Characteristics and distribution of endogenous RFamide-related peptide-1

We have recently identified RFamide-related peptide (RFRP) gene that would encode three peptides (i.e., RFRP-1, -2, and -3) in human and bovine, and demonstrated that synthetic RFRP-1 and -3 act as specific agonists for a G protein-coupled receptor OT7T022. However, molecular characteristics and tissue distribution of endogenous RFRPs have not been determined yet. In this study, we prepared a monoclonal antibody for the C-terminal portion of rat RFRP-1. As this antibody could recognize a consensus sequence among the C-terminal portions of rat, human, and bovine RFRP-1, we purified endogenous RFRP-1 from bovine hypothalamus on the basis of immunoreactivity to the antibody. The purified bovine endogenous RFRP-1 was found to have 35-amino-acid length that corresponds to 37-amino-acid length in human and rat. We subsequently constructed a sandwich enzyme immunoassay using the monoclonal antibody and a polyclonal antibody for the N-terminal portion of rat RFRP-1, and analyzed the tissue distribution of endogenous RFRP-1 in rats. Significant levels of RFRP-1 were detected only in the central nervous system, and the highest concentration of RFRP-1 was detected in the hypothalamus. RFRP-1-positive nerve cells were detected in the rat hypothalamus by immunohistochemical analyses using the monoclonal antibody. In culture, RFRP-1 lowered cAMP production in Chinese hamster ovary cells expressing OT7T022 and it was abolished by pre-treatment with pertussis toxin, suggesting that OT7T022 couples G(i)/G(o) in the signal transduction pathway.     

Receptor for the pain modulatory neuropeptides FF and AF is an orphan G protein-coupled receptor

Opiate tolerance and dependence are major clinical and social problems. The anti-opiate neuropeptides FF and AF (NPFF and NPAF) have been implicated in pain modulation as well as in opioid tolerance and may play a critical role in this process, although their mechanism of action has remained unknown. Here we describe a cDNA encoding a novel neuropeptide Y-like human orphan G protein-coupled receptor (GPCR), referred to as HLWAR77 for which NPAF and NPFF have high affinity. Cells transiently or stably expressing HLWAR77 bind and respond in a concentration-dependent manner to NPAF and NPFF and are also weakly activated by FMRF-amide (Phe-Met-Arg-Phe-amide) and a variety of related peptides. The high affinity and potency of human NPFF and human NPAF for HLWAR77 strongly suggest that these are the cognate ligands for this receptor. Expression of HLWAR77 was demonstrated in brain regions associated with opiate activity, consistent with the pain-modulating activity of these peptides, whereas the expression in adipose tissue suggests other physiological and pathophysiological activities for FMRF-amide neuropeptides. The discovery that the anti-opiate neuropeptides are the endogenous ligands for HLWAR77 will aid in defining the physiological role(s) of these ligands and facilitate the identification of receptor agonists and antagonists.     

New neuropeptides containing carboxy-terminal RFamide and their receptor in mammals

Only a few RFamide peptides have been identified in mammals, although they have been abundantly found in invertebrates. Here we report the identification of a human gene that encodes at least three RFamide-related peptides, hRFRP-1-3. Cells transfected with a seven-transmembrane-domain receptor, OT7T022, specifically respond to synthetic hRFRP-1 and hRFRP-3 but not to hRFRP-2. RFRP and OT7T022 mRNAs are expressed in particular regions of the rat hypothalamus, and intracerebroventricular administration of hRFRP-1 increases prolactin secretion in rats. Our results indicate that a variety of RFamide-related peptides may exist and function in mammals.     

