The saponin bomb: a nucleolar-localized β-glucosidase hydrolyzes triterpene saponins in Medicago truncatula

Plants often protect themselves from their own bioactive defense metabolites by storing them in less active forms. Consequently, plants also need systems allowing correct spatiotemporal reactivation of such metabolites, for instance under pathogen or herbivore attack. Via co-expression analysis with public transcriptomes, we determined that the model legume Medicago truncatula has evolved a two-component system composed of a β-glucosidase, denominated G1, and triterpene saponins, which are physically separated from each other in intact cells. G1 expression is root-specific, stress-inducible, and coregulated with that of the genes encoding the triterpene saponin biosynthetic enzymes. However, the G1 protein is stored in the nucleolus and is released and united with its typically vacuolar-stored substrates only upon tissue damage, partly mediated by the surfactant action of the saponins themselves. Subsequently, enzymatic removal of carbohydrate groups from the saponins creates a pool of metabolites with an increased broad-spectrum antimicrobial activity. The evolution of this defense system benefited from both the intrinsic condensation abilities of the enzyme and the bioactivity properties of its substrates. We dub this two-component system the saponin bomb, in analogy with the mustard oil and cyanide bombs, commonly used to describe the renowned β-glucosidase-dependent defense systems for glucosinolates and cyanogenic glucosides.

Bedaquiline- and clofazimine- selected Mycobacterium tuberculosis mutants: further insights on resistance driven largely by Rv0678

Drug-resistant tuberculosis is a serious global health threat. Bedaquiline (BDQ) is a relatively new core drug, targeting the respiratory chain in Mycobacterium tuberculosis (Mtb). While mutations in the BDQ target gene, atpE, are rare in clinical isolates, mutations in the Rv0678 gene, a transcriptional repressor regulating the efflux pump MmpS5-MmpL5, are increasingly observed, and have been linked to worse treatment outcomes. Nevertheless, underlying mechanisms of (cross)-resistance remain incompletely resolved. Our study aims to distinguish resistance associated variants from other polymorphisms, by assessing the in vitro onset of mutations under drug pressure, combined with their impact on minimum inhibitory concentrations (MICs) and on protein stability. For this purpose, isolates were exposed in vitro to sub-lethal concentrations of BDQ or clofazimine (CFZ). Selected colonies had BDQ- and CFZ-MICs determined on 7H10 and 7H11 agar. Sanger sequencing and additional Deeplex Myc-TB and whole genome sequencing (WGS) for a subset of isolates were used to search for mutations in Rv0678, atpE and pepQ. In silico characterization of relevant mutations was performed using computational tools. We found that colonies that grew on BDQ medium had mutations in Rv0678, atpE or pepQ, while CFZ-exposed isolates presented mutations in Rv0678 and pepQ, but none in atpE. Twenty-eight Rv0678 mutations had previously been described among in vitro selected mutants or in patients' isolates, while 85 were new. Mutations were scattered across the Rv0678 gene without apparent hotspot. While most Rv0678 mutations led to an increased BDQ- and/or CFZ-MIC, only a part of them surpassed the critical concentration (69.1% for BDQ and 87.9% for CFZ). Among the mutations leading to elevated MICs for BDQ and CFZ, we report a synonymous Val1Val mutation in the Rv0678 start codon. Finally, in silico characterization of Rv0678 mutations suggests that especially the C46R mutant may render Rv0678 less stable.

Transcriptome and metabolome profiling identify factors potentially involved in pro-vitamin A accumulation in cassava landraces

Cassava (Manihot esculenta Crantz) is a predominant food security crop in several developing countries. Its storage roots, rich in carbohydrate, are deficient in essential micronutrients, including provitamin A carotenoids. Increasing carotenoid content in cassava storage roots is important to reduce the incidence of vitamin A deficiency, a public health problem in sub-Saharan Africa. However, cassava improvement advances slowly, mainly due to limited information on the molecular factors influencing β-carotene accumulation in cassava. To address this problem, we performed comparative transcriptomic and untargeted metabolic analyses of roots and leaves of eleven African cassava landraces ranging from white to deep yellow colour, to uncover regulators of carotenoid biosynthesis and accumulation with conserved function in yellow cassava roots. Sequence analysis confirmed the presence of a mutation, known to influence β-carotene content, in PSY transcripts of deep yellow but not of pale yellow genotypes. We identified genes and metabolites with expression and accumulation levels significantly associated with β-carotene content. Particularly an increased activity of the abscisic acid catabolism pathway together with a reduced amount of L-carnitine, may be related to the carotenoid pathway flux, higher in yellow than in white storage roots. In fact, NCED_3.1 was specifically expressed at a lower level in all yellow genotypes suggesting that it could be a potential target for increasing carotenoid accumulation in cassava. These results expand the knowledge on metabolite compositions and molecular mechanisms influencing carotenoid biosynthesis and accumulation in cassava and provide novel information for biotechnological applications and genetic improvement of cassava with high nutritional values.

