Search

A high-throughput (HT) paradigm generating LC-MS-UV-ELSD-based natural product libraries to discover compounds with new bioactivities and or molecular structures is presented. To validate this methodology, an extract of the Indo-Pacific marine sponge Cacospongia mycofijiensis was evaluated using assays involving cytoskeletal profiling, tumor cell lines, and parasites. Twelve known compounds were identified including latrunculins (1-4, 10), fijianolides (5, 8, 9), mycothiazole (11), aignopsanes (6, 7), and sacrotride A (13). Compounds 1-5 and 8-11 exhibited bioactivity not previously reported against the parasite T. brucei, while 11 showed selectivity for lymphoma (U937) tumor cell lines. Four new compounds were also discovered including aignopsanoic acid B (13), apo-latrunculin T (14), 20-methoxy-fijianolide A (15), and aignopsane ketal (16). Compounds 13 and 16 represent important derivatives of the aignopsane class, 14 exhibited inhibition of T. brucei without disrupting microfilament assembly, and 15 demonstrated modest microtubule-stabilizing effects. The use of removable well plate libraries to avoid false positives from extracts enriched with only one or two major metabolites is also discussed. Overall, these results highlight the advantages of applying modern methods in natural products-based research to accelerate the HT discovery of therapeutic leads and/or new molecular structures using LC-MS-UV-ELSD-based libraries.

The cyclodepsipeptide jasplakinolide (1) (a.k.a. jaspamide), isolated previously from the marine sponge Jaspis splendens, is a unique cytotoxin and molecular probe that operates through stabilization of filamentous actin (F-actin). We have recently disclosed that two analogues of 1, jasplakinolides B (3) and E, were referred to the National Cancer Institute's (NCI) Biological Evaluation Committee and the objective of this study was to re-investigate a Fijian collection of J. splendens in an effort to find jasplakinolide congeners with similar biological properties. The current efforts have afforded six known jasplakinolide analogues (4 - 7, 9 - 10), two structures requiring revision (8 and 14) and four new congeners of 1 (11 - 13, 15) including open chain derivatives and structures with modified β-tyrosine residues. Compounds were evaluated for biological activity in the NCI's 60 cell line screen and in a microfilament disruption assay in both HCT-116 and HeLa cells. These two phenotypic screens provide evidence that each cytotoxic analogue, including jasplakinolide B (3), operates by modification of microfilaments. The new structure jasplakinolide V (13) has also been selected for study by the NCI's Biological Evaluation Committee. In addition, the results of a clonogenic dose response study on jasplakinolide are presented.

The world’s genomics data will never be stored in a single repository – rather, it will be distributed among many sites in many countries. No one site will have enough data to explain genotype to phenotype relationships in rare diseases; therefore, sites must share data. To accomplish this, the genetics community must forge common standards and protocols to make sharing and computing data among many sites a seamless activity. Through the Global Alliance for Genomics and Health, we are pioneering the development of shared application programming interfaces (APIs) to connect the world’s genome repositories. In parallel, we are developing an open source software stack (ADAM) that uses these APIs. This combination will create a cohesive genome informatics ecosystem. Using containers, we are facilitating the deployment of this software in a diverse array of environments. Through benchmarking efforts and big data driver projects, we are ensuring ADAM’s performance and utility.

In order to further explore the tolerance of proteins to amino acid substitutions within the interior, a series of core residues was replaced by methionine within the C-terminal domain of T4 lysozyme. By replacing leucine, isoleucine, valine and phenylalanine residues a total of 10 methionines could be introduced, which corresponds to a third of the residues that are buried in this domain. As more methionines are incorporated the protein gradually loses stability. This is attributed in part to a reduction in hydrophobic stabilization, in part to the increased entropic cost of localizing the long, flexible methionine sidechains, and in part to steric clashes. The changes in structure of the mutants relative to the wildtype protein are modest but tend to increase in an additive fashion as more methionines are included. In the most extreme case, namely the 10-methionine mutant, much of the C-terminal domain remains quite similar to wildtype (root-mean-square backbone shifts of 0.56 A), while the F and G helices undergo rotations of approximately 20° and center-of-mass shifts of approximately 1.4 A. For up to six methionine substitutions the changes in stability are additive. Beyond this point, however, the multiple mutants are somewhat more stable than suggested from the sum of their constituents, especially for those including the replacement Val111→Met. This is interpreted in terms of the larger structural changes associated with this substitution. The substituted sidechains in the mutant structures have somewhat higher crystallographic thermal factors than their counterparts in WT*. Nevertheless, the interiors of the mutant proteins retain a well-defined structure with little suggestion of molten-globule characteristics. Lysozymes in which selenomethionine has been incorporated rather than methionine tend to have increased stability. At the same time they also fold faster. This provides further evidence that, at the rate-limiting step in folding, the structure of the C-terminal domain of T4 lysozyme is similar to that of the fully folded protein.