Evolutionary origin and divergence of PQRFamide peptides and LPXRFamide peptides in the RFamide peptide family. Insights from novel lamprey RFamide peptides

Among the RFamide peptide groups, PQRFamide peptides, such as neuropeptide FF (NPFF) and neuropeptide AF (NPAF), share a common C-terminal Pro-Gln-Arg-Phe-NH(2) motif. LPXRFamide (X = L or Q) peptides, such as gonadotropin-inhibitory hormone (GnIH), frog growth hormone-releasing peptide (fGRP), goldfish LPXRFamide peptide and mammalian RFamide-related peptides (RFRPs), also share a C-terminal Leu-Pro-Leu/Gln-Arg-Phe-NH(2) motif. Such a similar C-terminal structure suggests that these two groups may have diverged from a common ancestral gene. In this study, we sought to clarify the evolutionary origin and divergence of these two groups, by identifying novel RFamide peptides from the brain of sea lamprey, one of only two extant groups of the oldest lineage of vertebrates, Agnatha. A novel lamprey RFamide peptide was identified by immunoaffinity purification using the antiserum against LPXRFamide peptide. The lamprey RFamide peptide did not contain a C-terminal LPXRFamide motif, but had the sequence SWGAPAEKFWMRAMPQRFamide (lamprey PQRFa). A cDNA of the precursor encoded one lamprey PQRFa and two related peptides. These related peptides, which also had the C-terminal PQRFamide motif, were further identified as mature endogenous ligands. Phylogenetic analysis revealed that lamprey PQRFamide peptide precursor belongs to the PQRFamide peptide group. In situ hybridization demonstrated that lamprey PQRFamide peptide mRNA is expressed in the regions predicted to be involved in neuroendocrine and behavioral functions. This is the first demonstration of the presence of RFamide peptides in the agnathan brain. Lamprey PQRFamide peptides are considered to have retained the most ancestral features of PQRFamide peptides.     

Cryptic peptide derived from the rat neuropeptide FF precursor affects G-proteins linked to opioid receptors in the rat brain

Recently, we reported the discovery of a novel amino acid sequence derived from the NPFF precursor NAWGPWSKEQLSPQA, which blocked the expression of conditioned place preference induced by morphine and reversed the antinociceptive activity of morphine (5mg/kg, s.c.) in the tail-immersion test in rats. Here, we name it as NPNA (Neuropeptide NA from its flanking amino acid residues). The synthetic peptide influenced the expression of mRNA coding for Galpha(i1), (i2), and (i3) subunits. The results provide further evidence that yet another bioactive sequence might be present within the NPFF precursor.     

Structure-activity relationships of neuropeptide FF: role of C-terminal regions

A structure-activity study was carried out to determine the importance of the C-terminal amino acids of the octapeptide Neuropeptide FF (NPFF) in binding and agonistic activity. Affinities of NPFF analogues were tested toward NPFF receptors of the rat spinal cord and the human NPFF2 receptors transfected in CHO cells. The activities of these analogues were evaluated by their ability to both inhibit adenylate cyclase in NPFF2 receptor transfected CHO cells and to reverse the effect of nociceptin on acutely dissociated rat dorsal raphe neurons. The substitutions of Phenylalanine8 by a tyrosine, phenylglycine or homophenylalanine were deleterious for high affinity. Similarly, the replacement of Arginine7 by a lysine or D.Arginine induces a loss in affinity. The pharmacological characterization showed that the presence of the amidated Phe8 and Arg7 residues are also extremely critical for activation of anti-opioid effects on dorsal raphe neurons. The sequence of the C-terminal dipeptide seems also to be responsible for the high affinity and the activity on human NPFF2 receptors. The results support the view that a code messaging the molecular interaction toward NPFF-receptors is expressed in the C-terminal region of these peptides but the N-terminal segment is important to gain very high affinity.     

