Application and utilization of chemoinformatics tools in lead generation and optimization.

The process of Drug Discovery is a complex and high risk endeavor that requires focused attention on experimental hypotheses, the application of diverse sets of technologies and data to facilitate high quality decision-making. All is aimed at enhancing the quality of the chemical development candidate(s) through clinical evaluation and into the market. In support of the lead generation and optimization phases of this endeavor, high throughput technologies such as combinatorial/high throughput synthesis and high throughput and ultra-high throughput screening, have allowed the rapid analysis and generation of large number of compounds and data. Today, for every analog synthesized 100 or more data points can be collected and captured in various centralized databases. The analysis of thousands of compounds can very quickly become a daunting task. In this article we present the process we have developed for both analyzing and prioritizing large sets of data starting from diversity and focused uHTS in support of lead generation and secondary screens supporting lead optimization. We will describe how we use informatics and computational chemistry to focus our efforts on asking relevant questions about the desired attributes of a specific library, and subsequently in guiding the generation of more information-rich sets of analogs in support of both processes.

Carbacyclic peptide mimetics as VCAM-VLA-4 antagonists.

Substitution of carbon for sulfur in a potent 13-membered cyclic disulfide containing peptide was accomplished via an intramolecular Wittig reaction and resulted in a series of 'carba' analogues. Potency in the VCAM-VLA-4 assay was sensitive to ring size and lower than that of the parent disulfide.

Cyclic thioether peptide mimetics as VCAM-VLA-4 antagonists.

Selective substitution of a sulfur atom by carbon in a highly potent 13-membered cyclic disulfide was accomplished by intramolecular displacement of a bromide. The potency of the resulting thioethers in the VCAM/VLA-4 assay was dependent on ring size and the position of the sulfur atom.

The design and synthesis of potent cyclic peptide VCAM-VLA-4 antagonists incorporating an achiral Asp-Pro mimetic.

The Asp-Pro sequence of the cyclic peptide Ac-HN-Tyr-Cys*-Asp-Pro-Cys*-OH (1) could be replaced with the achiral dipeptide mimetic 1-(2-aminoethyl)cyclpentylcarboxylic acid with retention of potent inhibition of the VCAM-VLA-4 interaction.

Effect of dietary fish oil, vitamin E, and probucol on renal injury in the rat.

Dietary fish oil, vitamin E, and probucol have been considered in a variety of human and experimental models of kidney disease. Using subtotal nephrectomized cholesterol-fed rats as a model for progressive kidney disease, we examined the effect of 5% dietary fish oil, or a combination of 5% dietary fish oil with 500 IU vitamin E/kg diet or 1% probucol on renal injury. Three-month-old Sprague Dawley rats were fed a control diet (C group) or a cholesterol supplemented (2%) diet (Ch group) containing either fish oil (FO group) or fish oil plus vitamin E (FO+E group) or fish oil plus probucol (FO+P group). After 4 weeks of dietary treatment, the right kidney was electrocoagulated and the left kidney nephrectomized. After 8 weeks, 24-hour urine was collected before sacrifice. No effect of the dietary treatments was noted on serum creatinine, blood urea nitrogen, or proteinuria, except that proteinuria was highest in FO+P group. Rats receiving the cholesterol diets had higher serum low density lipoprotein (LDL) + very low density lipoprotein (VLDL) cholesterol (P < 0.05). In contrast, rats in the FO+P group had the lowest serum total cholesterol and LDL+VLDL cholesterol among all groups. The FO group had 26% lower kidney alpha-tocopherol concentrations than the C group. However, inclusion of vitamin E in the diet (FO+E group) increased the kidney alpha-tocopherol status to a level comparable to that in the C group, whereas inclusion of probucol in fish oil diet (FO+P group) did not improve the kidney alpha-tocopherol status. Rats fed the cholesterol diet had a 2.5-fold higher glomerular segmental sclerosis (GSS) score and 1.5-fold higher glomerular macrophage (GM) subpopulation than the C group. These effects of the cholesterol diet were ameliorated by a fish oil diet (FO group: GSS by 30%, GM by 24%). The inclusion of vitamin E in the fish oil diet (FO+E group) did not further improve the GSS score or GM subpopulation. However, inclusion of probucol in fish oil diet (FO+P group) lowered the GSS score by 73% and reduced GM subpopulation by 83% compared with the Ch group. These remarkable changes can be attributed to the powerful hypocholesterolemic activity of probucol. Our findings indicate that progression of glomerular sclerosis in the rat remnant kidney model of progressive kidney disease can be significantly modulated with fish oil treatment.

