Biodiversity and chemodiversity: future perspectives in bioprospecting

Biological diversity and its constituent chemical diversity have served as one of the richest sources of bioprospecting leading to the discovery of some of the most important bioactive molecules for mankind. Despite this excellent record, in the recent past, however, bioprospecting of biological resources has met with little success; there has been a perceptible decline in the discovery of novel bioactive compounds. Several arguments have been proposed to explain the current poor success in bioprospecting. Among them, it has been argued that to bioprospect more biodiversity may not necessarily be productive, considering that chemical and functional diversity might not scale with biological diversity. In this paper, we offer a critique on the current perception of biodiversity and chemodiversity and ask to what extent it is relevant in the context of bioprospecting. First, using simple models, we analyze the relation among biodiversity, chemodiversity and functional redundancies in chemical plans of plants and argue that the biological space for exploration might still be wide open. Second, in the context of future bioprospecting, we argue that brute-force high throughput screening approaches alone are insufficient and cost ineffective in realizing bioprospecting success. Therefore, intelligent or non-random approaches to bioprospecting need to be adopted. We review here few examples of such approaches and show how these could be further developed and used in the future to accelerate the pace of discovery.

Controlled angiogenesis in the heart by cell-based expression of specific vascular endothelial growth factor levels

Vascular endothelial growth factor (VEGF) can induce normal angiogenesis or the growth of angioma-like vascular tumors depending on the amount secreted by each producing cell because it remains localized in the microenvironment. In order to control the distribution of VEGF expression levels in vivo, we recently developed a high-throughput fluorescence-activated cell sorting (FACS)-based technique to rapidly purify transduced progenitors that homogeneously express a specific VEGF dose from a heterogeneous primary population. Here we tested the hypothesis that cell-based delivery of a controlled VEGF level could induce normal angiogenesis in the heart, while preventing the development of angiomas. Freshly isolated human adipose tissue-derived stem cells (ASC) were transduced with retroviral vectors expressing either rat VEGF linked to a FACS-quantifiable cell-surface marker (a truncated form of CD8) or CD8 alone as control (CTR). VEGF-expressing cells were FACS-purified to generate populations producing either a specific VEGF level (SPEC) or uncontrolled heterogeneous levels (ALL). Fifteen nude rats underwent intramyocardial injection of 10(7) cells. Histology was performed after 4 weeks. Both the SPEC and ALL cells produced a similar total amount of VEGF, and both cell types induced a 50%-60% increase in both total and perfused vessel density compared to CTR cells, despite very limited stable engraftment. However, homogeneous VEGF expression by SPEC cells induced only normal and stable angiogenesis. Conversely, heterogeneous expression of a similar total amount by the ALL cells caused the growth of numerous angioma-like structures. These results suggest that controlled VEGF delivery by FACS-purified ASC may be a promising strategy to achieve safe therapeutic angiogenesis in the heart.

Echinococcus metacestodes as laboratory models for the screening of drugs against cestodes and trematodes

Among the cestodes, Echinococcus granulosus, Echinococcus multilocularis and Taenia solium represent the most dangerous parasites. Their larval stages cause the diseases cystic echinococcosis (CE), alveolar echinococcosis (AE) and cysticercosis, respectively, which exhibit considerable medical and veterinary health concerns with a profound economic impact. Others caused by other cestodes, such as species of the genera Mesocestoides and Hymenolepis, are relatively rare in humans. In this review, we will focus on E. granulosus and E. multilocularis metacestode laboratory models and will review the use of these models in the search for novel drugs that could be employed for chemotherapeutic treatment of echinococcosis. Clearly, improved therapeutic drugs are needed for the treatment of AE and CE, and this can only be achieved through the development of medium-to-high throughput screening approaches. The most recent achievements in the in vitro culture and genetic manipulation of E. multilocularis cells and metacestodes, and the accessability of the E. multilocularis genome and EST sequence information, have rendered the E. multilocularis model uniquely suited for studies on drug-efficacy and drug target identification. This could lead to the development of novel compounds for the use in chemotherapy against echinococcosis, and possibly against diseases caused by other cestodes, and potentially also trematodes.

A global ”imaging” view on systems approaches in immunology

The immune system exhibits an enormous complexity. High throughput methods such as the "-omic'' technologies generate vast amounts of data that facilitate dissection of immunological processes at ever finer resolution. Using high-resolution data-driven systems analysis, causal relationships between complex molecular processes and particular immunological phenotypes can be constructed. However, processes in tissues, organs, and the organism itself (so-called higher level processes) also control and regulate the molecular (lower level) processes. Reverse systems engineering approaches, which focus on the examination of the structure, dynamics and control of the immune system, can help to understand the construction principles of the immune system. Such integrative mechanistic models can properly describe, explain, and predict the behavior of the immune system in health and disease by combining both higher and lower level processes. Moving from molecular and cellular levels to a multiscale systems understanding requires the development of methodologies that integrate data from different biological levels into multiscale mechanistic models. In particular, 3D imaging techniques and 4D modeling of the spatiotemporal dynamics of immune processes within lymphoid tissues are central for such integrative approaches. Both dynamic and global organ imaging technologies will be instrumental in facilitating comprehensive multiscale systems immunology analyses as discussed in this review.

