Open PHACTS computational protocols for in silico target validation of cellular phenotypic screens: knowing the knowns.

Phenotypic screening is in a renaissance phase and is expected by many academic and industry leaders to accelerate the discovery of new drugs for new biology. Given that phenotypic screening is per definition target agnostic, the emphasis of in silico and in vitro follow-up work is on the exploration of possible molecular mechanisms and efficacy targets underlying the biological processes interrogated by the phenotypic screening experiments. Herein, we present six exemplar computational protocols for the interpretation of cellular phenotypic screens based on the integration of compound, target, pathway, and disease data established by the IMI Open PHACTS project. The protocols annotate phenotypic hit lists and allow follow-up experiments and mechanistic conclusions. The annotations included are from ChEMBL, ChEBI, GO, WikiPathways and DisGeNET. Also provided are protocols which select from the IUPHAR/BPS Guide to PHARMACOLOGY interaction file selective compounds to probe potential targets and a correlation robot which systematically aims to identify an overlap of active compounds in both the phenotypic as well as any kinase assay. The protocols are applied to a phenotypic pre-lamin A/C splicing assay selected from the ChEMBL database to illustrate the process. The computational protocols make use of the Open PHACTS API and data and are built within the Pipeline Pilot and KNIME workflow tools.

Hevin is down-regulated in many cancers and is a negative regulator of cell growth and proliferation.

We have cloned a human Hevin cDNA from omental adipose tissue of different patients by reverse transcription polymerase chain reaction and shown a sequence variation due to a possible polymorphism at amino acid position 161 (E/G). Hevin protein expressed in vitro showed molecular weights of approximately 75 kDa and 150 kDa, suggesting that Hevin may form a homodimer in vitro. Using Northern blots and a human expressed sequence tAg database analysis, Hevin was shown to be widely expressed in human normal or non-neoplastic diseased tissues with various levels. In contrast to this, its expression was strongly down-regulated in most neoplastic cells or tissues tested. However, neither the mechanism nor the physiological meaning of this down-regulation is known. As an initial step towards investigating the functional role of Hevin in cell growth and differentiation, we transiently or stably expressed this gene in cancer cells (HeLa 3S) that are devoid of endogenous Hevin and measured DNA synthesis (cell proliferation) by 5-bromo-2'-deoxyuridine incorporation. Hevin was shown to be a negative regulator of cell proliferation. Furthermore, we have shown that Hevin can inhibit progression of cells from G1 to S phase or prolong G1 phase. This is the first report which describes the function of Hevin in cell growth and proliferation. Through database analysis, Hevin was found to be located on chromosome 4 which contains loss of heterozygosity of many tumour suppressor genes. Taken together, these results suggest that Hevin may be a candidate for a tumour suppressor gene and a potential target for cancer diagnosis/therapy.

Isolation and expression of novel human glutamate carboxypeptidases with N-acetylated alpha-linked acidic dipeptidase and dipeptidyl peptidase IV activity.

Hydrolysis of the neuropeptide N-acetyl-L-aspartyl-L-glutamate (NAAG) by N-acetylated alpha-linked acidic dipeptidase (NAALADase) to release glutamate may be important in a number of neurodegenerative disorders in which excitotoxic mechanisms are implicated. The gene coding for human prostate-specific membrane antigen, a marker of prostatic carcinomas, and its rat homologue glutamate carboxypeptidase II have recently been shown to possess such NAALADase activity. In contrast, a closely related member of this gene family, rat ileal 100-kDa protein, possesses a dipeptidyl peptidase IV activity. Here, we describe the cloning of human ileal 100-kDa protein, which we have called a NAALADase- "like" (NAALADase L) peptidase based on its sequence similarity to other members of this gene family, and its inability to hydrolyze NAAG in transient transfection experiments. Furthermore, we describe the cloning of a third novel member of this gene family, NAALADase II, which codes for a type II integral membrane protein and which we have localized to chromosome 11 by fluorescent in situ hybridization analysis. Transient transfection of NAALADase II cDNA confers both NAALADase and dipeptidyl peptidase IV activity to COS cells. Expression studies using reverse transcription-polymerase chain reaction and Northern blot hybridization show that NAALADase II is highly expressed in ovary and testis as well as within discrete brain areas.

Structure of the human histamine H1 receptor gene.

