Determination of the functional epitopes of human interleukin-18-binding protein by site-directed mutagenesis

The human interleukin (IL)-18-binding protein (hIL-18BP) is a naturally occurring antagonist of IL-18, a proinflammatory cytokine that is related to IL-1beta and has an important role in defense against microbial invaders. As its name implies, the hIL-18BP binds to IL-18 with high affinity and prevents the interaction of IL-18 with its receptor. We genetically modified the C terminus of hIL-18BP by appending a 15-amino acid biotinylation recognition site and a six-histidine tag and then performed site-directed mutagenesis to determine the functional epitopes that mediate efficient binding to IL-18. The mutated IL-18BPs were secreted from mammalian cells, captured by metal affinity chromatography, biotinylated in situ, eluted, and immobilized on streptavidin-coated chips. Using surface plasmon resonance, we identified seven amino acids of hIL-18BP which, when changed individually to alanine, caused an 8-750-fold decrease in binding affinity, largely because of increased off-rates. These seven amino acids localized to the predicted beta-strand c and d of hIL-18BP immunoglobulin-like domain, and most had hydrophobic side chains. Just two amino acids, tyrosine 97 and phenylalanine 104, contributed approximately 50% of the binding free energy. Information obtained from these studies could contribute to the design of molecular antagonists of IL-18 for treatment of inflammatory diseases.     

Yaba monkey tumor virus encodes a functional inhibitor of interleukin-18

Interleukin-18 (IL-18) is a critical proinflammatory cytokine whose extracellular bioactivity is regulated by a cellular IL-18 binding protein (IL-18BP). Many poxviruses have acquired variants of this IL-18BP gene, some of which have been shown to act as viral virulence factors. Yaba monkey tumor virus (YMTV) encodes a related family member, 14L, which is similar to the orthopoxvirus IL-18BPs. YMTV 14L was expressed from a baculovirus system and tested for its ability to bind and inhibit IL-18. We found that YMTV 14L bound both human IL-18 (hIL-18) and murine IL-18 with high affinity, at 4.1 nM and 6.5 nM, respectively. YMTV 14L was able to fully sequester hIL-18 but could only partially inhibit the biological activity of hIL-18 as measured by gamma interferon secretion from KG-1 cells. Additionally, 17 hIL-18 point mutants were tested by surface plasmon resonance for their ability to bind to YMTV 14L. Two clusters of hIL-18 surface residues were found to be important for the hIL-18-YMTV 14L interaction, in contrast to results for the Variola virus IL-18BP, which has been shown to primarily interact with a single cluster of three amino acids. The altered binding specificity of YMTV 14L most likely represents an adaptation resulting in increased fitness of the virus and affirms the plasticity of poxviral inhibitor domains that target cytokines like IL-18.     

Molluscum contagiosum virus interleukin-18 (IL-18) binding protein is secreted as a full-length form that binds cell surface glycosaminoglycans through the C-terminal tail and a furin-cleaved form with only the IL-18 binding domain

Some poxviruses and their mammalian hosts encode homologous proteins that bind interleukin-18 (IL-18) with high affinity and inhibit IL-18-mediated immune responses. MC54L, the IL-18 binding protein of the human poxvirus that causes molluscum contagiosum, is unique in having a C-terminal tail of nearly 100 amino acids that is dispensable for IL-18 binding. When recombinant MC54L was expressed and purified via a C-terminal six-histidine tag, a shorter fragment was detected in addition to the full-length protein. This C-terminal fragment resulted from the cleavage of MC54L by cellular furin, as it was greatly diminished when furin was specifically inhibited or when a furin-deficient cell line was used for expression. Furthermore, the N- and C-terminal fragments of MC54L were generated by cleavage of the recombinant protein with furin in vitro. The furin cleavage site was mapped within a 32-amino-acid segment that is C terminal to the IL-18 binding domain. Full-length MC54L, but not the N-terminal IL-18 binding fragment, bound to cells and to purified heparin and other glycosaminoglycans that are commonly found on the cell surface and in the extracellular matrix. MC54L bound to heparin with a nanomolar K(d) and could simultaneously bind to IL-18. Their different glycosaminoglycan and cell binding properties may allow the long and short forms of MC54L to inactivate IL-18 near the site of infection and at more distal locations, respectively.     

