Search

We have computed the low energy minima for the two endomorphin peptides, N-acetyl-Tyr-Pro-Trp-Phe-NHCH 3 (endomorphin 1) and Tyr-Pro-Phe-Phe-NHCH 3 (endomorphin 2) in aqueous solution. These peptides block pain without inducing the harmful side effects of the opiates that bind to the same mu opiate receptor but have short half lives. From over 1000 starting conformations for each peptide, we find less than 200 low energy structures whose conformational energies were ≤ 5 kcal/mole of the energy of the global minimum. The most probable conformations calculated using the Boltzmann distribution for both peptides were similar to one another. Using the letter representation for backbone conformational states, these most probable structures were D A E E for endomorphin 1 and E A E E for endomorphin 2. Both of these structures form reverse turns at Pro 2-Trp (Phe) 3 resulting in the juxtaposition of the aromatic rings of Tyr 1 and Phe 4. The Trp residue of endomorphin 1 points to the back of the reverse turn. These features may be useful in the design of non-peptide analogues that will have longer half-lives than the peptides.

We have computed the low energy conformations of the negative regulatory domain of p53, residues 374-388 using Empirical Conformational Energies of Peptides Program including solvation and computed the statistical weights of distinct conformational states. We find that there are two high probability conformations, one an α-helix from Lys 374-Lys 381, followed by another helical structure involving Lys 382-Glu 388 (statistical weight of 0.48) and an all-α-helix for the entire sequence (statistical weight of 0.23). Both structure are superimposable on the NMR structure of this sequence bound to the S100 protein. The global minimum structure (statistical weight of 0.014) is a beta structure from Gly 374-Arg 379 followed by an α-helix from His 380-Glu 388. Based on these results, we propose a possible strategy for enhancement of p53 anti-tumor activity in cancer cells. Since the structure of this sequence bound to the sirtuin protein, Sir2, is a β-sheet, we further propose that the global minimum may be an intermediate on the α-β structure transition.

Using the chain-build-up method based on Empirical Conformational Energies of Peptides Program including solvation, we have computed, the low energy conformations of gonadotrpin-releasing hormone, GnRH, whose sequence is Pyro-Glu(PG)-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2. We have found 5,077 solvated conformations with conformational energies that were within 5 kcal/mole of that of the global minimum. These minima were found to occur in 802 distinct conformational classes of which 25 represented 70 % of the Boltzmann energy-weighted structures. Virtually all of these structures adopted bend conformations from Tyr 5-Leu 8, and 3,861 structures adopted bend conformations at residues 4-7. However, these structures differed significantly from one another, indicating that GnRH does not adopt a well-defined structure in aqueous solution consistent with the absence of a well-defined NMR structure of GnRH in water. A total of 300 of these structures were found to be superimposable on possible NMR structures for GnRH in DMSO with a combined statistical weight of 1.6 %. We found that Gly 6 adopts low energy "starred" states, e.g., C* and D*, that are energetically forbidden to L-amino acids but are low energy for D-amino acids, with a statistical weight of 43 %. This can explain why substitutions of L-amino acids for Gly 6 are known to inactivate GnRH while D-amino acid substitutions enhance its activity. Using these findings, in the accompanying manuscript, we compute the low energy conformations for the substituted GnRHs that enable inference of possible receptor-bound conformations.

In the preceding paper, using ECEPP, including the effects of water, and the chain build-up procedure, we computed the low energy structures for GnRH and found that there were no distinct low energy structures or structures with high statistical weights. To attempt to deduce possible structures of GnRH that may bind to the GnRH receptor, we computed the low energy structures for GnRH peptides that have L- and D-amino acids substituting for Gly 6. The L-amino acid-substituted peptides (L-Ala and L-Val) have very low or no affinity for the receptor and on activity (release of FSH and LH) while the D-Ala-, D-Leu-, D-Trp- and D-Phe-substituted peptides have significantly higher relative affinities and activities than those for native GnRH; the D-Val-substituted peptide has about one-third of the affinity and activity as native GnRH. Unlike native GnRH, our computations suggest that both sets of peptides form well-defined structures in water: the L-amino acid-substituted peptides are predominantly α-helical while the D-amino acid-substituted peptides adopted E*A A A E D*(C*) A E C A(C*) and minor variants of these structures. By eliminating structures that lay in common to the D-Ala and L-Val peptides and further eliminating structures that differed between the D-Ala and D-Leu peptides, we reduced the number of possible distinct binding conformations to 254. Searching for structures among these 254 conformations that had relative statistical weights that paralleled their relative affinities, we found two candidate structures: D*E A A E C*A E C A and D*G A A E D*A E C G*, both of which have conformations for residues 3-9 that are similar to the computed most probable structures for the D-amino acid-substituted GnRH peptides in water.

