基于蛋白质受体的药物分子计算机辅助设计策略

药物分子计算机辅助设计是一种在计算机或者理论上通过构建具有一定潜在药理活性的新化学实体的分子模拟方法。近十几年来,高通量组学技术的快速发展为生物和化学药物分子设计提供了良好的数据支撑和研究契机。另外,现代社会对生物制药合理性以及作用机理理解的要求越来越高,行业普遍要求药物需要有高效、无毒或者低毒以及靶向性强等特点。随着越来越多与药物靶点相关的蛋白质结构通过实验方法解析出来,基于蛋白质受体的药物分子设计方法可行性进一步提高,其方法也变得越来越重要。基于蛋白质受体的药物分子设计方法,一般是以蛋白质以及配体的三维结构出发进行分析,这让药物分子先导物的发现更加理性化。随着相关实验数据的积累以及深度学习等算法的发展,从而可以进行更加科学的药物分子设计,这在一定程度上加快了新药研发的进程,并更有利于探索相应的分子机理。本文对基于蛋白质受体的药物分子设计方法的常用策略进行系统的回顾、总结和展望。     

Biopolymer nanoparticle production for controlled release of biopharmaceuticals

Abstract

For drug applications, nanoparticles, used as drug carriers, offer the advantage of controlled release, therapeutic impact and targeted delivery. In drug delivery applications, biodegradable polymers can be extracted from natural sources or prepared synthetically by polymerization. Natural polymers typically have varying compositions and physiochemical properties. As a result, methods which utilize natural polymers to encapsulate drugs are more varied and polymer dependent. The following polymers are discussed in this review article: alginate, chitosan, gelatin, albumin, gliadin, pullulan, and dextran. Specialized encapsulation nanotechnologies will be discussed such as ionotropic gelation, complexation, the reverse microemulsion technique, cross-linking methods, emulsion-dependent methods, desolvation methods and self-assembly methods. For each biopolymer an overview of the structure is presented with the corresponding encapsulation techniques. Understanding the structure of the biopolymer is important as to not only understand the rational for current encapsulation techniques but to continue to develop new encapsulation techniques in pursuit of the ideal drug carrier for application in therapeutic treatments.     

New spectrophotometric methods for the determination of nifedipine in pharmaceutical formulations

Two simple, sensitive and economical spectrophotometric methods were developed for the determination of nifedipine in pharmaceutical formulations. Method A is based on the reaction of the nitro group of the drug with potassium hydroxide in dimethyl sulphoxide (DMSO) medium to form a coloured product, which absorbs maximally at 430 nm. Method B uses oxidation of the drug with ammonium molybdate and subsequently reduced molybdenum blue is measured at 830 nm. Beer's law is obeyed in the concentration range of 5.0-50.0 and 2.5-45.0 microg ml(-1) with methods A and B, respectively. Both methods have been successfully applied for the assay of the drug in pharmaceutical formulations. No interference was observed from common pharmaceutical adjuvants. The reliability and the performance of the proposed methods are established by point and interval hypothesis tests and through recovery studies.     

High-performance liquid chromatography determination of polydeoxyribonucleotides in plasma: its application to the determination of defibrotide's pharmacokinetics in the rabbit.

We describe an HPLC method for the determination of whole polydeoxyribonucleotides in animal plasma. This method was compared to a colorimetric method, which evaluates the sugar moiety of polydeoxyribonucleotides, and to an agarose gel electrophoresis method, which evaluates the whole polydeoxyribonucleotides as does the HPLC method, and was found to give results very close to those obtained with these two other methods. A pharmacokinetic study of the antithrombotic, profibrinolytic, polydeoxyribonucleotidic drug defibrotide was carried out by evaluating the plasma drug levels by these three methods. The pharmacokinetic parameters calculated from the data are very similar.     

The binding affinities of proteins interacting with the PDZ domain of PICK1

Protein interacting with C kinase (PICK1) is well conserved throughout evolution and plays a critical role in synaptic plasticity by regulating the trafficking and posttranslational modification of its interacting proteins. PICK1 contains a single PSD95/DlgA/Zo-1 (PDZ) protein-protein interaction domain, which is promiscuous and shown to interact with over 60 proteins, most of which play roles in neuronal function. Several reports have suggested the role of PICK1 in disorders such as epilepsy, pain, brain trauma and stroke, drug abuse and dependence, schizophrenia and psychosis. Importantly, lead compounds that block PICK1 interactions are also now becoming available. Here, a new modeling approach was developed to investigate binding affinities of PDZ interactions. Using these methods, the binding affinities of all major PICK1 interacting proteins are reported and the effects of PICK1 mutations on these interactions are described. These modeling methods have important implications in defining the binding properties of proteins interacting with PICK1 as well as the general structural requirements of PDZ interactions. The study also provides modeling methods to support in the drug design of ligands for PDZ domains, which may further aid in development of the family of PDZ domains as a drug target.     