Relationship between nitrogen mustard drug resistance in B-cell chronic lymphocytic leukemia (B-CLL) and protein expression of Bcl-2, Bax, Bcl-X and p53.

Several genes have been implicated in the regulation of apoptosis including bcl-2, bax, bcl-X and p53. These genes may be important in the development of nitrogen mustard (NM) drug resistance in B-cell chronic lymphocytic leukemia (B-CLL). Using Western blot analysis, we examined the levels of Bcl-2, Bax, Bcl-X and p53 protein expression and determined whether the levels of these proteins correlated with in vitro drug resistance in CLL patients' lymphocyte samples. Our investigations suggest that in CLL, NM drug resistance develops without any detectable alteration of Bcl-2, Bax or Bcl-X. In addition, we determined the presence of p53 mutations in 14 samples in order to assess if there is an association between in vitro drug resistance and the presence of p53 mutations. Using single-stranded conformational polymorphism (SSCP) and sequencing analysis, we observed a p53 mutation in two out of seven resistant samples. The mutation occurring in both cases was a G:C --> A:T transition at codon 273 (exon 8). One of these cases was de novo resistant to the nitrogen mustards. Only one of six samples with acquired resistance to the nitrogen mustards had a p53 mutation suggesting that p53 mutations are not a prominent feature of acquired NM resistance in CLL.

Natural killer cell activity in elderly men is enhanced by beta-carotene supplementation.

Natural killer (NK) cell activity has been postulated to be an immunologic link between beta-carotene and cancer prevention. In a cross-sectional, placebo-controlled, double-blind study we examined the effect of 10-12 y of beta-carotene supplementation (50 mg on alternate days) on NK cell activity in 59 (38 middle-aged men, 51-64 y; 21 elderly men, 65-86 y) Boston area participants in the Physicians' Health Study. No significant difference was seen in NK cell activity due to beta-carotene supplementation in the middle-aged group. The elderly men had significantly lower NK cell activity than the middle-aged men; however, there was no age-associated difference in NK cell activity in men supplemented with beta-carotene. beta-carotene-supplemented elderly men had significantly greater NK cell activity than elderly men receiving placebo. The reason for this is unknown; however, it was not due to an increase in the percentage of NK cells, nor to an increase in interleukin 2 (IL-2) receptor expression, nor to IL-2 production. beta-carotene may be acting directly on one or more of the lytic stages of NK cell cytotoxicity, or on NK cell activity-enhancing cytokines other than IL-2, such as IL-12. Our results show that long-term beta-carotene supplementation enhances NK cell activity in elderly men, which may be beneficial for viral and tumoral surveillance.

Carotenoid and tocopherol concentrations in plasma, peripheral blood mononuclear cells, and red blood cells after long-term beta-carotene supplementation in men.

To determine the effects of long-term beta-carotene supplementation on concentrations of carotenoids and tocopherols in plasma and in blood cells, fasting blood was collected from 73 randomly selected physicians from the Boston area who are participating in the Physicians Health Study (PHS). The PHS is a randomized, placebo-controlled, double-blind study. In 1982, 22,071 male physicians were assigned to one of four treatments (325 mg aspirin alone, 50 mg beta-carotene alone, both, or neither) every other day. Plasma, peripheral blood mononuclear cells (PBMCs), and red blood cells (RBCs) from physicians who have participated in the study for approximately 12 y were analyzed for carotenoids and tocopherols. Compared with the placebo group, the supplemented group had higher beta-carotene concentrations in plasma (1.73+/-0.16 compared with 0.54+/-0.06 micromol/L0, RBCs (91.5+/-9.7 compared with 31.2+/-4.2 pmol/g hemoglobin), and PBMCs (61.6+/-10.3 compared with 15.5+/-2.5 pmol/10(7) cells). There were no differences in other carotenoids or tocopherols in plasma, RBCs, and PBMCs between these two groups. The beta-carotene concentrations. Plasma cryptoxanthin correlated with both RBC and PBMC cryptoxanthin concentrations but plasma lycopene correlated only with PBMC lycopene concentrations. These data suggest that plasma may not be the best indicator of carotenoid status. Furthermore, long-term beta-carotene supplementation in men results in higher beta-carotene concentrations in plasma, RBCs and PBMCs without lowering concentrations of other carotenoids or tocopherols.