Using heavily methionine-substituted T4 lysozyme as an example, it is shown how the addition or deletion of a small number of methionines can simplify the location of selenium sites for use in MAD phasing. By comparing the X-ray data for a large number of singly substituted lysozymes, it is shown that the optimal amino acid to be substituted by methionine is leucine, followed, in order of preference, by phenylalanine, isoleucine and valine. The identification of leucine as the first choice agrees with the ranking suggested by the Dayhoff mutation probability, i.e. by the frequency of amino-acid substitutions in the sequences of related proteins. The ranking of the second and subsequent choices, however, differ significantly.

The availability of a series of phage T4 lysozymes with up to 14 methionine residues incorporated within the protein has made it possible to systematically compare the effect on protein stability of selenomethionine relative to methionine. Wild-type lysozyme contains two fully buried methionine residues plus three more on the surface. The substitution of these methionine residues with selenomethionine slightly stabilizes the protein. As more and more methionine residues are substituted into the protein, there is a progressive loss of stability. This is, however, increasingly offset in the selenomethionine variants, ultimately resulting in a differential increase in melting temperature of about 7°C. This increase, corresponding to about 0.25 kcal/mol per substitution, is in reasonable agreement with the difference in the solvent transfer free energy between the two amino acids.

In an attempt to identify a systematic relation between the structure of a protein and its folding kinetics, the rate of folding was determined for 20 mutants of T4 lysozyme in which a bulky, buried, nonpolar wild-type residue (Leu, Ile, Phe, Val, or Met) was substituted with alanine. Methionine, which approximated the size of the original side chain but which is of different shape and flexibility, was also substituted at most of the same sites. Mutations that substantially destabilize the protein and are located in the carboxy-terminal domain generally slow the rate of folding. Destabilizing mutations in the amino-terminal domain, however, have little effect on the rate of folding. Mutations that have little effect on stability tend to have little effect on the rate, no matter where they are located. These results suggest that, at the rate-limiting step, elements of structure in the C-terminal domain are formed and have a structure similar to that of the fully folded protein. Consistent with this, two variants that somewhat increase the rate of folding (Phe104 --> Met and Val149 --> Met) are located within the carboxy-terminal domain and maintain or improve packing with very little perturbation of the wild-type structure.

The substitution of methionines with leucines within the interior of a protein is expected to increase stability both because of a more favorable solvent transfer term as well as the reduced entropic cost of holding a leucine side chain in a defined position. Together, these two terms are expected to contribute about 1.4 kcal/mol to protein stability for each Met --> Leu substitution when fully buried. At the same time, this expected beneficial effect may be offset by steric factors due to differences in the shape of leucine and methionine. To investigate the interplay between these factors, all methionines in T4 lysozyme except at the amino-terminus were individually replaced with leucine. Of these mutants, M106L and M120L have stabilities 0.5 kcal/mol higher than wild-type T4 lysozyme, while M6L is significantly destabilized (-2.8 kcal/mol). M102L, described previously, is also destabilized (-0.9 kcal/mol). Based on this limited sample it appears that methionine-to-leucine substitutions can increase protein stability but only in a situation where the methionine side chain is fully or partially buried, yet allows the introduction of the leucine without concomitant steric interference. The variants, together with methionine-to-lysine substitutions at the same sites, follow the general pattern that substitutions at rigid, internal sites tend to be most destabilizing, whereas replacements at more solvent-exposed sites are better tolerated.

To test whether the structure of a protein is determined in a manner akin to the assembly of a jigsaw puzzle, up to 10 adjacent residues within the core of T4 lysozyme were replaced by methionine. Such variants are active and fold cooperatively with progressively reduced stability. The structure of a seven-methionine variant has been shown, crystallographically, to be similar to wild type and to maintain a well ordered core. The interaction between the core residues is, therefore, not strictly comparable with the precise spatial complementarity of the pieces of a jigsaw puzzle. Rather, a certain amount of give and take in forming the core structure is permitted. A simplified hydrophobic core sequence, imposed without genetic selection or computer-based design, is sufficient to retain native properties in a globular protein.

