Chandrika B-Rao

Managing protein flexibility in docking and its applications

Docking, virtual screening and structure-based drug design are routinely used in modern drug discovery programs. Although current docking methods deal with flexible ligands, managing receptor flexibility has proved to be challenging. In this brief review, we present the current state-of-the-art for computationally handling receptor flexibility, including a novel statistical computational approach published recently. We conclude, from a comparison of the different approaches, that a combination of methods is likely to provide the most reliable solution to the problem of finding the right protein conformation for a given ligand.

MLC1 gene is associated with schizophrenia and bipolar disorder in Southern India.

BACKGROUND Chromosome 22q13 has shown linkage with schizophrenia (SCZ) and bipolar affective disorder (BPAD). A missense mutation in MLC1 (putative cation-channel gene on 22q13) co-segregating with periodic catatonic schizophrenia has been reported. We have investigated the relationship of MLC1 with SCZ and BPAD in Southern India. METHODS All exons and flanking intronic sequences of MLC1 were screened for novel variations. Case-control (216 BPAD, 193 SCZ, 116 control subjects) and family-based analyses (113 BPAD, 107 SCZ families) were performed to evaluate association of MLC1 with these disorders. RESULTS We found 33 MLC1 sequence variations, including three novel mutations: Val210Ile, Leu308Gln, and Arg328His in six BPAD cases and Val210Ile in one control individual. Minor allele of a common variation, ss16339182 (in approximately 6 Kb Linkage-Disequilibrium [LD]-block) was associated with BPAD in case-control (p = .03) and family-based analyses (transmitted/nontransmitted [T/NT]-44/20; p = .003). Association was observed for rs2235349 and rs2076137 with SCZ and ss16339163 with BPAD in case-control study. Using Block 2 haplotype tagging single nucleotide polymorphisms (htSNPs), GC haplotype revealed association (p = .02) and excess transmission (p = .002) with BPAD. CONCLUSIONS Association of MLC1 with SCZ and BPAD suggests involvement of a common pathway. Rare missense mutations and common variants associated with BPAD favors hypothesis about likely involvement of both rare and common polymorphisms in etiology of this complex disorder.

β2-adrenergic receptor polymorphisms and response to salbutamol among Indian asthmatics∗

INTRODUCTION The beta(2)-adrenergic receptor (beta(2)AR or ADRbeta(2)) is the target for beta(2)-agonist drugs used for bronchodilation in asthma and other respiratory diseases. The aim of this study was to identify common single nucleotide polymorphisms (SNPs) and haplotypes in asthmatics and healthy individuals from an Indian population, and determine the influence of beta(2)AR SNPs in responsiveness to beta(2)-agonist therapy in asthma patients. METHODS Ten variable SNP sites within a span of 2.193 kb were identified in the beta(2)AR gene by sequencing and genotyping 374 bronchial asthma patients and healthy individuals from an Indian population. Spirometry tests were performed on 80 unrelated patients before and after administration of 200 microg of salbutamol. A post-bronchodilator forced expiratory volume in one second (FEV(1)) change of >or= 15.3% was considered a good response, and a change of<15.3% was defined as a poor response, to salbutamol. RESULTS The pattern of linkage disequilibrium between the ten SNPs showed a single, linked SNP block consisting of sites -468, -367, -47, -20, and 79 having strong linkage disequilibrium, while the SNPs at sites -1023, -654, 46, 252, and 523 showed very low linkage with one another and with the linked region. The SNPs were found to be organized into 16 haplotypes in the studied population. We found that patients with a homozygous Arg-16 form at nucleotide position 46 are poor responders with probability of 0.81, and patients with a homozygous Gly-16 form are good responders with a probability of 0.73. The responder status to salbutamol treatment and the genotype at nucleotide position 46 in beta(2)AR gene of an asthmatic patient are significantly associated in the studied Indian population (chi2=9.98, df=2, p=0.0068). Most importantly, this association for responsiveness to salbutamol at nucleotide position 46 is independent of other SNPs in the beta(2)AR gene. CONCLUSION This study suggests that the SNP at nucleotide position 46 has particular relevance to pharmacogenetics in the Indian population studied.

Association analysis of CAG repeats at the KCNN3 locus in Indian patients with bipolar disorder and schizophrenia.

Bipolar affective disorder and schizophrenia are severe behavioral disorders with a lifetime risk of approximately 1% in the population worldwide. There is evidence that these diseases may manifest the phenomenon of anticipation similar to that seen in diseases caused by trinucleotide repeat expansions. A recent report has implicated a potassium channel-coding gene, KCNN3, which contains a polymorphic CAG repeat in its coding region, in schizophrenia and bipolar disorder. We have tried to confirm these findings in Indian patients suffering from bipolar disorder and schizophrenia. No statistically significant evidence for the presence of an excess of longer alleles in the patient population, as compared to ethnically matched controls, was found. However, an analysis of the difference of allele sizes revealed a significantly greater number of patients with schizophrenia having differences of allele sizes > or = 5 when compared to normal controls. This finding may be of functional significance as the KCNN3 protein is thought to act as a tetramer, and a large difference in allele sizes would result in an asymmetric molecule with a different number of glutamine residues in each monomer. Am. J. Med. Genet. (Neuropsychiatr. Genet.) 96:744-748, 2000.

