Screening of selective histone deacetylase inhibitors by proteochemometric modeling

How to approach machine learning-based prediction of drug/compound-target interactions

The identification of drug/compound-target interactions (DTIs) constitutes the basis of drug discovery, for which computational predictive approaches have been developed. As a relatively new data-driven paradigm, proteochemometric (PCM) modeling utilizes both protein and compound properties as a pair at the input level and processes them via statistical/machine learning. The representation of input samples (i.e., proteins and their ligands) in the form of quantitative feature vectors is crucial for the extraction of interaction-related properties during the artificial learning and subsequent prediction of DTIs. Lately, the representation learning approach, in which input samples are automatically featurized via training and applying a machine/deep learning model, has been utilized in biomedical sciences. In this study, we performed a comprehensive investigation of different computational approaches/techniques for protein featurization (including both conventional approaches and the novel learned embeddings), data preparation and exploration, machine learning-based modeling, and performance evaluation with the aim of achieving better data representations and more successful learning in DTI prediction. For this, we first constructed realistic and challenging benchmark datasets on small, medium, and large scales to be used as reliable gold standards for specific DTI modeling tasks. We developed and applied a network analysis-based splitting strategy to divide datasets into structurally different training and test folds. Using these datasets together with various featurization methods, we trained and tested DTI prediction models and evaluated their performance from different angles. Our main findings can be summarized under 3 items: (i) random splitting of datasets into train and test folds leads to near-complete data memorization and produce highly over-optimistic results, as a result, should be avoided, (ii) learned protein sequence embeddings work well in DTI prediction and offer high potential, despite interaction-related properties (e.g., structures) of proteins are unused during their self-supervised model training, and (iii) during the learning process, PCM models tend to rely heavily on compound features while partially ignoring protein features, primarily due to the inherent bias in DTI data, indicating the requirement for new and unbiased datasets. We hope this study will aid researchers in designing robust and high-performing data-driven DTI prediction systems that have real-world translational value in drug discovery.

Emerging Promise of Computational Techniques in Anti-Cancer Research: At a Glance

Research on the immune system and cancer has led to the development of new medicines that enable the former to attack cancer cells. Drugs that specifically target and destroy cancer cells are on the horizon; there are also drugs that use specific signals to stop cancer cells multiplying. Machine learning algorithms can significantly support and increase the rate of research on complicated diseases to help find new remedies. One area of medical study that could greatly benefit from machine learning algorithms is the exploration of cancer genomes and the discovery of the best treatment protocols for different subtypes of the disease. However, developing a new drug is time-consuming, complicated, dangerous, and costly. Traditional drug production can take up to 15 years, costing over USD 1 billion. Therefore, computer-aided drug design (CADD) has emerged as a powerful and promising technology to develop quicker, cheaper, and more efficient designs. Many new technologies and methods have been introduced to enhance drug development productivity and analytical methodologies, and they have become a crucial part of many drug discovery programs; many scanning programs, for example, use ligand screening and structural virtual screening techniques from hit detection to optimization. In this review, we examined various types of computational methods focusing on anticancer drugs. Machine-based learning in basic and translational cancer research that could reach new levels of personalized medicine marked by speedy and advanced data analysis is still beyond reach. Ending cancer as we know it means ensuring that every patient has access to safe and effective therapies. Recent developments in computational drug discovery technologies have had a large and remarkable impact on the design of anticancer drugs and have also yielded useful insights into the field of cancer therapy. With an emphasis on anticancer medications, we covered the various components of computer-aided drug development in this paper. Transcriptomics, toxicogenomics, functional genomics, and biological networks are only a few examples of the bioinformatics techniques used to forecast anticancer medications and treatment combinations based on multi-omics data. We believe that a general review of the databases that are now available and the computational techniques used today will be beneficial for the creation of new cancer treatment approaches.