G. Makhatadze and P. Privalov [J. Mol. Biol., 226 (1995) 491] have recently measured the enthalpy of transfer of three proteins into urea and guanidinium chloride solutions as a function of concentration and temperature. The present paper applies the solvent-exchange model [J.A. Schellman, Biopolymers, (1994)] to the data and compares it with the binding model utilized in the original publication. Both calculations assume identical binding sites. It is found that the data may be fit tolerably well using either procedure, but that the parameters describing the binding vary considerably. Consideration of the transfer properties of amino acid moieties and small peptides leads to the conclusion that solvation sites are heterogeneous and that the quantities determined by both methods are statistical averages. The parameters describe an identical-site system that has (approximately) the same properties as the real heterogeneous system. The results have mainly heuristic and mechanistic value. One quantity determined with these simplified isotherms, Σk j Δh j , is a property of the real system and can serve as a measure of a thermal binding capacity for a protein. The appendices contain a resume of the solution theory required for the exchange model of solvation as well as the development of a number of empirical equations for the thermodynamic properties of urea and guanidinium chloride solutions.

A series of individual alanine mutations has been constructed in the helical region 115 to 123 in phage T4 lysozyme in order to evaluate the contribution to protein stability of the different side-chains within this region. Pairwise alanine mutations and a combination mutant with seven alanine substitutions were constructed to evaluate the additive effects upon structure and stability. Only three residues within this region (Ser117, Leu118 and Leu121) have a substantial influence upon stability (change in free energy of unfolding greater than 1.0 kcal/mol). Replacement of Ser117 with alanine results in an increase in protein stability of 1.27 kcal/mol, apparently due to the release of strain present in the wild-type protein. Replacement of the buried residues Leu118 and Leu121 is destabilizing. Substitution of the remaining six residues with alanine has relatively little effect on stability. This is consistent with prior studies showing that only 20 to 30% of the residues in amphipathic helices in T4 lysozyme are critical for stability. For some of the pairwise alanine mutants the effects on stability are additive. For most of these mutants, however, there is a slight (approximately 0.15 to 0.25 kcal/mol) non-additivity such that the double mutant is more stable than the sum of the constituent single mutants. This effect is consistently observed for residues with positions i, i +4; i.e. adjacent, but in consecutive turns of the helix, suggesting a weak but significant interaction between these amino acid residues. A more pronounced non-additivity (approximately 0.5 kcal/mol) is seen in the seven-alanine combination mutant. This non-additivity is due to a modest "collapse" or "repacking" that occurs for the combination mutants (especially the multiple alanine mutant) but is not possible for the single replacements. The truncation of some side-chains permits an increase in solvent accessibility of main-chain amide and carbonyl groups. This effect is most pronounced for the seven-alanine combination mutant, where two solvent molecules, not present in wild-type, hydrogen bond to main-chain carbonyl groups in the middle region of the helix. It has been suggested that the binding of such water molecules might represent the first step in solvent-mediated unfolding of an alpha-helix. The appearance of ordered solvent, however, appears to have very little effect on stability (approximately less than 0.2 kcal/mol).

A new 1-imidazoyl-3-carboxy-6-hydroxy-beta-carboline alkaloid, named hyrtiocarboline (1), was isolated from a Papua New Guinea marine sponge, Hyrtios reticulatus. The structure was elucidated from spectroscopic data, including (1)H-(15)N HMBC NMR experiments, which provided complementary (15)N chemical shift information in support of the structure. This compound showed selective antiproliferative activity against H522-T1 non-small cell lung, MDA-MB-435 melanoma, and U937 lymphoma cancer cell lines.

The goal of this study was to isolate and study additional jasplakinolide analogues from two taxonomically distinct marine sponges including two Auletta spp. and one Jaspis splendens. This led to the isolation of jasplakinolide (1) and eleven jasplakinolide analogues (3 – 13) including seven new analogues (6 – 10, 12, and 13). Structure elucidation of the new compounds was based on a combination of 1D and 2D NMR analysis, optical rotation, circular dichroism, and preparation of Mosher’s esters. Five of the new compounds are oxidized tryptophan derivatives of 1, including a unique quinazoline derivative (9). Compounds 1, 3, 5 – 8, and 11 were evaluated in the NCI 60 cell line screen and all compounds were tested in a microfilament disruption assay. Jasplakinolide B (11) exhibited potent cytotoxicity (GI50 < 1 nM vs. human colorectal adenocarcinoma (HCT-116) cells) but did not exhibit microfilament-disrupting activity at 80 nM.

