Current Topics in Medicinal Chemistry - Volume 14, Issue 3, 2014

The continued development of computational and synthetic methods has enabled the enumeration or preparation of a nearly endless universe of chemical structures. Nevertheless, the ability of this chemical universe to deliver small molecules that can both modulate biological targets and have drug-like physicochemical properties continues to be a topic of interest to the pharmaceutical industry and academic researchers alike. The chemical space described by public, commercial, in-house and virtual compound collections has been interrogated by multiple approaches including biochemical, cellular and virtual screening, diversity analysis, and in-silico profiling. However, current drugs and known chemical probes derived from these efforts are contained within a remarkably small volume of the predicted chemical space. Access to more diverse classes of chemical scaffolds that maintain the properties relevant for drug discovery is certainly needed to meet the increasing demands for pharmaceutical innovation. The Lilly Open Innovation Drug Discovery platform (OIDD) was designed to tackle barriers to innovation through the identification of novel molecules active in relevant disease biology models. In this article we will discuss several computational approaches towards describing novel, biologically active, drug-like chemical space and illustrate how the OIDD program may facilitate access to previously untapped molecules that may aid in the search for innovative pharmaceuticals.

This article describes the unique industrial - academic collaboration that has been running for four years between Janssen Pharmaceutica NV and the Vanderbilt Center for Neuroscience Drug Discovery (VCNDD) towards identifying the next generation of schizophrenia therapeutics. This was a true collaboration, with both entities engaged in chemistry, In vitro pharmacology, DMPK and In vivo behavioral pharmacology, and aligned to deliver a first-in-class clinical candidate (NME) and additional back-up molecules. Notably, a first NME was delivered in a rapid timeframe and targeted the novel mechanism of mGlu5 positive allosteric modulation. As with any true collaboration, both sides brought unique skills to the table - Janssen leveraged deep drug discovery expertise and infrastructure, while Vanderbilt brought deep knowledge of the chemistry and pharmacology of the target in addition to the ability to provide deep scientific insight into the mechanism behind target modulation. In this article, we will discuss the science which drove our collaboration as well as some key lessons learned.

The business model for medical therapy development has changed drastically. Large companies that once conducted their own Research and Development (R&D) and funded all the preclinical studies, all phases of clinical development and marketing of the products are increasingly turning to others for more and more of the earlier work in hopes of being able to in-license a de-risked program well downstream, take it through the final phases of clinical development and into the marketplace. This new paradigm has required patient-advocacy foundations, especially in the rare-disease space, to become far more effective in building relationships with all the players along the therapy-development pathway -- academic scientists, government agencies, other foundations with overlapping interests, biotechs, small biopharmaceutical entities and even the larger industry companies. From the perspective of the patient-advocacy community, these increasingly essential public-private partnerships have taken on the nature of what could be called joint-venture philanthropy and involve a broad spectrum of collaborations and financial relationships between foundations and industry partners that are not without concerns about potential conflicts of interest.

Partnerships between industry and academia are becoming increasingly complex and relevant in the drive to discover innovative new medicines. We describe the structure of the collaboration between the University of California – San Francisco – Small Molecule Discovery Center (UCSF-SMDC) and Genentech to develop chemical matter that inhibits the activity of caspase-6. We focus on the scientific basis for the partnership and how the orientation- and transactionrelated barriers were overcome. We describe the division of labor that allowed two groups to operate as a unified team to generate multiple chemical series with distinct mechanisms of action. The successful structure of the agreement serves as a model for future collaborations at both institutions.

The National Institutes of Health (NIH), academic medical centers and industry have a long and productive history in collaborating together. Decreasing R&D budgets in both the private and public sector have made the need for such collaborations paramount to reduce the risk of further declines in the number of innovative drugs reaching the market to address pressing public health needs. Doing more with less has forced both industry and public sector research institutions (PSRIs) to leverage resources and expertise in order to de-risk projects. In addition, it provides an opportunity to envision and implement new approaches to accomplish these goals. We discuss several of these innovative collaborations and partnerships at the NIH that demonstrate how the NIH and industry are working together to strengthen the drug development pipeline.