Using the chain build-up procedure based on the program ECEPP, we have computed the lowest energy structures for two terminally blocked subsequences from the antigenic circumsporozoite protein of Plasmodium berghei, that is known to cause malaria in animals. The full antigenic sequence is an octapeptide proline-rich tandem repeat, (Pro-Pro-Pro-Pro-Asn-Pro-Asn-Asp)(2). We computed the structures for the first octapeptide plus one Pro from the second octapeptide, terminally blocked CH(3)CO-Pro-Pro-Pro-Pro-Asn-Pro-Asn-Asp-Pro-NHCH(3) as well as the first octpeptide with an additional three Pro residues from the adjoining unit, i.e., CH(3)CO-Pro-Pro-Pro-Pro-Asn-Pro-Asn-Asp-Pro-Pro-Pro-NHCH(3). We find that the first sequence adopts a number of different low energy structures, the most probable of which has a probability of occurrence of 56 %. Addition of two more Pro residues results in the adoption a single, unique lowest energy structure that has a probability of occurrence of over 95 % without solvation effects and 86 % when solvation effects are included in the calculations. We predict that this structure may be the one recognized as a major antigenic determinant.

In the initial stages of the oxidative folding of both bovine pancreatic ribonuclease A (RNase A) and a 58-72 fragment thereof from the fully reduced, denatured state, the 65-72 correctly paired disulfide bond forms in preponderance over the incorrectly paired 58-65 disulfide bond. Since both disulfide-bonded loops contain the same number of amino acid residues, the question arises as to whether the native pairing results from interactions within the 58-72 segment that lead to a nativelike structure even in its fully reduced form. To answer this question, the chain buildup procedure, based on ECEPP, including a solvation treatment, was used to generate the low-energy structures for the 58-72 RNase segment, beginning with residue 72 and building back to residue 58; in this fragment, all three Cys residues (at positions 58, 65, and 72) initially exist in the reduced (CysH) state. After the open-chain energy minima of the 65-72 peptide were generated, these conformations were allowed to form the 65-72 disulfide bond, and the energies of the resulting oxidized conformations were reminimized and rehydrated. The global minimum of the loop-closed 65-72 structure and many of the low-lying loop-closed minima could be superimposed on the energy-minimized X-ray structure for residues 65-72. The low-energy structures for the full open chain 58-72 peptide were then computed and were allowed to form disulfide bonds either between residues 65 and 72 (native) or between residues 58 and 65 (non-native), and their energies were reminimized and rehydrated in the loop-closed state. Although the overall fold of the 65-72 loop-closed global minimum was the same as for the energy-minimized X-ray structure of these residues, the overall rms deviation was 3.9 A because of local deviations among residues 58-64. In contrast, the 65-72 segment of the global minimum of the 58-72 fragment could be superimposed on the corresponding residues of the energy-minimized X-ray structure. The lowest-energy structure for the 58-65 non-native paired 58-72 sequence was 6 kcal/mol higher in energy than that for the 58-72 peptide with the 65-72 disulfide bond formed. These results suggest that the native pairing of the 65-72 peptide arises from energetic determinants (adoption of left-handed single-residue conformations by Gly 68, and side chain interactions involving Gln 69) contained within this peptide sequence.

Crystals of bovine trypsin were acylated at the reactive residue, serine 195, to form the transiently stable p-guanidinobenzoate. Hydrolysis of this species was triggered in the crystals by a jump in pH. The hydrolysis was monitored by three-dimensional Laue crystallography, resulting in three x-ray diffraction structures, all from the same crystal and each representing approximately 5 seconds of x-ray exposure. The structures were analyzed at a nominal resolution of 1.8 angstroms and were of sufficient quality to reproduce subtle features in the electron-density maps for each of the structures. Comparison of the structures before and after the pH jump reveals that a water molecule has positioned itself to attack the acyl group in the initial step of the hydrolysis of this transient intermediate.

Platelet-type von Willebrand disease (PT-vWD) is an autosomal dominant bleeding disorder in which patient platelets exhibit an abnormally increased binding of circulating von Willebrand factor (vWF). We have recently shown that this abnormality is associated with a point mutation resulting in substitution of Val for Gly 233 in platelet membrane glycoprotein Ib alpha (GPIb alpha), a major component of the platelet GPIb/IX receptor for vWF. To investigate the effect of this substitution on the three-dimensional structure of this region of the protein, we have generated the allowed (low energy) conformations of the region of the GPI alpha protein containing residues 228-238 (with 5 residues on either side of the critical residue 233) with Gly 233 (wild type) and Val 233 (PT-vWD) using the computer program ECEPP (Empirical Conformational Energies of Peptides Program). The wild-type sequence is Tyr-Val-Trp-Lys-Gln-Gly-Val-Asp-Val-Lys-Ala. We find that the Gly 233-containing peptide can exist in two low energy conformers. The lowest energy conformer is a structure containing a beta-turn at Gln 232-Gly 233 while the alternative conformation is an amphipathic helical structure. Only the amphipathic helical structure is allowed for the Val 233-containing peptide which contains a hydrophobic 'face' consisting of Val 229, Val 233 and Val 236 and another hydrophilic surface composed of such residues as Lys 231 and Asp 235. No such surfaces exist for the lowest energy bend conformer for the Gly 233-containing peptide, but do exist in the higher energy helical structure. The amphipathic surfaces in the 228-238 region of the Val 233-containing GPIb alpha protein may associate strongly with complementary surfaces during vWF binding to the GPIb/IX receptor complex and may help explain heightened association of vWF with this receptor in PT-vWD.

