Interaction of insulin-like growth factor (IGF)-I and -II with IGF binding protein-2: mapping the binding surfaces by nuclear magnetic resonance.

The interaction of IGF binding protein-2 (IGFBP-2) with IGF-I and -II has been investigated in solution using nuclear magnetic resonance (NMR) spectroscopy. Chemical shift perturbations in 15N- and 2H/15N-labelled IGF-I or -II upon binding to unlabelled thioredoxin-tagged bovine IGFBP-2 (Trx(1-279)IGFBP-2) have been monitored to identify residues involved directly in the binding interaction as well as any affected by conformational changes associated with the interaction. A key step in obtaining high-quality spectra of the complexes was the use of transverse relaxation optimised spectroscopy (TROSY) methods with partially deuterated ligands. Indeed, because the effects of conformational averaging and aggregation are eliminated in IGF-I and -II bound to IGFBP-2, the spectra of the complexes are actually superior to those of the free ligands. Comparison of our results with the crystal structure of the complex between IGF-I and an N-terminal fragment of IGFBP-5 allowed identification of those residues perturbed by the C-domain of IGFBP-2. Other perturbations, such as those of Gly 19 and Asp 20 of IGF-I (and the corresponding residues in IGF-II) - which are located in a reverse turn linking N-domain and C-domain interactive surfaces - are due to local conformational changes in the IGF-I and -II. Our results confirm that the C-domain of IGFBP-2 plays a key role in binding regions of IGF-I and -II that are also involved in binding to the type-1 IGF receptor and thereby blocking ligand binding to this receptor.

Contribution of residues A54 and L55 of the human insulin-like growth factor-II (IGF-II) A domain to Type 2 IGF receptor binding specificity.

The underlying specificity of the interaction between insulin-like growth factor-II (IGF-II) and mammalian Type 2 insulin-like growth factor/cation-independent mannose 6 phosphate receptor (IGF2R) is not understood. We have mutated residues A54 and L55 of IGF-II in the second A domain helix to arginine (found in the corresponding positions of IGF-I) and measured IGF2R binding. There is a 4- and 3.3-fold difference in dissociation constants for A54R IGF-II and L55R IGF-II, respectively, and a 6.6-fold difference for A54R L55R IGF-II compared with IGF-II as measured by BlAcore analysis using purified rat IGF2R. This is also confirmed using cross-linking and soluble rat placental membrane receptor binding assays. Binding to the type I IGF receptor (IGF1R) and IGF binding protein-2 (IGFBP-2) is not altered. We can, therefore, conclude that residues at positions 54 and 55 in IGF-II are important for and equally contribute to IGF2R binding.

Localization of an insulin-like growth factor (IGF) binding site of bovine IGF binding protein-2 using disulfide mapping and deletion mutation analysis of the C-terminal domain.

We have investigated which region(s) of bovine insulin-like growth factor binding protein-2 (bIGFBP-2) interact with insulin-like growth factors (IGFs) using C-terminally truncated forms of bIGFBP-2. Initially to aid in mutant design, we defined the disulfide bonding pattern of bIGFBP-2 C-terminal region using enzymatic digestion. The pattern is Cys186-Cys220, Cys231-Cys242, and Cys244-Cys265. In addition, cyanogen bromide cleavage of bIGFBP-2 revealed that the N- and C-terminal cysteine-rich domains were not linked by disulfide bonds. Taking the disulfide bonding pattern into consideration, C-terminal truncation mutants were designed and expressed in COS-1 mammalian cells. Following IGF binding assays, a region between residues 222 and 236 was identified as important in IGF binding. Specifically, mutants truncated by 14, 36, and 48 residues from the C terminus bound IGFs to the same extent as wild type (WT) bIGFBP-2. Removal of 63 residues resulted in a greatly reduced (up to 80-fold) ability to bind IGF compared with WT bIGFBP-2. Interestingly this mutant lacked the IGF-II binding preference of WT bIGFBP-2. Residues 236-270 also appeared to play a role in determining IGF binding specificity as their removal resulted in mutants with higher IGF-II binding affinity.

