New approaches to open problems in gene expression microarray data

In the past decade, the advent of efficient genome sequencing tools and high-throughput experimental biotechnology has lead to enormous progress in the life science. Among the most important innovations is the microarray tecnology. It allows to quantify the expression for thousands of genes simultaneously by measurin the hybridization from a tissue of interest to probes on a small glass or plastic slide. The characteristics of these data include a fair amount of random noise, a predictor dimension in the thousand, and a sample noise in the dozens. One of the most exciting areas to which microarray technology has been applied is the challenge of deciphering complex disease such as cancer. In these studies, samples are taken from two or more groups of individuals with heterogeneous phenotypes, pathologies, or clinical outcomes. these samples are hybridized to microarrays in an effort to find a small number of genes which are strongly correlated with the group of individuals. Eventhough today methods to analyse the data are welle developed and close to reach a standard organization (through the effort of preposed International project like Microarray Gene Expression Data -MGED- Society [1]) it is not unfrequant to stumble in a clinician's question that do not have a compelling statistical method that could permit to answer it.The contribution of this dissertation in deciphering disease regards the development of new approaches aiming at handle open problems posed by clinicians in handle specific experimental designs. In Chapter 1 starting from a biological necessary introduction, we revise the microarray tecnologies and all the important steps that involve an experiment from the production of the array, to the quality controls ending with preprocessing steps that will be used into the data analysis in the rest of the dissertation. While in Chapter 2 a critical review of standard analysis methods are provided stressing most of problems that In Chapter 3 is introduced a method to adress the issue of unbalanced design of miacroarray experiments. In microarray experiments, experimental design is a crucial starting-point for obtaining reasonable results. In a two-class problem, an equal or similar number of samples it should be collected between the two classes. However in some cases, e.g. rare pathologies, the approach to be taken is less evident. We propose to address this issue by applying a modified version of SAM [2]. MultiSAM consists in a reiterated application of a SAM analysis, comparing the less populated class (LPC) with 1,000 random samplings of the same size from the more populated class (MPC) A list of the differentially expressed genes is generated for each SAM application. After 1,000 reiterations, each single probe given a "score" ranging from 0 to 1,000 based on its recurrence in the 1,000 lists as differentially expressed. The performance of MultiSAM was compared to the performance of SAM and LIMMA [3] over two simulated data sets via beta and exponential distribution. The results of all three algorithms over low- noise data sets seems acceptable However, on a real unbalanced two-channel data set reagardin Chronic Lymphocitic Leukemia, LIMMA finds no significant probe, SAM finds 23 significantly changed probes but cannot separate the two classes, while MultiSAM finds 122 probes with score >300 and separates the data into two clusters by hierarchical clustering. We also report extra-assay validation in terms of differentially expressed genes Although standard algorithms perform well over low-noise simulated data sets, multi-SAM seems to be the only one able to reveal subtle differences in gene expression profiles on real unbalanced data. In Chapter 4 a method to adress similarities evaluation in a three-class prblem by means of Relevance Vector Machine [4] is described. In fact, looking at microarray data in a prognostic and diagnostic clinical framework, not only differences could have a crucial role. In some cases similarities can give useful and, sometimes even more, important information. The goal, given three classes, could be to establish, with a certain level of confidence, if the third one is similar to the first or the second one. In this work we show that Relevance Vector Machine (RVM) [2] could be a possible solutions to the limitation of standard supervised classification. In fact, RVM offers many advantages compared, for example, with his well-known precursor (Support Vector Machine - SVM [3]). Among these advantages, the estimate of posterior probability of class membership represents a key feature to address the similarity issue. This is a highly important, but often overlooked, option of any practical pattern recognition system. We focused on Tumor-Grade-three-class problem, so we have 67 samples of grade I (G1), 54 samples of grade 3 (G3) and 100 samples of grade 2 (G2). The goal is to find a model able to separate G1 from G3, then evaluate the third class G2 as test-set to obtain the probability for samples of G2 to be member of class G1 or class G3. The analysis showed that breast cancer samples of grade II have a molecular profile more similar to breast cancer samples of grade I. Looking at the literature this result have been guessed, but no measure of significance was gived before.