Multidrug-resistant tuberculosis control in Rwanda overcomes a successful clone that causes most disease over a quarter century

SUMMARY BACKGROUND

Multidrug-resistant (MDR) tuberculosis (TB) poses an important challenge in TB management and control. Rifampicin resistance (RR) is a solid surrogate marker of MDR-TB. We investigated the RR-TB clustering rates, bacterial population dynamics to infer transmission dynamics, and the impact of changes to patient management on these dynamics over 27 years in Rwanda.

METHODS

We analysed whole genome sequences of a longitudinal collection of nationwide RR-TB isolates. The collection covered three important periods: before programmatic management of MDR-TB (PMDT; 1991-2005), the early PMDT phase (2006-2013), in which rifampicin drug-susceptibility testing (DST) was offered to retreatment patients only, and the consolidated phase (2014-2018), in which all bacteriologically confirmed TB patients had rifampicin DST done mostly via Xpert MTB/RIF assay. We constructed clusters based on a 5 SNP cut-off and resistance conferring SNPs. We used Bayesian modelling for dating and population size estimations, TransPhylo to estimate the number of secondary cases infected by each patient, and multivariable logistic regression to assess predictors of being infected by the dominant clone.

RESULTS

Of 308 baseline RR-TB isolates considered for transmission analysis, the clustering analysis grouped 259 (84.1%) isolates into 13 clusters. Within these clusters, a single dominant clone was discovered containing 213 isolates (82.2% of clustered and 69.1% of all RR-TB), which we named the "Rwanda Rifampicin-Resistant clone" (R3clone). R3clone isolates belonged to Ugandan sub-lineage 4.6.1.2 and its rifampicin and isoniazid resistance were conferred by the Ser450Leu mutation in rpoB and Ser315Thr in katG genes, respectively. All R3clone isolates had Pro481Thr, a putative compensatory mutation in the rpoC gene that likely restored its fitness. The R3clone was estimated to first arise in 1987 and its population size increased exponentially through the 1990s', reaching maximum size (∼84%) in early 2000 s', with a declining trend since 2014. Indeed, the highest proportion of R3clone (129/157; 82·2%, 95%CI: 75·3-87·8%) occurred between 2000 and 13, declining to 64·4% (95%CI: 55·1-73·0%) from 2014 onward. We showed that patients with R3clone detected after an unsuccessful category 2 treatment were more likely to generate secondary cases than patients with R3clone detected after an unsuccessful category 1 treatment regimen.

CONCLUSIONS

RR-TB in Rwanda is largely transmitted. Xpert MTB/RIF assay as first diagnostic test avoids unnecessary rounds of rifampicin-based TB treatment, thus preventing ongoing transmission of the dominant R3clone. As PMDT was intensified and all TB patients accessed rifampicin-resistance testing, the nationwide R3clone burden declined. To our knowledge, our findings provide the first evidence supporting the impact of universal DST on the transmission of RR-TB.

FASCICLIN-LIKE 18 Is a New Player Regulating Root Elongation in Arabidopsis thaliana

The plasticity of root development represents a key trait that enables plants to adapt to diverse environmental cues. The pattern of cell wall deposition, alongside other parameters, affects the extent, and direction of root growth. In this study, we report that FASCICLIN-LIKE ARABINOGALACTAN PROTEIN 18 (FLA18) plays a role during root elongation in Arabidopsis thaliana. Using root-specific co-expression analysis, we identified FLA18 to be co-expressed with a sub-set of genes required for root elongation. FLA18 encodes for a putative extra-cellular arabinogalactan protein from the FLA-gene family. Two independent T-DNA insertion lines, named fla18-1 and fla18-2, display short and swollen lateral roots (LRs) when grown on sensitizing condition of high-sucrose containing medium. Unlike fla4/salt overly sensitive 5 (sos5), previously shown to display short and swollen primary root (PR) and LRs under these conditions, the PR of the fla18 mutants is slightly longer compared to the wild-type. Overexpression of the FLA18 CDS complemented the fla18 root phenotype. Genetic interaction between either of the fla18 alleles and sos5 reveals a more severe perturbation of anisotropic growth in both PR and LRs, as compared to the single mutants and the wild-type under restrictive conditions of high sucrose or high-salt containing medium. Additionally, under salt-stress conditions, fla18sos5 had a small, chlorotic shoot phenotype, that was not observed in any of the single mutants or the wild type. As previously shown for sos5, the fla18-1 and fla18-1sos5 root-elongation phenotype is suppressed by abscisic acid (ABA) and display hypersensitivity to the ABA synthesis inhibitor, Fluridon. Last, similar to other cell wall mutants, fla18 root elongation is hypersensitive to the cellulose synthase inhibitor, Isoxaben. Altogether, the presented data assign a new role for FLA18 in the regulation of root elongation. Future studies of the unique vs. redundant roles of FLA proteins during root elongation is anticipated to shed a new light on the regulation of root architecture during plant adaptation to different growth conditions.

