Organic solvents identify specific ligand binding sites on protein surfaces

Over many years, a great deal of attention has focused on the growth regulatory effects of androgens in prostate cells. This, has also prompted widespread interest in the role of these steroid hormones in prostate cancer pathogenesis. Even so, no-one has so far been able to identify the exact relationship between androgenic hormone levels and the risk of these diseases though differences in hormonal patterns amongst racial ethnic groups has been reported to reflect diversities in prostate cancer incidence. One of the difficulties stems from the fact that serum hormone levels do not reflect the changes observed in prostate tissue androgen concentrations as the normal prostate progresses to a disease state. In this article efforts will be directed towards understanding some of the intra-prostate-specific mechanisms responsible for activating and/or repressing the androgen-dependent gene network associated with the gradual transition to a hormone refractive neoplastic state.     

Determination of the binding frame of the chaperone SecB within the physiological ligand oligopeptide-binding protein

Chaperone proteins demonstrate the paradoxical ability to bind ligands rapidly and with high affinity but with no apparent sequence specificity. To learn more about this singular property, we have mapped the binding frame of the chaperone SecB from E. coli on the oligopeptide-binding protein. Similar studies performed on the maltose-binding and galactose-binding proteins revealed centrally positioned binding frames of approximately 160 aminoacyl residues. The work described here shows that OppA, which is significantly longer than the previously studied ligands, has a binding frame that covers 460 amino acids, nearly the entire length of the protein. We propose modes of binding to account for the data.     

Topology of ligand binding sites on the nicotinic acetylcholine receptor

The nicotinic acetylcholine receptor (AChR) presents two very well differentiated domains for ligand binding that account for different cholinergic properties. In the hydrophilic extracellular region of both alpha subunits there exist the binding sites for agonists such as the neurotransmitter acetylcholine (ACh) and for competitive antagonists such as d-tubocurarine. Agonists trigger the channel opening upon binding while competitive antagonists compete for the former ones and inhibit its pharmacological action. Identification of all residues involved in recognition and binding of agonist and competitive antagonists is a primary objective in order to understand which structural components are related to the physiological function of the AChR. The picture for the localisation of the agonist/competitive antagonist binding sites is now clearer in the light of newer and better experimental evidence. These sites are mainly located on both alpha subunits in a pocket approximately 30-35 A above the surface membrane. Since both alpha subunits are sequentially identical, the observed high and low affinity for agonists on the receptor is conditioned by the interaction of the alpha subunit with the delta or the gamma chain, respectively. This relationship is opposite for curare-related drugs. This molecular interaction takes place probably at the interface formed by the different subunits. The principal component for the agonist/competitive antagonist binding sites involves several aromatic residues, in addition to the cysteine pair at 192-193, in three loops-forming binding domains (loops A-C). Other residues such as the negatively changed aspartates and glutamates (loop D), Thr or Tyr (loop E), and Trp (loop F) from non-alpha subunits were also found to form the complementary component of the agonist/competitive antagonist binding sites. Neurotoxins such as alpha-, kappa-bungarotoxin and several alpha-conotoxins seem to partially overlap with the agonist/competitive antagonist binding sites at multiple point of contacts. The alpha subunits also carry the binding site for certain acetylcholinesterase inhibitors such as eserine and for the neurotransmitter 5-hydroxytryptamine which activate the receptor without interacting with the classical agonist binding sites. The link between specific subunits by means of the binding of ACh molecules might play a pivotal role in the relative shift among receptor subunits. This conformational change would allow for the opening of the intrinsic receptor cation channel transducting the external chemical signal elicited by the agonist into membrane depolarisation. The ion flux activity can be inhibited by non-competitive inhibitors (NCIs). For this kind of drugs, a population of low-affinity binding sites has been found at the lipid-protein interface of the AChR. In addition, several high-affinity binding sites have been found to be located at different rings on the M2 transmembrane domain, namely luminal binding sites. In this regard, the serine ring is the locus for exogenous NCIs such as chlorpromazine, triphenylmethylphosphonium, the local anaesthetic QX-222, phencyclidine, and trifluoromethyliodophenyldiazirine. Trifluoromethyliodophenyldiazirine also binds to the valine ring, which is the postulated site for cembranoids. Additionally, the local anaesthetic meproadifen binding site seems to be located at the outer or extracellular ring. Interestingly, the M2 domain is also the locus for endogenous NCIs such as the neuropeptide substance P and the neurotransmitter 5-hydroxytryptamine. In contrast with this fact, experimental evidence supports the hypothesis for the existence of other NCI high-affinity binding sites located not at the channel lumen but at non-luminal binding domains. (ABSTRACT TRUNCATED)     