In 2010, the National Institutes of Health (NIH) established the Therapeutics for Rare and Neglected Diseases (TRND) program within the National Center for Advancing Translational Sciences (NCATS), which was created to stimulate drug discovery and development for rare and neglected tropical diseases through a collaborative model between the NIH, academic scientists, nonprofit organizations, and pharmaceutical and biotechnology companies. This paper describes one of the first TRND programs, the development of 2-hydroxypropyl-β-cyclodextrin (HP-β-CD) for the treatment of Niemann-Pick disease type C1 (NPC1). NPC is a neurodegenerative, autosomal recessive rare disease caused by a mutation in either the NPC1 (about 95% of cases) or the NPC2 gene (about 5% of cases). These mutations affect the intracellular trafficking of cholesterol and other lipids, which leads to a progressive accumulation of unesterified cholesterol and glycosphingolipids in the CNS and visceral organs. Affected individuals typically exhibit ataxia, swallowing problems, seizures, and progressive impairment of motor and intellectual function in early childhood, and usually die in adolescence. There is no disease modifying therapy currently approved for NPC1 in the US. A collaborative drug development program has been established between TRND, public and private partners that has completed the pre-clinical development of HP-β-CD through IND filing for the current Phase I clinical trial that is underway. Here we discuss how this collaborative effort helped to overcome scientific, clinical and financial challenges facing the development of new drug treatments for rare and neglected diseases, and how it will incentivize the commercialization of HP-β-CD for the benefit of the NPC patient community.

To catalyze research on historically untargeted protein kinases, we created the PKIS, an annotated set of 367 small molecule kinase inhibitors. The set has been widely distributed to academic collaborators as an open access tool. It has been used to identify chemical starting points for development of chemical probes for orphan kinases and to investigate kinase signaling in high content phenotypic assays. Access to the set comes with few restrictions other than the requirement that assay results be released into the public domain for the benefit of the entire research community. Examples from the efforts of several collaborators are summarized.

Biological systems are not only controlled by the abundance of individual proteins, but also by the formation of complexes and the dynamics of protein-protein interactions. The identification of the components of protein complexes can be obtained by shotgun proteomics using affinity purification coupled to mass spectrometry. Such studies include the analyses of several samples and experimental controls in order to discriminate true specific interactions from unspecific interactions and contaminants. However, shotgun proteomics have limited quantification capabilities for low abundant proteins on large sample sets due to the undersampling and the stochastic precursor ion selection. In this context, targeted proteomics constitutes a powerful analytical tool to systematically detect and quantify peptides in multiple samples, for instance those obtained from affinity purification experiments. Hypothesis-driven strategies have mainly relied on the selected reaction monitoring (SRM) technique performed on triple quadrupole instruments, which enables highly selective and sensitive measurements of peptides, acting as surrogates of the pre-selected proteins, over a wide range of concentrations. More recently, novel quantitative methods based on high resolution instruments, such as the parallel reaction monitoring (PRM) technique implemented on the quadrupole-orbitrap instrument, have arisen and provided alternatives to perform quantitative analyses with enhanced selectivity.The application of targeted proteomics to protein-protein interaction experiments from plasma and other physiological fluid samples and the inclusion of parallel reaction monitoring (PRM), combined with other recent technology developments opens a vast area for clinical application of proteomics. It is anticipated that it will reveal valuable information about specific, individual, responses against drugs, exogenous proteins or pathogens.

Interferons (IFNs) are proteins of the family of cytokines. Their antiproliferative function has been taken into account for several clinical therapies against malignant diseases. In this family, IFNs α and γ have demonstrated the highest antitumor effects. HerberPAG® is a new co-formulation with IFNs, α2b and γ. It has been obtained to increase the antiproliferative effect of individual IFNs and decrease their associated toxicity. Glioblastoma multiforme (GBM) is the most common primary brain tumor and one of the most deadly forms of cancer. The objective of the present work is to obtain insights into the regulation of Interferon-STAT-pathways and apoptosis in U87MG, at the transcriptional level. As a pharmacogenomic strategy we quantified mRNAs levels in vitro by quantitative PCR, using the cell line U87MG as a model. Some of the genes involved in the first steps of IFNs signaling pathways (stat1 and stat3) and apoptosis events (tp53, bax, bcl-2, bad, caspase3 (casp3), caspase8 (casp8) and caspase9 (casp9)) were studied. The detected mRNAs expression pattern for stat1and stat3 indicates a higher tumor suppressor activity of HerberPAG® compared to individuals IFNs. The up-regulation of tp53, bax, bad, casp3, casp8 and casp9 genes and the down regulation of bcl-2 gen, after the treatment with HerberPAG® show a pro-apoptotic function. HerberPAG® gene-induced profile shows an advantage in relation to IFN α2b and γ with a higher stat1 expression and a downregulation of bcl-2 which increases bax:bcl-2 ratio. The regulation of genes involved in IFN-STAT-pathways and apoptosis may be the first evidences to explain the increased antiproliferative properties of this co-formulation.