Characterizing the mu-conotoxin binding site on voltage-sensitive sodium channels with toxin analogs and channel mutations.

The three-dimensional organization of the domains of the rat skeletal muscle sodium channel subtype 1 (rSkM1) and the toxin-channel interaction surface have been explored by a complementary mutagenesis approach. This method involves probing mutant channels with analogs of the peptide toxin, mu-conotoxin (mu-CTX), for which the tertiary structure has been determined. mu-CTX has an overall net charge of +5. The blocking of Na+ currents of rSkM1 expressed in Xenopus oocytes by mu-CTX analogs in which negative charge had been removed by Asn substitution for Asp or positive charge had been decreased by Gln substitution for Arg or Lys was studied; the mu-CTX analogs exhibited decreased blocking potencies of up to 228-fold compared with an IC50 = 51.4 +/- 2.2 nM for native mu-CTX on wild-type rSkM1. Mutations at Arg 13 of mu-CTX were the most critical in decreasing potency and at Lys9 were the least critical. Charge alone, however, was not the essential factor in some toxin substitutions: the IC50 value for Asp12Asn showed little change while that for Asp12Glu was increased approximately 100-fold due to a change in conformation (revealed by NMR measurements of the toxin in solution). Focusing on the sites in the channel which might be involved in toxin binding, mutations were introduced involving substitutions at more than a dozen mostly anionic sites in putative extracellular residues of rSkM1. The toxin binding results indicate: firstly, many channel mutations at anionic sidechains on the putative extracellular surface of mu-CTX-sensitive channels, thought to be possible sites of interaction with toxin, have been shown to have no effect on toxin binding. Secondly, one channel mutation, rSkM1/Tyr401Cys, (in the loop between S5 and S6 of Domain 1), affected mu-CTX potency causing a 3.7-fold increase in IC50 value. The ratio of toxin blocking potencies was not significantly different when wild-type and the mutant (Tyr401Cys) rSkM1 channels were studied with two toxin analogs, Arg19Gln and Arg13Gln, in contrast to all other toxin derivatives examined. Since Tyr401 is known to be in the channel pore, these results suggest that either or both of the Arg residues at positions 13 and 19 of mu-CTX interact(s) with residue Tyr401 of rSkM1 and, therefore, indicate that mu-CTX extends into the pore region of the channel.

Potent peptide inhibitors of stromelysin based on the prodomain region of matrix metalloproteinases.

Stromelysin is secreted as an inactive zymogen that is activated in the extracellular space by cleavage of the His81-Phe82 bond with the release of the 81-amino acid propeptide domain. This segment contains a 12-amino acid sequence (MRKPRC75GVPDVG) that is highly conserved in all matrix metalloproteinases. Previous studies have shown that the hexapeptide, Ac-RCGVPD-NH2, and the pentapeptide, Ac-RCGVP-NH2, based on this region retain significant inhibitory activity. This new structure-activity relationship study of both peptides has shown that only Cys75 and Val77 are essential for inhibitory activity. Peptides based on this series inhibited stromelysin and collagenase with equal potency. Additional peptides spanning this region were synthesized in order to focus on these two sites. Significantly, isocysteine was substituted for Cys75 without significant loss of inhibitory activity. Tyr-(2,6-dichlorobenzyl) was substituted for Val77. The introduction of these 2 new residues into Ac-CGVP-NH2 produced a very potent inhibitor, Ac-isoCGY-(2,6 dichlorobenzyl)-P-NH2 with an IC50 of 3 microM. The following factors, acting in combination, determine the inhibitory activity of peptides in this series: distance between the sulfur atom and the peptide backbone, coordination geometry of the thiol side chain with the active-site zinc, and conformational flexibility of the side-chain.

