Heparin Binding by Fibronectin Module III-13 Involves Six Discontinuous Basic Residues Brought Together to Form a Cationic Cradle
1995; Elsevier BV; Volume: 270; Issue: 31 Linguagem: Inglês
10.1074/jbc.270.31.18558
ISSN1083-351X
AutoresThomas F. Busby, W. Scott Argraves, Shelesa A. Brew, Igor Pechik, Gary L. Gilliland, Kenneth C. Ingham,
Tópico(s)Peptidase Inhibition and Analysis
ResumoThe thirteenth type III domain of fibronectin binds heparin almost as well as fibronectin itself and contains a so-called heparin-binding consensus sequence, Arg6-Arg7-Ala8-Arg9 (residues 1697-1700 in plasma fibronectin). Barkalow and Schwarzbauer (Barkalow, F. J., and Schwarzbauer, J. E. (1991) J. Biol. Chem. 266, 7812-7818) showed that mutation of Arg6-Arg7 in domain III-13 of recombinant truncated fibronectins abolished their ability to bind heparin-Sepharose. However, synthetic peptides containing this sequence have negligible affinity for heparin (Ingham, K. C., Brew, S. A., Migliorini, M. M., and Busby, T. F.(1993) Biochemistry 32, 12548-12553). We generated a three-dimensional model of fibronectin type III-13 based on the structure of a homologous domain from tenascin. The model places Arg23, Lys25, and Arg54 parallel to and in close proximity to the Arg6-Arg7-Ala8-Arg9 motif, suggesting that these residues may also contribute to the heparin-binding site. Domain III-13 and six single-site mutants containing Ser in place of each of the above-mentioned basic residues were expressed in Escherichia coli. All of the purified mutant domains melted reversibly with a T m near that of the wild type indicating that they were correctly folded. When fluorescein-labeled heparin was titrated at physiological ionic strength, the wild type domain increased the anisotropy in a hyperbolic fashion with a Kd of 5-7 μM, close to that of the natural domain obtained by proteolysis of fibronectin. The R54S mutant bound 3-fold weaker and the remaining mutants bound at least 10-fold weaker than wild type. The results point out that the Arg6-Arg7-Ala8-Arg9 consensus sequence by itself has little affinity for heparin under physiological conditions, even when presented in the context of a folded domain. Thus, the heparin-binding site in fibronectin is more complex than previously realized. It is formed by a cluster of 6 positively charged residues that are remote in the sequence but brought together on one side of domain III-13 to form a "cationic cradle" into which the anionic heparin molecule could fit. The thirteenth type III domain of fibronectin binds heparin almost as well as fibronectin itself and contains a so-called heparin-binding consensus sequence, Arg6-Arg7-Ala8-Arg9 (residues 1697-1700 in plasma fibronectin). Barkalow and Schwarzbauer (Barkalow, F. J., and Schwarzbauer, J. E. (1991) J. Biol. Chem. 266, 7812-7818) showed that mutation of Arg6-Arg7 in domain III-13 of recombinant truncated fibronectins abolished their ability to bind heparin-Sepharose. However, synthetic peptides containing this sequence have negligible affinity for heparin (Ingham, K. C., Brew, S. A., Migliorini, M. M., and Busby, T. F.(1993) Biochemistry 32, 12548-12553). We generated a three-dimensional model of fibronectin type III-13 based on the structure of a homologous domain from tenascin. The model places Arg23, Lys25, and Arg54 parallel to and in close proximity to the Arg6-Arg7-Ala8-Arg9 motif, suggesting that these residues may also contribute to the heparin-binding site. Domain III-13 and six single-site mutants containing Ser in place of each of the above-mentioned basic residues were expressed in Escherichia coli. All of the purified mutant domains melted reversibly with a T m near that of the wild type indicating that they were correctly folded. When fluorescein-labeled heparin was titrated at physiological ionic strength, the wild type domain increased the anisotropy in a hyperbolic fashion with a Kd of 5-7 μM, close to that of the natural domain obtained by proteolysis of fibronectin. The R54S mutant bound 3-fold weaker and the remaining mutants bound at least 10-fold weaker than wild type. The results point out that the Arg6-Arg7-Ala8-Arg9 consensus sequence by itself has little affinity for heparin under physiological conditions, even when presented in the context