Co-expression for Genotype-Phenotype Function Annotation in Potato Research

Here we will show a recipe for running the online version of CoExpNetViz, which is available as a Cytoscape plug-in, and as a web tool. After choosing bait genes and transcriptome datasets in the Cytoscape plug-in, the analysis is run and the resulting network is displayed immediately. Using the web tool, the user can download the Cytoscape files and import them manually into the program.The easiest way to calculate correlations for your data is to use graphical interface online version of CoExpNetViz for the comparative co-expression construction; see http://bioinformatics.psb.ugent.be/webtools/coexpr/index.php .By providing a user-friendly web interface, CoExpNetViz makes comparative transcriptomics analysis accessible to plant researchers without specialized bioinformatics knowledge or programming skills.

Siliplant1 protein precipitates silica in sorghum silica cells

Silicon is absorbed by plant roots as silicic acid. The acid moves with the transpiration stream to the shoot, and mineralizes as silica. In grasses, leaf epidermal cells called silica cells deposit silica in most of their volume using an unknown biological factor. Using bioinformatics tools, we identified a previously uncharacterized protein in Sorghum bicolor, which we named Siliplant1 (Slp1). Slp1 is a basic protein with seven repeat units rich in proline, lysine, and glutamic acid. We found Slp1 RNA in sorghum immature leaf and immature inflorescence. In leaves, transcription was highest just before the active silicification zone (ASZ). There, Slp1 was localized specifically to developing silica cells, packed inside vesicles and scattered throughout the cytoplasm or near the cell boundary. These vesicles fused with the membrane, releasing their content in the apoplastic space. A short peptide that is repeated five times in Slp1 precipitated silica in vitro at a biologically relevant silicic acid concentration. Transient overexpression of Slp1 in sorghum resulted in ectopic silica deposition in all leaf epidermal cell types. Our results show that Slp1 precipitates silica in sorghum silica cells.

Biosynthesis and tissue-specific partitioning of camphor and eugenol in Ocimum kilimandscharicum

In Ocimum kilimandscharicum, the relative volatile composition of camphor in leaves was as high as 55%, while that of eugenol in roots was 57%. These metabolites were differentially partitioned between the aerial and root tissues. Global metabolomics revealed tissue-specific biochemical specialization, evident by the differential distribution of 2588 putative metabolites across nine tissues. Next-generation sequencing analysis indicated differential expression of 51 phenylpropanoid and 55 terpenoid pathway genes in aerial and root tissues. By integrating metabolomics with transcriptomics, the camphor biosynthesis pathway in O. kilimandscharicum was elucidated. In planta bioassays revealed the role of geranyl diphosphate synthase (gpps) and borneol dehydrogenase (bdh) in camphor biosynthesis. Further, the partitioning of camphor was attributed to tissue-specific gene expression of both the pathway entry point (gpps) and terminal (bdh) enzyme. Unlike camphor, eugenol accumulated more in roots; however, absence of the eugenol synthase gene in roots indicated long distance transport from aerial tissues. In silico co-expression analysis indicated the potential involvement of ATP-binding cassette, multidrug and toxic compound extrusion, and sugar transporters in eugenol transport. Similar partitioning was evident across five other Ocimum species. Overall, our work indicates that metabolite partitioning maybe a finely regulated process, which may have implications on plant growth, development, and defense.