The binding and alkylation rate constants for the reaction of four bromoacetamido pyrimidine nucleosides and bromoacetamide with bovine pancreatic ribonuclease A (RNase A) have been determined as a function of temperature. The four nucleoside derivatives react exclusively or preferentially with the NE2 atom of histidine-12 and include 2'-bromoacetamido-2'-deoxyuridine, 2'-bromoacetamido-2'deoxyxylofuranosyluracil, 3'-bromoacetamido-3'-deoxythymidine and 3'-bromoacetamido-3'-deoxyarabinofuranosyluracil. Transition-state parameters, delta H++ and delta S++, reveal that nucleosides with "up" OH groups experience relative rate enhancements which have been attributed to contacts between these groups and the enzyme in the transition state (1). Variations in alkylation rates are explained in terms of different degrees of entropic destabilization (2) of the nucleosides in the enzyme.affinity label complex.

The carboxyl-terminal region of the ras oncogene-encoded p21 protein is critical to the protein's function, since membrane binding through the C-terminus is necessary for its cellular activity. X-ray crystal structures for truncated p21 proteins are available, but none of these include the C-terminal region of the protein (from residues 172-189). Using conformational energy analysis, we determined the preferred three-dimensional structures for this C-terminal octadecapeptide of the H-ras oncogene p21 protein and generated these structures onto the crystal structure of the remainder of the protein. The results indicate that, like other membrane-associated proteins, the membrane-binding C-terminus of p21 assumes a helical hairpin conformation. In several low-energy orientations, the C-terminal structure is in close proximity to other critical locales of p21. These include the central transforming region (around Gln 61) and the amino terminal transforming region (around Gly 12), indicating that extracellular signals can be transduced through the C-terminal helical hairpin to the effector regions of the protein. This finding is consistent with the results of recent genetic experiments.

The low-energy conformations for a series of peptides based on the sequence of the ras P21 protein from position 55 to position 67 have been computed using conformational energy analysis. These sequences differed at position 61 and contained Gln, Pro, Leu, Lys, and Arg at this position. P21 proteins with Gln, Glu, or Pro at this position do not cause cell transformation at normal levels of expression; proteins with substitutions of at least 14 other amino acids at this position (Leu, Lys, and Arg having been found in tumors in place of the normally occurring Gln-61) do cause malignant transformation of cells in culture. We find that the segments of residues 55-67 from the nontransforming proteins (Gln- or Pro-61) adopt a structure that is energetically unfavorable for the same segment with Leu, Lys, or Arg at position 61. The critical feature of this structure is an alpha-helix from residues 62 to 68. Residue 61 (Gln or Pro) adopts an extended conformation. On the other hand, the segment from transforming proteins can adopt two structures, one all alpha-helical from residue 61 to residue 68 and the other a less-regular, higher-energy structure. The segments from the normal protein can adopt the all alpha-helical structure, a finding that can explain the fact that elevated intracellular levels of the normal protein also cause cell transformation. The results of the calculations suggest that specific changes in the structure of this region can account for the oncogenic effect of the proteins in which substitutions occur.

The biologically active conformations of a series of four peptides [four cholecystokinin (CCK)-related peptides and enkephalin] in their interactions with gastrointestinal receptors have been deduced using conformational computational analysis. The two peptides that interact exclusively with peripheral-type CCK receptors are the heptapeptide COOH-terminal fragment from CCK (CCK-7) and the analogous sequence from cerulein (CER-7) in which threonine replaces the methionine proximal to the NH2 terminus. The two peptides that interact exclusively with the gastrin receptor in the stomach are the active COOH-terminal fragment of little gastrin and the COOH-terminal tetrapeptide sequence common to all of these peptides, CCK-4. We find that preferred conformations for the peripherally active peptides CCK-7 and CER-7 are principally beta-bends, whereas little gastrin and CCK-4 are fundamentally helical. In the class of lowest energy structures for both CCK-7 and CER-7, the aromatic rings of the tyrosine and phenylalanine lie close to one another whereas the tryptophan indole ring points in the opposite direction. This structure is superimposable on the structures of a set of rigid indolyl benzodiazepine derivatives that interact with complete specificity and high affinity with peripheral CCK receptors further suggesting that the computed beta-bends are the biologically active conformation. The biologically active conformation for CCK-4 and the little gastrin hexapeptide has also been deduced. By excluding conformations common to CCK-7 and CCK-4, which do not bond to each other's receptors, and then by selecting conformations in common to CCK-4 and the gastrin-related hexapeptide, which do bind to each other's receptors, we deduce that the biologically active conformation at the gastrin receptor is partly helical and one in which the indole of tryptophan and the aromatic ring of phenylalanine are close to one another while the methionine and aspartic acid side chains point in the opposite direction. These major differences in preferred structures between the common CCK-7/CER-7 peptides and the common CCK-4/little gastrin peptides explain the mutually exclusive activities of these two sets of peptides. We have observed that [Met]enkephalin strongly antagonizes the action of the naturally occurring peripherally active CCK-8 (CCK-7 with an NH2-terminal aspartic acid residue added). The computed lowest energy structures for this opiate peptide closely resemble key features of the computed CCK-7/CER-7 structure, further supporting the proposed structure.