of a folded domain. Thus, the heparin-binding site in fibronectin is more complex than previously realized. It is formed by a cluster of 6 positively charged residues that are remote in the sequence but brought together on one side of domain III-13 to form a "cationic cradle" into which the anionic heparin molecule could fit. INTRODUCTIONThe interaction of heparin or heparan-sulfate glycosaminoglycans (GAGs) 1The abbreviations used are: GAGsglycosaminoglycansPCRpolymerase chain reaction. with proteins occurs in a variety of physiological processes including blood coagulation, lipoprotein metabolism, cell adhesion and migration, and regulation of growth factor activity(1Jackson R.L. Busch S.J. Cardin A.D. Physiol. Rev. 1991; 71: 481-539Crossref PubMed Scopus (955) Google Scholar, 2Bourin M. Lindahl U. Biochem. J. 1993; 289: 313-330Crossref PubMed Scopus (392) Google Scholar, 3Klagsbrun M. Baird A. Cell. 1991; 67: 229-231Abstract Full Text PDF PubMed Scopus (497) Google Scholar, 4Woods A. Couchman J.R. Adv. Exp. Med. Biol. 1992; 313: 87-96Crossref PubMed Google Scholar). Efforts to understand the nature of such interactions are hampered by the fact that GAGs are generally heterogeneous in their size and charge density(5Lindahl U. Kjelln L. Thromb. Hemostasis. 1991; 66: 44-48Crossref PubMed Scopus (70) Google Scholar). Furthermore, there is no example of a detailed structure of a GAG•protein complex determined by x-ray or NMR methods, although some attempts at molecular modeling have been made(6Grootenhuis P.D.J. van Boeckel C.A.A. J. Am. Chem. Soc. 1991; 113: 2743-2747Crossref Scopus (87) Google Scholar, 7Cardin A.D. Weintraub H.J.R. Arteriosclerosis. 1989; 9: 21-32Crossref PubMed Google Scholar, 8Mann D.M. Romm E. Migliorini M. J. Biol. Chem. 1994; 269: 23661-23667Abstract Full Text PDF PubMed Google Scholar, 9Stuckey J.A. St. Charles R. Edwards B.F.P. Proteins. 1992; 14: 277-287Crossref PubMed Scopus (89) Google Scholar). That electrostatic forces are involved is evident from the fact that the complexes can be disrupted by increasing ionic strength. It is clear that positively charged Arg and/or Lys residues on the protein play an important role because their modification by chemical (10Rosenberg R.D. Damus P.S. J. Biol. Chem. 1973; 248: 6490-6505Abstract Full Text PDF PubMed Google Scholar, 11Harper J.W. Lobb R.R. Biochemistry. 1988; 27: 671-678Crossref PubMed Scopus (58) Google Scholar) or recombinant (12Barkalow F.J. Schwarzbauer J.E. J. Biol. Chem. 1991; 266: 7812-7818Abstract Full Text PDF PubMed Google Scholar, 13Stone S.R. Brown-Luedi M.L. Rovelli G. Guidolin A. McGlynn E. Monard D. Biochemistry. 1994; 33: 7731-7735Crossref PubMed Scopus (52) Google Scholar, 14Ma Y. Henderson H.E. Liu M.-S. Zhang H. Forsythe I.J. Clarke-Lewis I. Hayden M.R. Brunzell J.D. J. Lipid Res. 1994; 35: 2049-2059Abstract Full Text PDF PubMed Google Scholar) means usually leads to a reduction in the affinity for heparin, the most widely studied GAG. This is supported by numerous studies with synthetic heparin-binding peptides. However, in cases where the affinities have been measured, the peptides rarely bind as tightly as the parent protein on which their sequence is based. This probably means that the tertiary structure of the protein holds the peptide in a conformation that is more complementary to the GAG or that other residues not contained in the peptide contribute to the binding site in the protein, or both.The present study concerns fibronectin, a large modular glycoprotein that interacts with a variety of macromolecules in the extracellular matrix and with cell-surface molecules such as integrins and heparan-sulfate proteoglycans. Such interactions are important in regulating cell behavior including growth, adhesion, spreading, migration, and differentiation(15Hynes R.O. Fibronectins. Springer-Verlag, New York1990Crossref Google Scholar). Recent studies have shown that certain cell types require both the GAG binding and cell binding regions of fibronectin for efficient spreading and formation of focal contacts(16Woods A. Couchman J.R. Johansson S. Hook M. EMBO J. 1986; 5: 665-670Crossref PubMed Scopus (322) Google Scholar, 17Woods A. McCarthy J.B. Furcht L.T. Couchman J.R. Mol. Biol. Cell. 1993; 4: 605-613Crossref PubMed Scopus (182) Google