(125)I]EYF: a new high affinity radioligand to neuropeptide FF receptors

[(125)I]EYF ([(125)I]EYWSLAAPQRFamide), a new radioiodinated probe derived from a peptide present in the rat Neuropeptide FF precursor (EFWSLAAPQRFamide, EFW-NPSF) was synthesized and its binding characteristics investigated on sections of the rat spinal cord and on membranes of mouse olfactory bulb. In both tissues, [(125)I]EYF binding was saturable and revealed a very high affinity interaction with a single class of binding sites in rat and mouse (K(D) = 0.041 and 0.019 nM, respectively).Competition studies showed that [(125)I]EYF bound to one class of binding sites exhibiting a high affinity for all the different peptides the precursor could generate (NPA-NPFF, SPA-NPFF, NPFF, EFW-NPSF, QFW-NPSF) with the exception of NPSF which displayed a low affinity.Autoradiographic studies demonstrated that [(125)I]EYF binding sites were fully inhibited by a synthetic Neuropeptide FF agonist (1DMe) in all areas of the rat brain. The density of [(125)I]EYF binding sites was high in the intralaminar thalamic nuclei, the parafascicular thalamic nucleus and in the superficial layers of the dorsal horn.Non specific binding reached 5-10% of the total binding in all brain areas. Similarly, in mouse brain experiments, the non-specific binding was never superior to 10%.These findings demonstrate that putative neuropeptides generated by the Neuropeptide FF precursor and containing the NPFF or NPSF sequences should bind to the same receptor. Furthermore, these data indicate that [(125)I]EYF is a useful radiolabeled probe to investigate the NPFF receptors; its major advantages being its high affinity and the very low non-specific binding it induces.     

Characterization of the NPGP receptor and identification of a novel short mRNA isoform in human hypothalamus

Recently, an orphan G protein coupled receptor (GPCR) termed NPGPR was described. A shorter variant of this receptor lacking exon 1 was shown to have subnanomolar affinity for neuropeptide FF (NPFF), a pain modulatory peptide, and therefore was named NPFF(2) receptor. Here, we characterize the full-length cloned NPGPR and identify a novel short form lacking exon 2 with a differential pattern of mRNA abundance in several tissues and organs. The NPGPR is most similar to the recently cloned neuropeptide FF (NPFF) receptor which lacks exon 1, but also shows high homology to the orexin and neuropeptide Y (NPY) receptor families, two neuropeptides involved in food intake regulation. Therefore, we used binding studies to examine the interaction of NPFF, orexin and NPY with the NPGPR. [125I] NPFF was displaced by NPFF with an IC(50) of 14.7 +/- 8.8 nM, whereas [125I] Orexin B was displaced by Orexin B with an IC(50) of 415 +/- 195 nM. We conclude that orexins interact with the NPGPR and that the affinity of NPFF for NPGPR is approximately 100-fold lower than for the NPFF2 receptor. We postulate that NPGPR is a splice variant of the family of NPFF receptors and displays a binding profile different from the other members of the NPFF receptor family due to the presence of exon 1. In order to evaluate whether NPGPR levels are affected by the feeding status, we examined the mRNA level using real-time PCR in two feeding models, i.e. before and after diet-induced body weight increase as well as after chronic food restriction in rats. However, hypothalamic NPGPR mRNA was unchanged in both models. Therefore, our evidence does not support the hypothesis that NPGPR is involved in feeding regulation.     

Evolutionary Origin of GnIH and NPFF in Chordates: Insights from Novel Amphioxus RFamide Peptides