Histamine H1 receptor expression has been reported to change in disorders such as allergic rhinitis, autoimmune myocarditis, rheumatoid arthritis and atherosclerosis. Here we report the isolation and characterization of genomic clones containing the 5' flanking (regulatory) region of the human histamine H1 receptor gene. An intron of approx. 5.8 kb was identified in the 5' untranslated region, which suggests that an entire subfamily of G-protein-coupled receptors may contain an intron immediately upstream of the start codon. The transcription initiation site was mapped by 5' rapid amplification of cDNA ends to a region 6.2 kb upstream of the start codon. Immediately upstream of the transcription start site a fragment of 1.85 kb was identified that showed promoter activity when placed upstream of a luciferase reporter gene and transiently transfected into cells expressing the histamine H1 receptor. The promoter sequence shares a number of characteristics with the promoter sequences of other G-protein-coupled receptor encoding genes, including binding sites for several transcription factors, and the absence of TATA and CAAT sequences at the appropriate locations. The promoter sequence described here differs from that reported previously [Fukui, Fujimoto, Mizuguchi, Sakamoto, Horio, Takai, Yamada and Ito (1994) Biochem. Biophys. Res. Commun. 201, 894-901] because the reported genomic clone was chimaeric. Furthermore our study provides evidence that the 3' untranslated region of the H1 receptor mRNA is much longer than previously accepted. Together, these findings provide a complete view of the structure of the human histamine H1 receptor gene. Both the coding region of the H1 receptor gene and its promoter region were independently mapped to chromosome 3p25.

Evolutionary affiliation of the marine nitrogen-fixing cyanobacterium Trichodesmium sp. strain NIBB 1067, derived by 16S ribosomal RNA sequence analysis.

The 16S rRNA sequence of Trichodesmium sp. strain NIBB 1067 was determined and used for the construction of a distance tree and bootstrap analysis. The tree shows that, among the available cyanobacterial 16S rRNA sequences, Trichodesmium NIBB 1067 has Oscillatoria PCC 7515 as its closest relative, presenting 94.9% of sequence similarity with the latter strain. This is in contrast to a difference of 9 mol% G+C in mean genomic DNA base composition between the two organisms. Nevertheless, the genotypic heterogeneity presented by a number of strains assigned to the genus Oscillatoria hinders a taxonomic decision on the separate existence of the genera Trichodesmium and Oscillatoria. The sequence of the internal transcribed spacer (ITS) between the 16S and 23S rRNA genes was also determined, as a possible marker to study inter- and intraspecific variability. The ITS contains the genes coding for tRNA(Ile) and tRNA(Ala) and its total length is 547 nucleotides. In six out of eight sequenced clones, there is a duplication of 29 nucleotides, surrounding the 5' end of the tRNA(Ile).

Reconstructing evolution from eukaryotic small-ribosomal-subunit RNA sequences: calibration of the molecular clock.

The detailed descriptions now available for the secondary structure of small-ribosomal-subunit RNA, including areas of highly variable primary structure, facilitate the alignment of nucleotide sequences. However, for optimal exploitation of the information contained in the alignment, a method must be available that takes into account the local sequence variability in the computation of evolutionary distance. A quantitative definition for the variability of an alignment position is proposed in this study. It is a parameter in an equation which expresses the probability that the alignment position contains a different nucleotide in two sequences, as a function of the distance separating these sequences, i.e., the number of substitutions per nucleotide that occurred during their divergence. This parameter can be estimated from the distance matrix resulting from the conversion of pairwise sequence dissimilarities into pairwise distances. Alignment positions can then be subdivided into a number of sets of matching variability, and the average variability of each set can be derived. Next, the conversion of dissimilarity into distance can be recalculated for each set of alignment positions separately, using a modified version of the equation that corrects for multiple substitutions and changing for each set the parameter that reflects its average variability. The distances computed for each set are finally averaged, giving a more precise distance estimation. Trees constructed by the algorithm based on variability calibration have a topology markedly different from that of trees constructed from the same alignments in the absence of calibration. This is illustrated by means of trees constructed from small-ribosomal-subunit RNA sequences of Metazoa. A reconstruction of vertebrate evolution based on calibrated alignments matches the consensus view of paleontologists, contrary to trees based on uncalibrated alignments. In trees derived from sequences covering several metazoan phyla, artefacts in topology that are probably due to a high clock rate in certain lineages are avoided.

Compilation of small ribosomal subunit RNA structures.

The database on small ribosomal subunit RNA structure contained 1804 nucleotide sequences on April 23, 1993. This number comprises 365 eukaryotic, 65 archaeal, 1260 bacterial, 30 plastidial, and 84 mitochondrial sequences. These are stored in the form of an alignment in order to facilitate the use of the database as input for comparative studies on higher-order structure and for reconstruction of phylogenetic trees. The elements of the postulated secondary structure for each molecule are indicated by special symbols. The database is available on-line directly from the authors by ftp and can also be obtained from the EMBL nucleotide sequence library by electronic mail, ftp, and on CD ROM disk.

The gene coding for small ribosomal subunit RNA in the basidiomycete Ustilago maydis contains a group I intron.