Scholar, 18Yoneda J. Saiki I. Igarashi Y. Kobayashi H. Fujii H. Ishizaki Y. Kimizuka F. Kato I. Azuma I. Exp. Cell Res. 1995; 217: 169-179Crossref PubMed Scopus (27) Google Scholar). The interaction of fibronectin with heparin is dominated by the COOH-terminal hep-2 region which can be isolated as a 30-kDa proteolytic fragment consisting of fibronectin type III domains 12 through 14. Previous work has shown that domain III-13, when isolated by further proteolysis or expressed in Escherichia coli as an independent domain, binds heparin in solution with almost the same affinity as the parent fragment(19Ingham K.C. Brew S.A. Migliorini M.M. Busby T.F. Biochemistry. 1993; 32: 12548-12553Crossref PubMed Scopus (45) Google Scholar). A cationic cluster, Arg6-Arg7-Ala8-Arg9, located near the NH2 terminus of III-13, matches one of two patterns that are commonly found in heparin-binding proteins, namely BBXB, where B represents a basic residue, Arg, Lys, or His, and X represents any residue(7Cardin A.D. Weintraub H.J.R. Arteriosclerosis. 1989; 9: 21-32Crossref PubMed Google Scholar). Arg6 and Arg7 were shown to be critical in that their simultaneous mutation abolished the ability of truncated recombinant fibronectins ("deminectins") to bind heparin-Sepharose under physiological conditions(12Barkalow F.J. Schwarzbauer J.E. J. Biol. Chem. 1991; 266: 7812-7818Abstract Full Text PDF PubMed Google Scholar). However, synthetic peptides containing this cluster have negligible affinity for heparin under physiological conditions, indicating that tertiary structure and/or other residues are important (19Ingham K.C. Brew S.A. Migliorini M.M. Busby T.F. Biochemistry. 1993; 32: 12548-12553Crossref PubMed Scopus (45) Google Scholar).Fibronectin type III domains are among the most ubiquitous of protein modules, occurring in about 2% of animal proteins(20Bork P. Doolittle R.F. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 8990-8994Crossref PubMed Scopus (195) Google Scholar). The three-dimensional structures of several type III domains have been elucidated, and they all show a similar β-sandwich fold which resembles that of the immunoglobulin C domain(21Leahy D.J. Hendrickson W.A. Aukhil I. Erickson H.P. Science. 1992; 258: 987-991Crossref PubMed Scopus (441) Google Scholar, 22Main A.L. Harvey T.S. Baron M. Boyd J. Campbell I.D. Cell. 1992; 71: 671-678Abstract Full Text PDF PubMed Scopus (420) Google Scholar, 23Dickinson C.D. Veerapandian B. Dai X. Hamlin R.C. Xuong N. Ruoslahti E. Ely K.R. J. Mol. Biol. 1994; 236: 1079-1092Crossref PubMed Scopus (183) Google Scholar, 24Huber A.H. Wang Y. Bieber A.J. Bjorkman P.J. Neuron. 1994; 12: 717-731Abstract Full Text PDF PubMed Scopus (102) Google Scholar). To gain insight into the three-dimensional arrangement of cationic residues in the folded structure of III-13, a molecular model was generated based on the known three-dimensional structure of a homologous type III domain in tenascin(21Leahy D.J. Hendrickson W.A. Aukhil I. Erickson H.P. Science. 1992; 258: 987-991Crossref PubMed Scopus (441) Google Scholar). The model predicted that in addition to the above-mentioned arginines 6, 7, and 9, which would be expected to fall close to each other, 3 additional cationic residues should be located nearby in which case they might also contribute to the heparin-binding site. Based on the model, all 6 residues were individually mutated, and the resulting expression products were tested for the presence of compact structure and their ability to bind heparin in the fluid and solid phases. The results are embodied in the title of this paper.MATERIALS AND METHODSMolecular ModelingThe initial model of fibronectin domain III-13 was generated using as a template the three-dimensional structure of a homologous domain III-3 from tenascin(21Leahy D.J. Hendrickson W.A. Aukhil I. Erickson H.P. Science. 1992; 258: 987-991Crossref PubMed Scopus (441) Google Scholar). The amino acid sequences of III-13 that differed from tenascin III-3 were replaced and adjusted manually to avoid stearic overlap using the program FRODO(25Jones T.A. J. Appl. Crystallogr. 