Inhibition of human stromelysin by peptides based on the N-terminal domain of tissue inhibitor of metalloproteinases-1.

The tissue inhibitors of metalloproteinases (TIMPs) represent a family of naturally occurring protein inhibitors of stromelysin and other members of the family of matrix metalloproteinases. A series of peptides based on the N-terminal sequence of natural TIMP-1 was synthesized and assessed for inhibitory activity against purified human stromelysin. Inhibitor peptides were identified in the loop (bounded by the disulfide bonds [C3-C99] and [C13-C124]), e.g., [C3(Acm)-C13], (IC50, 42 microM). It was established that inhibition was due to the free sulfhydryl group of either C13 or C124. However, peptides within [C70(Acm)-C98(Acm)] inhibited stromelysin independently of zinc co-ordination by cysteine. The binding epitope in TIMP-1 may be discontinuous and comprised of sequences from at least 2 loops.

Dietary soybean oil changes lipolytic rate and composition of fatty acids in plasma membranes of ovine adipocytes.

A study was conducted to determine which fatty acids in plasma membranes of adipose tissue from ruminants are changed when the diet is supplemented with unsaturated fatty acids and to determine the effect of the fat supplement on adipocyte metabolism. Ten sheep were randomly assigned to two isonitrogenous diets containing either no added fat (control) or 5 g soybean oil/100 g diet. Perirenal fat was removed at slaughter, adipocytes isolated by collagenase digestion, and plasma membranes prepared by centrifugation on a Percoll gradient. Feeding soybean oil to the sheep increased (P < 0.05) linoleic acid [18: 2(n-6)] concentration in subcutaneous fat and isolated adipocytes, suggesting partial escape of dietary unsaturated fatty acids from ruminal biohydrogenation. Soybean oil consumption also decreased (P < 0.05) concentrations of myristic acid, arachidonic acid [20: 4(n-6)] and anteiso 17:0 in plasma membranes, but increased (P < 0.05) trans 18:1. Lipogenesis was not affected by diet, but lipolysis tended to be greater (P = 0.07) in sheep fed the soybean oil-containing diet than in those fed the control diet. In ruminants, fatty acids of ruminal origin, namely trans intermediates of biohydrogenation or branched-chain fatty acids of microbial lipid, may account for as much change in the composition of plasma membranes and in cellular metabolism as do the small quantities of unsaturated fatty acids in the diet that escape biohydrogenation.

Production of monoclonal antibodies to prostromelysin (ProMMP-3) and establishment of a quantitative prostromelysin ELISA assay.

Human prostromelysin (59 kDa) was purified from the conditioned medium of IL-1-stimulated human dermal fibroblasts and anti-prostromelysin monoclonal antibodies were produced and identified by ELISA assay. Using prostromelysin, a C-terminally truncated recombinant form of prostromelysin consisting of amino acids 1-255, and their respective activated enzymes, we have begun mapping the epitopes recognized by these monoclonal antibodies. Various patterns of reactivity against the proenzymes and activated enzymes were observed. In further attempts to map the epitopes, we employed synthetic peptides representing hydrophilic regions of the primary amino acid sequence of prostromelysin. Our monoclonal antibodies did not recognize these peptides, suggesting that the antibodies may be recognizing conformational epitopes composed of non-linear portions of prostromelysin. Using these monoclonal antibodies, we have developed a quantitative prostromelysin sandwich ELISA assay.

Novel class of silicon-based protective groups for the side chain of tyrosine.

