The Receptor Binding Site of Human Interleukin-3 Defined by Mutagenesis and Molecular Modeling
1997; Elsevier BV; Volume: 272; Issue: 36 Linguagem: Inglês
10.1074/jbc.272.36.22630
ISSN1083-351X
AutoresBarbara K. Klein, Yiqing Feng, Charles A. McWherter, William F. Hood, Kumnan Paik, John P. McKearn,
Tópico(s)T-cell and B-cell Immunology
ResumoInterleukin-3 (IL-3) is a member of the cytokine superfamily that promotes multi-potential hematopoietic cell growth by interacting with a cell surface receptor composed of α and β chains. The newly available three-dimensional structure of a variant of human (h) IL-3 allowed us to evaluate new and existing mutagenesis data and to rationally interpret the structure-function relationship of hIL-3 on a structural basis. The amino acid residues that were identified to be important for hIL-3 activity are grouped into two classes. The first class consists of largely hydrophobic residues required for the structural integrity of the protein, including the residues in IL-3 that are largely conserved among 10 mammalian species. These residues form the core of a scaffold for the second class of more rapidly diverging solvent-exposed residues, likely to be required for interaction with the receptor. Ten important and solvent-exposed residues, Asp21, Gly42, Glu43, Gln45, Asp46, Met49, Arg94, Pro96, Phe113, and Lys116, map to one side of the protein and form a putative binding site for the α subunit of the receptor. A model of the IL-3·IL-3 receptor complex based on the human growth hormone (hGH)·hGH soluble receptor complex structure suggests that the interface between IL-3 and the IL-3 receptor α subunit consists of a cluster of hydrophobic residues flanked by electrostatic interactions. Although the IL-3/IL-3 receptor β subunit interface cannot be uniquely located due to the lack of sufficient experimental data, several residues of the β subunit that may interact with Glu22 of IL-3 are proposed. The role of these residues can be tested in future mutagenesis studies to define the interaction between IL-3 and IL-3 receptor β subunit. Interleukin-3 (IL-3) is a member of the cytokine superfamily that promotes multi-potential hematopoietic cell growth by interacting with a cell surface receptor composed of α and β chains. The newly available three-dimensional structure of a variant of human (h) IL-3 allowed us to evaluate new and existing mutagenesis data and to rationally interpret the structure-function relationship of hIL-3 on a structural basis. The amino acid residues that were identified to be important for hIL-3 activity are grouped into two classes. The first class consists of largely hydrophobic residues required for the structural integrity of the protein, including the residues in IL-3 that are largely conserved among 10 mammalian species. These residues form the core of a scaffold for the second class of more rapidly diverging solvent-exposed residues, likely to be required for interaction with the receptor. Ten important and solvent-exposed residues, Asp21, Gly42, Glu43, Gln45, Asp46, Met49, Arg94, Pro96, Phe113, and Lys116, map to one side of the protein and form a putative binding site for the α subunit of the receptor. A model of the IL-3·IL-3 receptor complex based on the human growth hormone (hGH)·hGH soluble receptor complex structure suggests that the interface between IL-3 and the IL-3 receptor α subunit consists of a cluster of hydrophobic residues flanked by electrostatic interactions. Although the IL-3/IL-3 receptor β subunit interface cannot be uniquely located due to the lack of sufficient experimental data, several residues of the β subunit that may interact with Glu22 of IL-3 are proposed. The role of these residues can be tested in future mutagenesis studies to define the interaction between IL-3 and IL-3 receptor β subunit. Interleukin-3 (IL-3) 1The abbreviations used are: IL-3, interleukin-3; hIL-3, human interleukin-3; IL-315–125, a deletion variant of IL-3 consisting of residues 15–125; SC-65369, a truncation variant of IL-3 mutated in 14 of 112 positions; IL-3R, interleukin-3 receptor complex; GM-CSF, granulocyte-macrophage colony-stimulating factor; IL-5, interleukin-5; GH, growth hormone; h, human; GHR, growth hormone receptor; SA, solvent-accessible surface area; CRM, cytokine receptor motif. is a multi-lineage hematopoietic growth factor that promotes the growth of most lineages of blood cell precursors (1Metcalf D. Science. 