
4
14
19
13
12
1
15
1
19
12
12 49
2.
4.
4.0
4.
4.
4.
4.
4.0
3.
4.0
4.
3.
3.
00
00
1
07
01
02
03
1
64
2
03
69
51 81
2.
3.
1.7
1.
2.
3.
3.
2.9
1.
1.8
2.
3.
3.
00
00
0
84
86
02
01
8
80
7
28
04
01 08
This article is protected by copyright. All rights reserved.
2.
3.
3.
4.
4.
2.
Asp4HN-
S
Accepted Article
Ser5HN Asp4HN-
W
Cys7HN Ser5HN-
W
Cys7HN Arg2H-
S
Gly3HN Asp4XH-
M
Cys7HN Ser5XH-
M
Cys7HN Pro6HCys7HN
M
2.
2.
2.1
2.
2.
1.
1.
1.6
2.
2.5
1.
1.
1.
00
50
4
21
11
49
48
9
00
1
84
92
99 40
2.
4.
3.0
2.
3.
4.
4.
1.8
3.
3.7
1.
4.
3.
00
00
3
24
80
01
02
3
74
2
70
01
55 72
2.
4.
2.8
3.
2.
3.
3.
2.4
2.
3.9
2.
4.
2.
00
00
8
36
91
78
67
6
98
0
80
00
60 01
2.
2.
2.9
3.
1.
1.
1.
3.4
2.
3.0
1.
1.
1.
00
50
7
12
29
37
41
0
97
5
28
42
45 91
2.
3.
3.0
1.
3.
1.
1.
3.0
3.
1.9
3.
1.
1.
00
00
7
91
06
43
71
4
04
6
02
75
20 26
2.
3.
3.0
3.
3.
3.
3.
3.0
3.
3.0
3.
3.
3.
00
00
9
09
09
07
05
5
09
6
09
02
10 68
2.
3.
2.8
3.
2.
2.
2.
3.0
2.
2.4
2.
3.
2.
00
00
8
02
87
79
90
2
84
3
83
02
84 23
*I = Intensity (i.e. S for Strong, M for Medium and W for Weak); * LB = Lower Bound; *UB = Upper Bound. # Atoms designated ‘X’ (e.g. XH) are pseudo-atoms representing the average location of the unresolved constituent atoms (XH1 and XH2 in this example). For peaks involving ambiguous prochiral hydrogens, the atom pair resulting in the shortest average distance was used for the comparison.
This article is protected by copyright. All rights reserved.
2.
1.
3.
1.
3.
3.
3.
Table 4: NOE based Distance Restraint Violations in Structures Obtained from Restrained Molecular Dynamics of 2000 ps Run
Accepted Article
From Dynamic Trajectories Distances
From Final Minimization
1
2
3
4
5
6
1
2
3
4
5
6
7
7
7
7
7
7
7
7
7
7
7
7
Lower Bound
-
-
1
1
1
-
-
-
1
1
1
-
Upper Bound
-
-
-
-
1
-
-
-
-
-
-
-
-
-
14.2
14.2
14.2
-
-
-
14.2
14.2
14.2
-
8%
8%
8%
8%
8%
8%
-
-
14.2
Short range Violations
(>0.3Å)
%
Violations Lower Bound
Upper Bound
-
-
-
-
-
-
-
-
-
8%
Long range
14
14
14
14
14
14
14
14
14
14
14
14
Lower Bound
-
2
1
3
2
1
-
-
1
1
2
-
Upper Bound
-
-
-
-
-
1
1
1
-
=
-
3
-
14.2
7.14
21.4
14.2
7.1
-
-
7.14
7.14
-
-
8%
%
2%
8%
4%
%
%
-
-
-
-
7.1
7.1
7.1
-
-
14.2
21.4
4%
4%
4%
8%
2%
Violations
(>0.3Å)
%
Violations Lower Bound
Upper Bound
-
This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
Accepted Article This article is protected by copyright. All rights reserved.