[Trichostatin A suppresses up-regulation of histone deacetylase 4 and reverses differential expressions of miRNAs in the spinal cord of rats with chronic constrictive injury]

OBJECTIVE

To explore the analgesic mechanism of intrathecal trichostatin A (TSA) injection in a rat model of neuropathic pain induced by chronic constrictive injury (CCI).

METHODS

Male SD rats were randomized into sham operation+ DMSO group (group S), CCI +DMSO group (group C), CCI +10 μg TSA group (group T), and in the latter two groups, rat models of neuropathic pain were established induced by CCI. The rats were given intrathecal injections of 10 μL 5% DMSO or 10 μg TSA (in 5% DMSO) once a day on days 7 to 9 after CCI or sham operation. The rats were euthanized after behavioral tests on day 10, and the lumbar segment of the spinal cord was sampled to determine the expression of histone deacetylase 4 (HDAC4) protein and mRNA and detect the differentially expressed miRNAs using a miRNA chip. MiR-190b-5p and miR-142-3p were selected for validation of the results using RT-qPCR.

RESULTS

Compared with those in group S, the rats in group C showed significantly decreased paw withdrawal mechanical threshold (PWMT) from day 3 to day 10 after CCI (P < 0.05); intrathecal injection of TSA significantly reversed the reduction of PWMT following CCI (P < 0.05). Positive HDAC4 expression was detected mainly in the cytoplasm of the neurons in the gray matter of the spinal cord, and was obviously up-regulated after CCI (Ρ < 0.05). Intrathecal injection of TSA significantly suppressed CCI-induced up-regulation of HDAC4 at 10 days after the operation (P < 0.05). Compared with the miRNA profile in group S, miRNA profiling identified 83 differentially expressed miRNAs in group C (fold change ≥2 or ≤0.5, P < 0.05); TSA treatment reversed the expressions of 58 of the differentially expressed miRNAs following CCI, including 41 miRNAs that were decreased after CCI but up-regulated following TSA treatment. The results of real-time PCR validated the changes in the expressions of miR-190b-5p and miR-142-3p.

CONCLUSIONS

TSA suppresses CCI-induced up-regulation of HDAC4 and reverses differential expressions of miRNAs in the spinal cord of rats, which may contribute to the analgesic effect of TSA on neuropathic pain.

Recent applications of deep learning and machine intelligence on in silico drug discovery: methods, tools and databases

The identification of interactions between drugs/compounds and their targets is crucial for the development of new drugs. In vitro screening experiments (i.e. bioassays) are frequently used for this purpose; however, experimental approaches are insufficient to explore novel drug-target interactions, mainly because of feasibility problems, as they are labour intensive, costly and time consuming. A computational field known as 'virtual screening' (VS) has emerged in the past decades to aid experimental drug discovery studies by statistically estimating unknown bio-interactions between compounds and biological targets. These methods use the physico-chemical and structural properties of compounds and/or target proteins along with the experimentally verified bio-interaction information to generate predictive models. Lately, sophisticated machine learning techniques are applied in VS to elevate the predictive performance. The objective of this study is to examine and discuss the recent applications of machine learning techniques in VS, including deep learning, which became highly popular after giving rise to epochal developments in the fields of computer vision and natural language processing. The past 3 years have witnessed an unprecedented amount of research studies considering the application of deep learning in biomedicine, including computational drug discovery. In this review, we first describe the main instruments of VS methods, including compound and protein features (i.e. representations and descriptors), frequently used libraries and toolkits for VS, bioactivity databases and gold-standard data sets for system training and benchmarking. We subsequently review recent VS studies with a strong emphasis on deep learning applications. Finally, we discuss the present state of the field, including the current challenges and suggest future directions. We believe that this survey will provide insight to the researchers working in the field of computational drug discovery in terms of comprehending and developing novel bio-prediction methods.