1991; 254: 529-533Crossref PubMed Scopus (332) Google Scholar, 2Schrader J.W. Annu. Rev. Immunol. 1986; 4: 205-230Crossref PubMed Google Scholar). It belongs to the helical cytokine superfamily and is further classified as a short chain cytokine similar to granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-5 (IL-5) (3Bazan J.F. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 6934-6938Crossref PubMed Scopus (1881) Google Scholar, 4Diederichs K. Boone T. Karplus P.A. Science. 1991; 254: 1779-1782Crossref PubMed Scopus (167) Google Scholar, 5Goodall G.J. Bagley C.J. Vadas M.A. Lopez A.F. Growth Factors. 1993; 8: 87-97Crossref PubMed Scopus (65) Google Scholar). Characteristic of the short chain helical cytokines, the structural core of the hIL-3 variant SC-65369 (6Feng Y. Klein B.K. Vu L. Aykent S. McWherter C.A. Biochemisty. 1995; 34: 6540-6551Crossref PubMed Scopus (24) Google Scholar, 7Feng Y. Klein B. McWherter C. J. Mol. Biol. 1996; 259: 524-541Crossref PubMed Scopus (66) Google Scholar) consists of an up-up-down-down four-helical bundle (designated as helix A through D) with a 30–40° packing angle. This helical bundle motif is completed by long overhand loops connecting the helices (loops AB and CD) and a type II turn (turn BC) between helices B and C. Distinct from the other short chain cytokines, the hIL-3 variant structure also revealed that loop AB contains an additional helix (helix A′) that is approximately parallel to helix D and interacts extensively with both helix D and loop CD. Furthermore, loop AB, which passes in front of helix D, has a threading topology that is more like that of the long chain cytokines such as growth hormone (8Abdel-Meguid S.S. Shieh H.-S. Smith W.W. Dayringer H.E. Violand B.N. Bentle L.A. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 6434-6437Crossref PubMed Scopus (434) Google Scholar, 9deVos A.M. Ultsch M. Kossiakoff A. Science. 1992; 255: 306-312Crossref PubMed Scopus (2029) Google Scholar). Binding to a cell surface receptor (IL-3R) composed of at least α and β chains is the initial event in the expression of IL-3's proliferative activity on a target cell. The α chain of this receptor (IL-3Rα) is specific for IL-3, whereas the β chain (βc) is shared with the related hematopoietic cytokines GM-CSF (10Kitamura T. Sato N. Arai K. Miyajima A. Cell. 1991; 66: 1165-1174Abstract Full Text PDF PubMed Scopus (512) Google Scholar) and IL-5 (11Tavernier J. Devos R. Cornelis S. Tuypens T. derHeyden J.V. Fiers W. Plaetinck G. Cell. 1991; 66: 1175-1184Abstract Full Text PDF PubMed Scopus (500) Google Scholar). IL-3 binds to IL-3Rα with low affinity and has no detectable affinity for βc in the absence of IL-3Rα; high affinity binding requires the presence of both chains. Following the formation of the high affinity IL-3·IL-3R complex, signal transduction is mediated by the β chain. Although the precise stoichiometry of the IL-3 receptor has not been established, we refer to the receptor as a heterodimer for simplicity. Several approaches have been undertaken to define the regions of hIL-3 that interact with the receptor, including constructing interspecies chimeras (12Kaushansky K. Shoemaker S.G. Broudy V.C. Lin N. Matous J.V. Alderman E.M. Aghajanian J.D. Szklut P.J. VanDyke R.E. Pearce M.K. Abrams J.S. J. Clin. Invest. 1992; 90: 1879-1888Crossref PubMed Scopus (24) Google Scholar, 13Dorssers L.C.J. Burger H. Wagemaker G. DeKoning J.P. Growth Factors. 1994; 11: 93-104Crossref PubMed Scopus (4) Google Scholar), mapping neutralizing and non-neutralizing anti-IL-3 monoclonal antibody epitopes (12Kaushansky K. Shoemaker S.G. Broudy V.C. Lin N. Matous J.V. Alderman E.M. Aghajanian J.D. Szklut P.J. VanDyke R.E. Pearce M.K. Abrams J.S. J. Clin. Invest. 