Gonadotropin-inhibitory hormone (GnIH) is a newly identified hypothalamic neuropeptide that inhibits pituitary hormone secretion in vertebrates. GnIH has an LPXRFamide (X = L or Q) motif at the C-terminal in representative species of gnathostomes. On the other hand, neuropeptide FF (NPFF), a neuropeptide characterized as a pain-modulatory neuropeptide, in vertebrates has a PQRFamide motif similar to the C-terminal of GnIH, suggesting that GnIH and NPFF have diverged from a common ancestor. Because GnIH and NPFF belong to the RFamide peptide family in vertebrates, protochordate RFamide peptides may provide important insights into the evolutionary origin of GnIH and NPFF. In this study, we identified a novel gene encoding RFamide peptides and two genes of their putative receptors in the amphioxus Branchiostoma japonicum. Molecular phylogenetic analysis and synteny analysis indicated that these genes are closely related to the genes of GnIH and NPFF and their receptors of vertebrates. We further identified mature RFamide peptides and their receptors in protochordates. The identified amphioxus RFamide peptides inhibited forskolin induced cAMP signaling in the COS-7 cells with one of the identified amphioxus RFamide peptide receptors expressed. These results indicate that the identified protochordate RFamide peptide gene is a common ancestral form of GnIH and NPFF genes, suggesting that the origin of GnIH and NPFF may date back to the time of the emergence of early chordates. GnIH gene and NPFF gene may have diverged by whole-genome duplication in the course of vertebrate evolution.     

Factor XI binding to activated platelets is mediated by residues R(250), K(255), F(260), and Q(263) within the apple 3 domain

To localize the platelet binding site on factor XI, rationally designed, conformationally constrained synthetic peptides were used to compete with [(125)I]factor XI binding to activated platelets. The major platelet binding energy resided within the sequence of amino acids T(249)-F(260). Homology scanning, using prekallikrein amino acid substitutions within the synthetic peptide T(249)-F(260), identified a major role for R(250) in platelet binding. Inhibition of [(125)I]factor XI binding to activated platelets by the recombinant Apple 3 domain of factor XI and inhibition by unlabeled factor XI were identical, whereas the recombinant Apple 3 domain of prekallikrein had little effect. A "gain-of-function" chimera in which the C-terminal amino acid sequence of the Apple 3 domain of prekallikrein was replaced with that of factor XI was as effective as the recombinant Apple 3 domain of factor XI and unlabeled factor XI in inhibiting [(125)I]factor XI binding to activated platelets. Alanine scanning mutagenic analysis of the recombinant Apple 3 domain of factor XI indicated that amino acids R(250), K(255), F(260), and Q(263) (but not K(252) or K(253)) are important for platelet binding. Thus, the binding energy mediating the interaction of factor XI with platelets is contained within the C-terminal amino acid sequence of the Apple 3 domain (T(249)-V(271)) and is mediated in part by amino acid residues R(250), K(255), F(260), and Q(263).     

The recognition specificity of the CHD1 chromodomain with modified histone H3 peptides

Histone tail peptides comprise the flexible portion of chromatin, the substance which serves as the packaging for the eukaryotic genome. According to the histone code hypothesis, reader protein domains (chromodomains) can recognize modifications of amino acid residues within these peptides, regulating the expression of genes. We have performed simulations on models of chromodomain helicase DNA-binding protein 1 complexed with a variety of histone H3 modifications. Binding free energies for both the overall complexes and the individual residues within the protein and peptides were computed with molecular mechanics-generalized Born surface area. The simulation results agree well with experimental data and identify several chromodomain helicase DNA-binding protein 1 residues that play key roles in the interaction with each of the H3 modifications. We identified one class of protein residues that bind to H3 in all of the complexes (generally interacting hydrophobically), and a second class of residues that bind only to particular H3 modifications (generally interacting electrostatically). Additionally, we found that modifications of H3R2 and H3T3 have a dominant effect on the binding affinity; methylation of H3K4 has little effect on the interaction strength when H3R2 or H3T3 is modified. Our findings with regard to the specificity shown by the latter class of protein residues in their binding affinity to certain modifications of H3 support the histone code hypothesis.     

Dimeric peptides of the C-terminal region of CXCL14 function as CXCL12 inhibitors

We recently reported that CXCL14 binds to CXCR4 with high affinity and inhibits CXCL12-mediated chemotaxis. Here we found that the C-terminal 51–77 amino acid residues of CXCL14 are responsible for CXCR4 binding. A disulfide dimer peptide of CXCL14(51–77) bound to CXCR4 with comparable affinity to full length CXCL14, and exhibited CXCL12 inhibitor activity. CXCR4 was efficiently internalized upon binding of dimeric CXCL14(51–77), thereby being reduced on the cell surface. Substitution of 5 amino acid residues in combination with the use of an oxime linker for dimerization increased the solubility and chemical stability of the dimeric CXCL14(51–77).     