Finding the molecular scaffold of nuclear receptor inhibitors through high-throughput screening based on proteochemometric modelling

Nuclear receptors (NR) are a class of proteins that are responsible for sensing steroid and thyroid hormones and certain other molecules. In that case, NR have the ability to regulate the expression of specific genes and associated with various diseases, which make it essential drug targets. Approaches which can predict the inhibition ability of compounds for different NR target should be particularly helpful for drug development. In this study, proteochemometric modelling was introduced to analysis the bioactivity between chemical compounds and NR targets. Results illustrated the ability of our PCM model for high-throughput NR-inhibitor screening after evaluated on both internal (AUC > 0.870) and external (AUC > 0.746) validation set. Moreover, in-silico predicted bioactive compounds were clustered according to structure similarity and a series of representative molecular scaffolds can be derived for five major NR targets. Through scaffolds analysis, those essential bioactive scaffolds of different NR target can be detected and compared. Generally, the methods and molecular scaffolds proposed in this article can not only help the screening of potential therapeutic NR-inhibitors but also able to guide the future NR-related drug discovery.

The nature of the GRE influences the screening for GR-activity enhancing modulators

Glucocorticoid resistance (GCR), i.e. unresponsiveness to the beneficial anti-inflammatory activities of the glucocorticoid receptor (GR), poses a serious problem in the treatment of inflammatory diseases. One possible solution to try and overcome GCR, is to identify molecules that prevent or revert GCR by hyper-stimulating the biological activity of the GR. To this purpose, we screened for compounds that potentiate the dexamethasone (Dex)-induced transcriptional activity of GR. To monitor GR transcriptional activity, the screen was performed using the lung epithelial cell line A549 in which a glucocorticoid responsive element (GRE) coupled to a luciferase reporter gene construct was stably integrated. Histone deacetylase inhibitors (HDACi) such as Vorinostat and Belinostat are two broad-spectrum HDACi that strongly increased the Dex-induced luciferase expression in our screening system. In sharp contrast herewith, results from a genome-wide transcriptome analysis of Dex-induced transcripts using RNAseq, revealed that Belinostat impairs the ability of GR to transactivate target genes. The stimulatory effect of Belinostat in the luciferase screen further depends on the nature of the reporter construct. In conclusion, a profound discrepancy was observed between HDACi effects on two different synthetic promoter-luciferase reporter systems. The favorable effect of HDACi on gene expression should be evaluated with care, when considering them as potential therapeutic agents. GEO accession number GSE96649.

A phase II trial of valproic acid in patients with advanced, radioiodine-resistant thyroid cancers of follicular cell origin

OBJECTIVE

Valproic acid (VA) is a histone deacetylase (HDAC) inhibitor that has antiproliferative effects on several types of cancer, including thyroid cancer. In addition, VA has been reported to upregulate the sodium-iodine symporter in thyroid cancer cells and increases radioiodine uptake in preclinical studies. The aim of this study was to assess the antiproliferative effects of VA and to evaluate if VA can increase the radioiodine uptake in patients with advanced, radioiodine-negative thyroid cancer.

DESIGN

An open-label Simon two-stage phase II trial.

PATIENTS AND MEASUREMENTS

Valproic acid was administered orally, and doses were adjusted to maintain serum trough levels between 50 and 100 mg/l for 10 weeks, followed by injections of recombinant human thyroid-stimulating hormone and a radioiodine uptake scan. Anatomical imaging studies were performed at week 16 to assess tumour response and radioiodine therapy in patients with increased radioiodine uptake.

RESULTS

Thirteen patients with a median age of 66 years (50-78 years) were enrolled and evaluated. Seven patients had papillary thyroid cancer (PTC), two had follicular variant PTC, two had follicular thyroid cancer, and two had Hürthle cell carcinoma. None of the 10 patients who completed the 10-week treatment had increased radioiodine uptake at their tumour sites. Three patients were taken off the study prior to the 10-week radioiodine uptake scan: one with grade-3 hepatic toxicity, one with disease progression and one for noncompliance. Four of 13 patients had decreased stimulated serum thyroglobulin with VA treatment. None of the patients had complete or partial responses based on Response Evaluation Criteria in Solid Tumors (RECIST), and six patients had disease progression.