Identification of a subdomain in the Moloney murine leukemia virus envelope protein involved in receptor binding.

We have mutated amino acids within the receptor-binding domain of Moloney murine leukemia virus envelope in order to identify residues involved in receptor binding. Analysis of mutations in the region of amino acids 81 to 88 indicates that this region is important for specific envelope-receptor interactions. None of the aspartate 84 (D-84) mutants studied bind measurably, although they are efficiently incorporated into particles. D-84 mutants have titers that correspond to the severity of the substitution. This observation suggests that D-84 may provide a direct receptor contact. Mutations in the other charged amino acids in this domain (R-83, E-86, and E-87) yield titers similar to those of wild-type envelope, but the affinity of the mutant envelope in the binding assay is decreased by nonconservative substitutions in parallel to the severity of the change. These other amino acids may either provide secondary receptor contacts or assist in maintaining a structure in the domain that favors efficient binding. We also studied other regions of high hydrophilicity. Our initial characterization indicates that amino acids 106 to 111 and 170 to 188 do not play a major role in receptor binding. Measurements of relative binding affinity and titer indicate that most mutations in the region of amino acids 120 to 131 did not significantly affect receptor binding. However, SU encoded by mutants H123V, R124L, and C131A as well as C81A could not be detected in particles and therefore did not bind measurably. Therefore, the region encompassed by amino acids 81 to 88 appears to be directly involved in receptor binding.     

Molecular characterization of human melanocortin-3 receptor ligand-receptor interaction

Melanocortin-3 receptor (MC3R), primarily expressed in the hypothalamus, plays an important role in the regulation of energy homeostasis. MC3R-deficient (MC3R(-)(/)(-)) mice demonstrate increased fat mass, higher feeding efficiency, hyperleptinaemia, and mild hyperinsulinism. At least one specific mutation of MC3R has been identified to be associated with human obesity. Functional analysis of this altered MC3R (I183N) has indicated that the mutation completely abolishes agonist-mediated receptor activation. However, the specific molecular determinants of MC3R responsible for ligand binding and receptor signaling are currently unknown. The present study is to determine the structural aspects of MC3R responsible for ligand binding and receptor signaling. On the basis of our theoretical model for MC1R, using mutagenesis, we have examined 19 transmembrane domain amino acids selected for these potential roles in ligand binding and receptor signaling. Our results indicate that (i) substitutions of charged amino acid residues E131 in transmembrane domain 2 (TM2), D154 and D158 in TM3, and H298 in TM6 with alanine dramatically reduced NDP-MSH binding affinity and receptor signaling, (ii) substitutions of aromatic amino acids F295 and F296 in TM6 with alanine also significantly decreased NDP-MSH binding and receptor activity, (iii) substitutions of D121in TM2 and D332 in TM7 with alanine resulted in the complete loss of ligand binding, ligand induced receptor activation, and cell surface protein expression, and (iv) interestingly, substitution of L165 in TM3 with methionine or alanine switched antagonist SHU9119 into a receptor agonist. In conclusion: Our results suggest that TM3 and TM6 are important for NDP-MSH binding, while D121 in TM2 and D332 in TM7 are crucial for receptor activity and signaling. Importantly, L165 in TM3 is critical for agonist or antagonist selectivity. These results provide important information about the molecular determinants of hMC3R responsible for ligand binding and receptor signaling.     

Functional studies of the bacterial avirulence protein AvrPto by mutational analysis

Pseudomonas syringae pathovars expressing avrPto are avirulent on plants expressing the resistance gene Pto. Over 85 mutants of avrPto were generated with multiple strategies, and several assays were used to characterize AvrPto function. Only a core of 95 amino acids of the 164 residues was sufficient for binding Pto in the yeast two-hybrid system. Only nine of 65 mutant proteins of AvrPto with amino acid substitutions, created in planta and in vitro, did not interact with Pto in the Gal4 yeast two-hybrid system, suggesting that AvrPto can tolerate many nonconservative substitutions and still interact with Pto. These nine and 12 additional substitution mutants of AvrPto were characterized further. The ability to elicit a hypersensitive response and the effect on pathogenesis in planta for these 21 mutants of AvrPto were strongly correlated with recognition by Pto in the yeast two-hybrid system. Analyses of two proteins with substitutions H54P or D52G/L65P indicated that these residues may be required for delivery into the host cell and protein stability in the bacterial cytoplasm, respectively. The mutants that no longer interacted with Pto and had modified activities in planta were predicted to have changes in their secondary structure.     