Expression profiling in developing peach seed and mesocarp as affected by growth regulators

Jasmonates (JAs) and spermidine (Sd) influence fruit (and seed) development and ripening. In order to unravel their effects in peach fruit, at molecular level, field applications of methyl jasmonate (MJ) and propyl dihydrojasmonate (PDJ), and Sd were performed at an early developmental stage (late S1). At commercial harvest, JA-treated fruit were less ripe than controls. Realtime RT-PCR analyses confirmed a down-regulation of ethylene biosynthetic, perception and signaling genes, and flesh softening-related genes. The expression of cell wall-related genes, of a sugar-transporter and hormone-related transcript levels was also affected by JAs. Seeds from JA-treated fruit showed a shift in the expression of developmental marker genes suggesting that the developmental program was probably slowed down, in agreement with the contention that JAs divert resources from growth to defense. JAs also affected phenolic content and biosynthetic gene expression in the mesocarp. Levels of hydroxycinnamic acids, as well as those of flavan-3-ols, were enhanced, mainly by MJ, in S2. Transcript levels of phenylpropanoid pathway genes were up-regulated by MJ, in agreement with phenolic content. Sd-treated fruits at harvest showed reduced ethylene production and flesh softening. Sd induced a short-term and long-term response patterns in endogenous polyamines. At ripening the up-regulation of the ethylene biosynthetic genes was dramatically counteracted by Sd, leading to a down-regulation of softening-related genes. Hormone-related gene expression was also altered both in the short- and long-term. Gene expression analyses suggest that Sd interfered with fruit development/ripening by interacting with multiple hormonal pathways and that fruit developmental marker gene expression was shifted ahead in accord with a developmental slowing down. 24-Epibrassinolide was applied to Flaminia peaches under field conditions early (S1) or later (S3) during development. Preliminary results showed that, at harvest, treated fruit tended to be larger and less mature though quality parameters did not change relative to controls.

Comparative expression profiling of E. coli and S. aureus inoculated primary mammary gland cells sampled from cows with different genetic predispositions for somatic cell score

BACKGROUND: During the past ten years many quantitative trait loci (QTL) affecting mastitis incidence and mastitis related traits like somatic cell score (SCS) were identified in cattle. However, little is known about the molecular architecture of QTL affecting mastitis susceptibility and the underlying physiological mechanisms and genes causing mastitis susceptibility. Here, a genome-wide expression analysis was conducted to analyze molecular mechanisms of mastitis susceptibility that are affected by a specific QTL for SCS on Bos taurus autosome 18 (BTA18). Thereby, some first insights were sought into the genetically determined mechanisms of mammary gland epithelial cells influencing the course of infection. METHODS: Primary bovine mammary gland epithelial cells (pbMEC) were sampled from the udder parenchyma of cows selected for high and low mastitis susceptibility by applying a marker-assisted selection strategy considering QTL and molecular marker information of a confirmed QTL for SCS in the telomeric region of BTA18. The cells were cultured and subsequently inoculated with heat-inactivated mastitis pathogens Escherichia coli and Staphylococcus aureus, respectively. After 1, 6 and 24 h, the cells were harvested and analyzed using the microarray expression chip technology to identify differences in mRNA expression profiles attributed to genetic predisposition, inoculation and cell culture. RESULTS: Comparative analysis of co-expression profiles clearly showed a faster and stronger response after pathogen challenge in pbMEC from less susceptible animals that inherited the favorable QTL allele 'Q' than in pbMEC from more susceptible animals that inherited the unfavorable QTL allele 'q'. Furthermore, the results highlighted RELB as a functional and positional candidate gene and related non-canonical Nf-kappaB signaling as a functional mechanism affected by the QTL. However, in both groups, inoculation resulted in up-regulation of genes associated with the Ingenuity pathways 'dendritic cell maturation' and 'acute phase response signaling', whereas cell culture affected biological processes involved in 'cellular development'. CONCLUSIONS: The results indicate that the complex expression profiling of pathogen challenged pbMEC sampled from cows inheriting alternative QTL alleles is suitable to study genetically determined molecular mechanisms of mastitis susceptibility in mammary epithelial cells in vitro and to highlight the most likely functional pathways and candidate genes underlying the QTL effect.