1978; 11: 268-272Crossref Google Scholar). This initial model was then subjected to energy minimization and molecular dynamics using the X-PLOR 3.1 program package(26Brunger A.T. X-PLOR: version 3.1. Yale University Press, New Haven, CT1992Google Scholar). In the first step of this process, the model was subjected to 150 cycles of conjugate gradient energy minimization to further reduce stearic contacts. During this initial minimization, coordinates of side chain atoms of conservative residues and all Ca atoms were held near their initial positions by applying additional harmonic potential restraints. The coordinates were modified by subjecting the model to a two-stage molecular dynamics simulation. In the first stage the model constrained as above was heated from 0 to 300 K in 5 ps and then equilibrated for 50 ps. To increase the mobility of the side chains in this simulation, the partial atomic charges were reduced to 0.5 of their initial values, and hydrogen bond energy terms were turned off. In the second stage of the simulation, the model was equilibrated for another 50 ps with the harmonic restraints of the model removed, the partial atomic charges restored to their original values, and the hydrogen bond energy terms turned on. The resulting model was then subjected to a final 200 cycles of conjugate gradient energy minimization to optimize the geometry and stereochemistry of the final structure.Expression of Recombinant DomainsExpression of recombinant III-13 and its mutants as fusion proteins with maltose-binding protein was accomplished in E. coli using a modified pMAL-p2 expression vector (New England Biolabs) in a manner similar to that previously described(27Matsuka Y.V. Medved L.V. Brew S.A. Ingham K.C. J. Biol. Chem. 1994; 269: 9539-9546Abstract Full Text PDF PubMed Google Scholar). cDNA fragments encoding residues 1-89 of domain III-13, i.e. Asn1692-Thr1780 of plasma fibronectin(28Skorstengaard K. Jensen M. Sahl P. Petersen T. Magnusson S. Eur. J. Biochem. 1986; 161: 441-453Crossref PubMed Scopus (151) Google Scholar), were prepared by PCR using 21-base synthetic primers flanking the desired regions and cDNA encoding human fibronectin kindly provided by Dufour et al.(29Dufour S. Gutman A. Bois F. Lamb N. Thiery J.P. Kornblihtt A.R. Exp. Cell Res. 1991; 193: 331-338Crossref PubMed Scopus (16) Google Scholar). The primers also contained the sequence for the BamHI and HindIII restriction sites for ligation into the pMAL-p2 expression vector. Approximately 20 ng of the ligated DNA was used to transform TB1 E. coli cells by electroporation. The electroporated cell/DNA mixture was grown in LB media at 37°C for 1 h. Serial dilutions were spread onto plates covered with LB media containing ampicillin and incubated at 37°C overnight. After screening for the presence of insert by restriction analysis and/or PCR analysis, individual colonies were grown overnight at 37°C with shaking at 225 revolutions/min, and diluted 100-fold with fresh ampicillin-containing media. After growing at 37°C for about 2.5 h or until the absorbance at 600 nm reached about 0.6, the cells were induced with isopropyl-1-thio-β-D-galactopyranoside at a final concentration of 0.3 mM and grown for an additional 3 h. Although the fusion proteins were preceded by a signal peptide sequence, the bulk of the product was found in the cytoplasm. The cells were harvested by centrifugation, lysed by sonication in 0.005% Triton X-100, and the fusion protein was purified immediately by affinity chromatography on immobilized amylose and/or heparin. The yield of fusion protein varied between 5 and 15 mg/liter. For unknown reasons, storage of the lysate at 4°C resulted in progressively lower yield of material binding to the amylose resin. Similarly, the efficiency of rebinding of maltose-binding protein and fusion protein to the amylose resin, even after exhaustive dialysis, was too low to be useful in further steps of purification. Therefore, after digestion of the fusion proteins with factor Xa(30Guan C.d. Li P. Riggs P.D. Inouye H. Gene (Amst.). 1988; 67: 21-30Crossref PubMed Scopus (543) Google Scholar), the liberated intact III-13 domains were purified to homogeneity by rechromatography on heparin-Sepharose and/or QMA anion-exchange media (Waters) and size-exclusion chromatography as needed. The amino terminus of the final products contained 2 additional residues, Ile and Leu, that are not part of the III-13 domain.MutagenesisArg → Ser and Lys → Ser mutations were prepared by the method of Ho et al.