Mutational analysis of a blood coagulation factor VIII-binding peptide.

A peptide screened from a combinatorial peptide library with the sequence EYKSWEYC performed best as a ligand for affinity chromatography of human blood coagulation factor VIII (FVIII). With this peptide immobilized on monolithic CIM columns via epoxy groups we were able to capture FVIII from diluted plasma. Rational substitution of amino acids by spot synthesis revealed that lysine and cysteine can be exchanged for almost all other proteinogenic amino acids without loss of affinity to FVIII. This offers the possibility of site-specific attachment via either one of these residues or the N- or C-terminus. The aliphatic positions O5 (tryptophan) and O7 (tyrosine), together with the charged position O6 (glutamic acid), seem to form the core of the binding unit. In the positions with aliphatic amino acids, substitution by tyrosine or phenylalanine, and in the positions with charged amino acids, substitution by aspartic acid or lysine, preserved the affinity to FVIII. The functionality of the selected peptides was confirmed by affinity chromatography. Selective binding and elution could be achieved.     

Identification of the LIV-I/LS system as the third phenylalanine transporter in Escherichia coli K-12

In Escherichia coli, the active transport of phenylalanine is considered to be performed by two different systems, AroP and PheP. However, a low level of accumulation of phenylalanine was observed in an aromatic amino acid transporter-deficient E. coli strain (DeltaaroP DeltapheP Deltamtr Deltatna DeltatyrP). The uptake of phenylalanine by this strain was significantly inhibited in the presence of branched-chain amino acids. Genetic analysis and transport studies revealed that the LIV-I/LS system, which is a branched-chain amino acid transporter consisting of two periplasmic binding proteins, the LIV-binding protein (LIV-I system) and LS-binding protein (LS system), and membrane components, LivHMGF, is involved in phenylalanine accumulation in E. coli cells. The K(m) values for phenylalanine in the LIV-I and LS systems were determined to be 19 and 30 micro M, respectively. Competitive inhibition of phenylalanine uptake by isoleucine, leucine, and valine was observed for the LIV-I system and, surprisingly, also for the LS system, which has been assumed to be leucine specific on the basis of the results of binding studies with the purified LS-binding protein. We found that the LS system is capable of transporting isoleucine and valine with affinity comparable to that for leucine and that the LIV-I system is able to transport tyrosine with affinity lower than that seen with other substrates. The physiological importance of the LIV-I/LS system for phenylalanine accumulation was revealed in the growth of phenylalanine-auxotrophic E. coli strains under various conditions.     

A simple novel method for the preparation of noncovalent homodimeric, biologically active human interleukin 10 in Escherichia coli-enhancing protein expression by degenerate PCR of 5' DNA in the open reading frame

DNA inserts encoding human interleukin 10 (hIL-10), optimized for codon usage and secondary RNA structure, were purchased from several commercial sources and subcloned into a pMon vector. Despite the optimization, protein expression was nil. We therefore subjected the 5′ segment of the cDNA encoding N-terminal amino acids 2–11 to degenerate PCR in order to create a small library of 130 K theoretical cDNA combinations that would not change the respective amino acid sequence and tested their expression. After screening over 320 colonies 10 hIL-10 clones encoding the original amino acid sequence were identified. Three nucleotide substitutions were sufficient to ensure reasonable protein expression. Subsequently, hIL-10 was expressed in Escherichia coli, refolded and purified to homogeneity, yielding over 95% electrophoretically pure noncovalent homodimeric protein, which was biologically active in MC/9 cells. The yield of recombinant hIL-10 from 10 L of fermentation culture was 60 mg and a protocol for its long-term storage as a carrier-free lyophilized powder at −20° was developed.     