Epitope analysis using kinetic measurements of antibody binding to synthetic peptides presenting single amino acid substitutions

The fine modulation of peptide–antibody interactions was investigated with anti-peptide monoclonal antibodies recognizing peptide 125–136 of the coat protein of tobacco mosaic virus. Nine synthetic peptides presenting single amino acid substitutions were selected for detailed analysis on the basis of their reactivity in ELISA. Kinetic measurements of the binding of four antibodies to these peptides performed with a biosensor instrument (BIAcore^TM, Pharmacia) were used to quantify the contribution of individual residues to antibody binding. The results showed that even conservative exchanges of some residues in the epitope results in a small but significant decrease of the equilibrium affinity constant due mostly to a higher dissociation rate constant of the monoclonal antibodies. Two amino acid residues directly adjacent to the epitope, which appeared to play no role when tested by ELISA, were shown to influence the kinetics of binding. These data should be useful for computer modelling of the peptide–antibody interactions.     

N-Glycans protect proteins from protease digestion through their binding affinities for aromatic amino acid residues

It was previously revealed [Yamaguchi, H., Nishiyama, T., and Uchida, M. (1999) J. Biochem. 126, 261-265] that N-glycans of both the high-mannose and complex types have binding affinity for aromatic amino acid residues. This study shows that free N-glycans protect proteins from protease digestion through their binding affinities for the aromatic amino acid residues exposed on protein molecules. Protease digestion of bovine pancreatic RNase A and bovine a-lactalbumin was depressed in solutions (1 mM or so) of free N-glycans of both the high-mannose and complex types. The increasing order of the protective effects of the N-glycans paralleled that of their affinities for aromatic amino acid residues; and the presence of aromatic amino acids practically abolished the protective effects of the N-glycans. The N-glycans also depressed the protease digestion of metallothionein, an aromatic amino acid-free protein, in agreement with the observation that the N-glycans also interact with the solvent-exposed aromatic amino acid residues of the proteases. Thus it seems probable that the N-glycans protect proteins from protease digestion by steric hindrance attributable to their binding affinity for the solvent-exposed aromatic amino acid residues of both substrate proteins and proteases.     

Identification of cultured cells selectively expressing Y1-, Y2-, or Y3-type receptors for neuropeptide Y/peptide YY.

Neuropeptide Y (NPY) and peptide YY (PYY) are homologous 36 amino acid amidated peptides that often, but not always, exert similar actions and binding profiles. The present study of cultured cells confirms that both peptides as well as radioiodinated analogs, i.e. 125I-Bolton-Hunter-NPY (125I-BH-NPY) and 125I-peptide YY (125I-PYY), show high affinity to binding sites/receptors of the previously proposed Y1- and Y2-subtypes, selectively expressed by the human neuroblastoma cell lines, SK-N-MC and SK-N-BE(2), respectively. In contrast, bovine adrenal chromaffin cells did not bind 125I-PYY, while displaying high affinity 125I-BH-NPY sites, and may therefore represent a cell type expressing a recently proposed Y3-type of (NPY-preferring) receptors. Several non-labeled fragments/analogs have been used in displacement experiments to further characterize the structural requirements for Y1-, Y2-, and Y3-type binding. In every instance, specific binding was reduced by addition of 5'-guanylylimidodiphosphate [Gpp(NH)p], indicating that the three receptor subtypes belong to the G-protein-coupled superfamily of receptors. Moreover, in both neuroblastoma cell lines, the peptides elicited, with appropriate orders of potency, reduction of forskolin-stimulated adenosine 3',5'-cyclic monophosphate (cAMP) accumulation. Finally, NPY-evoked 45Ca2+ influx was observed in SK-N-MC and in chromaffin cells. A common dual coupling mechanism of NPY/PYY receptors, i.e. to reduction of cAMP and to Ca2+ elevation, is therefore suggested to exist, although both phenomena could not be demonstrated in every cell type.     