Biochemistry for the benefit of humanity (practical achievements of my scientific work)

Science unites theory and practice, but theory is always in advance. Even our works (mentioned above) which are also important for practice and were awarded the State prizes could not be made without preliminary theoretical investigations. It should be said that our works with elaborated methods of therapy and drugs to treat chronic alcoholism, drug addiction, leucosis are rather of theoretical than of practical importance. Some our works which proved that carbon dioxide is the basis of life are also of especially great theoretical value. The paper deals with the investigations devoted to the problems of biochemistry in cattle breeding (the raising of fat content in milk; elaboration of the efficient method of fodder ensilage; raising of milk yield using the drug "Karboxilin"; development of the methods of isolation of crystalline glucose-oxidase and catalase used for clarifying blood) as well as to the problems of biochemistry in medicine (creation of the drug "Microcid", antileucosis drug "Corectin", drugs "Medichronal" and "Medicit" for treating alcoholism and drug addiction, drug "Namacit" for hindering the organism aging). Great attention is given to the problem of relations between the theoretical conception concerning the importance of CO2 in vital activity of human and animal organism and production of new drugs.     

Strategic genome-scale prioritization of unique drug targets: A case study of Streptococcus gordonii

The current reach of genomics extends facilitated identification of microbial virulence factors, a primary objective for antimicrobial drug and vaccine design. Many putative proteins are yet to be identified which can act as potent drug targets. There is lack and limitation of methods which appropriately combine several omics ways for putative and new drug target identification. The study emphasizes a combined bioinformatic and theoretical method of screening unique and putative drug targets, lacking similarity with experimentally reported essential genes and drug targets. Synteny based comparison was carried out with 11 streptococci considering S. gordonii as reference genome. It revealed 534 non-homologous genes of which 334 were putative. Similarity search against host proteome, metabolic pathway annotation and subcellular localization predication identified 16 potent drug targets. This is a first attempt of several combinational approaches of similarity search with target protein structural features for screening drug targets, yielding a pipeline which can be substantiated to other human pathogens.     

Use of gene networks for identifying and validating drug targets

We propose a new method for identifying and validating drug targets by using gene networks, which are estimated from cDNA microarray gene expression profile data. We created novel gene disruption and drug response microarray gene expression profile data libraries for the purpose of drug target elucidation. We use two types of microarray gene expression profile data for estimating gene networks and then identifying drug targets. The estimated gene networks play an essential role in understanding drug response data and this information is unattainable from clustering methods, which are the standard for gene expression analysis. In the construction of gene networks, we use the Bayesian network model. We use an actual example from analysis of the Saccharomyces cerevisiae gene expression profile data to express a concrete strategy for the application of gene network information to drug discovery.     

基于机器学习的药物-靶标相互作用预测*

近年来,随着计算机硬件、软件工具和数据丰度的不断突破,以机器学习为代表的人工智能技术在生物、基础医学和药学等领域的应用不断拓展和融合,极大地推动了这些领域的发展,尤其是药物研发领域的变革。其中,药物-靶标相互作用(drug-target interactions, DTI)的识别是药物研发领域中的重要难题和人工智能技术交叉融合的热门方向,研究人员在DTI预测方面做了大量的工作,构建了许多重要的数据库,开发或拓展了各类机器学习算法和工具软件。对基于机器学习的DTI预测的基本流程进行了介绍,并对利用机器学习预测DTI的研究进行了回顾,同时对不同的机器学习方法运用于DTI预测的优缺点进行了简单总结,以期对开发更加有效的预测算法和DTI预测的发展提供帮助。     

Enhanced Understanding of Pharmaceutical Materials Through Advanced Characterisation and Analysis

The impact of pharmaceutical materials properties on drug product quality and manufacturability is well recognised by the industry. An ongoing effort across industry and academia, the Manufacturing Classification System consortium, aims to gather the existing body of knowledge in a common framework to provide guidance on selection of appropriate manufacturing technologies for a given drug and/or guide optimization of the physical properties of the drug to facilitate manufacturing requirements for a given processing route. Simultaneously, material scientists endeavour to develop characterisation methods such as size, shape, surface area, density, flow and compactibility that enable a stronger understanding of materials powder properties. These properties are routinely tested drug product development and advances in instrumentation and computing power have enabled novel characterisation methods which generate larger, more complex data sets leading to a better understanding of the materials. These methods have specific requirements in terms of data management and analysis. An appropriate data management strategy eliminates time-consuming data collation steps and enables access to data collected for multiple methods and materials simultaneously. Methods ideally suited to extract information from large, complex data sets such as multivariate projection methods allow simpler representation of the variability contained within the data and easier interpretation of the key information it contains. In this review, an overview of the current knowledge and challenges introduced by modern pharmaceutical material characterisation methods is provided. Two case studies illustrate how the incorporation of multivariate analysis into the material sciences workflow facilitates a better understanding of materials.     