Site specific mutants of beta-galactosidase show that Tyr-503 is unimportant in Mg2+ binding but that Glu-461 is very important and may be a ligand to Mg2+

The Mg2+ concentrations required for half maximal activity, the dissociation constants, and the free energies of binding for Mg2+ bound to wild type beta-galactosidase and several site specific mutants are reported. The mutants have one of the following substitutions: Glu-461 substituted with Asp, Gln, Gly, His, or Lys; or Tyr-503 substituted with Phe, His or Cys. Substitutions for Tyr-503 had little effect on the affinity of the enzyme for Mg2+, implying that Tyr-503 is not involved in Mg2+ binding. Neutrally charged amino acids substituted for the negatively charged Glu-461 significantly decreased the affinity of the enzyme for Mg2+ and substitution of positively charged amino acids at this position further decreased the affinity. On the other hand, substitution by Asp (negative charge) at position 461 had no effect on the binding. Thus, the negatively charged side chain of Glu-461 is important for divalent cation binding to beta-galactosidase.     

Characterization of mutations in oligomerization domain of Lac repressor protein

A series of mutant lac repressor proteins at positions 281 or 282 was isolated for detailed characterization. Although Cys281 modification by sulfhydryl reagents abrogates pH effects on inducer binding and diminishes operator binding (Daly, T. J., Olson, J. S., and Matthews, K. S. (1986) Biochemistry 25, 5468-5474), substitution at this site by alanine, serine, phenylalanine, isoleucine, or methionine did not abolish completely the pH shift nor affect operator affinity. Thus, ionization of the sulfhydryl residue does not account fully for the alterations in inducer affinity and cooperativity of binding observed with elevated pH. Substitution for Cys281 did, however, alter the kinetic parameters for inducer association with the protein. The polarity of the side chain at 281 influenced the rates of sugar binding, presumably by altering the rate of opening/closing of the binding site. Furthermore, the presence of the branched side chain of isoleucine at position 281 disrupted oligomerization of the repressor. In contrast to the tolerance for substitution at 281, the only amino acid side chain exchanges for Tyr282 which yielded tetrameric protein with near normal operator binding characteristics were phenylalanine and leucine; this result is consistent with studies of suppressed nonsense mutations at position 282 which indicated repression occurred only for the corresponding substitutions (Kleina, L. G., and Miller, J. H. (1990) J. Mol. Biol. 221, 295-318). Despite the tetrameric character of the Y282F mutant protein, the pH dependence and cooperativity of inducer binding for this mutant protein were altered. All amino acid substitutions other than phenylalanine and leucine at this position resulted in either monomeric protein or no detectable repressor in the cell. Thus, the hydrophobic character of the side chain at position 282 is essential for tetramer formation, and the phenyl ring alone alters inducer binding parameters. The monomeric mutant proteins with substitutions for Tyr282 exhibited lower stability than their tetrameric counterparts, and the absence of dimer formation suggests alterations at this site affect both dimer and tetramer interfaces. Based on previous genetic studies and our detailed mutant characterization, the region encompassing 281 and 282, indicated by secondary structure prediction to be a turn or coil, is essential for oligomer formation and additionally exerts a strong influence on the dynamic properties of the protein, presumably mediated by interactions at the subunit interface which regulate the rate of opening and closing of the inducer binding cleft.     

Molecular evolution of a long wavelength-sensitive opsin in mimetic Heliconius butterflies (Lepidoptera: Nymphalidae)

This study examines the pattern of opsin nucleotide and amino acid substitution among mimetic species 'rings' of Heliconius butterflies that are characterized by divergent wing colour patterns. A long wavelength opsin gene, OPS1 , was sequenced from each of seven species of Heliconius and one species of Dryas (Lepidoptera: Nymphalidae). A parsimony analysis of OPS1 nucleotide and amino acid sequences resulted in a phylogeny that was consistent with that presented by Brower & Egan in 1997, which was based on mitochondrial cytochrome oxidase I and II as well as nuclear wingless genes. Nodes in the OPS1 phylogeny were well supported by bootstrap analysis and decay indices. An analysis of specific sites within the gene indicates that the accumulation of amino acid substitutions has occurred independently of the morphological diversification of Heliconius wing colour patterns. Amino acid substitutions were examined with respect to their location within the opsin protein and their possible interactions with the chromophore and the G-protein. Of the 15 amino acid substitutions identified among the eight species, one nonconservative replacement (A226Q) was identified in a position that may be involved in binding with the G-protein.     

Kinetic analysis of phenylalanine dehydrogenase mutants designed for aliphatic amino acid dehydrogenase activity with guidance from homology-based modelling.