Isocytosine-based inhibitors of xanthine oxidase: Design, synthesis, SAR, PK and in vivo efficacy in rat model of hyperuricemia

Structure-activity relationship studies were carried out for lead generation following structure-guided design approach from an isocytosine scaffold identified earlier for xanthine oxidase inhibition. A 470-fold improvement in in vitro IC(50) was obtained in the process. Five most potent compounds with nanomolar IC(50) values were selected for pharmacokinetics and in vivo experiments. The best compound showed good in vivo activity when administered intraperitoneally but was not active by oral route. The results suggest that improvement in oral exposure could improve the in vivo efficacy of this series.

Identification of novel isocytosine derivatives as xanthine oxidase inhibitors from a set of virtual screening hits.

In recent years, xanthine oxidase has emerged as an important target not only for gout but also for cardiovascular and metabolic disorders involving hyperuricemia. Contrary to popular belief, recent clinical trials with uricosurics have demonstrated that enhanced excretion of uric acid is, by itself, not adequate to treat hyperuricemia; simultaneous inhibition of production of uric acid by inhibition of xanthine oxidase is also important. Virtual screening of in-house synthetic library followed by in vitro and in vivo testing led to the identification of a novel scaffold for xanthine oxidase inhibition. In vitro activity results corroborated the results from molecular docking studies of the virtual screening hits. The isocytosine scaffold maintains key hydrogen bonding and pi-stacking interactions in the deep end of the xanthine-binding pocket, which anchors it in an appropriate pose to inhibit binding of xanthine and shows promise for further lead optimization using structure-based drug design approach.

Sample Size Considerations in Genetic Polymorphism Studies

Objectives: Molecular studies for genetic polymorphisms are being carried out for a number of different applications, such as genetic disorders in different populations, pharmacogenomics, genetic identification of ethnic groups for forensic and legal applications, genetic identification of breed/stock in animals and plants for commercial applications and conservation of germ plasm. In this paper, for a random sampling scheme, we address two questions: (A) What should be the minimum size of the sample so that, with a prespecified probability, all alleles at a given locus (or haplotypes at a given set of loci) are detected? (B) What should be the sample size so that the allele frequency distribution at a given locus (or haplotype frequency distribution at a given set of loci) is estimated reliably within permissible error limits? Methods: We have used combinatorial probabilistic arguments and Monte Carlo simulations to answer these questions. Results: We found that the minimum sample size required in case A depends mainly on the prespecified probability of detecting all alleles, while in case B, it varies greatly depending on the permissible error in estimation (which will vary with the application). We have obtained the minimum sample sizes for different degrees of polymorphism at a locus under high stringency, as well as a relaxed level of permissible error. We present a detailed sampling procedure for estimating allele frequencies at a given locus, which will be of use in practical applications. Conclusion: Since the sample size required for reliable estimation of allele frequency distribution increases with the number of alleles at the locus, there is a strong case for using biallelic markers (like single nucleotide polymorphisms) when the available sample size is about 800 or less.

Lead optimization of isocytosine-derived xanthine oxidase inhibitors

We report our attempts at improving the oral efficacy of low-nanomolar inhibitors of xanthine oxidase from isocytosine series through chemical modifications. Our lead compound had earlier shown good in vivo efficacy when administered intraperitoneally but not orally. Several modifications are reported here which achieved more than twofold improvement in exposure. A compound with significant improvement in oral efficacy was also obtained.

A_16_C haplotype in the FcεRIβ gene confers a higher risk for atopic asthma in the Indian population

We conducted a case‐control study aimed at identifying polymorphisms and haplotypes in the FcεRIβ gene associated with asthma and its associated phenotypes such as serum IgE levels. A G/A polymorphism in intron 2, a (CA)n repeat polymorphism in intron 5, and a C/T polymorphism in 3′‐UTR, encompassing a length of 8.74 Kb, were genotyped, and haplotypes were generated for unrelated patients and healthy volunteers in two independent cohorts. A significant association was observed at the level of alleles and genotypes studied with asthma independently in the two cohorts (p < 0.05). A change in CC_AA genotype combination to a non‐CC_AA genotype was associated with reduced risk for asthma. Interestingly, a three‐locus haplotype, A_16_C, was found to be significantly associated with asthma (p = 10−5 in cohort A and p = 0.004 in cohort B). On the other hand, the haplotype G_18_T was found to be negatively associated with asthma (p = 0.007 in cohort A and p = 0.0004 in cohort B). This is the first study identifying a haplotype A_16_C that predisposes individuals to asthma in the Indian population.

Modeling and Selection of Flexible Proteins for Structure-Based Drug Design: Backbone and Side Chain Movements in p38 MAPK

Receptor rearrangement upon ligand binding (induced fit) is a major stumbling block in docking and virtual screening. Even though numerous studies have stressed the importance of including protein flexibility in ligand docking, currently available methods provide only a partial solution to the problem. Most of these methods, being computer intensive, are often impractical to use in actual drug discovery settings. We had earlier shown that ligand‐induced receptor side‐chain conformational changes could be modeled statistically using data on known receptor–ligand complexes. In this paper, we show that a similar approach can be used to model more complex changes like backbone flips and loop movements. We have used p38 MAPK as a test case and have shown that a few simple structural features of ligands are sufficient to predict the induced variation in receptor conformations. Rigorous validation, both by internal resampling methods and on an external test set, corroborates this finding and demonstrates the robustness of the models. We have also compared our results with those from an earlier molecular dynamics simulation study on DFG loop conformations of p38 MAPK, and found that the results matched in the two cases. Our statistical approach enables one to predict the final ligand‐induced conformation of the active site of a protein, based on a few ligand properties, prior to docking the ligand. We can do this without having to trace the step‐by‐step process by which this state is arrived at (as in molecular dynamics simulations), thereby drastically reducing computational effort.