A high-throughput (HT) agar-based halo assay is described, which allows for rapid screening of chemical libraries for bioactivity in microorganisms such as yeast and bacteria. A pattern recognition algorithm was developed to identify halo-like shapes in plate reader optical density (OD) measurements. The authors find that the total growth inhibition within a detected halo provides an accurate estimate of a compound's potency measured in terms of its EC(50). The new halo recognition method performs significantly better than an earlier method based on single-point OD readings. An assay based on the halo algorithm was used to screen a 21,120-member library of drug-like compounds in Saccharomyces cerevisiae, leading to the identification of novel bioactive scaffolds containing derivatives of varying potencies. The authors also show that the HT halo assay can be performed with the pathogenic bacterium Vibrio cholerae and that liquid culture EC(50) values and halo scores show a good correlation in this organism. These results suggest that the HT halo assay provides a rapid and inexpensive way to screen for bioactivity in multiple microorganisms.

Chemogenomic screens were performed in both budding and fission yeasts, allowing for a cross-species comparison of drug–gene interaction networks. Drug–module interactions were more conserved than individual drug–gene interactions. Combination of data from both species can improve drug–module predictions and helps identify a compound's mode of action., Understanding the molecular effects of chemical compounds in living cells is an important step toward rational therapeutics. Drug discovery aims to find compounds that will target a specific pathway or pathogen with minimal side effects. However, even when an effective drug is found, its mode of action (MoA) is typically not well understood. The lack of knowledge regarding a drug's MoA makes the drug discovery process slow and rational therapeutics incredibly difficult. More recently, different high-throughput methods have been developed that attempt to discern how a compound exerts its effects in cells. One of these methods relies on measuring the growth of cells carrying different mutations in the presence of the compounds of interest, commonly referred to as chemogenomics (Wuster and Babu, 2008). The differential growth of the different mutants provides clues as to what the compounds target in the cell (Figure 2). For example, if a drug inhibits a branch in a vital two-branch pathway, then mutations in the second branch might result in cell death if the mutants are grown in the presence of the drug (Figure 2C). As these compound–mutant functional interactions are expected to be relatively rare, one can assume that the growth rate of a mutant–drug combination should generally be equal to the product of the growth rate of the untreated mutant with the growth rate of the drug-treated wild type. This expectation is defined as the neutral model and deviations from this provide a quantitative score that allow us to make informed predictions regarding a drug's MoA (Figure 2B; Parsons et al, 2006). The availability of these high-throughput approaches now allows us to perform cross-species studies of functional interactions between compounds and genes. In this study, we have performed a quantitative analysis of compound–gene interactions for two fungal species (budding yeast (S. cerevisiae) and fission yeast (S. pombe)) that diverged from each other approximately 500–700 million years ago. A collection of 2957 compounds from the National Cancer Institute (NCI) were screened in both species for inhibition of wild-type cell growth. A total of 132 were found to be bioactive in both fungi and 9, along with 12 additional well-characterized drugs, were selected for subsequent screening. Mutant libraries of 727 and 438 gene deletions were used for S. cerevisiae and S. pombe, respectively, and these were selected based on availability of genetic interaction data from previous studies (Collins et al, 2007; Roguev et al, 2008; Fiedler et al, 2009) and contain an overlap of 190 one-to-one orthologs that can be directly compared. Deviations from the neutral expectation were quantified as drug–gene interactions scores (D-scores) for the 21 compounds against the deletion libraries. Replicates of both screens showed very high correlations (S. cerevisiae r=0.72, S. pombe r=0.76) and reproduced well previously known compound–gene interactions (Supplementary information). We then compared the D-scores for the 190 one-to-one orthologs present in the data set of both species. Despite the high reproducibility, we observed a very poor conservation of these compound–gene interaction scores across these species (r=0.13, Figure 4A). Previous work had shown that, across these same species, genetic interactions within protein complexes were much more conserved than average genetic interactions (Roguev et al, 2008). Similarly we observed a higher cross-species conservation of the compound–module (complex or pathway) interactions than the overall compound–gene interactions. Specifically, the data derived from fission yeast were a poor predictor of S. cerevisaie drug–gene interactions, but a good predictor of budding yeast compound–module connections (Figure 4B). Also, a combined score from both species improved the prediction of compound–module interactions, above the accuracy observed with the S. cerevisae information alone, but this improvement was not observed for the prediction of drug–gene interactions (Figure 4B). Data from both species were used to predict drug–module interactions, and one specific interaction (compound NSC-207895 interaction with DNA repair complexes) was experimentally verified by showing that the compound activates the DNA damage repair pathway in three species (S. cerevisiae, S. pombe and H. sapiens). To understand why the combination of chemogenomic data from two species might improve drug–module interaction predictions, we also analyzed previously published cross-species genetic–interaction data. We observed a significant correlation between the conservation of drug–gene and gene–gene interactions among the one-to-one orthologs (r=0.28, P-value=0.0078). Additionally, the strongest interactions of benomyl (a microtubule inhibitor) were to complexes that also had strong and conserved genetic interactions with microtubules (Figure 4C). We hypothesize that a significant number of the compound–gene interactions obtained from chemogenomic studies are not direct interactions with the physical target of the compounds, but include many indirect interactions that genetically interact with the main target(s). This would explain why the compound interaction networks show similar evolutionary patterns as the genetic interactions networks. In summary, these results shed some light on the interplay between the evolution of genetic networks and the evolution of drug response. Understanding how genetic variability across different species might result in different sensitivity to drugs should improve our capacity to design treatments. Concretely, we hope that this line of research might one day help us create drugs and drug combinations that specifically affect a pathogen or diseased tissue, but not the host., We present a cross-species chemogenomic screening platform using libraries of haploid deletion mutants from two yeast species, Saccharomyces cerevisiae and Schizosaccharomyces pombe. We screened a set of compounds of known and unknown mode of action (MoA) and derived quantitative drug scores (or D-scores), identifying mutants that are either sensitive or resistant to particular compounds. We found that compound–functional module relationships are more conserved than individual compound–gene interactions between these two species. Furthermore, we observed that combining data from both species allows for more accurate prediction of MoA. Finally, using this platform, we identified a novel small molecule that acts as a DNA damaging agent and demonstrate that its MoA is conserved in human cells.