Through comparison with the high-resolution structure of Clostridium symbiosum glutamate dehydrogenase, the different substrate specificities of the homologous enzymes phenylalanine dehydrogenase and leucine dehydrogenase were attributed to two residues, glycine 124 and leucine 307, in Bacillus sphaericus phenylalanine dehydrogenase, which are replaced with alanine and valine in leucine dehydrogenases. As predicted, making these substitutions in phenylalanine dehydrogenase decreased the specific activity towards aromatic substrates and enhanced the activity towards some aliphatic amino acids in standard assays with fixed concentrations of both substrates. This study did not, however, distinguish effects on affinity from those on maximum catalytic rate. A fuller kinetic characterization of the single- and double-mutant enzymes now reveals that the extent of the shift in specificity was underestimated in the earlier study. The maximum catalytic rates for aromatic substrates are reduced for all the mutants, but, in addition, the apparent Km values are higher for the single-mutant G124A and double-mutant G124A/L307V compared with the wild-type enzyme. Conversely, specificity constants (kcat/Km) for the nonpolar aliphatic amino acids and the corresponding 2-oxoacids for the mutants are all markedly higher than for the wild type, with up to a 40-fold increase for l-norvaline and a 100-fold increase for its 2-oxoacid in the double mutant. In some cases a favourable change in Km was found to outweigh a smaller negative change in kcat. These results emphasize the risk of misjudging the outcome of protein engineering experiments through too superficial an analysis. Overall, however, the success of the predictions from molecular modelling indicates the usefulness of this strategy for engineering new specificities, even in advance of more detailed 3D structural information.     

Expression of human melanocortin 4 receptor in <Emphasis Type="Italic">Saccharomyces cerevisiae</Emphasis>

The melanocortin 4 receptor (MC4R) is involved in the regulation of energy homeostasis and is known as one of the major hypothalamic regulators of food intake. Several studies have shown that replacement of aspartic acid at position 126 of the MC4R abolishes the ligand binding. We used the modified yeast Saccharomyces cerevisiae strain MMY28 to functionally express the MC4R and characterise the importance of this amino acid for ligand based activation of the receptor. The efficiency of the functional expression system was estimated by activation with αMSH, ACTH and THIQ and compared with cAMP response in mammalian cells. We generated the library of MC4R mutants randomised at the amino acid position 126. Recombinant MC4R clones were screened for the αMSH induced activity in yeast. From 9 different amino acids obtained only the natural aspartic acid displayed the ligand dependent activity of MC4R. The MC4R variants with glutamic acid and leucine at position 126, however, displayed higher background activity than other amino acid substitutions. The results suggest that the yeast expression system is suitable for screening of the MC4R receptor ligands and that the substitution of aspartic acid at position 126 of MC4R by different amino acids functionally inactivates the receptor.     

V3 loop of the human immunodeficiency virus type 1 Env protein: interpreting sequence variability.

Two different states of human immunodeficiency virus type 1 are apparent in the asymptomatic and late stages of infection. Important determinants associated with these two states have been found within the V3 loop of the viral Env protein. In this study, two large data sets of published V3 sequences were analyzed to identify patterns of sequence variability that would correspond to these two states of the virus. We were especially interested in the pattern of basic amino acid substitutions, since the presence of basic amino acids in V3 has been shown to change virus tropism in cell culture. Four features of the sequence heterogeneity in V3 were observed: (i) approximately 70% of all nonconservative basic substitutions occur at four positions in V3, and V3 sequences with a basic substitution in at least one of these four positions contain approximately 95% of all nonconservative basic substitutions; (ii) substitution patterns within V3 are influenced by the identity of the amino acid at position 25; (iii) sequence polymorphisms account for a significant fraction of uncharged amino acid substitutions at several positions in V3, and sequence heterogeneity other than these polymorphisms is most significant at two positions near the tip of V3; and (iv) sequence heterogeneity in V3 (in addition to the basic amino acid substitutions) is approximately twofold greater in V3 sequences that contain basic amino acid substitutions. By using this sequence analysis, we were able to identify distinct groups of V3 sequences in infected patients that appear to correspond to these two virus states. The identification of these discrete sequence patterns in vivo demonstrates how the V3 sequence can be used as a genetic marker for studying the two states of human immunodeficiency virus type 1.     

Complete replacement set of amino acids at the C-terminus of thymidylate synthase: quantitative structure-activity relationship of mutants of an enzyme.