1992; 90: 1879-1888Crossref PubMed Scopus (24) Google Scholar, 14Lokker N.A. Strittmatter U. Steiner C. Fagg B. Graff P. Kocher H.P. Zenke G. J. Immunol. 1991; 146: 893-898PubMed Google Scholar), and site-specific mutagenesis (14Lokker N.A. Strittmatter U. Steiner C. Fagg B. Graff P. Kocher H.P. Zenke G. J. Immunol. 1991; 146: 893-898PubMed Google Scholar, 15Lokker N.A. Zenke G. Strittmatter U. Fagg B. Movva N.R. EMBO J. 1991; 10: 2125-2131Crossref PubMed Scopus (36) Google Scholar, 16Dorssers L.C.J. Mostert M.C. Burger H. Janssen C. Lemson P.J. van Lambalgen R. Wagemaker G. van Leen R.W. J. Biol. Chem. 1991; 266: 21310-21317Abstract Full Text PDF PubMed Google Scholar, 17Lopez A.F. Shannon M.F. Barry S. Phillips J.A. Cambareri B. Dottore M. Simmons P. Vadas M.A. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11842-11846Crossref PubMed Scopus (37) Google Scholar, 18Barry S.C. Bagley C.J. Phillips J. Dottore M. Cambareri B. Moretti P. D'Andrea R. Goodall G.J. Shannon M.F. Vadas M.A. Lopez A.F. J. Biol. Chem. 1994; 269: 8488-8492Abstract Full Text PDF PubMed Google Scholar, 19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar, 20Bagley C.J. Phillips J. Cambareri B. Vadas M.A. Lopez A.F. J. Biol. Chem. 1996; 271: 31922-31928Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar). These studies identified sequences important for activity in both N- and C-terminal regions of IL-3 and revealed many specific sites where amino acid substitutions lead to significant changes in receptor binding affinity and growth-promoting activity. Targeted saturation mutagenesis was employed to produce over 700 single point mutants of hIL-315–125, which were then evaluated for expression and biological activity (19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). This study identified several discrete regions of hIL-3 that are important for activity. To date, the relationships of these discontinuous regions to one another could only be inferred on the basis of secondary structure prediction and homology modeling of hIL-3 (3Bazan J.F. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 6934-6938Crossref PubMed Scopus (1881) Google Scholar, 12Kaushansky K. Shoemaker S.G. Broudy V.C. Lin N. Matous J.V. Alderman E.M. Aghajanian J.D. Szklut P.J. VanDyke R.E. Pearce M.K. Abrams J.S. J. Clin. Invest. 1992; 90: 1879-1888Crossref PubMed Scopus (24) Google Scholar, 17Lopez A.F. Shannon M.F. Barry S. Phillips J.A. Cambareri B. Dottore M. Simmons P. Vadas M.A. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 11842-11846Crossref PubMed Scopus (37) Google Scholar, 19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). The recently available three-dimensional structure of an hIL-3 variant (7Feng Y. Klein B. McWherter C. J. Mol. Biol. 1996; 259: 524-541Crossref PubMed Scopus (66) Google Scholar) provides for the first time an experimentally derived model to understand the effect of these mutations. In this paper, the activities of hIL-3 mutants are interpreted in terms of IL-3's three-dimensional structure. Together these data define an IL-3Rα binding site on IL-3 consisting of the solvent-exposed surface of 10 residues located on helix A, helix A′, loop CD, and helix D. These findings are exploited to construct a molecular model of the IL-3·IL-3R complex structure. This model provides explanations of the mutagenesis results on the structural basis, suggests alternative ligand binding sites on IL-3Rβ, and proposes additional mutations for further investigation of the cytokine-receptor interaction. Recombinant DNA procedures used to generate the IL-3 variants have been described previously (19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). The plasmids were transformed into theEscherichia coli strain JM101 (ATCC 33876), and the variants were purified from cytoplasmic inclusion bodies (19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar) to greater than 90% homogeneity as determined by SDS-polyacrylamide gel electrophoresis. Circular dichroism (CD) measurements were conducted on a Jasco J500C at 20 °C. Samples for the CD experiments contained between 0.08 and 0.42 mg/ml protein in 20 mm ammonium bicarbonate at pH 4. The percentage of helical conformation was estimated from the signal at 222 nm according to Chen et al.