Measuring substrate binding and affinity of purified membrane transport proteins using the scintillation proximity assay

The scintillation proximity assay (SPA) is a rapid radioligand binding assay. Upon binding of radioactively labeled ligands (here L-[(3)H]arginine or D-[(3)H]glucose) to acceptor proteins immobilized on fluoromicrospheres (containing the scintillant), a light signal is stimulated and measured. The application of SPA to purified, detergent-solubilized membrane transport proteins allows substrate-binding properties to be assessed (e.g., substrate specificity and affinity), usually within 1 d. Notably, the SPA makes it possible to study specific transporters without interference from other cellular components, such as endogenous transporters. Reconstitution of the target transporter into proteoliposomes is not required. The SPA procedure allows high sample throughput and simple sample handling without the need for washing or separation steps: components are mixed in one well and the signal is measured directly after incubation. Therefore, the SPA is an excellent tool for high-throughput screening experiments, e.g., to search for substrates and inhibitors, and it has also recently become an attractive tool for drug discovery.

Modulation of L-α-lysophosphatidylinositol/GPR55 mitogen-activated protein kinase (MAPK) signaling by cannabinoids

GPR55 is activated by l-α-lysophosphatidylinositol (LPI) but also by certain cannabinoids. In this study, we investigated the GPR55 pharmacology of various cannabinoids, including analogues of the CB1 receptor antagonist Rimonabant®, CB2 receptor agonists, and Cannabis sativa constituents. To test ERK1/2 phosphorylation, a primary downstream signaling pathway that conveys LPI-induced activation of GPR55, a high throughput system, was established using the AlphaScreen® SureFire® assay. Here, we show that CB1 receptor antagonists can act both as agonists alone and as inhibitors of LPI signaling under the same assay conditions. This study clarifies the controversy surrounding the GPR55-mediated actions of SR141716A; some reports indicate the compound to be an agonist and some report antagonism. In contrast, we report that the CB2 ligand GW405833 behaves as a partial agonist of GPR55 alone and enhances LPI signaling. GPR55 has been implicated in pain transmission, and thus our results suggest that this receptor may be responsible for some of the antinociceptive actions of certain CB2 receptor ligands. The phytocannabinoids Δ9-tetrahydrocannabivarin, cannabidivarin, and cannabigerovarin are also potent inhibitors of LPI. These Cannabis sativa constituents may represent novel therapeutics targeting GPR55.

An integrated nano-scale approach to profile miRNAs in limited clinical samples

Profiling miRNA expression in cells that directly contribute to human disease pathogenesis is likely to aid the discovery of novel drug targets and biomarkers. However, tissue heterogeneity and the limited amount of human diseased tissue available for research purposes present fundamental difficulties that often constrain the scope and potential of such studies. We established a flow cytometry-based method for isolating pure populations of pathogenic T cells from bronchial biopsy samples of asthma patients, and optimized a high-throughput nano-scale qRT-PCR method capable of accurately measuring 96 miRNAs in as little as 100 cells. Comparison of circulating and airway T cells from healthy and asthmatic subjects revealed asthma-associated and tissue-specific miRNA expression patterns. These results establish the feasibility and utility of investigating miRNA expression in small populations of cells involved in asthma pathogenesis, and set a precedent for application of our nano-scale approach in other human diseases. The microarray data from this study (Figure 7) has been submitted to the NCBI Gene Expression Omnibus (GEO; http://ncbi.nlm.nih.gov/geo) under accession no. GSE31030.

Pancreatic carcinoma, pancreatitis, and healthy controls: metabolite models in a three-class diagnostic dilemma

Metabolomics as one of the most rapidly growing technologies in the "-omics" field denotes the comprehensive analysis of low molecular-weight compounds and their pathways. Cancer-specific alterations of the metabolome can be detected by high-throughput mass-spectrometric metabolite profiling and serve as a considerable source of new markers for the early differentiation of malignant diseases as well as their distinction from benign states. However, a comprehensive framework for the statistical evaluation of marker panels in a multi-class setting has not yet been established. We collected serum samples of 40 pancreatic carcinoma patients, 40 controls, and 23 pancreatitis patients according to standard protocols and generated amino acid profiles by routine mass-spectrometry. In an intrinsic three-class bioinformatic approach we compared these profiles, evaluated their selectivity and computed multi-marker panels combined with the conventional tumor marker CA 19-9. Additionally, we tested for non-inferiority and superiority to determine the diagnostic surplus value of our multi-metabolite marker panels. Compared to CA 19-9 alone, the combined amino acid-based metabolite panel had a superior selectivity for the discrimination of healthy controls, pancreatitis, and pancreatic carcinoma patients [Formula: see text] We combined highly standardized samples, a three-class study design, a high-throughput mass-spectrometric technique, and a comprehensive bioinformatic framework to identify metabolite panels selective for all three groups in a single approach. Our results suggest that metabolomic profiling necessitates appropriate evaluation strategies and-despite all its current limitations-can deliver marker panels with high selectivity even in multi-class settings.