The two mannose 6-phosphate binding sites of the insulin-like growth factor-II/mannose 6-phosphate receptor display different ligand binding properties

The two mannose 6-phosphate (Man-6-P) binding sites of the insulin-like growth factor-II/mannose 6-phosphate receptor (IGF-II/MPR) have been localized to domains 1-3 and 7-9, and studies have shown that Arg435 in domain 3 and Arg 1334 in domain 9 are essential for Man-6-P binding. To determine whether the IGF-II/MPR containing a single Man-6-P binding site is functional, clonal mouse L cell lines stably transfected with either mutant bovine IGF-II/MPR cDNA, containing substitutions at position 435 and/or 1334, or the wild type receptor cDNA were assayed for their ability to sort lysosomal enzymes to the lysosome. Mutant receptors containing a single Man-6-P binding site were approximately 50% less efficient than the wild type receptor in the overall targeting of lysosomal enzymes to the lysosome. Mutant receptors containing a substitution at Arg1334 (Dom9(Ala)), in contrast to those containing a substitution at Arg435 (Dom3(Ala)), were unable to target cathepsin D and beta-hexosaminidase to the lysosome. Equilibrium binding assays using 125I-labeled beta-glucuronidase demonstrated that Dom3(Ala) and Dom9(Ala) had a Kd of 2.0 and 4.3 nM, respectively. In addition, Dom3(Ala), unlike Dom9(Ala), was unable to completely dissociate from ligand under acidic pH conditions. These data indicate that the two Man-6-P binding sites of the IGF-II/MPR are not functionally equivalent.     

Identification of two classes of lipid molecule binding sites on the microsomal triglyceride transfer protein

The gene for the microsomal triglyceride transfer protein (MTP) is defective in subjects with the genetic disease abetalipoproteinemia, indicating that MTP is essential for the assembly of apolipoprotein B containing lipoproteins. In vitro, MTP is a lipid molecule binding protein that catalyzes lipid transport between membranes by a shuttle mechanism. In this study, the lipid binding properties of MTP were examined. MTP was incubated with donor phosphatidylcholine vesicles of varying neutral lipid composition. MTP was subsequently reisolated by ultracentrifugation, and MTP-bound lipid was quantitated. When the triolein content of the vesicles was increased up to 4 mol %, neutral lipid binding to MTP increased proportionately, while phosphatidylcholine binding appeared to remain constant around two molecules per MTP. Using phosphatidylcholine emulsions containing 60 mol % triolein as the donor particles resulted in only a slight increase in triolein binding to MTP. The highest triolein:MTP ratio observed was (0.20-0.25):1. Differences in the neutral and phospholipid binding properties of MTP were observed by measuring the transport of lipid from MTP to acceptor vesicles. Transport of triolein was rapid and complete, while phosphatidylcholine transport was biphasic, containing rapid and slow phases. These results indicated that MTP contains more than one class of lipid molecule binding site. Measurements of fluorescent lipid transport from donor vesicles to MTP supported this hypothesis. The transport of pyrene-labeled triglyceride from donor particles to MTP was rapid, while phosphatidylcholine transfer had fast and slow phases. From these data, we propose that MTP contains at least two distinct classes of lipid molecule binding sites that differ in function. The fast site or sites are responsible for lipid transport.     

Imidazole binding to Rhodobacter capsulatus cytochrome c2. Effect of site-directed mutants on ligand binding