Introduction of spacer peptides N-terminal to a cleavage recognition motif in recombinant fusion proteins can improve site-specific cleavage.

To improve site-specific cleavage of a methionyl porcine growth hormone [[Met1]-pGH(1-46)-IGF-II] fusion protein by the enzyme H64A subtilisin, a series of flexible, unstructured spacer peptides were introduced N-terminal to the cleavage site. When enzymatic digestion preceded refolding of the fusion proteins, IGF-II could only be liberated from substrates which contained spacer peptides. Compared with the parent construct, the yield of IGF-II from refolded fusion proteins containing spacers was improved up to two-fold. Furthermore, this cleavage rate was improved by removing a competing protease recognition motif from the fusion partner. These data show that fusion partners can influence site-specific proteolysis of fusion proteins. Introduction of flexible spacers between the moieties can alleviate these interactions.

Production of a human epidermal growth factor fusion protein and its degradation in rat gastrointestinal flushings.

This study describes the biosynthesis of a human epidermal growth factor fusion protein, Long EGF, that has a 53 amino acid extension peptide derived from the 46 N-terminal amino acids of porcine GH. The approach allowed the production of Long EGF at high efficiency due to the expression of the fusion protein in high yield as inclusion bodies in Escherichia coli. Long EGF had a slightly lower potency compared with native EGF in a range of assays, including binding to anti-EGF antibodies or the EGF receptor, stimulation of Balb/3T3 fibroblast and rat intestinal epithelial cell growth, as well as counteracting the inhibition of mink lung epithelial cell proliferation by transforming growth factor-beta 1. Degradation of Long EGF and native EGF was compared in gastrointestinal flushings as an indication of whether the EGF domain of the fusion protein would be protected from proteolytic cleavage and be useful as a trophic agent in the gut. Incubation with flushings from the stomach or jejunum of rats caused rapid cleavage of the extension peptide, releasing native EGF. A C-terminal truncation of Arg53 in the stomach and a removal of the C-terminal pentapeptide (49 Trp-Trp-Glu-Leu-Arg53) in the small bowel was demonstrated by N-terminal sequencing and mass spectrometry. The degradation patterns were reflected by changes in migration of products on SDS-PAGE and in subsequent binding activities to the EGF receptor and anti-EGF antibodies. The data show that a human EGF fusion protein can be produced efficiently in a bacterial expression system and that it retains biological activity in vitro. Although the extension peptide was rapidly cleaved from Long EGF in both stomach and small bowel producing similar biological activity to native EGF, it could not prevent subsequent degradation of the EGF domain. Other strategies are being investigated to develop an effective oral form of EGF that resists digestion by proteases in the gastrointestinal tract.

Insulin-like growth factor (IGF)-II binding to IGF-binding proteins and IGF receptors is modified by deletion of the N-terminal hexapeptide or substitution of arginine for glutamate-6 in IGF-II.

Recombinant insulin-like growth factor-II (IGF-II) and two structural analogues, des(1-6)IGF-II and [Arg6]-IGF-II, were produced to investigate the role of N-terminal residues in binding to IGF-binding proteins (IGFBPs) and hence the biological properties of the modified peptides. The growth factors were modelled on two previously characterized variants of IGF-I, des(1-3)IGF-I and [Arg3]-IGF-I, which both show substantially decreased binding to IGFBPs and were expressed as fusion proteins in Escherichia coli. The biological activities of the corresponding analogues of IGF-I and IGF-II were compared in rat L6 myoblasts and H35B hepatoma cells. In the L6-myoblast protein-synthesis assay, the IGF-II analogues, des(1-6)IGF-II and [Arg6]-IGF-II, were slightly more potent than IGF-II but about 10-fold less potent than IGF-I and 100-fold less potent than the respective IGF-I analogues, des(1-3)IGF-I and [Arg3]IGF-I. In H35 hepatoma cells the anabolic response measured was the inhibition of protein breakdown, and the potency order was insulin >>> [Arg3]-IGF-I > des(1-3)IGF-I > [Arg6]-IGF-II > des(1-6)IGF-II > IGF-I > IGF-II. Binding of the IGFs and their analogues to the type 1 IGF receptor in L6 myoblasts and to the insulin receptor in H35 hepatoma cells did not fully explain the observed anabolic potency differences. Moreover, binding of all four analogues to the IGFBPs secreted by L6 myoblasts and H35B hepatoma cells was greatly decreased compared with the parent IGF. We conclude that the observed anabolic response to each IGF was determined by their relative binding to the competing cell receptor and IGFBP binding sites present.