2'(3')-O-Bromoacetyluridine reacts rapidly and selectively with bovine pancreatic ribonuclease A at pH 5.5 and 25 degrees. Under conditions of high molar ratios of nucleoside derivative to enzyme, the only derivative is N-3-carboxymethylhistidine-12 ribonuclease A. The reaction occurs almost exclusively with the histidine-12 residue at the active site inactivation of the enzyme is accompanied by the stoichiometric disappearance of unmodified ribonuclease A and appearance of the product, N-3-carboxymethylhistidine-12 ribonuclease A. Kinetic studies indicate a mechanism involving saturation of the enzyme by the nucleoside derivative. The inhibitor constant, Kb, is 0.087 M and k3 is 35.1 times 10(-4) sec minus 1. The reaction of 2'(3')-O-bromoacetyluridine with the enzyme occurs at a rate approximately 3100 times greater than that corresponding to the reaction with L-histidine. The alkylation reaction is inhibited competitively by uridine with a Ki of 0.013 M. 2'(3')-O-Bromoacetyluridine inactivates ribonuclease A 4.5 times faster than bromoacetic acid and the specificity for alkylation of active-site histidine residues is different. 2'(3')-O-Bromoacetyluridine reacts 1000 times more rapidly with ribonuclease A than iodoacetamide. The contribution of nucleoside binding to the overall rate of alkylation is discussed.

Nucleoside triphosphate pyrophosphohydrolase (EC 3.6.1.8) activity is associated with matrix vesicles purified from collagenase digests of fetal calf epiphyseal cartilage. This enzyme hydrolyzes nucleoside triphosphates to nucleotides and PPi, the latter inducing precipitation in the presence of Ca2+ and Pi. An assay for matrix vesicle nucleoside triphosphate pyrophosphohydrolase is developed using beta, gamma-methylene ATP as substrate. The assay is effective in the presence of matrix vesicle-associated ATPase, pyrophosphatase, and alkaline phosphatase activities. A soluble nucleoside triphosphate pyrophosphohydrolase is obtained from matrix vesicles by treatment with 5 mM sodium deoxycholate. The solubilized enzyme induced the precipitation of calcium phosphate in the presence of ATP, Ca2+, and Pi. Extraction of deoxycholate-solubilized enzymes from matrix vesicles with 1-butanol destroys nucleoside triphosphate pyrophosphohydrolase activity while enhancing the specific activities of ATPase, pyrophosphatase, and alkaline phosphatase. In solutions devoid of ATP and matrix vesicles, concentrations of PPi between 10 and 100 microM induce calcification in mixtures containing initial Ca2+ X P ion products of 3.5 to 7.9 mM2. This finding plus the discovery of nucleoside triphosphate pyrophosphohydrolase in matrix vesicles supports the view that these extracellular organelles induce calcium precipitation by the enzymatic production of PPi. Nucleoside triphosphate pyrophosphohydrolase is more active against pyrimidine nucleoside triphosphates than the corresponding purine derivatives. The pH optimum is 10.0 and the enzyme is neither activated nor inhibited by Mg2+ or Ca2+ ions or mixtures of the two. Vmax at pH 7.5 for beta, gamma-methylene ATP is 0.012 mumol of substrate hydrolyzed per min per mg of protein and Km is below 10 microM. The enzyme is irreversibly destroyed at pH 4 and is stable at pH 10.5.

The three-dimensional structures of the transforming region of the product of the EJ/T24 human bladder oncogene and of the c-Ha ras-1 gene product have been calculated by using conformational energy calculations. These two genes, representing a transforming oncogene and its normal cellular homologue, encode 21,000-dalton peptides that differ by one amino acid at position 12. We therefore examined the energetically allowed conformations of the hydrophobic decapeptide surrounding this substitution site. The calculations show that the most favorable form of the c-Ha ras-1 gene product exists when glycine-12 is in a left-handed bend conformation. No other amino acid can adopt this conformation and thus the bladder oncogene peptide containing valine at position 12 has a markedly different three-dimensional structure. A simple model is proposed to account for the consequences of a position 12 mutation.