(21Chen Y.H. Yang Y.T. Martinez H.M. Biochemistry. 1972; 11: 4120-4131Crossref PubMed Scopus (1913) Google Scholar). Receptor binding affinity and cell proliferation activity of IL-3 variants were measured essentially as described previously (22Thomas J.W. Baum C.M. Hood W.F. Klein B. Monahan J.B. Paik K. Staten N. Abrams M. Donnelly A. McKearn J.P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3779-3783Crossref PubMed Scopus (51) Google Scholar, 23Santoli D. Yang Y.C. Clark S.C. Kreider B.L. Caracciolo D. Rovera G. J. Immunol. 1987; 139: 3348-3354PubMed Google Scholar). Activity for each variant was reported as the concentration that gave 50% of the maximal response by fitting a four-parameter logistic regression model to the data. Relative potencies (EC50 of hIL-3 divided by EC50 of variant) are the mean of at least three independent assays. A model for the wild-type hIL-315–125 was generated from the energy-minimized average NMR structure of the hIL-3 variant SC-65369 (Protein Data Bank entry 1JLI). The mutated residues in the structure of SC-65369 were replaced with their wild-type counterparts using INSIGHTII (Biosym Technologies, San Diego, CA). These replacements were well-tolerated because nearly all of them occur on the surface of the molecule. The resultant structure was exported into XPLOR 3.1 (24Brunger A.T. J. Mol. Biol. 1988; 203: 803-816Crossref PubMed Scopus (361) Google Scholar, 25Brunger A.T. A System for X-ray Crystallography and NMR, X-PLOR Version 3.1. Yale University Press, New Haven, CT1992Google Scholar), and 1000 energy minimization steps were performed while fixing the backbone N, Cα, C′, and O atoms. During the energy minimization, the electrostatic energy was included in the total energy, but the charges at the side chains of the ionizable residues, as well as at the termini, were excluded. Solvent-accessible surface area was calculated for each residue using the method of Lee and Richards (26Lee B. Richards F.M. J. Mol. Biol. 1971; 55: 379-400Crossref PubMed Scopus (5360) Google Scholar) as implemented in XPLOR using a solvent probe radius of 1.4 Å. Sequence alignment of the cytokine receptor motif (CRM) domains of the receptors was performed initially using the BESTFIT routine in the UWGCG Software Package (50UWGCG Software Package (1991) Version 7.3, Genetics Computer Group, Madison, WI.Google Scholar) and subsequently manually refined using the same package. Consideration was given to previously identified consensus sequence motifs (3Bazan J.F. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 6934-6938Crossref PubMed Scopus (1881) Google Scholar), secondary structure predictions, the locations of β strands and loops of growth hormone receptor (GHR) in the GH·GHR complex structure (Protein Data Bank entry 3HHR) (9deVos A.M. Ultsch M. Kossiakoff A. Science. 1992; 255: 306-312Crossref PubMed Scopus (2029) Google Scholar), and the locations of residues that contribute to the hydrophobic cores of the GHR domains. Specific reference points employed included the locations of conserved disulfide bridges in the N-terminal domain of GHR and the “WSXWS box” in the C-terminal domain. The initial coordinates of the CRM domains of IL-3Rα (residues 106–294) and βc (residues 244–439) were constructed by substituting the IL-3R sequences into the corresponding GHR structures in INSIGHTII. Extended loops were modeled from fragments of highly refined protein crystal structures in the Protein Data Bank and spliced into the receptor structures. The initial models for each subunit were exported into SYBYL (Tripos Associates, St. Louis, MO) where the side chain torsion angles were iteratively scanned for combinations that would relieve van der Waals violations. The structures were then exported to CHAIN (Baylor College of Medicine, Houston, TX) to improve the main chain geometry of the spliced loops. The resulting structures were