High-affinity AKAP7delta-protein kinase A interaction yields novel protein kinase A-anchoring disruptor peptides

PKA (protein kinase A) is tethered to subcellular compartments by direct interaction of its regulatory subunits (RI or RII) with AKAPs (A kinase-anchoring proteins). AKAPs preferentially bind RII subunits via their RII-binding domains. RII-binding domains form structurally conserved amphipathic helices with unrelated sequences. Their binding affinities for RII subunits differ greatly within the AKAP family. Amongst the AKAPs that bind RIIalpha subunits with high affinity is AKAP7delta [AKAP18delta; K(d) (equilibrium dissociation constant) value of 31 nM]. An N-terminally truncated AKAP7delta mutant binds RIIalpha subunits with higher affinity than the full-length protein presumably due to loss of an inhibitory region [Henn, Edemir, Stefan, Wiesner, Lorenz, Theilig, Schmidtt, Vossebein, Tamma, Beyermann et al. (2004) J. Biol. Chem. 279, 26654-26665]. In the present study, we demonstrate that peptides (25 amino acid residues) derived from the RII-binding domain of AKAP7delta bind RIIalpha subunits with higher affinity (K(d)=0.4+/-0.3 nM) than either full-length or N-terminally truncated AKAP7delta, or peptides derived from other RII binding domains. The AKAP7delta-derived peptides and stearate-coupled membrane-permeable mutants effectively disrupt AKAP-RII subunit interactions in vitro and in cell-based assays. Thus they are valuable novel tools for studying anchored PKA signalling. Molecular modelling indicated that the high affinity binding of the amphipathic helix, which forms the RII-binding domain of AKAP7delta, with RII subunits involves both the hydrophobic and the hydrophilic faces of the helix. Alanine scanning (25 amino acid peptides, SPOT technology, combined with RII overlay assays) of the RII binding domain revealed that hydrophobic amino acid residues form the backbone of the interaction and that hydrogen bond- and salt-bridge-forming amino acid residues increase the affinity of the interaction.     

Specific recognition of the collagen triple helix by chaperone HSP47. II. The HSP47-binding structural motif in collagens and related proteins

The endoplasmic reticulum-resident chaperone heat-shock protein 47 (HSP47) plays an essential role in procollagen biosynthesis. The function of HSP47 relies on its specific interaction with correctly folded triple-helical regions comprised of Gly-Xaa-Yaa repeats, and Arg residues at Yaa positions have been shown to be important for this interaction. The amino acid at the Yaa position (Yaa(-3)) in the N-terminal-adjoining triplet containing the critical Arg (defined as Arg(0)) was also suggested to be directly recognized by HSP47 (Koide, T., Asada, S., Takahara, Y., Nishikawa, Y., Nagata, K., and Kitagawa, K. (2006) J. Biol. Chem. 281, 3432-3438). Based on this finding, we examined the relationship between the structure of Yaa(-3) and HSP47 binding using synthetic collagenous peptides. The results obtained indicated that the structure of Yaa(-3) determined the binding affinity for HSP47. Maximal binding was observed when Yaa(-3) was Thr. Moreover, the required relative spatial arrangement of these key residues in the triple helix was analyzed by taking advantage of heterotrimeric collagen-model peptides, each of which contains one Thr(-3) and one Arg(0). The results revealed that HSP47 recognizes the Yaa(-3) and Arg(0) residues only when they are on the same peptide strand. Taken together, the data obtained led us to define the HSP47-binding structural epitope in the collagen triple helix and also define the HSP47-binding motif in the primary structure. A motif search against human protein database predicted candidate clients for this molecular chaperone. The search result indicated that not all collagen family proteins require the chaperoning by HSP47.     

