Revealing the dynamics and mechanisms of bacterial interactions in cheese production with metabolic modelling.

Cheese taste and flavour properties result from complex metabolic processes occurring in microbial communities. A deeper understanding of such mechanisms makes it possible to improve both industrial production processes and end-product quality through the design of microbial consortia. In this work, we caracterise the metabolism of a three-species community consisting of Lactococcus lactis, Lactobacillus plantarum and Propionibacterium freudenreichii during a seven-week cheese production process. Using genome-scale metabolic models and omics data integration, we modeled and calibrated individual dynamics using monoculture experiments, and coupled these models to capture the metabolism of the community. This model accurately predicts the dynamics of the community, enlightening the contribution of each microbial species to organoleptic compound production. Further metabolic exploration revealed additional possible interactions between the bacterial species. This work provides a methodological framework for the prediction of community-wide metabolism and highlights the added value of dynamic metabolic modeling for the comprehension of fermented food processes.

Translatome analysis at the egg-to-embryo transition in sea urchin.

Early embryogenesis relies on the translational regulation of maternally stored mRNAs. In sea urchin, fertilization triggers a dramatic rise in translation activity, necessary for the onset of cell division. Here, the full spectrum of the mRNAs translated upon fertilization was investigated by polysome profiling and sequencing. The translatome of the early sea urchin embryo gave a complete picture of the polysomal recruitment dynamics following fertilization. Our results indicate that only a subset of maternal mRNAs were selectively recruited onto polysomes, with over-represented functional categories in the translated set. The increase in translation upon fertilization depends on the formation of translation initiation complexes following mTOR pathway activation. Surprisingly, mTOR pathway inhibition differentially affected polysomal recruitment of the newly translated mRNAs, which thus appeared either mTOR-dependent or mTOR-independent. Therefore, our data argue for an alternative to the classical cap-dependent model of translation in early development. The identification of the mRNAs translated following fertilization helped assign translational activation events to specific mRNAs. This translatome is the first step to a comprehensive analysis of the molecular mechanisms governing translation upon fertilization and the translational regulatory networks that control the egg-to-embryo transition as well as the early steps of embryogenesis.

Elucidating the response of Kluyveromyces lactis to arsenite and peroxide stress and the role of the transcription factor KlYap8.

All organisms need to sense and respond to a range of stress conditions. In this study, we used transcriptional profiling to identify genes and cellular processes that are responsive during arsenite and tert-butyl hydroperoxide exposure in Kluyveromyces lactis. Many arsenite-responsive genes encode proteins involved in redox processes, protein folding and stabilization, and transmembrane transport. The majority of peroxide-responsive genes encode functions related to transcription, translation, redox processes, metabolism and transport. A substantial number of these stress-regulated genes contain binding motifs for the AP-1 like transcription factors KlYap1 and KlYap8. We demonstrate that KlYap8 binds to and regulates gene expression through a 13 base-pair promoter motif, and that KlYap8 provides protection against arsenite, antimonite, cadmium and peroxide toxicity. Direct transport assays show that Klyap8Δ cells accumulate more arsenic and cadmium than wild type cells and that the Klyap8Δ mutant is defective in arsenic and cadmium export. KlYap8 regulates gene expression in response to both arsenite and peroxide, and might cooperate with KlYap1 in regulation of specific gene targets. Comparison of KlYap8 with its Saccharomyces cerevisiae orthologue ScYap8 indicates that KlYap8 senses and responds to multiple stress signals whereas ScYap8 is only involved in the response to arsenite and antimonite. Thus, our data suggest that functional specialization of ScYap8 has occurred after the whole genome duplication event. This is the first genome-wide stress response analysis in K. lactis and the first demonstration of KlYap8 function.

A comprehensive evaluation of normalization methods for Illumina high-throughput RNA sequencing data analysis.

During the last 3 years, a number of approaches for the normalization of RNA sequencing data have emerged in the literature, differing both in the type of bias adjustment and in the statistical strategy adopted. However, as data continue to accumulate, there has been no clear consensus on the appropriate normalization method to be used or the impact of a chosen method on the downstream analysis. In this work, we focus on a comprehensive comparison of seven recently proposed normalization methods for the differential analysis of RNA-seq data, with an emphasis on the use of varied real and simulated datasets involving different species and experimental designs to represent data characteristics commonly observed in practice. Based on this comparison study, we propose practical recommendations on the appropriate normalization method to be used and its impact on the differential analysis of RNA-seq data.