further refined by energy minimization using XPLOR (24Brunger A.T. J. Mol. Biol. 1988; 203: 803-816Crossref PubMed Scopus (361) Google Scholar, 25Brunger A.T. A System for X-ray Crystallography and NMR, X-PLOR Version 3.1. Yale University Press, New Haven, CT1992Google Scholar). Initial energy minimizations were performed with all non-hydrogen atoms fixed and then repeated with all backbone atoms fixed. Subsequent full atom energy minimization employed distance and backbone dihedral angle constraints for β sheets to preserve secondary structure while still permitting side chain and segmental movements. The side chains were subject to 10 ps of molecular dynamics at 100 K and subsequently to energy minimizations with full electrostatic potentials for all atoms except for side chain atoms of ionizable residues and the termini. The model for the structure of the complex was generated through several stages. Initially, key mutagenesis data were used to identify putative complementary residues as tether points to place IL-3 in the ligand binding pocket. IL-3 and the receptor were then subject to rigid body movement tethered with distances from the side chain carboxyl group of IL-3 Asp21to the center of the charged side chain atoms of IL-3Rα Arg234, Lys235, and Arg277, and from the carboxyl group of IL-3 Glu43 to the center of the charged side chain atoms of IL-3Rα Arg145 and Arg146 constrained to be ≤5.0 Å. There is sufficient space between the two subunits in the receptor model so that no movement of the βc subunit was required when IL-3 is docked to the α subunit. Next, a full atom energy minimization of the complex (except for backbone atoms of IL-3Rα residues 211–294 and βc residues 343–439, sequences that are involved primarily in the intersubunit interface) was carried out using the same ligand-receptor distance constraints as described above. This minimization excluded electrostatic interactions, and it used distance and backbone dihedral angle constraints to maintain secondary structures in both IL-3 and its receptor. A subsequent round of energy minimization included electrostatic contributions for all atoms except for charged atoms of ionizable side chains and termini. At the final stage, the minimization was carried out with full electrostatic potentials for all atoms except for chain termini. The solvent-accessible areas buried upon complex formation were calculated in XPLOR using a probe size of 1.4 Å. The electrostatic potentials surrounding hIL-315–125 and the receptor were calculated using GRASP (27Nicholls A. Sharp K. Honig B. Proteins Struct. Funct. Genet. 1991; 11: 281-296Crossref PubMed Scopus (5318) Google Scholar). The Amber parameter set was used to assign partial charges. Dielectric constants of 80 and 2 were assigned to the solvent and the solute, respectively. The ionic strength of the solvent was set to 145 mm, the water probe radius to 1.4 Å, and the ionic radius to 2 Å. A charge of 0.5 unit was assigned to histidine side chains. Because only the CRM domains of the receptor sequences were used in modeling, the beginning and end of the receptor chains were left uncharged. An extensive screening mutagenesis study was carried out using hIL-315–125 as the template (19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar) to identify amino acid residues important for biological activity. Libraries of single point mutants were created at nearly 90% of the positions using polymerase chain reaction mutagenesis methods. The resulting mutants were expressed for secretion in E. coli, and the secreted proteins were recovered after osmotic shock. Their bioactivity was measured using an IL-3-dependent cell proliferation assay, and protein concentrations were determined by enzyme-linked immunosorbent assay. The activities were evaluated for mutations at 97 of the 111 sites (with an average of eight amino acid substitutions at each position) to evaluate their tolerance to substitution. The results indicated that for >70% of the residues evaluated, substitution by other amino acids with a variety of size, polarity, and charge maintained