The 'TranSeq' 3'-end sequencing method for high-throughput transcriptomics and gene space refinement in plant genomes

High-throughput RNA sequencing has proven invaluable not only to explore gene expression but also for both gene prediction and genome annotation. However, RNA sequencing, carried out on tens or even hundreds of samples, requires easy and cost-effective sample preparation methods using minute RNA amounts. Here, we present TranSeq, a high-throughput 3'-end sequencing procedure that requires 10- to 20-fold fewer sequence reads than the current transcriptomics procedures. TranSeq significantly reduces costs and allows a greater increase in size of sample sets analyzed in a single experiment. Moreover, in comparison with other 3'-end sequencing methods reported to date, we demonstrate here the reliability and immediate applicability of TranSeq and show that it not only provides accurate transcriptome profiles but also produces precise expression measurements of specific gene family members possessing high sequence similarity. This is difficult to achieve in standard RNA-seq methods, in which sequence reads cover the entire transcript. Furthermore, mapping TranSeq reads to the reference tomato genome facilitated the annotation of new transcripts improving >45% of the existing gene models. Hence, we anticipate that using TranSeq will boost large-scale transcriptome assays and increase the spatial and temporal resolution of gene expression data, in both model and non-model plant species. Moreover, as already performed for tomato (ITAG3.0; www.solgenomics.net), we strongly advocate its integration into current and future genome annotations.

Network Modeling Unravels Mechanisms of Crosstalk between Ethylene and Salicylate Signaling in Potato

To develop novel crop breeding strategies, it is crucial to understand the mechanisms underlying the interaction between plants and their pathogens. Network modeling represents a powerful tool that can unravel properties of complex biological systems. In this study, we aimed to use network modeling to better understand immune signaling in potato (Solanum tuberosum). For this, we first built on a reliable Arabidopsis (Arabidopsis thaliana) immune signaling model, extending it with the information from diverse publicly available resources. Next, we translated the resulting prior knowledge network (20,012 nodes and 70,091 connections) to potato and superimposed it with an ensemble network inferred from time-resolved transcriptomics data for potato. We used different network modeling approaches to generate specific hypotheses of potato immune signaling mechanisms. An interesting finding was the identification of a string of molecular events illuminating the ethylene pathway modulation of the salicylic acid pathway through Nonexpressor of PR Genes1 gene expression. Functional validations confirmed this modulation, thus supporting the potential of our integrative network modeling approach for unraveling molecular mechanisms in complex systems. In addition, this approach can ultimately result in improved breeding strategies for potato and other sensitive crops.

Deciphering genetic factors that determine melon fruit-quality traits using RNA-Seq-based high-resolution QTL and eQTL mapping

Combined quantitative trait loci (QTL) and expression-QTL (eQTL) mapping analysis was performed to identify genetic factors affecting melon (Cucumis melo) fruit quality, by linking genotypic, metabolic and transcriptomic data from a melon recombinant inbred line (RIL) population. RNA sequencing (RNA-Seq) of fruit from 96 RILs yielded a highly saturated collection of > 58 000 single-nucleotide polymorphisms, identifying 6636 recombination events that separated the genome into 3663 genomic bins. Bin-based QTL analysis of 79 RILs and 129 fruit-quality traits affecting taste, aroma and color resulted in the mapping of 241 QTL. Thiol acyltransferase (CmThAT1) gene was identified within the QTL interval of its product, S-methyl-thioacetate, a key component of melon fruit aroma. Metabolic activity of CmThAT1-encoded protein was validated in bacteria and in vitro. QTL analysis of flesh color intensity identified a candidate white-flesh gene (CmPPR1), one of two major loci determining fruit flesh color in melon. CmPPR1 encodes a member of the pentatricopeptide protein family, involved in processing of RNA in plastids, where carotenoid and chlorophyll pigments accumulate. Network analysis of > 12 000 eQTL mapped for > 8000 differentially expressed fruit genes supported the role of CmPPR1 in determining the expression level of plastid targeted genes. We highlight the potential of RNA-Seq-based QTL analysis of small to moderate size, advanced RIL populations for precise marker-assisted breeding and gene discovery. We provide the following resources: a RIL population genotyped with a unique set of SNP markers, confined genomic segments that harbor QTL governing 129 traits and a saturated set of melon eQTLs.

MorphDB: Prioritizing Genes for Specialized Metabolism Pathways and Gene Ontology Categories in Plants