A new compound of mixed polyketide synthase-nonribosomal peptide synthetase (PKS/NRPS) origin, 11- O-methylpseurotin A ( 1), was identified from a marine-derived Aspergillus fumigatus. Bioassay-guided fractionation using a yeast halo assay with wild-type and cell cycle-related mutant strains of Saccharomyces cerevisiae resulted in the isolation of 1, which selectively inhibited a Hof1 deletion strain. Techniques including 1D and 2D NMR, HRESIMS, optical rotation, J-based analysis, and biosynthetic parallels were used in the elucidation of the planar structure and absolute configuration of 1. A related known compound, pseurotin A ( 2), was also isolated and found to be inactive in the yeast screen.

Human lipoxygenases (hLO) have been implicated in a variety of diseases and cancers and each hLO isozyme appears to have distinct roles in cellular biology. This fact emphasizes the need for discovering selective hLO inhibitors for both understanding the role of specific lipoxygenases in the cell and developing pharmaceutical therapeutics. To this end, we have modified a known lipoxygenase assay for high-throughput (HTP) screening of both the National Cancer Institute (NCI) and the UC Santa Cruz marine extract library (UCSC-MEL) in search of platelet-type 12-hLO (12-hLO) selective inhibitors. The HTP screen led to the characterization of five novel 12-hLO inhibitors from the NCI repository. One is the potent but non-selective michellamine B, a natural product, anti-viral agent. The other four compounds were selective inhibitors against 12-hLO, with three being synthetic compounds and one being alpha-mangostin, a natural product, caspase-3 pathway inhibitor. In addition, a selective inhibitor was isolated from the UCSC-MEL (neodysidenin), which has a unique chemical scaffold for a hLO inhibitor. Due to the unique structure of neodysidenin, steady-state inhibition kinetics were performed and its mode of inhibition against 12-hLO was determined to be competitive (K(i)=17microM) and selective over reticulocyte 15-hLO-1 (K(i) 15-hLO-1/12-hLO>30).