Although ligand binding in c-type cytochromes is not directly related to their physiological function, it has the potential to provide valuable information on protein stability and dynamics, particularly in the region of the methionine sixth heme ligand and the nearby peptide chain that has been implicated in electron transfer. Thus, we have measured the equilibrium and kinetics of binding of imidazole to eight mutants of Rhodobacter capsulatus cytochrome c2 that differ in overall protein stability. We found that imidazole binding affinity varies 70-fold, but does not correlate with overall protein stability. Instead, each mutant exerts an effect at the local level, with the largest change due to mutant G95E (glycine substituted by glutamate), which shows 30-fold stronger binding as compared with the wild-type protein. The kinetics of imidazole binding are monophasic and reach saturation at high ligand concentrations for all the mutants and wild-type protein, which is attributed to a rate-limiting conformational change leading to breakage of the iron-methionine bond and providing a binding site for imidazole. The mutants show as much as an 18-fold variation in the first-order rate constant for the conformational change, with the largest effect found with mutant G95E. The kinetics also show a lack of correlation with overall protein stability, but are consistent with localized effects on the dynamics of hinge region 88-102 of the protein, which changes conformation to permit ligand binding. These results are consistent with R. capsulatus cytochrome c2 stabilizing the complex through hydrogen bonding to the imidazole. The larger effects of mutant G95E on equilibrium and kinetics are likely to be due to its location within the hinge region adjacent to heme ligand methionine 96, which is displaced by imidazole.     

Predicting ligand binding to proteins by affinity fingerprinting

BACKGROUND: There are many ways to represent a molecule's properties, including atomic-connectivity drawings, NMR spectra, and molecular orbital models. Prior methods for predicting the biological activity of compounds have largely depended on these physical representations. Measuring a compound's binding potency against a small reference panel of diverse proteins defines a very different representation of the molecule, which we call an affinity fingerprint. Statistical analysis of such fingerprints provides new insights into aspects of binding interactions that are shared among a wide variety of proteins. These analyses facilitate prediction of the binding properties of these compounds assayed against new proteins. RESULTS: Affinity fingerprints are reported for 122 structurally-diverse compounds using a reference panel of eight proteins that collectively are able to generate unique fingerprints for about 75% of the small organic compounds tested. Application of multivariate regression techniques to this database enables the creation of computational surrogates to represent new proteins that are surprisingly effective at predicting binding potencies. We illustrate this for two enzymes with no previously recognizable similarity to each other or to any of the reference proteins. Fitting of analogous computational surrogates to four other proteins confirms the generality of the method; when applied to a fingerprinted library of 5000 compounds, several sub-micromolar hits were correctly predicted. CONCLUSIONS: An affinity fingerprint database, which provides a rich source of data defining operational similarities among proteins, can be used to test theories of cryptic homology unexpected from current understanding of protein structure. Practical applications to drug design include efficient pre-screening of large numbers of compounds against target proteins using fingerprint similarities, supplemented by a small number of empirical measurements, to select promising compounds for further study.     

Differential proximity probing of two DNA binding sites in the Escherichia coli recA protein using photo-cross-linking methods

The DNA strand-exchange reaction catalyzed by the Escherichia coli RecA protein occurs between the two DNA binding sites that are functionally distinct. Site I is the site to which a DNA molecule (normally single-stranded DNA) binds first; this first binding makes site II available for additional DNA-binding (normally double- stranded DNA). Photo-cross linking was employed to identify the amino acid residues located close to the bound DNA molecule(s). A ssDNA oligo containing multiple 5-iodouracil residues (IdU) was cross-linked to RecA by irradiation with a XeC1 pulse laser (308 nm), and the cross-linked peptides were purified and sequenced. To differentiate the two DNA binding sites, we used two protocols for making RecA-ssDNA complexes: (1) IdU-containing oligo was mixed with a stoichiometric excess of RecA, a condition which favors the binding of the oligo to site I, and (2) RecA was first allowed to bind to a nonphotoreactive oligo and then chased with the IdU-containing oligo, a condition which favors the binding of the IdU-oligo to site II. We observed that when RecA was in excess (site I probing), cross-linking occurred to Met-164 which is located in the disordered loop 1 of the RecA crystal structure [Story, R.M., Weber, I.T., & Steitz, T.A. (1992) Nature 355, 318-325]. When site II was probed, the majority of cross-linking occurred to Met-202 or Phe-203, located in loop 2. These results support the idea that, as predicted by Story and co-workers (1992), the disordered loops are involved in DNA binding. The results also suggest that the two sites are not only functionally but also physically distinct.     