CONCLUSIONS

Valproic acid does not increase radioiodine uptake and does not have anticancer activity in patients with advanced, radioiodine-negative thyroid cancer of follicular cell origin.

Proteochemometric Modeling of the Interaction Space of Carbonic Anhydrase and its Inhibitors: An Assessment of Structure-based and Sequence-based Descriptors

Due to its physiological and clinical roles, carbonic anhydrase (CA) is one of the most interesting case studies. There are different classes of CAinhibitors including sulfonamides, polyamines, coumarins and dithiocarbamates (DTCs). However, many of them hardly act as a selective inhibitor against a specific isoform. Therefore, finding highly selective inhibitors for different isoforms of CA is still an ongoing project. Proteochemometrics modeling (PCM) is able to model the bioactivity of multiple compounds against different isoforms of a protein. Therefore, it would be extremely applicable when investigating the selectivity of different ligands towards different receptors. Given the facts, we applied PCM to investigate the interaction space and structural properties that lead to the selective inhibition of CA isoforms by some dithiocarbamates. Our models have provided interesting structural information that can be considered to design compounds capable of inhibiting different isoforms of CA in an improved selective manner. Validity and predictivity of the models were confirmed by both internal and external validation methods; while Y-scrambling approach was applied to assess the robustness of the models. To prove the reliability and the applicability of our findings, we showed how ligands-receptors selectivity can be affected by removing any of these critical findings from the modeling process.

The recent progress in proteochemometric modelling: focusing on target descriptors, cross-term descriptors and application scope

As an extension of the conventional quantitative structure activity relationship models, proteochemometric (PCM) modelling is a computational method that can predict the bioactivity relations between multiple ligands and multiple targets. Traditional PCM modelling includes three essential elements: descriptors (including target descriptors, ligand descriptors and cross-term descriptors), bioactivity data and appropriate learning functions that link the descriptors to the bioactivity data. Since its appearance, PCM modelling has developed rapidly over the past decade by taking advantage of the progress of different descriptors and machine learning techniques, along with the increasing amounts of available bioactivity data. Specifically, the new emerging target descriptors and cross-term descriptors not only significantly increased the performance of PCM modelling but also expanded its application scope from traditional protein-ligand interaction to more abundant interactions, including protein-peptide, protein-DNA and even protein-protein interactions. In this review, target descriptors and cross-term descriptors, as well as the corresponding application scope, are intensively summarized. Additionally, we look forward to seeing PCM modelling extend into new application scopes, such as Target-Catalyst-Ligand systems, with the further development of descriptors, machine learning techniques and increasing amounts of available bioactivity data.

Proteochemometric modeling of the antigen-antibody interaction: new fingerprints for antigen, antibody and epitope-paratope interaction

Despite the high specificity between antigen and antibody binding, similar epitopes can be recognized or cross-neutralized by paratopes of antibody with different binding affinities. How to accurately characterize this slight variation which may or may not change the antigen-antibody binding affinity is a key issue in this area. In this report, by combining cylinder model with shell structure model, a new fingerprint was introduced to describe both the structural and physical-chemical features of the antigen and antibody protein. Furthermore, beside the description of individual protein, the specific epitope-paratope interaction fingerprint (EPIF) was developed to reflect the bond and the environment of the antigen-antibody interface. Finally, Proteochemometric Modeling of the antigen-antibody interaction was established and evaluated on 429 antigen-antibody complexes. By using only protein descriptors, our model achieved the best performance (R2=0.91,Qtest2=0.68) among peers. Further, together with EPIF as a new cross-term, our model (R2=0.92,Qtest2=0.74) can significantly outperform peers with multiplication of ligand and protein descriptors as a cross-term (R2≤0.81,Qtest2≤0.44). Results illustrated that: 1) our newly designed protein fingerprints and EPIF can better describe the antigen-antibody interaction; 2) EPIF is a better and specific cross-term in Proteochemometric Modeling for antigen-antibody interaction. The fingerprints designed in this study will provide assistance to the description of antigen-antibody binding, and in future, it may be valuable help for the high-throughput antibody screening. The algorithm is freely available on request.