Engineering multi-degrading bacterial communities to bioremediate soils contaminated with pesticides residues.

Parallel to the important use of pesticides in conventional agriculture there is a growing interest for green technologies to clear contaminated soil from pesticides and their degradation products. Bioaugmentation i. e. the inoculation of degrading micro-organisms in polluted soil, is a promising method still in needs of further developments. Specifically, improvements in the understanding of how degrading microorganisms must overcome abiotic filters and interact with the autochthonous microbial communities are needed in order to efficiently design bioremediation strategies. Here we designed a protocol aiming at studying the degradation of two herbicides, glyphosate (GLY) and isoproturon (IPU), via experimental modifications of two source bacterial communities. We used statistical methods stemming from genomic prediction to link community composition to herbicides degradation potentials. Our approach proved to be efficient with correlation estimates over 0.8 - between model predictions and measured pesticide degradation values. Multi-degrading bacterial communities were obtained by coalescing bacterial communities with high GLY or IPU degradation ability based on their community-level properties. Finally, we evaluated the efficiency of constructed multi-degrading communities to remove pesticide contamination in a different soil. While results are less clear in the case of GLY, we showed an efficient transfer of degrading capacities towards the receiving soil even at relatively low inoculation levels in the case of IPU. Altogether, we developed an innovative protocol for building multi-degrading simplified bacterial communities with the help of genomic prediction tools and coalescence, and proved their efficiency in a contaminated soil.

New insights into sulfur metabolism in yeasts as revealed by studies of Yarrowia lipolytica.

Yarrowia lipolytica, located at the frontier of hemiascomycetous yeasts and fungi, is an excellent candidate for studies of metabolism evolution. This yeast, widely recognized for its technological applications, in particular produces volatile sulfur compounds (VSCs) that fully contribute to the flavor of smear cheese. We report here a relevant global vision of sulfur metabolism in Y. lipolytica based on a comparison between high- and low-sulfur source supplies (sulfate, methionine, or cystine) by combined approaches (transcriptomics, metabolite profiling, and VSC analysis). The strongest repression of the sulfate assimilation pathway was observed in the case of high methionine supply, together with a large accumulation of sulfur intermediates. A high sulfate supply seems to provoke considerable cellular stress via sulfite production, resulting in a decrease of the availability of the glutathione pathway's sulfur intermediates. The most limited effect was observed for the cystine supply, suggesting that the intracellular cysteine level is more controlled than that of methionine and sulfate. Using a combination of metabolomic profiling and genetic experiments, we revealed taurine and hypotaurine metabolism in yeast for the first time. On the basis of a phylogenetic study, we then demonstrated that this pathway was lost by some of the hemiascomycetous yeasts during evolution.

Accumulation of intracellular glycogen and trehalose by Propionibacterium freudenreichii under conditions mimicking cheese ripening in the cold.

Seven Propionibacterium freudenreichii strains exhibited similar responses when placed at 4°C. They slowed down cell machinery, displayed cold stress responses, and rerouted their carbon metabolism toward trehalose and glycogen synthesis, both accumulated in cells. These results highlight the molecular basis of long-term survival of P. freudenreichii in the cold.

Specific and extensive endometrial deregulation is present before conception in IVF/ICSI repeated implantation failures (IF) or recurrent miscarriages.