Gene expression variation between distinct areas of breast cancer measured from paraffin-embedded tissue cores

BACKGROUND: Diagnosis and prognosis in breast cancer are mainly based on histology and immunohistochemistry of formalin-fixed, paraffin-embedded (FFPE) material. Recently, gene expression analysis was shown to elucidate the biological variance between tumors and molecular markers were identified that led to new classification systems that provided better prognostic and predictive parameters. Archived FFPE samples represent an ideal source of tissue for translational research, as millions of tissue blocks exist from routine diagnostics and from clinical studies. These should be exploited to provide clinicians with more accurate prognostic and predictive information. Unfortunately, RNA derived from FFPE material is partially degraded and chemically modified and reliable gene expression measurement has only become successful after implementing novel and optimized procedures for RNA isolation, demodification and detection. METHODS: In this study we used tissue cylinders as known from the construction of tissue microarrays. RNA was isolated with a robust protocol recently developed for RNA derived from FFPE material. Gene expression was measured by quantitative reverse transcription PCR. RESULTS: Sixteen tissue blocks from 7 patients diagnosed with multiple histological subtypes of breast cancer were available for this study. After verification of appropriate localization, sufficient RNA yield and quality, 30 tissue cores were available for gene expression measurement on TaqMan(R) Low Density Arrays (16 invasive ductal carcinoma (IDC), 8 ductal carcinoma in situ (DCIS) and 6 normal tissue), and 14 tissue cores were lost. Gene expression values were used to calculate scores representing the proliferation status (PRO), the estrogen receptor status and the HER2 status. The PRO scores measured from entire sections were similar to PRO scores determined from IDC tissue cores. Scores determined from normal tissue cores consistently revealed lower PRO scores than cores derived from IDC or DCIS of the same block or from different blocks of the same patient. CONCLUSION: We have developed optimized protocols for RNA isolation from histologically distinct areas. RNA prepared from FFPE tissue cores is suitable for gene expression measurement by quantitative PCR. Distinct molecular scores could be determined from different cores of the same tumor specimen.

Gene expression in cultured endometrium from women with different outcomes following IVF

Estradiol and progesterone are crucial for the acquisition of receptivity and the change in transcriptional activity of target genes in the implantation window. The aim of this study was to differentiate the regulation of genes in the endometrium of patients with recurrent implantation failure (IF) versus those who became pregnant after in vitro fertilization (IVF) treatment. Moreover, the effect of embryo-derived factors on endometrial transcriptional activity was studied. Nine women with known IVF outcome (IF, M, miscarriage, OP, ongoing pregnancy) and undergoing hysteroscopy with endometrial biopsy were enrolled. Biopsies were taken during the midluteal phase. After culture in the presence of embryo-conditioned IVF media, total RNA was extracted and submitted to reverse transcription, target cDNA synthesis, biotin labelling, fragmentation and hybridization using the Affymetrix Human Genome U133A 2.0 Chip. Differential expression of selected genes was re-analysed by quantitative PCR, in which the results were calculated as threshold cycle differences between the groups and normalized to Glyceraldehyde phosphate dehydrogenase and beta-actin. Differences were seen for several genes from endometrial tissue between the IF and the pregnancy groups, and when comparing OP with M, 1875 up- and 1807 down-regulated genes were returned. Real-time PCR analysis confirmed up-regulation for somatostatin, PLAP-2, mucin 4 and CD163, and down-regulation of glycodelin, IL-24, CD69, leukaemia inhibitory factor and prolactin receptor between Op and M. When the different embryo-conditioned media were compared, no significant differential regulation could be demonstrated. Although microarray profiling may currently not be sensitive enough for studying the effects of embryo-derived factors on the endometrium, the observed differences in gene expression between M and OP suggest that it will become an interesting tool for the identification of fertility-relevant markers produced by the endometrium.

UPLC-ESI-TOFMS-based metabolomics and gene expression dynamics inspector self-organizing metabolomic maps as tools for understanding the cellular response to ionizing radiation

Global transcriptomic and proteomic profiling platforms have yielded important insights into the complex response to ionizing radiation (IR). Nonetheless, little is known about the ways in which small cellular metabolite concentrations change in response to IR. Here, a metabolomics approach using ultraperformance liquid chromatography coupled with electrospray time-of-flight mass spectrometry was used to profile, over time, the hydrophilic metabolome of TK6 cells exposed to IR doses ranging from 0.5 to 8.0 Gy. Multivariate data analysis of the positive ions revealed dose- and time-dependent clustering of the irradiated cells and identified certain constituents of the water-soluble metabolome as being significantly depleted as early as 1 h after IR. Tandem mass spectrometry was used to confirm metabolite identity. Many of the depleted metabolites are associated with oxidative stress and DNA repair pathways. Included are reduced glutathione, adenosine monophosphate, nicotinamide adenine dinucleotide, and spermine. Similar measurements were performed with a transformed fibroblast cell line, BJ, and it was found that a subset of the identified TK6 metabolites were effective in IR dose discrimination. The GEDI (Gene Expression Dynamics Inspector) algorithm, which is based on self-organizing maps, was used to visualize dynamic global changes in the TK6 metabolome that resulted from IR. It revealed dose-dependent clustering of ions sharing the same trends in concentration change across radiation doses. "Radiation metabolomics," the application of metabolomic analysis to the field of radiobiology, promises to increase our understanding of cellular responses to stressors such as radiation.