A novel class of silyl-based protective groups compatible with the Bpoc/t-Bu strategy has been developed for the side chain of tyrosine. Carbobenzyloxy (CBZ) and biphenylisopropyloxy (Bpoc)-O-beta-trimethylsilylethyl-tyrosine (10 and 12) and CBZ-O-beta-dimethylphenylsilylethyl-tyrosine 14 were prepared in reasonable yields and in very high purity. The trimethylsilylethyl (TMSE) group proved to be 3-4 times more stable than the tert-butyl ether group towards 0.5% TFA. The latter is removed up to 4% during the acidolysis of the Bpoc group. As expected, the dimethylphenylsilylethyl (DMPSE) group was even more resistant towards 0.5% TFA (five time greater than the TMSE analog). Both silyl protective groups were found to be resistant towards a variety of reagents used in peptide synthesis, such as trialkylamines, hydroxybenzotriazole, trialkylphosphine and nucleophiles. They are readily removed in neat TFA in 5-20 min in the absence of cation scavengers, without any detectable alkylation of the phenolic ring. The application of the new silyl-based protective group was demonstrated by the synthesis of the C-terminal 29 amino acid peptide of the basic pancreatic trypsin inhibitor by the prior thiol capture methodology. The protected octapeptide BocC(Acm)QT)(tBu)FVY(TMSE)GG-PO-dibenzofuranthiol++ + was synthesized by solid-phase peptide synthesis using Bpoc-(O-TMSE)-Tyr-OH in greater than 90% yield and coupled to an unprotected 21-mer. The partially blocked, purified peptide was deprotected quantitatively in neat TFA in 1 h.

Peptides based on the conserved predomain sequence of matrix metalloproteinases inhibit human stromelysin and collagenase.

Prostromelysin, a member of the family of matrix metalloproteinases, is secreted as a zymogen which is activated after cleavage of the His81-Phe82 bond. The 82 amino acid propeptide that is removed during activation contains 12 amino acids, MRKPRC75GVPDVG, that are highly conserved in all MMPs. We evaluated a series of peptides that span this region for their ability to inhibit stromelysin. The hexapeptide, Ac-RCGVPD, and the pentapeptide, Ac-RCGVP had IC50 values of approx. 10 microM. The tetrapeptide, Ac-RCGV, was somewhat less potent with an IC50 of 60 microM. Smaller peptides, e.g. Ac-RCG, were significantly less potent as inhibitors. Substitutions of Cys75 with Ser resulted in a complete loss of inhibitory activity. The peptides in this series also inhibited human fibroblast collagenase.

Characterization of a tight-binding MMP-3 inhibitor using improved fluorescence spectroscopy techniques.

Accurate kinetic characterization of stromelysin (MMP-3) inhibitors is critical in the design of potent inhibitors of this enzyme. We have successfully modified a previously described assay [1] which used an internally quenched peptide substrate (Dnp-PYAYWMR) that, upon cleavage by MMP-3, produces the products, Dnp-PYA (quiet) and YWMR (a fluorophore at 360 nm). This improved assay uses purified human MMP-3 in the presence of either 5% methanol or 5% DMSO. Fluorescence intensities associated with total hydrolysis of substrate by enzyme have been successfully mimicked using a combination of the product peptides as a standard. We have determined a Km of 39.2 microM and Kcat/Km of 4.6 microM/h for MMP-3 (in 5% MeOH) using this peptide substrate. This assay was used successfully to characterize Ro 31-4724 ((N-[(N-[2-[(N-hydroxycarbamoyl)methyl]-4-methyl-valeryl]-L-leucyl ] - L-alanine ethyl ester) as a reversible, tightly binding, inhibitor with a Ki of 26 nm.

Identification of growth hormone DNA polymorphisms which respond to divergent selection for abdominal fat content in chickens.

Two strains of meat-type chickens which had been derived from the same genetic base, but were selected for high or low abdominal fat content, respectively, were analyzed for polymorphisms in the growth hormone gene (GH). A total of four DNA polymorphisms were identified, one at a SacI restriction site and three at MspI restriction sites. Restriction mapping indicated that all polymorphisms were in exons and/or introns and not in flanking regions of the gene. The incidence of GH polymorphisms was determined in 20 chickens from each strain and significant differences were observed for two of the four polymorphisms. Analysis by DNA fingerprinting using (CAC)5 as a probe indicated that the inbreeding coefficient was 0.1 in both strains and that random genetic drift was minimal. Thus, the selection for abdominal fat appears to have affected the frequency of alleles of the growth hormone gene. Whether this is the direct consequence of an altered growth hormone gene on fat metabolism or reflects linkage to an allele of a neighbouring gene remains to be determined.