The objective was to examine if IVF/ICSI repeated implantation failures (IF) or recurrent miscarriages (RM) could be related to preconceptional endometrial deregulations. IF was defined as the absence of pregnancy despite the transfer of at least ten IVF/ICSI good quality embryos, and RM as having at least three unexplained miscarriages. Fertile controls (FC) were women who had given birth at least once. Endometrial biopsy was performed in the mild luteal phase of a non-conceptual cycle (five women were selected in each group). Affymetrix chips (GeneChip Human Genome U133 Plus2.0 Array) were used for hybridization. Data were normalized by the gcRMA method, and raw p values adjusted by the Bonferroni procedure (1%). Differential expression of selected genes was analysed using real-time PCR. Gene networks and biological functions were explored using the Ingenuity Pathways Analysis software. Endometrial gene expression profiles at the time of uterine receptivity differ dramatically in the endometrium among FC, RM, and IF patients. Compared to FC, 2126 and 2477 genes are differentially expressed in IF and RM groups, respectively, and 2363 between IF and RM. In both conditions, differential gene expression referred mainly to DNA transcription and expression. Other main cellular functions deregulated in IF conditions correspond to cell morphology, cellular development, cell cycle, and cellular assembly, while in RM conditions, deregulated cellular functions relate to cell signalling (degradation of cyclic AMP and calcium metabolism) and cellular maintenance. In both conditions, there is an over-representation of deregulations related to the haematological system. In the IF condition, cell-mediated immune response and nervous system development and function are highly deregulated, while in RM patients, main deregulations are in organ and tissue development, humoral immune response, and muscular system development and function. Extensive endometrial deregulations are present before conception in patients who experienced IF or RM with both distinct and common deregulation.

Endometrium as an early sensor of in vitro embryo manipulation technologies.

Implantation is crucial for placental development that will subsequently impact fetal growth and pregnancy success with consequences on postnatal health. We postulated that the pattern of genes expressed by the endometrium when the embryo becomes attached to the mother uterus could account for the final outcome of a pregnancy. As a model, we used the bovine species where the embryo becomes progressively and permanently attached to the endometrium from day 20 of gestation onwards. At that stage, we compared the endometrial genes profiles in the presence of an in vivo fertilized embryo (AI) with the endometrial patterns obtained in the presence of nuclear transfer (SCNT) or in vitro fertilized embryos (IVF), both displaying lower and different potentials for term development. Our data provide evidence that the endometrium can be considered as a biological sensor able to fine-tune its physiology in response to the presence of embryos whose development will become altered much later after the implantation process. Compared with AI, numerous biological functions and several canonical pathways with a major impact on metabolism and immune function were found to be significantly altered in the endometrium of SCNT pregnancies at implantation, whereas the differences were less pronounced with IVF embryos. Determining the limits of the endometrial plasticity at the onset of implantation should bring new insights on the contribution of the maternal environment to the development of an embryo and the success of pregnancy.

Unraveling negative biotic interactions determining soil microbial community assembly and functioning.

Microbial communities play important roles in all ecosystems and yet a comprehensive understanding of the ecological processes governing the assembly of these communities is missing. To address the role of biotic interactions between microorganisms in assembly and for functioning of the soil microbiota, we used a top-down manipulation approach based on the removal of various populations in a natural soil microbial community. We hypothesized that removal of certain microbial groups will strongly affect the relative fitness of many others, therefore unraveling the contribution of biotic interactions in shaping the soil microbiome. Here we show that 39% of the dominant bacterial taxa across treatments were subjected to competitive interactions during soil recolonization, highlighting the importance of biotic interactions in the assembly of microbial communities in soil. Moreover, our approach allowed the identification of microbial community assembly rule as exemplified by the competitive exclusion between members of Bacillales and Proteobacteriales. Modified biotic interactions resulted in greater changes in activities related to N- than to C-cycling. Our approach can provide a new and promising avenue to study microbial interactions in complex ecosystems as well as the links between microbial community composition and ecosystem function.

Global regulation of the response to sulfur availability in the cheese-related bacterium Brevibacterium aurantiacum.

In this study, we combined metabolic reconstruction, growth assays, and metabolome and transcriptome analyses to obtain a global view of the sulfur metabolic network and of the response to sulfur availability in Brevibacterium aurantiacum. In agreement with the growth of B. aurantiacum in the presence of sulfate and cystine, the metabolic reconstruction showed the presence of a sulfate assimilation pathway, thiolation pathways that produce cysteine (cysE and cysK) or homocysteine (metX and metY) from sulfide, at least one gene of the transsulfuration pathway (aecD), and genes encoding three MetE-type methionine synthases. We also compared the expression profiles of B. aurantiacum ATCC 9175 during sulfur starvation or in the presence of sulfate. Under sulfur starvation, 690 genes, including 21 genes involved in sulfur metabolism and 29 genes encoding amino acids and peptide transporters, were differentially expressed. We also investigated changes in pools of sulfur-containing metabolites and in expression profiles after growth in the presence of sulfate, cystine, or methionine plus cystine. The expression of genes involved in sulfate assimilation and cysteine synthesis was repressed in the presence of cystine, whereas the expression of metX, metY, metE1, metE2, and BL613, encoding a probable cystathionine-γ-synthase, decreased in the presence of methionine. We identified three ABC transporters: two operons encoding transporters were transcribed more strongly during cysteine limitation, and one was transcribed more strongly during methionine depletion. Finally, the expression of genes encoding a methionine γ-lyase (BL929) and a methionine transporter (metPS) was induced in the presence of methionine in conjunction with a significant increase in volatile sulfur compound production.