Novel recombinant fusion protein analogues of insulin-like growth factor (IGF)-I indicate the relative importance of IGF-binding protein and receptor binding for enhanced biological potency.

An efficient expression system in Escherichia coli for several biologically active insulin-like growth factor-I (IGF-I) fusion peptide analogues is described. These novel IGF-I fusion protein analogues have properties that make them very useful reagents in the investigation of IGF-I action. The analogues comprise an IGF-I sequence and the first 11 amino acids of methionyl porcine growth hormone (pGH) and include [Met1]-pGH(1-11)-Val-Asn-IGF-I, which contains the authentic IGF-I sequence, and two analogues, [Met1]-pGH(1-11)-Val-Asn-[Gly3]-IGF-I and [Met1]-pGH(1-11)-Val-Asn-[Arg3]-IGF-I, where Glu-3 in the human IGF-I sequence has been replaced by Gly or Arg respectively. The three peptides are referred to as Long IGF-I, Long [Gly3]-IGF-I or Long [Arg3]-IGF-I depending on the IGF-I sequence present. Production of the purified fusion peptides was aided by folding the reduced and denatured fusion peptide sequence under conditions that gave very high yields of biologically active product. Introduction of a hydrophobic N-terminal extension peptide appears to facilitate the correct folding of the IGF-I analogues compared with that obtained previously when folding normal-length IGFs. The biological activities of the IGF-I fusion peptides were compared with authentic IGF-I and the truncated analogue, des(1-3)IGF-I. In L6 rat myoblasts, all the analogues were more potent than authentic IGF-I in their abilities to stimulate protein and DNA synthesis and inhibit protein breakdown. In H35 hepatoma cells, where the IGFs act through the insulin receptor, the Long IGF-I analogues maintained a similar potency relative to IGF-I as was observed in the L6 myoblasts. The order of biological potency in cell lines secreting IGF-binding proteins (IGFBPs) into the medium was Long [Arg3]-IGF-I-des(1-3)IGF-I greater than Long [Gly3]-IGF-I greater than Long IGF-I greater than IGF-I. In chicken embryo fibroblasts, a cell line that does not secrete detectable IGFBPs into the medium, Long [Arg3]-IGF-I, was less potent than IGF-I. Investigation of receptor and IGFBP association by these analogues reinforced our previous findings that N-terminal analogues of IGF-I show increased biological potency due to changes in the degree of their IGFBP interactions.

Purification, amino acid sequences and assay cross-reactivities of porcine insulin-like growth factor-I and -II.

Porcine insulin-like growth factor-I (IGF-I) and IGF-II have been characterized to help define the roles of these peptides in the growth process. The amino acid sequence of porcine IGF-I was found to be identical to the human and bovine peptides. Porcine IGF-II was more similar to human IGF-II than to forms of this growth factor in other mammalian species, differing only in the replacement of asparagine for serine at residue 36. In a biological assay that measures the stimulation of protein synthesis in rat L6 myoblasts, porcine IGF-I was approximately ninefold more potent than porcine IGF-II or bovine IGF-II, while recombinant human IGF-I and IGF-II had half the potency of the respective natural peptides. Porcine and recombinant human IGF-I showed essentially equal competition for binding in a human IGF-I radioimmunoassay while between 0.6 and 1.5% cross-reactivity was observed with human, bovine or porcine IGF-II. A receptor assay for IGF-II demonstrated similar potencies for the three IGF-II peptides, while the cross-reactivity of recombinant human IGF-I was only 0.05%. Porcine IGF-I exhibited a higher cross-reactivity, presumably due to very slight contamination with IGF-II.