Using conformational energy calculations, we previously predicted that there are two distinct binding modes for hexasaccharide substrates of hen egg white lysozyme (HEWL), a "left-sided" binding mode and a "right-sided" one. The former involves such residues as Arg-45, Asn-46, and Thr-47, while the latter involves such residues as Asn-113 and Arg-114. The left-sided binding mode was predicted to predominate for (GlcNAc)6. We now present two lines of experimental evidence that indicate that left-sided binding occurs for this substrate. First, we show that ring-necked pheasant lysozyme (RNPL), in which Lys and His replace Asn and Arg at positions 113 and 114, respectively, has the same affinity for (GlcNAc)6 as does HEWL, indicating that the "right" side is not involved in equilibrium binding to the substrate. Second, we show that a monoclonal antibody, HyHEL-5, which binds specifically to an epitope including residues Arg-45, Asn-46, Thr-47, Asp-48, and Arg-68 on the far "left" side of HEWL, is competitively displaced by (GlcNAc)5 and (GlcNAc)6 but not by GlcNAc, (GlcNAc)2, or (GlcNAc)4. Only the former two substrates can bind in site F in the lower active site. Since these two substrates are the only ones that competitively displace HyHEL-5, our results suggest that the terminal saccharide residues of these substrates bind to the left side of the active site cleft, as predicted from theory.

Possible biologically active (receptor-bound) conformations of peptides derived from cholecystokinin (CCK) have been deduced using conformational analysis combined with comparative studies of their biological specificities. Two peptides, the completely active carboxyl terminal heptapeptide from CCK (CCK-7), whose sequence is Tyr-Met-Gly-Trp-Met-Asp-Phe-NH2, and the carboxyl terminal heptapeptide from cerulein (CER-7) which has the same sequence as for CCK-7 except for replacement of Met 2 with a Thr 2, both stimulate peripheral receptors in gall bladder, pancreas, and pylorus in the gastrointestinal system. In contrast, two other very similar peptides, the last four residues of CCK (CCK-4) whose sequence is Trp-Met-Asp-Phe-NH2, and the carboxyl terminal hexapeptide of little gastrin (LGA-6, Tyr-Gly-Trp-Met-Asp-Phe-NH2, i.e., residue 2 deleted relative to CCK-7 and CER-7 sequences), interact specifically with gastrin receptors and not at all or very weakly with peripheral receptors. All of these peptides react with CCK receptors in the central nervous system, especially in forebrain. The results in the GI tract suggest that the peptides active on peripheral receptors adopt structures that are significantly different from those of the peptides that interact with gastrin receptors. We have generated all of the many low energy conformations for each of these peptides. By retaining only the conformations that are the same for peptides within the same group and then rejecting those resulting conformations that are the same for the peptides in the two different groups, we can greatly reduce the possible active conformations for the peptides within each class.(ABSTRACT TRUNCATED AT 250 WORDS)

The binding of cancer cells to the basement membrane glycoprotein laminin appears to be a critical step in the metastatic process. This binding can be inhibited competitively by a specific pentapeptide sequence (Tyr-Ile-Gly-Ser-Arg) of the laminin B1 chain, and this peptide can prevent metastasis formation in vivo. However, other similar pentapeptide sequences (e.g., Tyr-Ile-Gly-Ser-Glu) have been found to be much less active in metastasis inhibition, raising the possibility that such amino acid substitutions produce structural changes responsible for altering binding to the laminin receptor. In this study, conformational energy analysis has been used to determine the three-dimensional structures of these peptides. The results indicate that the substitution of Glu for the terminal Arg produces a significant conformational change in the peptide backbone at the middle Gly residue. These results have important implications for the design of drugs that may be useful in preventing metastasis formation and tumor spread.

The effect of the substitution of Arg for Gly 13 on the structure of the transforming region decapeptide (Leu 6-Gly 15) of the ras oncogene encoded P21 protein has been investigated using conformational energy analysis. A human malignancy has been identified that contains a ras gene with a single mutation in the thirteenth codon such that the encoded protein would have Arg substituted for Gly at this position, and transfection of cells in culture with this gene results in malignant transformation. Conformational analysis demonstrates that the Arg 13 decapeptide adopts a conformation identical to that for other peptides with substitutions at position 13 (Asp 13, Val 13) from transforming proteins that is distinctively different from that for peptides (Gly 13, Ser 13) from normal, nontransforming proteins. This is found to be an indirect effect resulting from changes in the conformation of Gly 12 produced by substitutions at position 13. These results are consistent with recent analysis of crystallographic data of proteins on conformational preferences for glycine in tripeptide sequences.