Advances in computational methods for ligand binding kinetics

Binding kinetic parameters can be correlated with drug efficacy, which in recent years led to the development of various computational methods for predicting binding kinetic rates and gaining insight into protein–drug binding paths and mechanisms. In this review, we introduce and compare computational methods recently developed and applied to two systems, trypsin–benzamidine and kinase–inhibitor complexes. Methods involving enhanced sampling in molecular dynamics simulations or machine learning can be used not only to predict kinetic rates, but also to reveal factors modulating the duration of residence times, selectivity, and drug resistance to mutations. Methods which require less computational time to make predictions are highlighted, and suggestions to reduce the error of computed kinetic rates are presented.     

Recognizing novel chemicals/drugs for anatomical therapeutic chemical classes with a heat diffusion algorithm

Drug Anatomical Therapeutic Chemical (ATC) classification system is a widely used and accepted drug classification system. It is recommended and maintained by World Health Organization (WHO). Each drug in this system is assigned one or more ATC codes, indicating which classes it belongs to in each of five levels. Given a chemical/drug, correct identification of its ATC codes in such system can be helpful to understand its therapeutic effects. Several computational methods have been proposed to identify the first level ATC classes for any drug. Most of them built multi-label classifiers in this regard. One previous study proposed a quite different scheme, which contained two network methods, based on shortest path (SP) and random walk with restart (RWR) algorithms, respectively, to infer novel chemicals/drugs for each first level class. However, due to the limitations of SP and RWR algorithms, there still exist lots of hidden chemicals/drugs that above two methods cannot discover. This study employed another classic network algorithm, Laplacian heat diffusion (LHD) algorithm, to construct a new computational method for recognizing novel latent chemicals/drugs of each first level ATC class. This algorithm was applied on a chemical network, which containing lots of chemical interaction information, to evaluate the associations of candidate chemicals/drugs and each ATC class. Three screening tests, which measured the specificity and association to one ATC class, followed to yield more reliable potential members for each class. Some hidden chemicals/drugs were recognized, which cannot be found out by previous methods, and they were extensively analyzed to confirm that they can be novel members in the corresponding ATC class.     

Rapid measurement of drug release from temperature-sensitive liposomes by electron paramagnetic resonance and radioisotope techniques

We have developed two new methods for quantifying drug release from temperature-sensitive liposomes. Large unilamellar vesicles were made by the reverse phase evaporation process. They contained a water-soluble electron paramagnetic resonance probe, trimethyl-4-amino-2,2,6,6-tetramethyl piperidine N-oxyl and the radioisotope cytosine-[3H]1-beta-D-arabinofuranoside in their aqueous compartment. Release of the electron paramagnetic resonance probe was measured by placing the liposomes in a solution of a spin label quenching agent, potassium ferricyanide, and monitoring the reduction in signal strength. The measurement of radioisotope released involved rapid ultracentrifugation of the liposomes after which the supernatant was tested for the presence of radioactivity. Both methods were found to be rapid and convenient ways of measuring drug release from temperature-sensitive liposomes and both methods gave comparable results. The radioisotope assay provides a direct measurement of drug leakage, whereas the electron spin resonance assay provides a continuous marker for liposome stability as a function of temperature.     

The onset of a revolution in drug discovery

Cell electrophysiology is extremely important to cell biology, medicine and drug development, and micro-electro-mechanical (MEMS)-based systems hold great promise to alleviate several problems associated with the current methods used by the pharmaceutical industry to probe drug development. Porterfield et al. at Purdue University have developed an integrated MEMS-based system, which offers considerable advantages compared with current electrophysiological systems and might have greater implications as an advanced method for drug development.     