Exploration of sulfur metabolism in the yeast Kluyveromyces lactis.

Hemiascomycetes are separated by considerable evolutionary distances and, as a consequence, the mechanisms involved in sulfur metabolism in the extensively studied yeast, Saccharomyces cerevisiae, could be different from those of other species of the phylum. This is the first time that a global view of sulfur metabolism is reported in the biotechnological yeast Kluyveromyces lactis. We used combined approaches based on transcriptome analysis, metabolome profiling, and analysis of volatile sulfur compounds (VSCs). A comparison between high and low sulfur source supplies, i.e., sulfate, methionine, or cystine, was carried out in order to identify key steps in the biosynthetic and catabolic pathways of the sulfur metabolism. We found that sulfur metabolism of K. lactis is mainly modulated by methionine. Furthermore, since sulfur assimilation is highly regulated, genes coding for numerous transporters, key enzymes involved in sulfate assimilation and the interconversion of cysteine to methionine pathways are repressed under conditions of high sulfur supply. Consequently, as highlighted by metabolomic results, intracellular pools of homocysteine and cysteine are maintained at very low concentrations, while the cystathionine pool is highly expandable. Moreover, our results suggest a new catabolic pathway for methionine to VSCs in this yeast: methionine is transaminated by the ARO8 gene product into 4-methylthio-oxobutyric acid (KMBA), which could be exported outside of the cell by the transporter encoded by PDR12 and demethiolated by a spontaneous reaction into methanethiol and its derivatives.

Fine-Tuning of Process Parameters Modulates Specific Metabolic Bacterial Activities and Aroma Compound Production in Semi-Hard Cheese.

The formation of cheese flavor mainly results from the production of volatile compounds by microorganisms. We investigated how fine-tuning cheese-making process parameters changed the cheese volatilome in a semi-hard cheese inoculated with Lactococcus (L.) lactis, Lactiplantibacillus (L.) plantarum, and Propionibacterium (P.) freudenreichii. A standard (Std) cheese was compared with three variants of technological itineraries: a shorter salting time (7 h vs 10 h, Salt7h), a shorter stirring time (15 min vs 30 min, Stir15min), or a higher ripening temperature (16 °C vs 13 °C, Rip16°C). Bacterial counts were similar in the four cheese types, except for a 1.4 log10 reduction of L. lactis counts in Rip16°C cheeses after 7 weeks of ripening. Compared to Std, Stir15min and Rip16°C increased propionibacterial activity, causing higher concentrations of acetic, succinic, and propanoic acids and lower levels of lactic acid. Rip16°C accelerated secondary proteolysis and volatile production. We thus demonstrated that fine-tuning process parameters could modulate the cheese volatilome by influencing specific bacterial metabolisms.

Gene expression profiles of bovine caruncular and intercaruncular endometrium at implantation.