Determination of calcium binding sites in gas-phase small peptides by tandem mass spectrometry

Low-energy (LE) and high-energy (HE) collisionally activated decompositions (CAD) of calcium/peptide complexes of the form [M - H + Ca]+ and [M + Ca]2+ reflect the site of calcium binding in various gas-phase peptides that are models of the calcium binding site III of rabbit skeletal troponin C. The Ca2+ binding sites involve an aspartic acid, glutamic acid, and asparagine, which are in the metal-binding loops of calcium-binding proteins. Both fast atom bombardment (FAB) and electrospray ionization (ESI) were used to generate the metal/peptide complexes. When submitted to LE CAD, ESI-produced Ca2+/peptide complexes undergo fragmentations that are controlled by Ca2+ binding and provide information on the Ca2+ binding site. The LE CAD spectra are simple, indicating that Ca2+ binding involves specific oxygen ligands including acidic side chains and that only a few low-energy fragmentation channels exist. The HE CAD spectra of FAB-produced Ca2+/peptide complexes are more complex, owing to the introduction of high internal energy into the precursor ion. Interactions of the other alkaline-earth metal ions Mg2+ and Ba2+ with these peptides reveal that the ligand preferences of these metal ions are slightly different than those of Ca2+.     

Molecular analysis of ligand binding to the second cluster of complement-type repeats of the low density lipoprotein receptor-related protein. Evidence for an allosteric component in receptor-associated protein-mediated inhibition of ligand binding

The low density lipoprotein receptor-related protein (LRP), a member of the low density lipoprotein receptor gene family, mediates the cellular uptake of a diversity of ligands. A folding chaperone, the 39-kDa receptor-associated protein (RAP) that resides in the early compartments of the secretory pathway inhibits the binding of all ligands to the receptor and may serve to prevent premature binding of ligands to the receptor during the trafficking to the cell surface. To elucidate the molecular interactions that underlie the interplay between the receptor, RAP, and the ligands, we have analyzed and delineated the binding sites of plasminogen activator inhibitor-1 (PAI-1), tissue-type plasminogen activator (t-PA).PAI-1 complexes, RAP, and the anti-LRP Fab fragment Fab A8. To that end, we have generated a series of soluble recombinant fragments spanning the second cluster of complement-type repeats (C3-C10) and the amino-terminal flanking epidermal growth factor repeat (E4) of LRP (E4-C10; amino acids 787-1165). All fragments were expressed by stably transfected baby hamster kidney cells and purified by affinity chromatography. A detailed study of ligand binding to the fragments using surface plasmon resonance revealed the presence of three distinct, Ca2+-dependent ligand binding sites in the cluster II domain (Cl-II) of LRP. t-PA.PAI-1 complexes as well as PAI-1 bind to a domain located in the amino-terminal portion of Cl-II, spanning repeats E4-C3-C7. Adjacent to this site and partially overlapping is a high affinity RAP-binding site located on repeats C5-C7. Fab A8, a pseudo-ligand of the receptor, binds to a third Ca2+-dependent binding site on repeats C8-C10 at the carboxyl-terminal end of Cl-II. Next, we studied the RAP-mediated inhibition of ligand binding to LRP and to Cl-II. As expected, we observed a strong inhibition of t-PA.PAI-1 complex and Fab A8 binding to LRP by RAP (IC50 congruent with 0.3 nM), whereas in the reverse experiment, competition of t-PA. PAI-1 complexes and Fab A8 for RAP binding to LRP could only be shown at high concentrations of competitors (>/=1 microM). Interestingly, even though the equilibrium dissociation constants for the binding of RAP to LRP and to Cl-II are similar, the binding of the ligands to Cl-II is only prevented by RAP at concentrations that are at least 2 orders of magnitude higher than those required for inhibition of ligand binding to LRP. Our results favor models that propose RAP-induced allosteric inhibition of ligand binding to LRP that may require LRP moieties that are located outside Cl-II of the receptor.     

Protein footprinting approach to mapping DNA binding sites of two archaeal homing enzymes: evidence for a two-domain protein structure

The archaeal intron-encoded homing enzymes I-PorI and I-DmoI belong to a family of endonucleases that contain two copies of a characteristic LAGLIDADG motif. These endonucleases cleave their intron- or intein- alleles site-specifically, and thereby facilitate homing of the introns or inteins which encode them. The protein structure and the mechanism of DNA recognition of these homing enzymes is largely unknown. Therefore, we examined these properties of I-PorI and I-DmoI by protein footprinting. Both proteins were susceptible to proteolytic cleavage within regions that are equidistant from each of the two LAGLIDADG motifs. When complexed with their DNA substrates, a characteristic subset of the exposed sites, located in regions immediately after and 40-60 amino acids after each of the LAGLIDADG motifs, were protected. Our data suggest that the enzymes are structured into two, tandemly repeated, domains, each containing both the LAGLIDADG motif and two putative DNA binding regions. The latter contains a potentially novel DNA binding motif conserved in archaeal homing enzymes. The results are consistent with a model where the LAGLIDADG endonucleases bind to their non-palindromic substrates as monomeric enzymes, with each of the two domains recognizing one half of the DNA substrate.     