(31Ho S.N. Hunt H.D. Horton R.M. Pullen J.K. Pease L.R. Gene (Amst.). 1989; 77: 51-59Crossref PubMed Scopus (6797) Google Scholar). Synthetic complementary 21-base primers containing the desired mutation and spanning the region to be mutated, as well as two primers exterior to the type III-13 insert, were annealed to the wild type III-13 cDNA template. PCR extension of these primers resulted in two cDNAs that were then annealed to each other through their 21-base overlapping region (containing the mutation) and, using the primers exterior to the III-13 insert, extended again to produce the complete mutated inserts. The 6-fold degeneracy in the codon for serine was exploited, when possible, to introduce unique restriction sites in the mutated insert. In those cases, digestion with the unique restriction enzyme followed by electrophoresis could be used as an initial test for the presence of the mutation before the insert was ligated into the plasmid. The R6S mutation (TCG for AGA) introduced a second TaqI site (T-CGA), the R7S mutation (TCG for AGG) added a third MboI site (-GATC), the R9S mutation (TCT for CGT) inserted a unique XmnI site (GAANN-NNTTC), the R23S mutation (TCG for AGA) produced a unique SalI site (G-TCGAC), the K25S mutation (TCG for AAG) introduced a second TaqI site (T-CGA), the R47S mutation added a unique SalI site (G-TCGAC), and the R54S mutation inserted a unique XhoI site (C-TCGAG). After initial confirmation of the mutation by digestion of the PCR products at these unique sites, the mutated inserts were ligated into the pMAL-p2 plasmid, transformed, expressed, and purified as outlined above. Repeated attempts to generate the K25S mutation using an AGT codon for the serine (unique ScaI site) were unsuccessful; only about 25% of the plasmid was cleavable, and induced cells produced no fusion protein. This was not a problem when the TCG codon was used.With the exception of the R54S mutant, all fusion proteins were soluble and monomeric. All cleaved and purified III-13 domains were homogeneous by analytical size-exclusion chromatography on Superdex-75 (Pharmacia) and/or SDS-polyacrylamide gel electrophoresis in 8-25% gradient polyacrylamide gels (Pharmacia Phast system). All mutations were confirmed by sequencing at both the DNA and the protein level. Protein sequencing was done with a Hewlett Packard G1000S protein sequencing system. With the Arg6, Arg7, Arg9, Arg23, and Lys25 mutants, this was done on the intact domains because the mutations were close enough to the NH2 terminus of the domain. With the Arg47 and Arg54 mutants, it was necessary to first cleave chemically or enzymatically, sequence the resulting mixture of peptides, and compare the results with the known sequence of domain III-13 using custom software developed by G. Argraves (Shelton, CT).Thermal StabilityThe structural integrity of the recombinant III-13 domains was assessed by heating a solution of the domain at a concentration of ~0.1 mg/ml at ~1°C/min in the SLM 8000C spectrofluorometer while monitoring the ratio of fluorescence at 350 nm to that at 320 nm with excitation at 280 nm(32Novokhatny V. Schwarz F. Atha D. Ingham K. J. Mol. Biol. 1992; 227: 1182-1191Crossref PubMed Scopus (20) Google Scholar). The fluorescence ratio provides a convenient and sensitive means of detecting the spectral shift that accompanies denaturation. Changes in the fluorescence ratio permitted detection of melting transitions and assessment of relative stability and the degree of reversibility of the denaturation upon cooling.Fluorescence AnisotropyMeasurements were made with the SLM-8000C spectrofluorometer in the T format with excitation and emission wavelengths of 493 and 524 nm, respectively. All experiments described here were performed with Sephadex G-100 fraction no. 4 of fluoresceinamine-labeled heparin (FA-heparin) having an average molecular mass of ~15,000 daltons(33Ingham K.C. Brew S.B. Atha D.A. Biochem J. 1990; 272: 605-611Crossref PubMed Scopus (83) Google Scholar). Titrations of 0.1 μM FA-heparin with recombinant III-13 domains were performed in 0.02 M Tris buffer, pH 7.4, containing 0.02% NaN3 and no NaCl (TB) or 0.15 M NaCl (TBS). Small amounts of a stock solution