Recent times have seen an enormous growth of "omics" data, of which high-throughput gene expression data are arguably the most important from a functional perspective. Despite huge improvements in computational techniques for the functional classification of gene sequences, common similarity-based methods often fall short of providing full and reliable functional information. Recently, the combination of comparative genomics with approaches in functional genomics has received considerable interest for gene function analysis, leveraging both gene expression based guilt-by-association methods and annotation efforts in closely related model organisms. Besides the identification of missing genes in pathways, these methods also typically enable the discovery of biological regulators (i.e., transcription factors or signaling genes). A previously built guilt-by-association method is MORPH, which was proven to be an efficient algorithm that performs particularly well in identifying and prioritizing missing genes in plant metabolic pathways. Here, we present MorphDB, a resource where MORPH-based candidate genes for large-scale functional annotations (Gene Ontology, MapMan bins) are integrated across multiple plant species. Besides a gene centric query utility, we present a comparative network approach that enables researchers to efficiently browse MORPH predictions across functional gene sets and species, facilitating efficient gene discovery and candidate gene prioritization. MorphDB is available at http://bioinformatics.psb.ugent.be/webtools/morphdb/morphDB/index/. We also provide a toolkit, named "MORPH bulk" (https://github.com/arzwa/morph-bulk), for running MORPH in bulk mode on novel data sets, enabling researchers to apply MORPH to their own species of interest.

T-DNA-genome junctions form early after infection and are influenced by the chromatin state of the host genome

Agrobacterium tumefaciens mediated T-DNA integration is a common tool for plant genome manipulation. However, there is controversy regarding whether T-DNA integration is biased towards genes or randomly distributed throughout the genome. In order to address this question, we performed high-throughput mapping of T-DNA-genome junctions obtained in the absence of selection at several time points after infection. T-DNA-genome junctions were detected as early as 6 hours post-infection. T-DNA distribution was apparently uniform throughout the chromosomes, yet local biases toward AT-rich motifs and T-DNA border sequence micro-homology were detected. Analysis of the epigenetic landscape of previously isolated sites of T-DNA integration in Kanamycin-selected transgenic plants showed an association with extremely low methylation and nucleosome occupancy. Conversely, non-selected junctions from this study showed no correlation with methylation and had chromatin marks, such as high nucleosome occupancy and high H3K27me3, that correspond to three-dimensional-interacting heterochromatin islands embedded within euchromatin. Such structures may play a role in capturing and silencing invading T-DNA.

Agrobacterium tumefaciens mediated T-DNA integration is an important tool for genetic engineering in plants. This work compares the genetic and epigenetic landscapes of T-DNA-genome junctions under selective and non-selective conditions. Under selection, preferential junctions in low-nucleosome occupancy and hypomethylated regions were found. In the absence of selection, these biases disappeared and T-DNA-genome junctions were uniformly distributed with a preference for 3D-interacting heterochromatin islands embedded within euchromatin, suggesting that many integration events become transcriptionally inactive.

MYB107 and MYB9 Homologs Regulate Suberin Deposition in Angiosperms

Suberin, a polymer composed of both aliphatic and aromatic domains, is deposited as a rough matrix upon plant surface damage and during normal growth in the root endodermis, bark, specialized organs (e.g., potato [Solanum tuberosum] tubers), and seed coats. To identify genes associated with the developmental control of suberin deposition, we investigated the chemical composition and transcriptomes of suberized tomato (Solanum lycopersicum) and russet apple (Malus x domestica) fruit surfaces. Consequently, a gene expression signature for suberin polymer assembly was revealed that is highly conserved in angiosperms. Seed permeability assays of knockout mutants corresponding to signature genes revealed regulatory proteins (i.e., AtMYB9 and AtMYB107) required for suberin assembly in the Arabidopsis thaliana seed coat. Seeds of myb107 and myb9 Arabidopsis mutants displayed a significant reduction in suberin monomers and altered levels of other seed coat-associated metabolites. They also exhibited increased permeability, and lower germination capacities under osmotic and salt stress. AtMYB9 and AtMYB107 appear to synchronize the transcriptional induction of aliphatic and aromatic monomer biosynthesis and transport and suberin polymerization in the seed outer integument layer. Collectively, our findings establish a regulatory system controlling developmentally deposited suberin, which likely differs from the one of stress-induced polymer assembly recognized to date.

Characterization of a New Pink-Fruited Tomato Mutant Results in the Identification of a Null Allele of the SlMYB12 Transcription Factor