The association rate constant for heme binding to globin is independent of protein structure

Rate constants for CO-heme binding to 35 different recombinant apomyoglobins and several other apoproteins were measured in an effort to understand the factors governing heme affinity and the velocity of the association reaction. Surprisingly, the rate constant for the binding of monomeric heme is approximately 1 x 10(8) M-1 s-1 regardless of the structure or overall affinity of the apoprotein for iron-porphyrin. Major differences between the proteins are reflected primarily in the rates of dissociation of the prosthetic group. Slow phases observed in the reaction of CO heme with excess apomyoglobin result from formation of nonspecific heme-protein complexes which must dissociate before heme can bind specifically in the heme pocket. Once the specific heme-globin complex is formed, the heme pocket rapidly collapses around the porphyrin, simultaneously forming the bond between the proximal His93 and the heme iron atom. The overall affinity of sperm whale apomyoglobin for hemin is approximately 1 x 10(14) M-1. Nonspecific hydrophobic interactions between the porphyrin and the apolar heme cavity account for a factor of 10(5)-10(7). Covalent bond formation between Fe3+ and His93(F8) provides an additional factor of 10(3)-10(4). Specific interactions with conserved amino acids in the heme pocket contribute the final factor of 10(3)-10(4).     

A new model of dual interacting ligand binding sites on integrin alphaIIbbeta3

The platelet integrin alphaIIbbeta3 mediates platelet aggregation and platelet adhesion. This integrin is the key to hemostasis and also to pathologic vascular occlusion. A key domain on alphaIIbbeta3 is the ligand binding site, which can bind to plasma fibrinogen and to a number of Arg-Gly-Asp (RGD)-type ligands. However, the nature and function of the ligand binding pocket on alphaIIbbeta3 remains controversial. Some studies suggest the presence of two ligand binding pockets, whereas other reports indicate a single binding pocket. Here we use surface plasmon resonance to show that alphaIIbbeta3 contains two distinct ligand binding pockets. One site binds to fibrinogen, and a separate site binds to RGD-type ligands. More importantly, however, the two ligand binding pockets are interactive. RGD-type ligands are capable of binding to alphaIIbbeta3 even when it is already occupied by fibrinogen. Once bound, RGD-type ligands induce the dissociation of fibrinogen from alphaIIbbeta3. This allosteric cross-talk has important implications for anti-platelet therapy because it suggests a novel approach for the dissolution of existing platelet thrombi.     

Differences in the binding sites of two site-3 sodium channel toxins

Site-3 toxins from scorpion and sea anemone bind to Na channels and selectively inhibit current decay. Anthopleurins A and B (ApA and ApB, respectively), toxins found in the venom of the sea anemone Anthopleura xanthogrammica, bind to closed states of mammalian skeletal and cardiac Na channels with differing affinities which arise from differences in first-order toxin/channel dissociation rate constants, koff. Using chimera comprising domain interchanges between channel isoforms, we examined the structural basis of this differential affinity. Toxin/channel association rates, kon, were similar for both toxins and both parental channels. Domain 4 determined koff for ApA, while ApB dissociated from all tested chimera in a cardiac-like manner. To probe this surprising difference between two such closely related toxins, we examined the interaction of chimeric channels with a form of ApB in which the two nonconserved basic residues, Arg-12 and Lys-49, were converted to the corresponding neutral amino acids from ApA. In the chimera comprising domain 1 from the cardiac muscle isoform and domains 2-4 from the skeletal muscle isoform, toxin dissociated at a rate intermediate between those of the parental channels. We conclude that the differential component of ApA binding is controlled by domain 4 and that some component of ApB binding is not shared by ApA. This additional component probably binds to an interface between channel domains and is partly mediated by toxin residues Arg-12 and Lys-49.     