When drug discovery meets web search: Learning to Rank for ligand-based virtual screening

Background

The rapid increase in the emergence of novel chemical substances presents a substantial demands for more sophisticated computational methodologies for drug discovery. In this study, the idea of Learning to Rank in web search was presented in drug virtual screening, which has the following unique capabilities of 1). Applicable of identifying compounds on novel targets when there is not enough training data available for these targets, and 2). Integration of heterogeneous data when compound affinities are measured in different platforms.

Results

A standard pipeline was designed to carry out Learning to Rank in virtual screening. Six Learning to Rank algorithms were investigated based on two public datasets collected from Binding Database and the newly-published Community Structure-Activity Resource benchmark dataset. The results have demonstrated that Learning to rank is an efficient computational strategy for drug virtual screening, particularly due to its novel use in cross-target virtual screening and heterogeneous data integration.

Conclusions

To the best of our knowledge, we have introduced here the first application of Learning to Rank in virtual screening. The experiment workflow and algorithm assessment designed in this study will provide a standard protocol for other similar studies. All the datasets as well as the implementations of Learning to Rank algorithms are available at http://www.tongji.edu.cn/~qiliu/lor_vs.html.

Graphical Abstract

Polypharmacology modelling using proteochemometrics (PCM): recent methodological developments, applications to target families, and future prospects

Proteochemometric (PCM) modelling is a computational method to model the bioactivity of multiple ligands against multiple related protein targets simultaneously.

Proteochemometric modeling in a Bayesian framework

Proteochemometrics (PCM) is an approach for bioactivity predictive modeling which models the relationship between protein and chemical information. Gaussian Processes (GP), based on Bayesian inference, provide the most objective estimation of the uncertainty of the predictions, thus permitting the evaluation of the applicability domain (AD) of the model. Furthermore, the experimental error on bioactivity measurements can be used as input for this probabilistic model.

In this study, we apply GP implemented with a panel of kernels on three various (and multispecies) PCM datasets. The first dataset consisted of information from 8 human and rat adenosine receptors with 10,999 small molecule ligands and their binding affinity. The second consisted of the catalytic activity of four dengue virus NS3 proteases on 56 small peptides. Finally, we have gathered bioactivity information of small molecule ligands on 91 aminergic GPCRs from 9 different species, leading to a dataset of 24,593 datapoints with a matrix completeness of only 2.43%.

GP models trained on these datasets are statistically sound, at the same level of statistical significance as Support Vector Machines (SVM), with R02 values on the external dataset ranging from 0.68 to 0.92, and RMSEP values close to the experimental error. Furthermore, the best GP models obtained with the normalized polynomial and radial kernels provide intervals of confidence for the predictions in agreement with the cumulative Gaussian distribution. GP models were also interpreted on the basis of individual targets and of ligand descriptors. In the dengue dataset, the model interpretation in terms of the amino-acid positions in the tetra-peptide ligands gave biologically meaningful results.

Histone deacetylase inhibitory effect of Brazilian propolis and its association with the antitumor effect in Neuro2a cells