Polyarginine Peptides As a New Class of Ligands of Nicotinic Acetylcholine Receptors

Arginine-containing peptides R3, R8, and R16 were obtained by solid-phase peptide synthesis, and their binding to nicotinic acetylcholine receptors (nAChRs) of muscle and neuronal (α7) types was studied by competitive radioligand assay with the use of ¹²⁵I-α-bungarotoxin. The resulting peptides exhibited a significantly greater binding activity with respect to the muscle-type nAChRs than to the α7 receptor. Thus, we have discovered a new class of nAChR ligands. The affinity of the synthesized oligoarginines for nAChR depended on the number of amino acid residues in the chain. The highest affinity was exhibited by the R16 peptide, which contained 16 arginine residues.     

Characterization of the cocaine binding site on the sigma-1 receptor

The cocaine photoaffinity label 3-iodo-4-azidococaine ([125I]IACoc) binds to the sigma-1 receptor with an affinity that is 2-3 orders of magnitude higher than the parent compound cocaine [Kahoun, J. R., and Ruoho, A. E. (1992) Proc. Natl. Acad. Sci. U.S.A. 89, 1393-1397]. In the present study, the binding properties of several cocaine derivatives to the guinea pig liver sigma-1 receptor were determined. The results from assessing the affinity of various derivatives of cocaine which were substituted on the phenyl ring indicated that an important determinant of binding to the guinea pig sigma-1 receptor binding site may be the development of a dipole in the ring in which the pi electron density of the phenyl ring is reduced. This implies that an electron-rich source is present in the sigma-1 receptor binding site, such as the pi system of an aromatic ring or other electron-rich side chains, which interact with the phenyl ring of cocaine. The precise [125I]IACoc derivatization site in the guinea pig sigma-1 receptor was identified using chemical cleavage and purification of the resulting labeled peptides. Cyanogen bromide cleavage of the [125I]IACoc photolabeled sigma-1 receptor followed by radiosequencing identified Asp188, which is located in the putative steroid binding domain-like II (SBDL II) near the carboxyl terminus, as the site of [125I]IACoc insertion. Systematic truncation of the C-terminus indicated the requirement for the last 15 amino acid residues of the receptor for [125I]IACoc photolabeling.     

Identification of peptide epitopes of MAGE-1, -2, -3 that demonstrate HLA-A3-specific binding

 The MAGE gene family of tumour antigens are expressed in a wide variety of human cancers. We have identified 43 nonamer peptide sequences, from MAGE-1, -2 and -3 proteins that contain binding motifs for HLA-A3 MHC class I molecules. The T2 cell line, transfected with the cDNA for the HLA-A3 gene, was used in a MHC class I stabilisation assay performed at 37°C and 26°C. At 37°C, 2 peptides were identified that stabilised HLA-A3 with high affinity (fluorescence ratio, FR >1.5), 4 peptides with low affinity (FR 1.11 – 1.49) and 31 peptides that did not stabilise this HLA haplotype (FR <1.1). At 26°C, 12 peptides were identified that stabilised HLA-A3 with high affinity, 8 peptides with low affinity and 17 peptides that did not stabilise this HLA haplotype. Two peptides stabilised HLA-A3 at both temperatures. Small changes in one to three amino acids at positions distinct from the anchor residues altered peptide affinity. Data were compared to a similar study in which a peptide competition assay was used to investigate MAGE-1 peptide binding to several HLA haplotypes. This study demonstrates that anchor residues do not accurately predict peptide binding to specific HLA haplotypes, changes in one to three amino acids at positions distinct from anchor residues influence peptide binding and alternative methods of determining peptide binding yield different results. We are currently investigating the ability of these peptides to induce antitumour cytotoxic T lymphocyte activity as they may be of potential therapeutic value. Received: 4 January 1996 / Accepted: 20 March 1996