At implantation the endometrium undergoes modifications necessary for its physical interactions with the trophoblast as well as the development of the conceptus. We aim to identify endometrial factors and pathways essential for a successful implantation in the caruncular (C) and the intercaruncular (IC) areas in cattle. Using a 13,257-element bovine oligonucleotide array, we established expression profiles at day 20 of the estrous cycle or pregnancy (implantation), revealing 446 and 1,295 differentially expressed genes (DEG) in C and IC areas, respectively (false discovery rate = 0.08). The impact of the conceptus was higher on the immune response function in C but more prominent on the regulation of metabolism function in IC. The C vs. IC direct comparison revealed 1,177 and 453 DEG in cyclic and pregnant animals respectively (false discovery rate = 0.05), with a major impact of the conceptus on metabolism and cell adhesion. We selected 15 genes including C11ORF34, CXCL12, CXCR4, PLAC8, SCARA5, and NPY and confirmed their differential expression by quantitative RT-PCR. The cellular localization was analyzed by in situ hybridization and, upon pregnancy, showed gene-specific patterns of cell distribution, including a high level of expression in the luminal epithelium for C11ORF34 and MX1. Using primary cultures of bovine endometrial cells, we identified PTN, PLAC8, and CXCL12 as interferon-tau (IFNT) target genes and MSX1 and CXCR7 as IFNT-regulated genes, whereas C11ORF34 was not an IFNT-regulated gene. Our transcriptomic data provide novel molecular insights accounting for the biological functions related to the C or IC endometrial areas and may contribute to the identification of potential biomarkers for normal and perturbed early pregnancy.

Variational Inference for Coupled Hidden Markov Models Applied to the Joint Detection of Copy Number Variations.

Hidden Markov models provide a natural statistical framework for the detection of the copy number variations (CNV) in genomics. In this context, we define a hidden Markov process that underlies all individuals jointly in order to detect and to classify genomics regions in different states (typically, deletion, normal or amplification). Structural variations from different individuals may be dependent. It is the case in agronomy where varietal selection program exists and species share a common phylogenetic past. We propose to take into account these dependencies inthe HMM model. When dealing with a large number of series, maximum likelihood inference (performed classically using the EM algorithm) becomes intractable. We thus propose an approximate inference algorithm based on a variational approach (VEM), implemented in the CHMM R package. A simulation study is performed to assess the performance of the proposed method and an application to the detection of structural variations in plant genomes is presented.

Subclinical endometritis in dairy cattle is associated with distinct mRNA expression patterns in blood and endometrium.

Cattle with subclinical endometritis (SCE) are sub-fertile and diagnosing subclinical uterine disease remains a challenge. The hypothesis for this study was that endometrial inflammation is reflected in mRNA expression patterns of peripheral blood leucocytes. Transcriptome profiles were evaluated in healthy cows and in cows with SCE using circulating white blood cells (WBC) and endometrial biopsy samples collected from the same animals at 45-55 days postpartum. Bioinformatic analyses of microarray-based transcriptional data identified gene profiles associated with distinct biological functions in circulating WBC and endometrium. In circulating WBC, SCE promotes a pro-inflammatory environment, whereas functions related to tissue remodeling are also affected in the endometrium. Nineteen differentially expressed genes associated with SCE were common to both circulating WBC and the endometrium. Among these genes, transcript abundance of immune factors C3, C2, LTF, PF4 and TRAPPC13 were up-regulated in SCE cows at 45-55 days postpartum. Moreover, mRNA expression of C3, CXCL8, LTF, TLR2 and TRAPPC13 was temporally regulated during the postpartum period in circulating WBC of healthy cows compared with SCE cows. This observation might indicate an advantageous modulation of the immune system in healthy animals. The transcript abundance of these genes represents a potential source of indicators for postpartum uterine health.

Statistical methodology for the analysis of dye-switch microarray experiments.

BACKGROUND: In individually dye-balanced microarray designs, each biological sample is hybridized on two different slides, once with Cy3 and once with Cy5. While this strategy ensures an automatic correction of the gene-specific labelling bias, it also induces dependencies between log-ratio measurements that must be taken into account in the statistical analysis. RESULTS: We present two original statistical procedures for the statistical analysis of individually balanced designs. These procedures are compared with the usual ML and REML mixed model procedures proposed in most statistical toolboxes, on both simulated and real data. CONCLUSION: The UP procedure we propose as an alternative to usual mixed model procedures is more efficient and significantly faster to compute. This result provides some useful guidelines for the analysis of complex designs.

Normalization for triple-target microarray experiments.