Sheep insulin-like growth factors I and II: sequences, activities and assays.

This report describes the purification, sequences, and activities of insulin-like growth factors (IGFs) from adult and fetal sheep plasma. IGF-1 from adult sheep is identical to human and bovine IGF-I, except for substitution in the sheep of Ala at residue 66 for Pro in the human and bovine polypeptides. IGF-II from adult sheep differs from bovine IGF-II also by a single amino acid, with residue 62 being Ala in ovine and Thr in bovine IGF-2. The first 10 amino-terminal residues of fetal sheep plasma IGF-I and 92% of the amino acids of fetal IGF-II were identified and found to be the same as those of the corresponding IGFs isolated from adult sheep. Ovine IGF-I was virtually equipotent with human IGF-I in growth-related bioassays and in a RIA for human and bovine IGF-I and inhibited the binding of radiolabeled human IGF-I to type I IGF receptors and to a pure IGF-binding protein. Ovine and bovine IGF-II were also found to be similar to each other in biological and immunochemical activities, and in their binding to type I and II IGF receptors and IGF-binding protein. As observed with human and bovine IGF-I and IGF-II, ovine IGF-I bound slightly better to type I IGF receptors than ovine IGF-II, but bound very poorly to type II IGF receptors. This study shows that IGFs from sheep are very similar to those of human and bovine in structure and activity and defines sensitive radioligand assays specific for ovine IGF-I and ovine IGF-II.

Insulin-like growth factors 1 and 2 in bovine colostrum. Sequences and biological activities compared with those of a potent truncated form.

1. Insulin-like growth factors 1 and 2 (IGF-1 and IGF-2) together with a truncated form of IGF-1 were purified to homogeneity from bovine colostrum. 2. Two forms of IGF-1 were totally resolved from IGF-2 in the purification by h.p.l.c. involving cation-exchange and reverse-phase columns. 3. The complete amino acid sequences for all three forms of IGF were determined. The sequence of bovine IGF-1 was found to be identical with that of human IGF-1, and that of the variant lacked the N-terminal tripeptide Gly-Pro-Glu (-3N:IGF-1). Bovine IGF-2 was found to differ in three residues of the C-domain compared with human IGF-2, with serine, isoleucine and asparagine substituted for alanine, valine and serine respectively at positions 32, 35 and 36. 4. Protein synthesis in L6 rat myoblasts was stimulated and protein degradation inhibited in a co-ordinate response with all three IGFs. The relative potency in both processes was -3N:IGF-1 greater than IGF-1 greater than IGF-2. A similar order of potency was obtained for the stimulation of DNA synthesis by -3N:IGF-1 and IGF-1. The approximately 10-fold effect on biological activity of removing the N-terminal tripeptide is unexpected in view of current information on IGF-1 structure and function.

BRIEF REPORT: HIGHLY CORRELATED PREDICTOR VARIABLES IN MULTIPLE REGRESSION MODELS.

The inclusion in a multiple regression model of a predictor variable which is highly correlated with other prediotor variables is usually not recommended. The argument is that the new predictor variable is accouihing for variance which has already been accounted for in the model. The following discussion is a defense for the procedure of including highly correlated predictor variables under certain circumstances. First, highly correlated variables can be used when there is a requirement that the predictor variables account for a certain number of group membership vectors. A second and more important situation occurs when there is theoretical or empirical justification for the inclusion of such a variable. The following discussion is limited to a very specific kind of highly correlated variable-one that contains the squared elements of one of the original variables.

MEETING THE GOALS OF RESEARCH WITH MULTIPLE LINEAR REGRESSION.

Multiple linear regression is discussed as it relates to several goals of research: Predictability, Parsimony, Replication, and Validity Generalization. law. The emphasis i s upon the per cent of variance accounted for in the criterion under investigation, rather than on statistical significance from random events. Additional remarks concerning curvilinear relationships and data snooping are also presented. Dingman's Canons of Repmducibility are discussed within the framework of multiple linear regression and the goals of research. I These goals are presented with the development of a well established physical.