Emdogain-regulated gene expression in palatal fibroblasts requires TGF-βRI kinase signaling.

Genome-wide microarrays have suggested that Emdogain regulates TGF-β target genes in gingival and palatal fibroblasts. However, definitive support for this contention and the extent to which TGF-β signaling contributes to the effects of Emdogain has remained elusive. We therefore studied the role of the TGF-β receptor I (TGF-βRI) kinase to mediate the effect of Emdogain on palatal fibroblasts. Palatal fibroblasts were exposed to Emdogain with and without the inhibitor for TGF-βRI kinase, SB431542. Emdogain caused 39 coding genes to be differentially expressed in palatal fibroblasts by microarray analysis (p<0.05; >10-fold). Importantly, in the presence of the TGF-βRI kinase inhibitor SB431542, Emdogain failed to cause any significant changes in gene expression. Consistent with this mechanism, three independent TGF-βRI kinase inhibitors and a TGF-β neutralizing antibody abrogated the increased expression of IL-11, a selected Emdogain target gene. The MAPK inhibitors SB203580 and U0126 lowered the impact of Emdogain on IL-11 expression. The data support that TGF-βRI kinase activity is necessary to mediate the effects of Emdogain on gene expression in vitro.

Insulin, CCAAT/enhancer-binding proteins and lactate regulate the human 11β-hydroxysteroid dehydrogenase type 2 gene expression in colon cancer cell lines.

11β-Hydroxysteroid dehydrogenases (11beta-HSD) modulate mineralocorticoid receptor transactivation by glucocorticoids and regulate access to the glucocorticoid receptor. The isozyme 11beta-HSD2 is selectively expressed in mineralocorticoid target tissues and its activity is reduced in various disease states with abnormal sodium retention and hypertension, including the apparent mineralocorticoid excess. As 50% of patients with essential hypertension are insulin resistant and hyperinsulinemic, we hypothesized that insulin downregulates the 11beta-HSD2 activity. In the present study we show that insulin reduced the 11beta-HSD2 activity in cancer colon cell lines (HCT116, SW620 and HT-29) at the transcriptional level, in a time and dose dependent manner. The downregulation was reversible and required new protein synthesis. Pathway analysis using mRNA profiling revealed that insulin treatment modified the expression of the transcription factor family C/EBPs (CCAAT/enhancer-binding proteins) but also of glycolysis related enzymes. Western blot and real time PCR confirmed an upregulation of C/EBP beta isoforms (LAP and LIP) with a more pronounced increase in the inhibitory isoform LIP. EMSA and reporter gene assays demonstrated the role of C/EBP beta isoforms in HSD11B2 gene expression regulation. In addition, secretion of lactate, a byproduct of glycolysis, was shown to mediate insulin-dependent HSD11B2 downregulation. In summary, we demonstrate that insulin downregulates HSD11B2 through increased LIP expression and augmented lactate secretion. Such mechanisms are of interest and potential significance for sodium reabsorption in the colon.

Bone-Conditioned Medium Changes Gene Expression in Bone-Derived Fibroblasts.

PURPOSE Autologous bone is used for augmentation in the course of oral implant placement. Bone grafts release paracrine signals that can modulate mesenchymal cell differentiation in vitro. The detailed genetic response of the bone-derived fibroblasts to these paracrine signals has remained elusive. Paracrine signals accumulate in bone-conditioned medium (BCM) prepared from porcine cortical bone chips. MATERIALS AND METHODS In this study, bone-derived fibroblasts were exposed to BCM followed by a whole genome expression profiling and downstream quantitative reverse transciptase polymerase chain reaction of the most strongly regulated genes. RESULTS The data show that ADM, IL11, IL33, NOX4, PRG4, and PTX3 were differentially expressed in response to BCM in bone-derived fibroblasts. The transforming growth factor beta (TGF-β) receptor 1 antagonist SB431542 blocked the effect of BCM on the expression of the gene panel, except for IL33. CONCLUSION These in vitro results extend existing evidence that cortical bone chips release paracrine signals that provoke a robust genetic response in mesenchymal cells that is not exclusively mediated via the TGF-β receptor. The present data provide further insights into the process of graft consolidation.