The nucleotide sequence of the gene coding for small ribosomal subunit RNA in the basidiomycete Ustilago maydis was determined. It revealed the presence of a group I intron with a length of 411 nucleotides. This is the third occurrence of such an intron discovered in a small subunit rRNA gene encoded by a eukaryotic nuclear genome. The other two occurrences are in Pneumocystis carinii, a fungus of uncertain taxonomic status, and Ankistrodesmus stipitatus, a green alga. The nucleotides of the conserved core structure of 101 group I intron sequences present in different genes and genome types were aligned and their evolutionary relatedness was examined. This revealed a cluster including all group I introns hitherto found in eukaryotic nuclear genes coding for small and large subunit rRNAs. A secondary structure model was designed for the area of the Ustilago maydis small ribosomal subunit RNA precursor where the intron is situated. It shows that the internal guide sequence pairing with the intron boundaries fits between two helices of the small subunit rRNA, and that minimal rearrangement of base pairs suffices to achieve the definitive secondary structure of the 18S rRNA upon splicing.

Phylogenetic analysis of five medically important Candida species as deduced on the basis of small ribosomal subunit RNA sequences.

The classification of species belonging to the genus Candida Berkhout is problematic. Therefore, we have determined the small ribosomal subunit RNA (srRNA) sequences of the type strains of three human pathogenic Candida species; Candida krusei, C. lusitaniae and C. tropicalis. The srRNA sequences were aligned with published eukaryotic srRNA sequences and evolutionary trees were inferred using a matrix optimization method. An evolutionary tree comprising all available eukaryotic srRNA sequences, including two other pathogenic Candida species, C. albicans and C. glabrata, showed that the yeasts diverge rather late in the course of eukaryote evolution, namely at the same depth as green plants, ciliates and some smaller taxa. The cluster of the higher fungi consists of 10 ascomycetes and ascomycete-like species with the first branches leading to Neurospora crassa, Pneumocystis carinii, Candida lusitaniae and C. krusei, in that order. Next there is a dichotomous divergence leading to a group consisting of Torulaspora delbrueckii, Saccharomyces cerevisiae, C. glabrata and Kluyveromyces lactis and a smaller group comprising C. tropicalis and C. albicans. The divergence pattern obtained on the basis of srRNA sequence data is also compared to various other chemotaxonomic data.

A proposal for the secondary structure of a variable area of eukaryotic small ribosomal subunit RNA involving the existence of a pseudoknot.

Eukaryotic small ribosomal subunit RNAs contain an area of variable structure, V4, which comprises about 250 nucleotides in most species, whereas the corresponding area in bacterial small ribosomal subunit RNAs consists of about 64 nucleotides folded into a single hairpin. There is no consensus on the secondary structure of area V4 in eukaryotes, about 10 different models having been proposed. The prediction of a model on a comparative basis poses special problems because, due to the variability of the area in length as well as sequence, a dependable alignment is very difficult to achieve. A new model was derived by systematic examination of all combinations of helices that have been hitherto proposed, plus some new ones. The following properties of the helices were examined: transposability to all presently known sequences, presence of compensating substitutions, and thermodynamic stability. A model was selected by ranking all possible combinations of transposable helices according to the number of compensating substitutions scored. The optimal model comprises a pseudoknot and four hairpin structures. Certain species contain additional hairpins inserted between these structural elements, while in others the structure is partially or entirely deleted.

The 18S ribosomal RNA sequence of the sea anemone Anemonia sulcata and its evolutionary position among other eukaryotes.

Evolutionary trees based on partial small ribosomal subunit RNA sequences of 22 metazoa species have been published [(1988) Science 239, 748-753]. In these trees, cnidarians (Radiata) seemed to have evolved independently from the Bilateria, which is in contradiction with the general evolutionary view. In order to further investigate this problem, the complete srRNA sequence of the sea anemone Anemonia sulcata was determined and evolutionary trees were constructed using a matrix optimization method. In the tree thus obtained the sea anemone and Bilateria together form a monophyletic cluster, with the sea anemone forming the first line of the metazoan group.

Small ribosomal subunit RNA sequences, evolutionary relationships among different life forms, and mitochondrial origins.

A tree was constructed from a structurally conserved area in an alignment of 83 small ribosomal subunit sequences of eukaryotic, archaebacterial, eubacterial, plastidial, and mitochondrial origin. The algorithm involved computation and optimization of a dissimilarity matrix. According to the tree, only plant mitochondria belong to the eubacterial primary kingdom, whereas animal, fungal, algal, and ciliate mitochondria branch off from an internal node situated between the tree primary kingdoms. This result is at variance with a parsimony tree of similar size published by Cedergren et al. (J Mol Evol 28:98-112, 1988), which postulates the mitochondria to be monophyletic and to belong to the eubacterial primary kingdom. The discrepancy does not follow from the use of conflicting sequence alignments, hence it must be due to the use of different treeing algorithms. We tested our algorithm on a set of sequences resulting from a simulated evolution and found it capable of faithfully reconstructing a branching topology that involved very unequal evolutionary rates. The use of more limited or more extended areas of the complete sequence alignment, comprising only very conserved or also more variable portions of the small ribosomal subunit structure, does have some influence on the tree topology. In all cases, however, the nonplant mitochondria seem to branch off before the emergence of eubacteria, and the differences are limited to the branching pattern among different types of mitochondria.