Determining the number of essential ligand binding sites in enzymes using the slopes of the Wang-Srivastava plots

Systemic necrotizing vasculitis is uncommon in children and may be rarely associated with gangrene. We describe a 3-yr-old girl with parvovirus B19-induced necrotizing vasculitis whose digital gangrene was successfully treated with iloprost, a prostacyclin analogue.     

Identification of ligand binding sites on integrin alpha4beta1 through chemical cross-linking

We have used chemical cross-linking to identify sequences in integrin alpha4beta1 that are involved in its interactions with ligands. A recently described leucine-aspartic acid-valine (LDV)-based small molecule inhibitor of alpha4beta1 (BIO-1494), that contained a single reactive amino group for targeting the cross-linking, was used for these studies. The specificity of the interaction was defined by (i) the ability to block the interaction with a competitive inhibitor lacking the reactive group, (ii) the absolute requirement of divalent cations for cross-linking, and (iii) the lack of cross-linking to the functionally related integrin alpha4beta7. With ANB-NOS as the cross-linker, only the beta1 chain was labeled with BIO-1494, while with the more flexible cross-linker DSS both the alpha4 and beta1 chains were modified. Similar results were obtained when cross-linking was performed on K562 cells expressing alpha4beta1 but not on K562 cells expressing alpha2beta1. The site of cross-linking on the beta1 chain was localized by CNBr peptide mapping within residues 130-146, a region that contains the putative metal binding site DXSXS and for which analogous data had been generated with RGD binding to integrin alphaIIbbeta3. The striking similarity between the data we generated for an LDV ligand and published data for the RGD family supports the notion of a common ligand binding pocket formed by both integrin chains. The cross-linking strategy developed here should serve as a useful tool for studying alpha4beta1 function.     

Determination of protein binding of gyrase inhibitors by means of continuous ultrafiltration

In order to characterize the protein binding of a drug, it is necessary to have a method which is close to in vivo conditions and fast in the course of measurement. The continuous ultrafiltration fulfils both requirements for substances with a high extent of protein binding. In this study, 18 gyrase inhibitors in clinical practice, characterized by a lower extent of protein binding, were subjected to the titration procedure of the continuous ultrafiltration using bovine and human serum albumin (BSA, HSA), and human plasma. The results of the continuous ultrafiltration were found to be similar to those obtained by means of the 'classical' discontinuous ultrafiltration using plasma (correlation between continuous and discontinuous ultrafiltration r2 = 0.87). In the cases of pipemidic acid, enoxacin and rufloxacin, the continuous method gave approximately 20% lower degrees of protein binding than the discontinuous procedure, which utilizes plasma having the full range of proteins. It is likely that these drugs bind mainly to other proteins in plasma than HSA. This finding proves that this fast method is worthwhile in the whole range of protein binding.     

Mutations in the carboxyl terminus of the agouti protein decrease agouti inhibition of ligand binding to the melanocortin receptors

Several mutations that cause ectopic expression of the agouti gene result in obesity, hyperinsulinemia, and yellow coat color. A candidate pathway for agouti induced obesity and hyperinsulinemia is through altered signaling by melanocortin receptors, as agouti normally regulates coat coloration through antagonism of melanocortin receptor 1. Furthermore, melanocortin peptides mediate functions including steroidogenesis, lipolysis, and thermoregulation. We report apparent inhibition dissociation constants for mouse and human agouti protein inhibition of ligand binding to the melanocortin receptors, to determine which of these receptors might be involved in agouti induced diabetes. The similarity in the apparent K(I) values for agouti inhibition of ligand binding to the brain melanocortin receptors 3 and 4 (mouse: K(I) app = 190 +/- 74 and 54 +/- 18 nM; human: K(I) app = 140 +/- 56 and 70 +/- 18 nM, respectively) suggests that the MC3-R is a potential candidate for a receptor mediating the effects of agouti protein overexpression. Agouti residues important for melanocortin receptor inhibition were identified through the analysis of deletion constructs and site-specific variants. Val83 is important for inhibition of binding to MC1-R (K(I) app for Val83Ala agouti increased 13-fold relative to wild-type protein). Arg85, Pro86, and Pro89 are important for selective inhibition of binding between MC1-R and MC3-R and MC4-R as their apparent K(I) values are essentially unchanged at MC1-R, while they have increased 6-10-fold relative to wild-type protein at MC3-R and MC4-R.