Point-of-care testing in the cardiovascular operating theatre

Point-of-care testing (POCT) remains under scrutiny by healthcare professionals because of its ill-tried, young history. POCT methods are being developed by a few major equipment companies based on rapid progress in informatics and nanotechnology. Issues as POCT quality control, comparability with standard laboratory procedures, standardisation, traceability and round robin testing are being left to hospitals. As a result, the clinical and operational benefits of POCT were first evident for patients on the operating table. For the management of cardiovascular surgery patients, POCT technology is an indispensable aid. Improvement of the technology has meant that clinical laboratory pathologists now recognise the need for POCT beyond their high-throughput areas.

Classification and selection of biomarkers in genomic data using LASSO

High-throughput gene expression technologies such as microarrays have been utilized in a variety of scientific applications. Most of the work has been on assessing univariate associations between gene expression with clinical outcome (variable selection) or on developing classification procedures with gene expression data (supervised learning). We consider a hybrid variable selection/classification approach that is based on linear combinations of the gene expression profiles that maximize an accuracy measure summarized using the receiver operating characteristic curve. Under a specific probability model, this leads to consideration of linear discriminant functions. We incorporate an automated variable selection approach using LASSO. An equivalence between LASSO estimation with support vector machines allows for model fitting using standard software. We apply the proposed method to simulated data as well as data from a recently published prostate cancer study.

In-frame triplet deletions in RHD alter the D antigen phenotype

BACKGROUND: The deletion of three adjacent nucleotides in an exon may cause the lack of a single amino acid, while the protein sequence remains otherwise unchanged. Only one such in-frame deletion is known in the two RH genes, represented by the RHCE allele ceBP expressing a "very weak e antigen." STUDY DESIGN AND METHODS: Blood donor samples were recognized because of discrepant results of D phenotyping. Six samples came from Switzerland and one from Northern Germany. The molecular structures were determined by genomic DNA nucleotide sequencing of RHD. RESULTS: Two different variant D antigens were explained by RHD alleles harboring one in-frame triplet deletion each. Both single-amino-acid deletions led to partial D phenotypes with weak D antigen expression. Because of their D category V-like phenotypes, the RHD(Arg229del) allele was dubbed DVL-1 and the RHD(Lys235del) allele DVL-2. These in-frame triplet deletions are located in GAGAA or GAAGA repeats of the RHD exon 5. CONCLUSION: Partial D may be caused by a single-amino-acid deletion in RhD. The altered RhD protein segments in DVL types are adjacent to the extracellular loop 4, which constitutes one of the most immunogenic parts of the D antigen. These RhD protein segments are also altered in all DV, which may explain the similarity in phenotype. At the nucleotide level, the triplet deletions may have resulted from replication slippage. A total of nine amino acid positions in an Rhesus protein may be affected by this mechanism.

A graph theoretic approach to testing associations between disparate sources of functional genomic data

The last few years have seen the advent of high-throughput technologies to analyze various properties of the transcriptome and proteome of several organisms. The congruency of these different data sources, or lack thereof, can shed light on the mechanisms that govern cellular function. A central challenge for bioinformatics research is to develop a unified framework for combining the multiple sources of functional genomics information and testing associations between them, thus obtaining a robust and integrated view of the underlying biology. We present a graph theoretic approach to test the significance of the association between multiple disparate sources of functional genomics data by proposing two statistical tests, namely edge permutation and node label permutation tests. We demonstrate the use of the proposed tests by finding significant association between a Gene Ontology-derived "predictome" and data obtained from mRNA expression and phenotypic experiments for Saccharomyces cerevisiae. Moreover, we employ the graph theoretic framework to recast a surprising discrepancy presented in Giaever et al. (2002) between gene expression and knockout phenotype, using expression data from a different set of experiments.

Classification Using Generalized Partial Least Squares

The advances in computational biology have made simultaneous monitoring of thousands of features possible. The high throughput technologies not only bring about a much richer information context in which to study various aspects of gene functions but they also present challenge of analyzing data with large number of covariates and few samples. As an integral part of machine learning, classification of samples into two or more categories is almost always of interest to scientists. In this paper, we address the question of classification in this setting by extending partial least squares (PLS), a popular dimension reduction tool in chemometrics, in the context of generalized linear regression based on a previous approach, Iteratively ReWeighted Partial Least Squares, i.e. IRWPLS (Marx, 1996). We compare our results with two-stage PLS (Nguyen and Rocke, 2002A; Nguyen and Rocke, 2002B) and other classifiers. We show that by phrasing the problem in a generalized linear model setting and by applying bias correction to the likelihood to avoid (quasi)separation, we often get lower classification error rates.