Breast cancer is the leading cause of cancer related deaths in women. Most breast cancers stem from mammary ductal cells that secrete nipple aspirate fluid (NAF), a biological sample that contains proteins associated with the tumor microenvironment. In this study, NAF samples from both breasts of 7 Brazilian patients with unilateral breast cancer were analyzed. These samples were systematically compared using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and two-dimensional fluorescence difference gel electrophoresis (2D-DIGE); substantial qualitative individual differences were observed. In general, when NAF samples were compared from both breasts within the same patient their electrophoretic patterns were very similar, regardless of their cancer status. A comparison of all patients identified 2 main NAF protein profiles. The HomEP, homogeneous expression profile, was characterized by typical SDS-PAGE and 2D-DIGE protein patterns that were observed in patients with a good breast cancer prognosis and were similar to previous Type I NAF classifications that used one-dimensional electrophoresis. The HetEP, heterogeneous expression profile, was characterized by distinct protein patterns that have not been reported in previous studies and have been primarily observed in breast cancer patients with a poor prognosis. The NAF samples were rich in metal-dependent proteolytic enzymes, as visualized by SDS-PAGE zymography. They varied qualitatively with respect to their gelatinolytic band distribution. However, there were no correlations between these characteristics and the pathologic features of these tumors. A comparative analysis of NAF samples taken from each breast in a single patient showed conserved zymographic patterns. In conclusion, the present study highlights important distinctions in the protein content of individual NAF samples and provides insight into the composition of the tumor microenvironment. These data reinforce breast cancer as a heterogeneous disease with a diverse natural history, which is becoming increasingly evident through other recent studies.

The numerous efforts invested in the identification of biomarkers for neurodegenerative and neuropsychiatric disorders, such as Parkinson’s disease and schizophrenia, are justified because these disorders affect several million people worldwide. Although genetic implications and the role of the environment have been shown in the progression of those disorders, together with anatomical and neurochemical characteristics, an integrated view of the biochemical pathways involved in the pathophysiology of these disorders is still being unraveled. The use of proteomic methodologies, molecular mechanisms and potential biomarker candidates for the prognosis, diagnosis and treatment of brain disorders has been discussed. Similar methodologies can be applied for the large-scale identification of peptides to characterize the brain peptidome with the aim of closing the knowledge gaps that remain. Brain cells contain a large number of peptides that play pivotal roles in cell communication. Peptidome studies have recently identified more than 800 peptides in mouse brain extracts, with half of them derived from secretory pathways. For example, several of these peptides were identified as bioactive neuropeptides that activate G-coupled receptors. In addition, intracellular peptides derived from nuclear, cytosolic and mitochondrial proteins have been identified, including the hemopressins, which act with high selectivity for the cannabinoid receptor type 1. Considering the importance of peptides in cell signaling, the present review intends to discuss the recent findings of the peptidome field, focusing on Parkinson’s disease and schizophrenia. New approaches to evaluate intracellular peptide signaling at the protein-protein interaction level and the future perspectives of peptides as intracellular modulators of signal transduction are explored.

Melanoma is the third most common brain metastasis cause in the United States as it has a relatively high susceptibility to metastasize to the central nervous system. Among the different origins for brain metastasis, those originating from primary gastric melanomas are extremely rare. Here, we compare protein profiles obtained from formalin-fixed paraffin- embedded (FFPE) tissues of a primary gastric melanoma with its meningeal metastasis. For this, the contents of a microscope slide were scraped and ultimately analyzed by nano-chromatography coupled online with tandem mass spectrometry using an Orbitrap XL. Our results disclose 184 proteins uniquely identified in the primary gastric melanoma, 304 in the meningeal metastasis, and 177 in common. Notably, we indentified several enzymes related to changes in the metabolism that are linked to producing energy by elevated rates of glycolysis in a process called the Warburg effect. Moreover, we show that our FFPE proteomic approach allowed identification of key biological markers such as the S100 protein that we further validated by immunohistochemistry for both, the primary and metastatic tumor samples. That said, we demonstrated a powerful strategy to retrospectively mine data for aiding in the understanding of metastasis, biomarker discovery, and ultimately, diseases. To our knowledge, these results disclose for the first time a comparison of the proteomic profiles of gastric melanoma and its corresponding meningeal metastasis.