The C-terminal residue of thymidylate synthase (TS) is highly conserved and has been implicated in cofactor binding, catalysis, and a conformational change. The codon for the C-terminal valine of Lactobacillus casei TS has been replaced with those for 19 other amino acids and the amber stop codon. Fourteen of the resulting mutant proteins were active by genetic complementation using a Thy- strain of Escherichia coli, and 18 mutants were active by in vitro assay. Only the aspartate and amber mutations had undetectable activity. All of the mutants were expressed at high levels (5-30% of soluble protein) and were purified by phosphocellulose chromatography. In general, the alterations at position 316 led to little effect on the Km for dUMP, an increase in Km for the folate cofactor, and a decrease in kcat. The observations show that TS can tolerate the substitution of most amino acids for valine at the C-terminus without a complete loss of activity, that hydrophobic substitutions are preferred, and that the C-terminal side chain is involved in both cofactor binding and catalysis. There was an excellent correlation between log kcat and hydrophobicity of the side chain at position 316 and an inverse correlation between log Km and the hydrophobicity of this residue. Kinetic parameters of the cofactor-independent TS-catalyzed dehalogenation of BrdUMP showed no variation with the side chain at position 316. In context of the structure of TS, it is proposed that binding of the cofactor triggers a conformational change in which the C-terminal side chain undergoes hydrophobic interactions that stabilize a rate-limiting transition state of the TS reaction.     

The structure of a doripenem‐bound OXA‐51 class D β‐lactamase variant with enhanced carbapenemase activity

OXA‐51 is a class D β‐lactamase that is thought to be the native carbapenemase of Acinetobacter baumannii. Many variants of OXA‐51 containing active site substitutions have been identified from A. baumannii isolates, and some of these substitutions increase hydrolytic activity toward carbapenem antibiotics. We have determined the high‐resolution structures of apo OXA‐51 and OXA‐51 with one such substitution (I129L) with the carbapenem doripenem trapped in the active site as an acyl‐intermediate. The structure shows that acyl‐doripenem adopts an orientation very similar to carbapenem ligands observed in the active site of OXA‐24/40 (doripenem) and OXA‐23 (meropenem). In the OXA‐51 variant/doripenem complex, the indole ring of W222 is oriented away from the doripenem binding site, thereby eliminating a clash that is predicted to occur in wildtype OXA‐51. Similarly, in the OXA‐51 variant complex, L129 adopts a different rotamer compared to I129 in wildtype OXA‐51. This alternative position moves its side chain away from the hydroxyethyl moiety of doripenem and relieves another potential clash between the enzyme and carbapenem substrates. Molecular dynamics simulations of OXA‐51 and OXA‐51 I129L demonstrate that compared to isoleucine, a leucine at this position greatly favors a rotamer that accommodates the ligand. These results provide a molecular justification for how this substitution generates enhanced binding affinity for carbapenems, and therefore helps explain the prevalence of this substitution in clinical OXA‐51 variants.     

Substitutions for glycine alpha 1-637 and glycine alpha 2-694 of type I procollagen in lethal osteogenesis imperfecta. The conformational strain on the triple helix introduced by a glycine substitution can be transmitted along the helix

Two substitutions for glycine in the triple-helical domain were found in type I procollagen synthesized by skin fibroblasts from two probands with lethal osteogenesis imperfecta. One was a substitution of valine for glycine alpha 1-637, and the other was a substitution of arginine for glycine alpha 2-694. The effects of the mutations on the zipper-like folding of the collagen triple helix were similar, since there was post-translational overmodification of the collagenase A fragments (amino acids 1-775) but not of more COOH-terminal fragments of the protein. The mutations differed markedly, however, on their effects on thermal unfolding of the triple helix. The collagenase A fragment from the collagen containing the arginine alpha 2-694 substitution was cleaved at about amino acid 700 when incubated with trypsin at 30-35 degrees C. Therefore, there was micro-unfolding of the triple helix at a site close to the glycine substitution. Surprisingly, however, the collagenase A fragment with the valine alpha 1-637 substitution was also cleaved at about amino acid 700 under the same conditions. The results, therefore, demonstrated that although most glycine substitutions delay folding of the triple helix in regions that are NH2-terminal to the site of the substitution, the effects on unfolding can be transmitted to regions that are COOH-terminal to the site of the glycine substitution.