BACKGROUND: Most microarray studies are made using labelling with one or two dyes which allows the hybridization of one or two samples on the same slide. In such experiments, the most frequently used dyes are Cy3 and Cy5. Recent improvements in the technology (dye-labelling, scanner and, image analysis) allow hybridization up to four samples simultaneously. The two additional dyes are Alexa488 and Alexa494. The triple-target or four-target technology is very promising, since it allows more flexibility in the design of experiments, an increase in the statistical power when comparing gene expressions induced by different conditions and a scaled down number of slides. However, there have been few methods proposed for statistical analysis of such data. Moreover the lowess correction of the global dye effect is available for only two-color experiments, and even if its application can be derived, it does not allow simultaneous correction of the raw data. RESULTS: We propose a two-step normalization procedure for triple-target experiments. First the dye bleeding is evaluated and corrected if necessary. Then the signal in each channel is normalized using a generalized lowess procedure to correct a global dye bias. The normalization procedure is validated using triple-self experiments and by comparing the results of triple-target and two-color experiments. Although the focus is on triple-target microarrays, the proposed method can be used to normalize p differently labelled targets co-hybridized on a same array, for any value of p greater than 2. CONCLUSION: The proposed normalization procedure is effective: the technical biases are reduced, the number of false positives is under control in the analysis of differentially expressed genes, and the triple-target experiments are more powerful than the corresponding two-color experiments. There is room for improving the microarray experiments by simultaneously hybridizing more than two samples.

Transcriptomic response of Debaryomyces hansenii during mixed culture in a liquid model cheese medium with Yarrowia lipolytica.

Yeasts play a crucial role in cheese ripening. They contribute to the curd deacidification, the establishment of acid-sensitive bacterial communities, and flavour compounds production via proteolysis and catabolism of amino acids (AA). Negative yeast-yeast interaction was observed between the yeast Yarrowia lipolytica 1E07 (YL1E07) and the yeast Debaryomyces hansenii 1L25 (DH1L25) in a model cheese but need elucidation. YL1E07 and DH1L25 were cultivated in mono and co-cultures in a liquid synthetic medium (SM) mimicking the cheese environment and the growth inhibition of DH1L25 in the presence of YL1E07 was reproduced. We carried out microbiological, biochemical (lactose, lactate, AA consumption and ammonia production) and transcriptomic analyses by microarray technology to highlight the interaction mechanisms. We showed that the DH1L25 growth inhibition in the presence of YL1E07 was neither due to the ammonia production nor to the nutritional competition for the medium carbon sources between the two yeasts. The transcriptomic study was the key toward the comprehension of yeast-yeast interaction, and revealed that the inhibition of DH1L25 in co-culture is due to a decrease of the mitochondrial respiratory chain functioning.

Profiling at mRNA, protein, and metabolite levels reveals alterations in renal amino acid handling and glutathione metabolism in kidney tissue of Pept2-/- mice.

PEPT2 is an integral membrane protein in the apical membrane of renal epithelial cells that operates as a rheogenic transporter for di- and tripeptides and structurally related drugs. Its prime role is thought to be the reabsorption of filtered di- and tripeptides contributing to amino acid homeostasis. To elucidate the role of PEPT2 in renal amino acid metabolism we submitted kidney tissues of wild-type and a Pept2(-/-) mouse line to a comprehensive transcriptome, proteome and metabolome profiling and analyzed urinary amino acids and dipeptides. cDNA microarray analysis identified 147 differentially expressed transcripts in transporter-deficient animals, and proteome analysis by 2D-PAGE and MALDI-TOF-MS identified 37 differentially expressed proteins. Metabolite profiling by GC-MS revealed predominantly altered concentrations of amino acids and derivatives. Urinary excretion of amino acids demonstrated increased glycine and cysteine/cystine concentrations and dipeptides in urine were assessed by amino acid analysis of urine samples before and after in vitro dipeptidase digestion. Dipeptides constituted a noticeable fraction of urinary amino acids in Pept2(-/-) animals, only, and dipeptide-bound glycine and cystine were selectively increased in Pept2(-/-) urine samples. These findings were confirmed by a drastically increased excretion of cysteinyl-glycine (cys-gly). Urinary loss of cys-gly together with lower concentrations of cysteine, glycine, and oxoproline in kidney tissue and altered expression of mRNA and proteins involved in glutathione (GSH) metabolism suggests that PEPT2 is predominantly a system for reabsorption of cys-gly originating from GSH break-down, thus contributing to resynthesis of GSH.