The identification and characterization of new tomato (Solanum lycopersicum) mutants affected in fruit pigmentation and nutritional content can provide valuable insights into the underlying biology, as well as a source of new alleles for breeding programs. To date, all characterized pink-pigmented tomato fruit mutants appear to result from low SlMYB12 transcript levels in the fruit skin. Two new mutant lines displaying a pink fruit phenotype (pf1 and pf2) were characterized in this study. In the pf mutants, SlMYB12 transcripts accumulated to wild-type levels but exhibited the same truncation, which resulted in the absence of the essential MYB activation domain coding region. Allelism and complementation tests revealed that both pf mutants were allelic to the y locus and showed the same recessive null allele in homozygosis: Δy A set of molecular and metabolic effects, reminiscent of those observed in the Arabidopsis (Arabidopsis thaliana) myb11 myb12 myb111 triple mutant, were found in the tomato Δy mutants. To our knowledge, these have not been described previously, and our data support the idea of their being null mutants, in contrast to previously described transcriptional hypomorphic pink fruit lines. We detected a reduction in the expression of several flavonol glycosides and some associated glycosyl transferases. Transcriptome analysis further revealed that the effects of the pf mutations extended beyond the flavonoid pathway into the interface between primary and secondary metabolism. Finally, screening for Myb-binding sites in the candidate gene promoter sequences revealed that 141 of the 152 co-down-regulated genes may be direct targets of SlMYB12 regulation.

hfAIM: A reliable bioinformatics approach for in silico genome-wide identification of autophagy-associated Atg8-interacting motifs in various organisms

Most of the proteins that are specifically turned over by selective autophagy are recognized by the presence of short Atg8 interacting motifs (AIMs) that facilitate their association with the autophagy apparatus. Such AIMs can be identified by bioinformatics methods based on their defined degenerate consensus F/W/Y-X-X-L/I/V sequences in which X represents any amino acid. Achieving reliability and/or fidelity of the prediction of such AIMs on a genome-wide scale represents a major challenge. Here, we present a bioinformatics approach, high fidelity AIM (hfAIM), which uses additional sequence requirements-the presence of acidic amino acids and the absence of positively charged amino acids in certain positions-to reliably identify AIMs in proteins. We demonstrate that the use of the hfAIM method allows for in silico high fidelity prediction of AIMs in AIM-containing proteins (ACPs) on a genome-wide scale in various organisms. Furthermore, by using hfAIM to identify putative AIMs in the Arabidopsis proteome, we illustrate a potential contribution of selective autophagy to various biological processes. More specifically, we identified 9 peroxisomal PEX proteins that contain hfAIM motifs, among which AtPEX1, AtPEX6 and AtPEX10 possess evolutionary-conserved AIMs. Bimolecular fluorescence complementation (BiFC) results verified that AtPEX6 and AtPEX10 indeed interact with Atg8 in planta. In addition, we show that mutations occurring within or nearby hfAIMs in PEX1, PEX6 and PEX10 caused defects in the growth and development of various organisms. Taken together, the above results suggest that the hfAIM tool can be used to effectively perform genome-wide in silico screens of proteins that are potentially regulated by selective autophagy. The hfAIM system is a web tool that can be accessed at link: http://bioinformatics.psb.ugent.be/hfAIM/.

CoExpNetViz: Comparative Co-Expression Networks Construction and Visualization Tool

MOTIVATION

Comparative transcriptomics is a common approach in functional gene discovery efforts. It allows for finding conserved co-expression patterns between orthologous genes in closely related plant species, suggesting that these genes potentially share similar function and regulation. Several efficient co-expression-based tools have been commonly used in plant research but most of these pipelines are limited to data from model systems, which greatly limit their utility. Moreover, in addition, none of the existing pipelines allow plant researchers to make use of their own unpublished gene expression data for performing a comparative co-expression analysis and generate multi-species co-expression networks.

RESULTS

We introduce CoExpNetViz, a computational tool that uses a set of query or "bait" genes as an input (chosen by the user) and a minimum of one pre-processed gene expression dataset. The CoExpNetViz algorithm proceeds in three main steps; (i) for every bait gene submitted, co-expression values are calculated using mutual information and Pearson correlation coefficients, (ii) non-bait (or target) genes are grouped based on cross-species orthology, and (iii) output files are generated and results can be visualized as network graphs in Cytoscape.

AVAILABILITY

The CoExpNetViz tool is freely available both as a PHP web server (link: http://bioinformatics.psb.ugent.be/webtools/coexpr/) (implemented in C++) and as a Cytoscape plugin (implemented in Java). Both versions of the CoExpNetViz tool support LINUX and Windows platforms.

The MORPH-R web server and software tool for predicting missing genes in biological pathways

A biological pathway is the set of molecular entities involved in a given biological process and the interrelations among them. Even though biological pathways have been studied extensively, discovering missing genes in pathways remains a fundamental challenge. Here, we present an easy-to-use tool that allows users to run MORPH (MOdule-guided Ranking of candidate PatHway genes), an algorithm for revealing missing genes in biological pathways, and demonstrate its capabilities. MORPH supports the analysis in tomato, Arabidopsis and the two new species: rice and the newly sequenced potato genome. The new tool, called MORPH-R, is available both as a web server (at http://bioinformatics.psb.ugent.be/webtools/morph/) and as standalone software that can be used locally. In the standalone version, the user can apply the tool to new organisms using any proprietary and public data sources.