Influence of Incorporation Methods on Partitioning Behavior of Lipophilic Drugs into Various Phases of a Parenteral Lipid Emulsion

The purpose of this study was to investigate the effect of drug incorporation methods on the partitioning behavior of lipophilic drugs in parenteral lipid emulsions. Four lipophilic benzodiazepines, alprazolam, clonazepam, diazepam, and lorazepam, were used as model drugs. Two methods were used to incorporate drugs into an emulsion: dissolving the compound in the oil phase prior to emulsification (de novo emulsification), and directly adding a concentrated solution of drug in a solubilizer to the emulsion base (extemporaneous addition). Based on the molecular structures and determination of the oil and aqueous solubilities and the partition coefficients of the drugs, the lipophilicity was ranked as diazepam > clonazepam > lorazepam > alprazolam. Ultracentrifugation was used to separate the emulsion into four phases, the oil phase, the phospholipid-rich phase, the aqueous phase and the mesophase, and the drug content in each phase was determined. Partitioning of diazepam, which has the highest lipophilicity and oil solubility among the four drugs, was unaffected by the drug incorporation method, with both methods giving a high proportion of drug in the inner oil phase and the phospholipid-rich phase, compared to the aqueous phase and mesophase. Partitioning of the less lipophilic drugs (alprazolam, clonazepam, and lorazepam) in the phases of the emulsion system was dependent on the method of incorporation and the drug solubility properties. Emulsions of the three drugs prepared by de novo emulsification exhibited higher drug localization in the phospholipid-rich phase compared to those made by extemporaneous addition. With the latter method, the drugs tended to localize in the outer aqueous phase and mesophase, with less deposition in the phospholipid-rich phase and no partitioning into the inner oil phase.     

Molecular tools for the rapid detection of drug resistance in animal trypanosomes

There are currently 17 African countries in which animal trypanocidal drug resistance has been reported. Large-scale surveys were carried out in only ten of them. The lack of baseline information is mainly due to the fact that the methods currently available for the detection of drug resistance are laborious, expensive and time consuming. In this review the mechanisms involved in resistance to isometamidium and diminazene will be discussed, together with some new molecular detection tools that have been developed recently enabling faster diagnosis of drug resistance than conventional laboratory or field tests.     

Prediction of chemical-protein interactions network with weighted network-based inference method

Chemical-protein interaction (CPI) is the central topic of target identification and drug discovery. However, large scale determination of CPI is a big challenge for in vitro or in vivo experiments, while in silico prediction shows great advantages due to low cost and high accuracy. On the basis of our previous drug-target interaction prediction via network-based inference (NBI) method, we further developed node- and edge-weighted NBI methods for CPI prediction here. Two comprehensive CPI bipartite networks extracted from ChEMBL database were used to evaluate the methods, one containing 17,111 CPI pairs between 4,741 compounds and 97 G protein-coupled receptors, the other including 13,648 CPI pairs between 2,827 compounds and 206 kinases. The range of the area under receiver operating characteristic curves was 0.73 to 0.83 for the external validation sets, which confirmed the reliability of the prediction. The weak-interaction hypothesis in CPI network was identified by the edge-weighted NBI method. Moreover, to validate the methods, several candidate targets were predicted for five approved drugs, namely imatinib, dasatinib, sertindole, olanzapine and ziprasidone. The molecular hypotheses and experimental evidence for these predictions were further provided. These results confirmed that our methods have potential values in understanding molecular basis of drug polypharmacology and would be helpful for drug repositioning.     

深度学习在药物活性预测研究中的应用

药物从研发到临床应用需要耗费较长的时间，研发期间的投入成本可高达十几亿元。而随着医药研发与人工智能的结合以及生物信息学的飞速发展，药物活性相关数据急剧增加，传统的实验手段进行药物活性预测已经难以满足药物研发的需求。借助算法来辅助药物研发，解决药物研发中的各种问题能够大大推动药物研发进程。传统机器学习方法尤其是随机森林、支持向量机和人工神经网络在药物活性方面能够达到较高的预测精度。深度学习由于具有多层神经网络，模型可以接收高维的输入变量且不需要人工限定数据输入特征，可以拟合较为复杂的函数模型，应用于药物研发可以进一步提高各个环节的效率。在药物活性预测中应用较为广泛的深度学习模型主要是深度神经网络（deep neural networks，DNN）、循环神经网络（recurrent neural networks，RNN）和自编码器（auto encoder，AE），而生成对抗网络（generative adversarial networks，GAN）由于其生成数据的能力常常被用来和其他模型结合进行数据增强。近年来深度学习在药物分子活性预测方面的研究和应用综述表明，深度学习模型的准确度和效率均高于传统实验方法和传统机器学习方法。因此，深度学习模型有望成为药物研发领域未来十年最重要的辅助计算模型。     

Model of distribution of anticancer agents transplanted into the brain tissue using the random walk method.

A Monte-Carlo approach to analysis of dispersion in the tissue of a locally administered drug is presented. The distribution of a drug in the tissue is simulated as a distribution of randomly walking particles. The approach is demonstrated on a simple situation for which both experimental results and an analytical solution are known. The approach can be used in situations, where common numeric methods are difficult to use, especially for analyses of drug transport in an inhomogeneous space, and problems with complex boundary conditions, e.g. in analyses of dispersion of anticancer agents locally applied into tumours.