Four new bromoacetamido pyrimidine nucleosides have been synthesized and are affinity labels for the active site of bovine pancreatic ribonuclease A (RNase A). All bind reversibly to the enzyme and react covalently with it, resulting in inactivation. The binding constants Kb and the first-order decomposition rate constants k3 have been determined for each derivative. They are the following: 3'-(bromoacetamido)-3'-deoxyuridine, Kb = 0.062 M, k3 = 3.3 X 10(-4) s-1; 2'-(bromoacetamido)-2'-deoxyxylofuranosyluracil, Kb = 0.18 M, k3 = 1700 X 10(-4) s-1; 3'-(bromoacetamido)-3'-deoxyarabinofuranosyluracil, Kb = 0.038 M, k3 = 6.6 X 10(-4) s-1; and 3'-(bromoacetamido)-3'-deoxythymidine, Kb = 0.094 M, k3 = 2.7 X 10(-4) s-1. 3'-(Bromoacetamido)-3'-deoxyuridine reacts exclusively with the histidine-119 residue, giving 70% of a monoalkylated product substituted at N-1, 14% of a monoalkylated derivative substituted at N-3, and 16% of a dialkylated species substituted at both N-1 and N-3. Both 2'-(bromoacetamido)-2'-deoxyxylofuranosyluracil and 3'-(bromoacetamido)-3'-deoxyarabinofuranosyluracil react with absolute specificity at N-3 of the histidine-12 residue. 3'-(Bromoacetamido)-3'-deoxythymidine alkylates histidines-12 and -119. The major product formed in 57% yield is substituted at N-3 of histidine-12. A monoalkylated derivative, 8% yield, is substituted at N-1 of histidine-119. A disubstituted species is formed in 14% yield and is alkylated at both N-3 of histidine-12 and N-1 of histidine-119. A specific interaction of the "down" 2'-OH group, unique to 3'-(bromoacetamido)-3'-deoxyuridine, serves to orient the 3'-bromoacetamido residue close to the imidazole ring of histidine-119. The 2'-OH group of 3',5'-dinucleoside phosphate substrates may serve a similar role in the catalytic mechanism, allowing histidine-119 to protonate the leaving group in the transphosphorylation step. (Bromoacetamido)nucleosides are bound in the active site of RNase A in a variety of distinct conformations which are responsible for the different specificities and alkylation rates.

A homology has been found between an octapeptide involved in attachment of the human immunodeficiency virus to helper/inducer T cells and an octapeptide segment of bovine pancreatic ribonuclease A. This segment (residues 19-26) contains the sites for subtilisin cleavage of this enzyme into the S-peptide and S-protein. From the X-ray crystal structure of ribonuclease, this sequence is known to be exposed to solvent and interacts little with the rest of the protein. A structure for the human immunodeficiency virus attachment peptide can be deduced from this homology, as a well-defined structure has been determined for this sequence in ribonuclease. This can be readily accomplished using previously developed computer methods based upon conformational energy calculations. The calculated structure for human immunodeficiency virus peptide is identical to the ribonuclease segment (19-26) in backbone conformation. It is stabilized by internal interactions of nonpolar residues, and by exposure of polar hydroxyl groups. The results suggest that the T-cell human immunodeficiency virus receptor may be hydrophilic in nature and that conservation of the sequence in two presumably functionally unrelated proteins is related to the need for conservation of exposed structure.

The P-glycoprotein of the mdr 1 gene is responsible for the phenomenon of multidrug resistance in human cells. The presumed drug-binding site of the wild-type P-glycoprotein contains a glycine at position 185. A mutant P-glycoprotein which contains valine at this position causes cells to retain resistance to colchicine, but to lose cross-resistance to other drugs such as the chemotherapeutic agents vinblastine and Adriamycin. This has been hypothesized to be due to a conformational change in the protein induced by the amino acid substitution. Using conformational energy analysis, we have determined the allowed three-dimensional structures for the wild-type and mutant proteins in the region of position 185. The results indicate that the wild-type protein adopts a unique left-handed conformation at position 185 which is energically unfavorable for the protein with L-amino acids (including valine) at this position. This conformational change induced by amino acid substitutions for Gly 185 could explain the differences in binding to the P-glycoprotein of various drugs and, hence, the differences in drug resistance exhibited by various cell lines expressing these proteins.

Conformational analysis, based on ECEPP (Empirical Conformational Energy Program for Peptides) using the chain build-up procedure, was applied to determine the low-energy conformations for a series of tetrapeptides. The tetrapeptides are components of larger peptides which have been found to bind to the CD4 receptor of monocytes. Several previous studies have implicated the tetrapeptide units investigated here as being critical to the biological activities of the full peptides. Five such tetrapeptides were studied: Ser-Ser-Asn-Tyr (from ribonuclease A), Thr-Thr-Asn-Tyr (from peptide T, known to block human immunodeficiency virus from attaching to CD4+ T cells), Thr-Ile-Asn-Tyr (from polio virus coat protein, which is less active than the other peptides in binding to CD4 receptors), Ser-Ser-Ala-Tyr (from the gp 120 coat protein of human immunodeficiency virus, a variant of the peptide T sequence, active in blocking viral attachment to CD4+ cells), and the tetrapeptide from an active synthetic pentapeptide, Asn-Thr-Lys-Tyr (from Asn-Thr-Lys-Tyr-Thr). Using a 7 kcal/mol cutoff, the low-energy conformations for each peptide were computed. Approximately 20,000 conformations were computed for each tetrapeptide. Residue probability profiles were determined for each tetrapeptide. All tetrapeptides except for the polio sequence showed flexibility in the sense that many low-energy conformations were possible. In previous studies, it was postulated that the critical tetrapeptide units would adopt conformations similar to the one observed in a segment of ribonuclease A, residues 22-25, a beta-bend, which is part of an octapeptide segment (residues 19-26) that is homologous to the sequence of peptide T.(ABSTRACT TRUNCATED AT 250 WORDS)