A re-collection of Plakortis quasiamphiaster from Vanuatu in 2003 resulted in the isolation of three known compounds, plakinidine A (1) and amphiasterins B1 (6) and B2 (7). Also isolated was a new bis-oxygenated pyrroloacridine alkaloid, plakinidine E (8), with a unique O-substitution versus N-substitution at position C-12 in 1. The biological evaluation of the active compounds in two assays provided complementary data. Plakinidine A (1) exhibited cytotoxicity against human colon H-116 cells with an IC50 of 0.23 microg/mL, but there were no effects against the yeast Saccharomyces cerevisiae diploid homozygous deletion strain of topoisomerase I (top1Delta). By contrast, 8 was inactive against H-116 cells but was potent in the yeast halo screen.

Tail-associated lysozyme of bacteriophage T4 (tail lysozyme), the product of gene 5 (gp 5), is an essential structural component of the hub of the phage baseplate. It is synthesized as a 63-kDa precursor, which later cleaves to form mature gp 5 with a molecular weight of 43,000. To elucidate the role of the C-terminal region of the precursor protein, gene 5 was cloned and overexpressed and the product was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblotting, analytical ultracentrifugation, and circular dichroism. It was shown that the precursor protein tends to be cleaved into two fragments during expression and that the cleavage site is close to or perhaps identical to the cleavage site in the infected cell. The two fragments, however, remained associated. The lysozyme activity of the precursor or the nicked protein is about 10% of that of mature gp 5. Both the N-terminal mature tail lysozyme and the C-terminal fragment were then isolated and characterized by far-UV circular dichroism and analytical ultracentrifugation. The latter remained trimeric after dissociation from the N-terminal fragment and is rich in β-structure as predicted by an empirical method. To trace the fate of the C-terminal fragment, antiserum was raised against a synthesized peptide of the last 12 C-terminal residues. Surprisingly, the C-terminal fragment was found in the tail and the phage particle by immunoblotting. The significance of this finding is discussed in relation to the molecular assembly and infection process.

Bacterial biofilms are assemblages of bacterial cells and extracellular matrix that result in the creation of surface-associated macrocolony formation. Most bacteria are capable of forming biofilms under suitable conditions. Biofilm formation by pathogenic bacteria on medical implant devices has been linked to implant rejection in up to 10% of cases, due to biofilm-related secondary infections. In addition, biofilm formation has been implicated in both bacterial persistence and antibiotic resistance. In this study, a method has been developed for the discovery of small molecule inhibitors of biofilm formation in Vibrio cholerae, through the use of high-throughput epifluorescence microscopy imaging. Adaptation of a strategy for the growth of bacterial biofilms in wellplates, and the subsequent quantification of biofilm coverage within these wells, provides the first example of an image-based 384-well format system for the evaluation of biofilm inhibition in Vibrio cholerae. Application of this method to the high-throughput screening of small molecule libraries has lead to the discovery of 29 biofilm lead structures, many of which eliminate biofilm formation without altering bacterial cell viability.

In order to determine the thermodynamic cost of introducing a polar group within the core of a protein, a series of nine Ala-->Ser and 3 Val-->Thr substitutions was constructed in T4 lysozyme. The sites were all within alpha-helices but ranged from fully solvent-exposed to totally buried. The range of destabilization incurred by the Ala-->Ser substitutions was found to be very similar to that for the Val-->Thr replacements. For the solvent-exposed and partly exposed sites the destabilization was modest (approximately less than 0.5 kcal/mol). For the completely buried sites the destabilization was larger, but variable (approximately 1-3 kcal/mol). Crystal structure determinations showed that the Ala-->Ser mutant structures were, in general, very similar to their wild-type counterparts, even though the replacements introduce a hydroxyl group. This is in part because the introduced serines are all within alpha-helices and at congested sites can avoid steric clashes with surrounding atoms by making a hydrogen bond to a backbone carbonyl oxygen in the preceding turn of the helix. The three substituted threonine side chains essentially superimpose on their valine counterparts but display somewhat larger conformational adjustments. The results illustrate how a protein structure will adapt in different ways to avoid the presence of an unsatisfied hydrogen bond donor or acceptor. In the most extreme case, Val 149-->Thr, which is also the most destabilizing variant (delta delta G = 2.8 kcal/mol), a water molecule is incorporated in the mutant structure in order to provide a hydrogen-bonding partner. The results are consistent with the view that many hydrogen bonds within proteins contribute only marginally to stability but that noncharged polar groups that lack a hydrogen-bonding partner are very destabilizing (delta delta G approximately greater than 3 kcal/mol). Supportive of other studies, the alpha-helix propensity of alanine is seen to be higher than that of serine (delta delta G = 0.46 +/- 0.04 kcal/mol), while threonine and valine are similar in alpha-helix propensity.