of the recombinant peptides were added continuously with a motorized syringe controlled by the same computer that controls the fluorometer. The change in anisotropy, ΔA, as a function of titrant concentration was fitted to a single class of equivalent binding sites on the FA-heparin by using the following equation: ΔA=ΔAmax*[titrant]/(Kd+[titrant])(Eq. 1) where [titrant] is the free concentration of fragment (or peptide), ΔAmax is the maximum anisotropy change that would be produced at saturating concentrations of titrant, and Kd is the apparent dissociation constant of the heparin-fragment complex(19Ingham K.C. Brew S.A. Migliorini M.M. Busby T.F. Biochemistry. 1993; 32: 12548-12553Crossref PubMed Scopus (45) Google Scholar). Since in all cases the concentration of FA-heparin was low compared to the range of concentration of fragment, the free fragment concentration was taken as the total. The concentrations of the fragments were determined from the absorbance at 280 nm, using a molar extinction coefficient, ϵ = 10,800 M-1 cm-1.Analytical Affinity ChromatographyHeparin-Sepharose was prepared as described(33Ingham K.C. Brew S.B. Atha D.A. Biochem J. 1990; 272: 605-611Crossref PubMed Scopus (83) Google Scholar). The recombinant peptides or fusion proteins were applied at 1 ml/min to a 1.7-ml column of heparin-Sepharose in either TB or TBS using a Pharmacia fast protein liquid chromatography system. A linear gradient to 0.6 M NaCl was used for elution, which was monitored by fluorescence at 340 nm with excitation at 280 nm.RESULTSMolecular Modeling of Domain III-13The fibronectin type III domains form a β sandwich with a folding topology similar to that of the immunoglobulin C domain(21Leahy D.J. Hendrickson W.A. Aukhil I. Erickson H.P. Science. 1992; 258: 987-991Crossref PubMed Scopus (441) Google Scholar, 22Main A.L. Harvey T.S. Baron M. Boyd J. Campbell I.D. Cell. 1992; 71: 671-678Abstract Full Text PDF PubMed Scopus (420) Google Scholar, 23Dickinson C.D. Veerapandian B. Dai X. Hamlin R.C. Xuong N. Ruoslahti E. Ely K.R. J. Mol. Biol. 1994; 236: 1079-1092Crossref PubMed Scopus (183) Google Scholar, 24Huber A.H. Wang Y. Bieber A.J. Bjorkman P.J. Neuron. 1994; 12: 717-731Abstract Full Text PDF PubMed Scopus (102) Google Scholar). Domain III-13 of fibronectin was modeled after domain III-3 from tenascin(21Leahy D.J. Hendrickson W.A. Aukhil I. Erickson H.P. Science. 1992; 258: 987-991Crossref PubMed Scopus (441) Google Scholar). The results are illustrated in Fig. 1 where the charged side chains are highlighted. Arginines 6, 7, and 9 conform to one of the sequence patterns identified in a number of heparin-binding proteins(1Jackson R.L. Busch S.J. Cardin A.D. Physiol. Rev. 1991; 71: 481-539Crossref PubMed Scopus (955) Google Scholar, 7Cardin A.D. Weintraub H.J.R. Arteriosclerosis. 1989; 9: 21-32Crossref PubMed Google Scholar). Arginines 6 and 7 are the ones whose mutation by Barkalow and Schwarzbauer (12Barkalow F.J. Schwarzbauer J.E. J. Biol. Chem. 1991; 266: 7812-7818Abstract Full Text PDF PubMed Google Scholar) abolished binding of truncated fibronectins to heparin-Sepharose at physiological ionic strength. The model suggests that additional positively charged residues Arg23, Lys25, and Arg54 might also contribute to heparin binding since they lie in close proximity to arginines 6, 7, and 9. Although the positions of the side chains are not precisely determined by the model, it is clear that 6 of the 10 positive charges in this domain lie within a few nanometers of each other on one side of the structure, in a region devoid of negative charge. A similar structure has been identified as the GAG-binding site in human lactoferrin and has been termed a "cationic cradle"(8Mann D.M. Romm E. Migliorini M. J. Biol. Chem. 1994; 269: 23661-23667Abstract Full Text PDF PubMed Google Scholar). In an effort to evaluate the relative importance of all 6 of these residues, each of them was mutated separately to a serine residue. The resulting constructs were expressed in E. coli and evaluated for folding integrity and heparin binding. Arginine 47, which in this model lies on the opposite side of the domain and is not predicted to be part of the binding site, was also mutated as a control for the possibility that simply reducing the net positive