significant level of biological activity. For the purpose of the following discussion, a residue is considered to be intolerant to substitution when, of at least three substitutions tested, no more than one had greater than 5% of the proliferation activity of the parental molecule. Residues that were found to have approximately half active (≥5% of hIL-3 activity) and half inactive ( 5% activity are considered tolerant. Using these criteria, the results reported previously by Olins et al. (19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar) are summarized in Table I. A total of 16 amino acid residues are classified as intolerant, and another 12 are classified as partially tolerant. Substitution-sensitive residues, consisting of both intolerant and partially tolerant residues, are distributed throughout the linear sequence of the protein.Table IAmino acid residues sensitive to substitution (see text) and conserved in mammalian IL-3 sequencesPositionClass1-aSensitivity to amino acid substitution: I, intolerant; PT, partially tolerant; T, tolerant; ND, not determined.SA area1-bIndicates the solvent-accessible area (in Å2) for the given residue in the three-dimensional model of hIL-315–125.Conservation1-cExtent of conservation at a given position: ch, chimpanzee; gi, gibbon; rh, rhesus; ov, ovine; bo, bovine; ra, rat; mu, mouse; ex, except.Cys16I20.6ex, ov, boIle20PT0.0ch, gi, rhAsp21PT68.4ch, gi, rhGlu22I81.9allIle23T0.7allIle24PT26.2ch, gi, rh, muLeu27I2.2allGlu43I128.4primateAsp44I22.7primateIle47PT7.2primate, ovLeu48I0.0allMet49PT111.6ch, gi, ta, maAsn52PT8.5primateLeu53I0.6ex, muArg54PT1.2primate, muAsn57ND0.0allLeu58I0.0allPhe61I0.0allAla64I12.1ex, ra, muLeu68PT3.3primateIle74ND10.4allLeu78ND0.3allLeu81I0.0ex, ov, boCys84I45.4ex, ov, boPro86T45.4allArg94I100.3ch, gi, rhPro96I54.0primateIle97PT8.9ex, muIle99PT7.9allGly102PT10.1primatePhe107ND1.3allLys110I2.5allLeu111ND0.3allPhe113ND49.6ex, ov, boLeu115I0.0ex, mu, raThr117PT16.7ex, bo1-a Sensitivity to amino acid substitution: I, intolerant; PT, partially tolerant; T, tolerant; ND, not determined.1-b Indicates the solvent-accessible area (in Å2) for the given residue in the three-dimensional model of hIL-315–125.1-c Extent of conservation at a given position: ch, chimpanzee; gi, gibbon; rh, rhesus; ov, ovine; bo, bovine; ra, rat; mu, mouse; ex, except. Open table in a new tab A model for hIL-315–125 was derived from the recently determined NMR structure of the hIL-3 variant SC-65369 (7Feng Y. Klein B. McWherter C. J. Mol. Biol. 1996; 259: 524-541Crossref PubMed Scopus (66) Google Scholar). In addition to N- and C-terminal truncations of 13 and 8 residues, respectively, SC-65369 contains 14 amino acid substitutions (V14A, N18I, T25H, Q29R, L32N, F37P, G42S, Q45M, N51R, R55T, E59L, N62V, S67H, and Q69E) that were found to be localized at the solvent-exposed surface of the protein. Despite the number of amino acid substitutions, both biophysical and biochemical data (7Feng Y. Klein B. McWherter C. J. Mol. Biol. 1996; 259: 524-541Crossref PubMed Scopus (66) Google Scholar) indicate that the structural features of SC-65369 are relevant for wild-type hIL-3. In comparison to native IL-3, SC-65369 is fully active in both cell proliferation and cell surface IL-3Rα binding assays, and it has a similar helical content as determined by CD spectroscopy. 2R. Schilling, unpublished results. Residues that are sensitive to substitution may be so either because they are important for receptor contact and their substitution diminishes the complementarity required for optimal receptor recognition, or because they are important for structural integrity and their substitution deranges the structure, thereby indirectly distorting the recognition site. Correlating solvent accessibility with sensitivity to substitution is one approach to assess the role of an amino acid residue in protein-protein interaction. Solvent-accessible surface areas (SA) for individual residues in hIL-315–125 are presented in Fig.1. Fifteen out of the total of 28 substitution-sensitive residues in Table I are found to be hydrophobic (Ala, Phe, Ile, Leu, Val, Met, Trp). Met49 is the only hydrophobic and substitution-sensitive residue with a large solvent exposure (SA = 112 Å2). The remainder are largely buried inside the structure (SA ≤30 Å2). Among the non-hydrophobic residues classified as intolerant to substitution, Cys16 and Cys84 form the only disulfide bridge in hIL-3. Studies on both hIL-3 variants 3R. McKinnie and B. Klein, unpublished results. and mouse IL-3 (28Clark-Lewis I. Hood L. Kent S.B.H. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 7897-7901Crossref PubMed Scopus (36) Google Scholar) indicated that this disulfide bond is required for bioactivity and may do so by stabilizing the tertiary structure. In addition to the above residues, five of the substitution-sensitive yet hydrophilic residues (Asp44, Asn52, Arg54, Lys110, and Thr117) are also buried in the structure (Fig. 1 and Table I). These residues are likely to stabilize the structure through internal salt bridging and/or hydrogen bonding interactions (see below). While greater than two-thirds of the substitution-sensitive residues are involved in maintaining the three-dimensional structure of hIL-3, six (Asp21, Glu22, Glu43, Met49, Arg94, and Pro96) have significant solvent-accessible area (>50 Å2) and are good candidates for being involved in direct contacts with the receptor. To confirm the results of the screening mutagenesis studies and to test the effect of the mutations on receptor binding properties, several point mutants were selected for purification and full characterization. Mutants were chosen on the basis of their sensitivity to substitution, solvent accessibility, and location in the three-dimensional structure (TableII). The genes encoding the point mutants were transferred to an E. coli cytoplasmic expression system, expressed, and purified to homogeneity. To rule out the possibility that structural perturbation leads to the loss of activity, seven of the mutants (G42A, G42D, E43N, D44A, E50D, F113Y, and K116W), along with the parental molecule hIL-315–125, were characterized using far UV circular dichroism spectroscopy. All of the variants had a helical content of ∼40% that was indistinguishable from that of hIL-315–125 and was also consistent with the solution structure of SC-65369 (6Feng Y. Klein B.K. Vu L. Aykent S. McWherter C.A. Biochemisty. 1995; 34: 6540-6551Crossref PubMed Scopus (24) Google Scholar). These data indicate that the mutants are properly folded with no large scale changes in global conformation.Table IIRelative activities of a selected set of purified hIL-3 variantsPositionSubstitutionRelative binding2-aLow affinity binding (α receptor subunit only) measured relative to hIL-31–133.Relative AML activity2-bCell proliferation activity expressed relative to hIL-31–133.hIL-315–1251.72.52-cData from Olins et al. (19).Glu22Gly1.30.008Gly42Asp1692-cData from Olins et al. (19).Gly42Ala113.2Glu43Asn<0.20.1Asp44Ala1.10.2Gln45Val1882-cData from Olins et al. (19).Asp46Ser7102-cData from Olins et al. (19).Glu50Asp772-cData from Olins et al. (19).Phe113Tyr<0.20.15Leu115Met<0.20.04Lys116Val282-cData from Olins et al. (19).Lys116Trp50262-cData from Olins et al. (19).Gln122Phe1.462-cData from Olins et al. (19).Summary of cell proliferation activity and low affinity receptor binding activity of selected single point variants in hIL-315–125. Assays are described under “Experimental Procedures.”2-a Low affinity binding (α receptor subunit only) measured relative to hIL-31–133.2-b Cell proliferation activity expressed relative to hIL-31–133.2-c Data from Olins et al. (19Olins P.O. Bauer S.C. Braford-Goldberg S. Sterbenz K. Polazzi J.O. Caparon M.H. Klein B.K. Easton A.M. Paik K. Klover J.A. Thiele B.R. McKearn J.P. J. Biol. Chem. 1995; 270: 23754-23760Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). Open table in a new tab Summary of cell proliferation activity and low affinity receptor binding activity of selected single point variants in hIL-315–125. Assays are described under “Experimental Procedures.” The cell proliferation activity and
Referência(s)