Combined correlation-based network and mQTL analyses efficiently identified loci for branched-chain amino acid, serine to threonine, and proline metabolism in tomato seeds

Correlation-based network analysis (CNA) of the metabolic profiles of seeds of a tomato introgression line mapping population revealed a clique of proteinogenic amino acids: Gly, Ile, Pro, Ser, Thr, and Val. Correlations between profiles of these amino acids exhibited a statistically significant average correlation coefficient of 0.84 as compared with an average correlation coefficient of 0.39 over the 16 119 other metabolite cliques containing six metabolites. In silico removal of cliques was used to quantify their importance in determining seminal network properties, highlighting the strong effects of the amino acid clique. Quantitative trait locus analysis revealed co-localization for the six amino acids on chromosome 2, 4 and 10. Sequence analysis identified a unique set of 10 genes on chromosome 2 only, which were associated with amino acid metabolism and specifically the metabolism of Ser-Gly and their conversion into branched-chain amino acids. Metabolite profiling of a set of sublines, with introgressions on chromosome 2, identified a significant change in the abundance of the six amino acids in comparison with M82. Expression analysis of candidate genes affecting Ser metabolism matched the observation from the metabolite data, suggesting a coordinated behavior of the level of these amino acids at the genetic level. Analysis of transcription factor binding sites in the promoter regions of the identified genes suggested combinatorial response to light and the circadian clock.

Evaluation and integration of functional annotation pipelines for newly sequenced organisms: the potato genome as a test case

BACKGROUND

For most organisms, even if their genome sequence is available, little functional information about individual genes or proteins exists. Several annotation pipelines have been developed for functional analysis based on sequence, 'omics', and literature data. However, researchers encounter little guidance on how well they perform. Here, we used the recently sequenced potato genome as a case study. The potato genome was selected since its genome is newly sequenced and it is a non-model plant even if there is relatively ample information on individual potato genes, and multiple gene expression profiles are available.

RESULTS

We show that the automatic gene annotations of potato have low accuracy when compared to a "gold standard" based on experimentally validated potato genes. Furthermore, we evaluate six state-of-the-art annotation pipelines and show that their predictions are markedly dissimilar (Jaccard similarity coefficient of 0.27 between pipelines on average). To overcome this discrepancy, we introduce a simple GO structure-based algorithm that reconciles the predictions of the different pipelines. We show that the integrated annotation covers more genes, increases by over 50% the number of highly co-expressed GO processes, and obtains much higher agreement with the gold standard.

CONCLUSIONS

We find that different annotation pipelines produce different results, and show how to integrate them into a unified annotation that is of higher quality than each single pipeline. We offer an improved functional annotation of both PGSC and ITAG potato gene models, as well as tools that can be applied to additional pipelines and improve annotation in other organisms. This will greatly aid future functional analysis of '-omics' datasets from potato and other organisms with newly sequenced genomes. The new potato annotations are available with this paper.

Rewiring Host Lipid Metabolism by Large Viruses Determines the Fate of Emiliania huxleyi, a Bloom-Forming Alga in the Ocean

Marine viruses are major ecological and evolutionary drivers of microbial food webs regulating the fate of carbon in the ocean. We combined transcriptomic and metabolomic analyses to explore the cellular pathways mediating the interaction between the bloom-forming coccolithophore Emiliania huxleyi and its specific coccolithoviruses (E. huxleyi virus [EhV]). We show that EhV induces profound transcriptome remodeling targeted toward fatty acid synthesis to support viral assembly. A metabolic shift toward production of viral-derived sphingolipids was detected during infection and coincided with downregulation of host de novo sphingolipid genes and induction of the viral-encoded homologous pathway. The depletion of host-specific sterols during lytic infection and their detection in purified virions revealed their novel role in viral life cycle. We identify an essential function of the mevalonate-isoprenoid branch of sterol biosynthesis during infection and propose its downregulation as an antiviral mechanism. We demonstrate how viral replication depends on the hijacking of host lipid metabolism during the chemical "arms race" in the ocean.