charge might affect heparin binding.Reversible Unfolding of Recombinant III-13 and Its MutantsThe thermal stability of the recombinant III-13 domains was assessed as a measure of their structural integrity. The various recombinant peptides were heated at 1°C/min in the fluorometer while monitoring the ratio of fluorescence at 350 nm to that at 320 nm as a measure of the spectral shift that accompanies unfolding(32Novokhatny V. Schwarz F. Atha D. Ingham K. J. Mol. Biol. 1992; 227: 1182-1191Crossref PubMed Scopus (20) Google Scholar). As shown in Fig. 2, all of the products underwent a cooperative sigmoidal unfolding transition similar to that observed previously with natural fragments derived from the parent protein by proteolysis(32Novokhatny V. Schwarz F. Atha D. Ingham K. J. Mol. Biol. 1992; 227: 1182-1191Crossref PubMed Scopus (20) Google Scholar). The transitions were highly reversible in that the fluorescence ratio returned to a value close to the original upon cooling. The Tm values varied between 60 and 71°C (Table 1). These results show that all of the recombinant III-13 domains, as isolated, were folded into compact structures with stabilities similar to one another and to that of the natural domain.Figure 2:Melting of recombinant fibronectin domain III-13 and its mutants. Samples were equilibrated in TBS at ~0.1 mg/ml and heated at 1°C/min while monitoring the ratio of fluorescence intensity at 350 nm to that at 320 nm with excitation at 280 nm. The dashed curves indicate reversibility on cooling.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Tabled 1 Open table in a new tab Analytical Affinity Chromatography on Heparin-SepharoseAffinity chromatography was used as a qualitative test of the ability of the recombinant III-13 domains to bind heparin(19Ingham K.C. Brew S.A. Migliorini M.M. Busby T.F. Biochemistry. 1993; 32: 12548-12553Crossref PubMed Scopus (45) Google Scholar). All of the products bound in high yield to the heparin-Sepharose column at room temperature, whether applied in the presence (TBS) or absence (TB) of 0.15 M NaCl. Natural III-13, derived from plasma fibronectin by proteolysis(19Ingham K.C. Brew S.A. Migliorini M.M. Busby T.F. Biochemistry. 1993; 32: 12548-12553Crossref PubMed Scopus (45) Google Scholar), and wild type rIII-13 eluted similarly between 0.43 and 0.45 M in a gradient of NaCl (Table 1) indicating that the recombinant and natural domains are functionally similar. The remaining mutants all eluted earlier, between 0.25 and 0.38 M NaCl. The final concentration of NaCl required for elution was similar whether starting from 0.0 M (TB, Table 1) or 0.15 M (TBS, not shown). The R47S control mutant also bound and was eluted at the same salt concentration as the natural and recombinant wild type fragments.Titration of Fluorescent-labeled HeparinA fluorescence polarization anisotropy assay was used to obtain a more quantitative estimate of the effect of the various mutations on the affinity for heparin in the fluid phase. The results are presented in Fig. 3 where solid curves represent best fits to. In TB, all of the products caused a dose-dependent increase in fluorescence anisotropy of FA-heparin. Dissociation constants ranged from 0.5 to 1.4 μM for the natural, wild type and R47S control domains and all of the mutants fell within this same range (Table 1). At physiological ionic strength, in TBS, the Kd values of the wild type and R47S control domains were close to each other and to that of the natural domain although the values for all three were approximately 10-fold higher than at low ionic strength. In contrast, the binding of the R6S, R7S, R9S, and R23S mutants was too weak to cause a significant increase in the anisotropy at the concentrations of the domains that were achieved (Fig. 3). The Kd for these mutants is conservatively estimated at ≥100 μM, based on the reasonable assumption that the change in anisotropy caused by saturation with these mutants would be similar to that caused by the wild type. The same assumption was used in fitting the data for the K25S mutant, which caused a slight increase in anisotropy but still failed to achieve a substantial fractional saturation of the response at the highest concentration employed. The R54S mutant had the highest
Referência(s)