The field of proteomics has grown vertiginously in the last years. This has been due fundamentally to technological improvements in the instrumentation, methods, and easy-to-use software, thereby making it possible to address a large number of biological questions and to deepen the study of the proteome of several organisms. The development in the field has imposed a challenge in the computational analysis of the commonly obtained large datasets generated in a single proteomics experiment, which still remains. An alternative to tackle this general issue has been the use of auxiliary information generated during the proteomics experiment to validate the confidence of the identifications. In this manuscript we review the main molecular descriptors used for building predictor models for estimating retention time, isoelectric point and peptide “detectability”, which are key tools in the design of several validation strategies based in these criteria. We also give an overview of the main open source tools and libraries used for computing molecular descriptors.

The emergence of new DNA sequencing technologies has allowed an exponential growth of genomic and transcriptomic data that ultimately yielded important results to several areas such as medicine and biology. This continuous technological progress presents several advantages and caveats that have to be considered for each new method. In this review, we describe the so-called second and third generation DNA sequencing technologies, how they changed the study of genomes and transcriptomes, and most importantly, what are the key factors that should be considered in a sequencing project. Taken together, we present a “sequencing project map” that includes a practical and graphical cost-benefit analysis for genome and transcriptome projects which allows scientist to easily classify their workflow into one of our proposed templates according to the goals and experimental design of the project at hand. In all, this review reflects the pros and cons of the most widely adopted experimental designs, sequencing technologies, and exposes them to help scientists interested in these tools to choose the best strategy for their project.

The right estimation of the evolutionary distance between DNA or protein sequences is the cornerstone of the current phylogenetic analysis based on distance methods. Herein, it is demonstrated that the Manhattan distance (dw), weighted by the evolutionary importance of the nucleotide bases in the codon, is a naturally derived metric in the standard genetic code cube inserted into the three-dimensional Euclidean space. Based on the application of distance dw, a novel evolutionary model is proposed. This model includes insertion/deletion mutations that are very important for cancer studies, but usually discarded in classical evolutionary models. In this study, the new evolutionary model was applied to the phylogenetic analysis of the DNA protein-coding regions of 13 mammal mitochondrial genomes and of four cancer genetic- susceptibility genes (ATM, BRCA1, BRCA2 and p53) from nine mammals. The opossum (a marsupial) was used as an out-group species for both sets of sequences. The new evolutionary model yielded the correct topology, while the current models failed to separate the evolutionarily distant species of mouse and opossum.

Over the past decades DNA sequencing technologies have been improving in aspects like quality, read length, runtimes and yields, all at a lower cost. Despite these improvements, genome assembly remains a challenge in genome sequencing projects, especially when different sequencing platforms are used. At the present, there is no program that can handle and assemble different sequencing technologies better than specialized software for each of them. Also, very few protocols are available for merging results from different algorithms, technologies or both. We present GARM, (Genome Assembler, Reconciliation and Merging) a pipeline to merge assemblies from different algorithms or sequencing technologies.

Since the advent of high-throughput DNA sequencing technologies, the ever-increasing rate at which genomes have been published has generated new challenges notably at the level of genome annotation. Even if gene predictors and annotation softwares are more and more efficient, the ultimate validation is still in the observation of predicted gene product( s). Mass-spectrometry based proteomics provides the necessary high throughput technology to show evidences of protein presence and, from the identified sequences, confirmation or invalidation of predicted annotations. We review here different strategies used to perform a MS-based proteogenomics experiment with a bottom-up approach. We start from the strengths and weaknesses of the different database construction strategies, based on different genomic information (whole genome, ORF, cDNA, EST or RNA-Seq data), which are then used for matching mass spectra to peptides and proteins. We also review the important points to be considered for a correct statistical assessment of the peptide identifications. Finally, we provide references for tools used to map and visualize the peptide identifications back to the original genomic information.

Proteins are the principal mediators of the functions in the cell; therefore, any abnormal variations on their abundance levels may reflect the presence of pathological processes. In this sense, many researchers rely on the functional interpretation of protein lists generated by quantitative proteomics experiments to analyze, for instance, these variations in the context of diseases´ molecular basis and drug discovery. Since no analytical strategy or bioinformatics tool by itself is capable of extract all the information covered by a single experiment; herein we seek to provide the biologists with four groups of different but complementary bioinformatics tools for the functional interpretation of quantitative proteomics results. To this end we will review the basic concepts of a set of different bioinformatics approaches and we will give examples of freely available tools for each one of these approaches.