Propolis is a resinous product produced by honey bees and is known to have antitumor functions. On the other hand, histone deacetylase (Hdac) inhibitors have recently attracted attention for their antitumor effects. In this study, we examined whether Brazilian green propolis has an Hdac inhibitory activity and its contribution on antitumor effects. By in vitro Hdac activity assay, Brazilian propolis extract (BPE) significantly inhibited the enzyme activity. Actually, BPE treatment increased the intracellular histone acetylation in Neuro2a cells. Regarding antitumor effect in Neuro2a cells, BPE treatment significantly decreased cell viability. An Hdac activator theophylline significantly attenuated the effect. Then, we analyzed whether the decreasing effect on cell number was caused by cell death or growth retardation. By live/dead cell staining, BPE treatment significantly increased the dead cell number. By cell cycle analysis, BPE treatment retarded cell cycle at the M-phase. Both of these cellular effects were suppressed by addition of theophylline. These data indicate that BPE induced both cell death and growth retardation via Hdac inhibitory activity. We demonstrated that Brazilian propolis bears regulatory functions on histone acetylation via Hdac inhibition, and the effect contributes antitumor functions. Our data suggest that intake of Brazilian propolis shows preventing effects against cancer.

Energy-optimised pharmacophore approach to identify potential hotspots during inhibition of Class II HDAC isoforms

Histone deacetylases (HDACs) are conjugated enzymes that modulate chromatin architecture by deacetylating lysine residues on the histone tails leading to transcriptional repression. Pharmacological interventions of these enzymes with small molecule inhibitors called Histone deacetylase inhibitors (HDACi) have shown enhanced acetylation of the genome and are hence emerging as potential targets at the clinic. Type-specific inhibition of Class II HDACs has shown enhanced therapeutic benefits against developmental and neurodegenerative disorders. However, the structural identity of class-specific isoforms limits the potential of their inhibitors in precise targeting of their enzymes. Diverse strategies have been implemented to recognise the features in HDAC enzymes which may help in identifying isoform specificity factors. This work attempts a computational approach that combines in silico docking and energy-optimised pharmacophore (E-pharmacophore) mapping of 18 known HDAC inhibitors and has identified structural variations that regulate their interactions against the six Class II HDAC enzymes considered for the study. This combined approach establishes that inhibitors possessing higher number of aromatic rings in different structural regions might function as potent inhibitors, while inhibitors with scarce ring structures might point to compromised potency. This would aid the rationale for chemical optimisation and design of isoform selective HDAC inhibitors with enhanced affinity and therapeutic efficiency.

Histone deacetylase (HDAC) inhibitory and antiproliferative activities of phenolic-rich extracts derived from the rhizome of Hydnophytum formicarum Jack.: sinapinic acid acts as HDAC inhibitor

Background

The rhizome of Hydnophytum formicarum Jack., a medicinal plant known in Thai as Hua-Roi-Roo, has been used in Thai traditional herbal medicine for treatment of cancer. We assessed the ability of its ethanolic and phenolic-rich extracts and its major phenolic compound, sinapinic acid, possessing histone deacetylase (HDAC) inhibitory activity to inhibit proliferation of 5 human cancer cell lines.

Methods

HeLa cells were used to study HDAC inhibitory activity of the extracts, sinapinic acid, and a well-known HDAC inhibitor sodium butyrate. Five human cancer cell lines and one non-cancer cell line were used to study antiproliferative activities of the plant extracts, sinapinic acid and sodium butyrate, comparatively.

Results

Results indicated that ethanolic and phenolic-rich extracts of H. formicarum Jack. rhizome possessed both antiproliferative activity and HDAC inhibitory activity in HeLa cells. Sinapinic acid, despite its lower HDAC inhibitory activity than the well-known HDAC inhibitor sodium butyrate, inhibited the growth of HeLa and HT29 cells more effectively than sodium butyrate. However, sinapinic acid inhibited the growth of HCT116 and Jurkat cells less effectively than sodium butyrate. The non-cancer cell line (Vero cells) and breast cancer cell line (MCF-7 cells) appeared to be resistant to both sinapinic acid and sodium butyrate. The growth inhibitory effects of the ethanolic and phenolic-rich extracts and sinapinic acid in HeLa cells were mediated by induction of apoptosis.

Conclusions

The results of this study support the efficacy of H. formicarum Jack. rhizome ethanolic and phenolic-rich extracts for the treatment of cervical cancer, colon cancer, and T- cell leukemia in an alternative medicine. Further studies of other active ingredients from this plant are needed.