Ruminal biohydrogenation of linoleoyl methionine and calcium linoleate in sheep.

Four ruminally and duodenally cannulated Hampshire wethers were used in a 4 x 4 Latin square experiment to determine whether linoleoyl methionine and calcium linoleate would increase duodenal flow of unsaturated fatty acids (C18:2 + cis C18:1). All animals received the same basal diet plus a treatment enclosed in gelatin capsules that were placed directly in the rumen. Of the four experimental treatments, one was a control (empty capsules) and three were 5 g of fatty acid equivalent as either free linoleic acid, calcium linoleate, or linoleoyl methionine. Linoleoyl methionine had the lowest ruminal disappearance of C18:2 + cis C18:1. Ruminal loss of unsaturated fatty acids from each supplement exclusive of feed unsaturated fatty acids was 69.8, 92.9, and 94.6% for linoleoyl methionine, free linoleic acid, and calcium linoleate, respectively. Duodenal flow of methionine also was higher for linoleoyl methionine than for control, free linoleic acid, or calcium linoleate (2.5, 1.7, 2.0, and 2.5 g/d, respectively). Plasma linoleic acid was higher for linoleoyl methionine than for control or free linoleic acid but was not different from calcium linoleate (22.0, 17.8, 18.9, and 20.2% of total fatty acids, respectively). Plasma methionine levels were not different among treatments. Intestinal disappearance of unsaturated fatty acids did not differ among treatments. Linoleoyl methionine resisted ruminal biohydrogenation and was digested normally in the intestine. Calcium linoleate did not escape biohydrogenation by ruminal bacteria.

Resolution of proline acylation problem for thiol capture strategy by use of a chloro-dibenzofuran template.

The acyl transfer rate for proline, in the prior thiol capture strategy, was enhanced by changing the electronic character of the dibenzofuran template. The rate of amide bond formation between proline and cysteine by the 1-chloro-4-hydroxy-6-mercaptodibenzofuran was measured to be 0.012 min-1, which translates to a half-life of 53 min. Further enhancement of the reaction rate was accomplished by the use of a 1,3-dichloro-dibenzofuran template. The k1 for the reaction was measured to be 0.093 min-1, and the half-life was calculated to be 7 min. To test the applicability of the activated template, 1-chloro-4-hydroxy-6-mercaptodibenzofuran, in peptide synthesis, the 34 amino acid long peptide, H-RPDFCLEPPYTGPCRKARNNFKSADECMRTCGGA-OH, was synthesized. This peptide represents the condensation of the N-terminal 13-mer and the C-terminal 21-mer of the basic pancreatic trypsin inhibitor.

Resistance of fatty acyl amides to degradation and hydrogenation by ruminal microorganisms.

Two in vitro trials were conducted to determine whether fatty acyl amides are degraded and hydrogenated by ruminal microorganisms. The treatments consisted of ground hay supplemented with either no lipid, linoleoyl Met ethyl ester, or free linoleic acid plus Met ethyl ester. Incubations were carried out in Erlenmeyer flasks at 39 degrees C under CO2. Cultures were sampled at predetermined times and analyzed for long-chain fatty acids, Met, and VFA. In trial 1, the rate of disappearance of linoleic acid was lower for the amide than for the FFA (.004 and -.047/h, respectively). In trial 2, there were no differences in the rate of disappearance of linoleic acid from 0 to 6 h (-.237 and -.357/h for amide and FFA, respectively), but the rates from 6 to 48 h (-.003 and -.027/h for amide and FFA, respectively) were different. Linoleoyl Met cultures also had higher acetate to propionate ratio and lower loss of Met compared with free linoleic acid cultures. There was no loss of radioactivity from [14C]stearoyl Met after 24 h of incubation, indicating its resistance to bacterial breakdown. The results showed that fatty acyl amides resist bacterial breakdown and prevent loss of double bonds by microbial biohydrogenation.