The enzymatic activities and in vitro calcification properties of matrix vesicle fractions isolated from normal and osteoarthritic (OA) human articular cartilage were compared to determine the essential conditions for calcification in these tissues. Four groups of human cartilage were examined, I, normal articular cartilage from aged, nonOA joints; II, discolored or fibrillated cartilage from OA joints; III, osteophytic cartilage from OA joints; IV, loose body cartilage from OA joints. Fetal bovine growth plate cartilage was also studied. Both ATP- and 5'-AMP-dependent in vitro matrix vesicle calcification occurs in all cartilage groups examined and, for human articular cartilage, these activities increase progressively from Groups I to II to III. Calcification does not occur in the absence of either phosphate or pyrophosphate. Alkaline phosphatase, 5'-AMPase, and ATP:pyrophosphohydrolase activities are increased in Groups III and IV cartilage compared with Group I and are detected at high levels in fetal bovine growth plate cartilage. Pyrophosphatase activity occurs in only those cartilage groups juxtaposed to areas of new bone formation (osteophytic, loose body, and bovine growth plate). These results suggest that OA, growth plate, and even normal articular cartilage all have the potential to undergo calcification as long as both phosphate and pyrophosphate ions can be generated at sufficiently high levels. However, the capacity for cartilage to deposit hydroxyapatite, as it does during bone formation, may depend on the presence of pyrophosphatase activity.

Substitutions of amino acids for Gly 12 or Gly 13 in the ras oncogene-encoded P21 proteins have been demonstrated to produce unique structural changes in these proteins that correlate with their ability to produce cell transformation. For example, the P21 proteins with Arg 12 or Val 13 are both known to be actively transforming. Recent site-specific mutagenesis experiments on the transforming Arg 12 protein have found that the substitution of Val for Gly 10 has no effect on transforming activity whereas the substitution of Val for Gly 13 led to a loss of transforming activity. In this study, we examine the structural effects of these substitutions on the amino terminal hydrophobic decapeptide (Leu 6-Gly 15) of P21 using conformational energy analysis. The results show that the transforming proteins with Gly 10 and Arg 12 or Val 10 and Arg 12 can both adopt the putative malignancy-causing conformation, whereas, for the nontransforming protein with Arg 12 and Val 13, this conformation is energetically disallowed. These results further support the theory that due to structural changes the transforming P21 proteins are unable to bind to some regulatory cellular element which may be the recently identified binding protein responsible for the induction of increased GTPase activity in normal P21 compared with transforming mutants.

The 4-arsono-2-nitrophenyl chromophore can serve as a versatile spectrophotometric probe of the surface structure of proteins. Values of pK1' and pK2' for the arsonic acid ionizations are near 3 and 8, respectively, and the presence of nearby positive and negative charges produces substantial alterations in the spectral response of the probe. Changes in the extinction at the wavelength of maximum difference are 30-50% of the extinction coefficients, epsilonmax, for each ionization of the arsonic acid moiety. The titration of 41-(4-arsono-2-nitrophenyl)ribonuclease A indicates that the arsonate dianion binds near the active-site histidine residues. With protonation of a carboxylate side chain in the acidic region, presumably aspartic acid-121, the active site is disrupted. The 41-(4-arsono-2-nitrophenyl) group interacts to a greater degree with the histidine-119 side chain than it does with the histidine-12 residue. Interactions of uridine or 3'-cytidylic acid with the ligand-binding region of 41-(4-arsono-2-nitrophenyl) ribonuclease A modify the spectrophotometric response extensively. 3'-Cytidylic acid binds 41-(4-arsono-2-nitrophenyl) ribonuclease A with an affinity 300 times less than that for native ribonuclease A and 17 times lower than that for 41-(2,4-dinitrophenyl) ribonuclease A. The arsononitrophenyl chromophore is responsive to changes in the active site of ribonuclease A induced by such perturbants as ligand binding, chemical modification, and both acid and thermal denaturation.

4-Arsono-2-nitrofluorobenzene reacts selectively at the anion binding site of bovine pancreatic ribonuclease A. The major derivative is the inactive 41-(4-arsono-2-nitrophenyl) ribonuclease A (45% yield). Additional products are 1-alpha-(4-arsono-2-nitrophenyl) ribonuclease A (11% yield) which is enzymatically active and the disubstituted, inactive 1,41-bis-(4-arsono-2-nitrophenyl) ribonuclease A (25% yield). 2' (3')-O-Bromoacetyluridine reacts with 41-(4-arsono-2-nitrophenyl) ribonuclease A exclusively at the histidine-12 residue at a rate which is approximately one-fourth the rate observed with the unmodified enzyme. Saturation kinetics are observed and the dissociation constant for the protein-inhibitor complex is 0.096 +/- 0.023 M. The first-order unimolecular decomposition constant for complex breakdown is 8.9 +/- 2.9 X 10(-4) s-1. 2'-Bromoacetamido-2'-deoxyuridine reacts with 41-(4-arsono-2-nitrophenyl) ribonuclease A 25 times more slowly than 2'(3')-O-bromoacetyluridine. Bromoacetate reacts with 41-(4-arsono-2-nitrophenyl) ribonuclease A predominantly at the histidine-119 residue at a rate 45 times less than that found for the unmodified enzyme. The results of the alkylation studies imply that the dianionic arsonate does not occupy the phosphate binding site in the enzyme but is sufficiently proximate to account for a decrease in bromoacetate binding as well as a reduction in the nucleophilic reactivity of histidine-12 and -119. All these effects may be accounted for in terms of a local electrostatic perturbation of the active site region by the arsononitrophenyl group.