The Arabidopsis exocyst subcomplex subunits involved in a golgi-independent transport into the vacuole possess consensus autophagy-associated atg8 interacting motifs

The exocyst complex is a multi-subunits evolutionary conserved complex, which was originally shown to be primarily associated with vesicular transport to the plasma membrane. A recent report (Kulich et al., 2013 Traffic; In Press) revealed that AtEXO70B1, one of the multiple subunits of the exocyst complex of Arabidopsis thaliana plants, is co-transported with the autophagy-associated Atg8f protein to the vacuole. This pathway does not involve the Golgi apparatus. The co-localization of AtEXO70B1 and Atg8f suggests either that both of these proteins are co-transported together to the vacuole or, alternatively, that Atg8 binds to a putative Atg8 interacting motif (AIM) located within the AtEXO70B1 polypeptide, apparently forming a tethering complex for an autophagic complex that is transported to the vacuole. In the present addendum, by tooling a bioinformatics approach, we show that AtEXO70B1 as well as the additional 20 paralogs of Arabidopsis EXO70 exocyst subunits each possess one or more AIMs whose consensus sequence implies their high fidelity binding to Atg8. This indicates that the autophagy machinery is strongly involved in the assembly, transport, and apparently also the function of AtEXO70B1 as well as the exocyst sub complex.

Biosynthesis of antinutritional alkaloids in solanaceous crops is mediated by clustered genes

Steroidal glycoalkaloids (SGAs) such as α-solanine found in solanaceous food plants--as, for example, potato--are antinutritional factors for humans. Comparative coexpression analysis between tomato and potato coupled with chemical profiling revealed an array of 10 genes that partake in SGA biosynthesis. We discovered that six of them exist as a cluster on chromosome 7, whereas an additional two are adjacent in a duplicated genomic region on chromosome 12. Following systematic functional analysis, we suggest a revised SGA biosynthetic pathway starting from cholesterol up to the tetrasaccharide moiety linked to the tomato SGA aglycone. Silencing GLYCOALKALOID METABOLISM 4 prevented accumulation of SGAs in potato tubers and tomato fruit. This may provide a means for removal of unsafe, antinutritional substances present in these widely used food crops.

The MORPH algorithm: ranking candidate genes for membership in Arabidopsis and tomato pathways

Closing gaps in our current knowledge about biological pathways is a fundamental challenge. The development of novel computational methods along with high-throughput experimental data carries the promise to help in the challenge. We present an algorithm called MORPH (for module-guided ranking of candidate pathway genes) for revealing unknown genes in biological pathways. The method receives as input a set of known genes from the target pathway, a collection of expression profiles, and interaction and metabolic networks. Using machine learning techniques, MORPH selects the best combination of data and analysis method and outputs a ranking of candidate genes predicted to belong to the target pathway. We tested MORPH on 230 known pathways in Arabidopsis thaliana and 93 known pathways in tomato (Solanum lycopersicum) and obtained high-quality cross-validation results. In the photosynthesis light reactions, homogalacturonan biosynthesis, and chlorophyll biosynthetic pathways of Arabidopsis, genes ranked highly by MORPH were recently verified to be associated with these pathways. MORPH candidates ranked for the carotenoid pathway from Arabidopsis and tomato are derived from pathways that compete for common precursors or from pathways that are coregulated with or regulate the carotenoid biosynthetic pathway.

Lycopene cyclase paralog CruP protects against reactive oxygen species in oxygenic photosynthetic organisms

In photosynthetic organisms, carotenoids serve essential roles in photosynthesis and photoprotection. A previous report designated CruP as a secondary lycopene cyclase involved in carotenoid biosynthesis [Maresca J, et al. (2007) Proc Natl Acad Sci USA 104:11784-11789]. However, we found that cruP KO or cruP overexpression plants do not exhibit correspondingly reduced or increased production of cyclized carotenoids, which would be expected if CruP was a lycopene cyclase. Instead, we show that CruP aids in preventing accumulation of reactive oxygen species (ROS), thereby reducing accumulation of β-carotene-5,6-epoxide, a ROS-catalyzed autoxidation product, and inhibiting accumulation of anthocyanins, which are known chemical indicators of ROS. Plants with a nonfunctional cruP accumulate substantially higher levels of ROS and β-carotene-5,6-epoxide in green tissues. Plants overexpressing cruP show reduced levels of ROS, β-carotene-5,6-epoxide, and anthocyanins. The observed up-regulation of cruP transcripts under photoinhibitory and lipid peroxidation-inducing conditions, such as high light stress, cold stress, anoxia, and low levels of CO(2), fits with a role for CruP in mitigating the effects of ROS. Phylogenetic distribution of CruP in prokaryotes showed that the gene is only present in cyanobacteria that live in habitats characterized by large variation in temperature and inorganic carbon availability. Therefore, CruP represents a unique target for developing resilient plants and algae needed to supply food and biofuels in the face of global climate change.