A sequence similarity has been found between two segments of endothiapepsin (acid proteinase, 2APE), bovine pancreatic ribonuclease A, and peptide T, a segment of the gp120 protein of human immune deficiency virus (HIV), which has been implicated in blocking viral attachment to the T4 receptor. The two similar sequences of the acid proteinase enzyme are Leu-Ile-Asp-Ser-Ser-Ala-Tyr-Thr (residues 169-176) and Tyr-Thr-Gly-Ser-Leu-Asn-Tyr-Thr (residues 175-182). Since the X-ray crystallographic structures of the acid proteinase and ribonuclease are known, it has been possible to determine whether the three-dimensional structures of the segments are similar. Portions of both the segments of acid proteinase are directly superimposable on the structure of the RNase A 19-26 segment. The fact that the three similar sequences from two completely unrelated proteins give rise to almost identical structures raises the possibility that these segments may be involved in nucleating the folding of these proteins. In addition, this provides further support for the concept that the octapeptide sequence of peptide T of HIV, which is also similar in sequence to the 19-26 sequence of RNase A, is also structurally similar to these residues, which adopt a beta-bend conformation. Furthermore, comparison of similarities and differences in the structure of these similar sequences provides an explanation for alterations in the biological activity of various truncated or substituted derivatives of peptide T and additional confirmation of the structural requirements for peptide T in T4-receptor recognition.

The three-dimensional structures of the carboxyl-terminal regions of the P21 protein products of the human Harvey (Ha), Kirsten (KiA and KiB), and neuroblastoma (N) RAS oncogenes and various mutants have been determined by using conformational energy analysis. The carboxyl-terminal region of P21 has been strongly implicated in the binding of the protein to the inner surface of the plasma membrane without which the protein is inactive. The only invariant residue in this region is Cys-186, which is necessary for the post-translational addition of palmitic acid. The surrounding sequences of the active native proteins differ considerably. Nevertheless, certain amino acid substitutions in this region are known to eliminate membrane binding and protein activity, suggesting that there is a conserved common structural feature in this region in the native proteins that is disrupted in the mutant proteins. Conformational energy analysis shows that the four native P21 proteins have a common structure in the form of an alpha-helix for the terminal pentapeptide. A mutant, pBW277, that fails to bind to the membrane and is inactive cannot adopt an alpha-helical structure in this region because of a proline at position 188. Another mutant, pBW766, that retains membrane binding and activity, on the other hand, retains the preference for an alpha-helical conformation in the terminal pentapeptide. These findings suggest that, despite various amino acid sequences in this region, the carboxyl-terminal pentapeptides of the P21 proteins form a distinctive structural domain that must have an alpha-helical structure for membrane binding and intracellular activity.

The GTP-binding p21 protein encoded by the ras-oncogene can be activated to cause malignant transformation of cells by substitution of a single amino acid at critical positions along the polypeptide chain. Substitution of any non-cyclic L-amino acid for Gly 12 in the normal protein results in a transforming protein. This substitution occurs in a hydrophobic sequence (residues 6-15) which is known to be involved in binding the phosphate moities of GTP (and GDP). We find, using conformational energy calculations, that the 6-15 segment of the normal protein (with Gly 12) adopts structures that contain a bend at residues 11 and 12 with the Gly in the D* conformation, not allowed energetically for L-amino acids. Substitution of non-cyclic L-amino acids for Gly 12 results in shifting this bend to residues 12 and 13. We show that many computed structures for the Gly 12-containing phosphate binding loop, segment 9-15, are superimposable on the corresponding segment of the recently determined X-ray crystallographic structure for residues 1-171 of the p21 protein. All such structures contain bends at residues 11 and 12 and most of these contain Gly 12 in the C* or D* conformational state. Other computed conformations for the 9-15 segment were superimposable on the structure of the corresponding 18-23 segment of EFtu, the bacterial chain elongation factor having structural similarities to the p21 protein in the phosphate-binding regions. This segment contains a Val residue where a Gly occurs in the p21 protein. As previously predicted, all of these superimposable conformations contain a bend at positions 12 and 13, not 11 and 12. If these structures that are superimposable on EFtu are introduced into the p21 protein structure, bad contacts occur between the sidechain of the residue (here Val) at position 12 and another phosphate binding loop region around position 61. These bad contacts between the two segments can be removed by changing the conformation of the 61 region in the p21 protein to the corresponding position of the homologous region in EFtu. In this new conformation, a large site becomes available for the binding of phosphate residues. In addition, such phenomena as autophosphorylation of the p21 protein by GTP can be explained with this new model structure for the activated protein which cannot be explained by the structure for the non-activated protein.