Integrated mRNA and miRNA expression profiling in blood reveals candidate biomarkers associated with endurance exercise in the horse.

The adaptive response to extreme endurance exercise might involve transcriptional and translational regulation by microRNAs (miRNAs). Therefore, the objective of the present study was to perform an integrated analysis of the blood transcriptome and miRNome (using microarrays) in the horse before and after a 160 km endurance competition. A total of 2,453 differentially expressed genes and 167 differentially expressed microRNAs were identified when comparing pre- and post-ride samples. We used a hypergeometric test and its generalization to gain a better understanding of the biological functions regulated by the differentially expressed microRNA. In particular, 44 differentially expressed microRNAs putatively regulated a total of 351 depleted differentially expressed genes involved variously in glucose metabolism, fatty acid oxidation, mitochondrion biogenesis, and immune response pathways. In an independent validation set of animals, graphical Gaussian models confirmed that miR-21-5p, miR-181b-5p and miR-505-5p are candidate regulatory molecules for the adaptation to endurance exercise in the horse. To the best of our knowledge, the present study is the first to provide a comprehensive, integrated overview of the microRNA-mRNA co-regulation networks that may have a key role in controlling post-transcriptomic regulation during endurance exercise.

The molecular basis for beta-lactamase gene expression in B. anthracis, B. cereus and B. thuringiensis

The susceptibility of most Bacillus anthracis strains to β-lactam antibiotics is intriguing considering that the B. anthracis genome harbors two β-lactamase genes, bla1 and bla2, and closely-related species, Bacillus cereus and Bacillus thuringiensis, typically produce β-lactamases. This work demonstrates that B. anthracis bla expression is affected by two genes, sigP and rsp, predicted to encode an extracytoplasmic function sigma factor and an antisigma factor, respectively. Deletion of the sigP/rsp locus abolished bla expression in a penicillin-resistant clinical isolate and had no effect on bla expression in a prototypical penicillin-susceptible strain. Complementation with sigP/rsp from the penicillin-resistant strain, but not the penicillin-susceptible strain, conferred β-lactamase activity upon both mutants. These results are attributed to a nucleotide deletion near the 5' end of rsp in the penicillin-resistant strain that is predicted to result in a nonfunctional protein. B. cereus and B. thuringiensis sigP and rsp homologues are required for inducible penicillin resistance in those species. Expression of the B. cereus or B. thuringiensis sigP and rsp genes in a B. anthracis sigP/rsp-null mutant confers resistance to β-lactam antibiotics, suggesting that while B. anthracis contains the genes necessary for sensing β-lactam antibiotics, the B. anthracis sigP/rsp gene products are insufficient for bla induction. ^ Because alternative sigma factors recognize unique promoter sequence, direct targets can be elucidated by comparing transcriptional profiling results with an in silico search using the sigma factor binding sequence. Potential σP -10 and -35 promoter elements were identified upstream from bla1 bla2 and sigP. Results obtained from searching the B. anthracis genome with the conserved sequences were evaluated against transcriptional profiling results comparing B. anthracis 32 and an isogenic sigP/rsp -null strain. Results from these analyses indicate that while the absence of the sigP gene significantly affects the transcript levels of 16 genes, only bla1, bla2 and sigP are directly regulated by σP. The genomes of B. cereus and B. thuringiensis strains were also analyzed for the potential σP binding elements. The sequence was located upstream from the sigP and bla genes, and previously unidentified genes predicted to encode a penicillin-binding protein (PBP) and a D-alanyl-D-alanine carboxypeptidase, indicating that the σ P regulon in these species responds to cell-wall stress caused by β-lactam antibiotics. ^ β-lactam antibiotics prevent attachment of new peptidoglycan to the cell wall by blocking the active site of PBPs. A B. cereus and B. thuringiensis pbp-encoding gene located near bla1 contains a potential σP recognition sequence upstream from the annotated translational start. Deletion of this gene abolished β-lactam resistance in both strains. Mutations in the active site of the PBP were detrimental to β-lactam resistance in B. cereus, but not B. thuringiensis, indicating that the transpeptidase activity is only important in B. cereus. I also found that transcript levels of the PBP-encoding gene are not significantly affected by the presence of β-lactam antibiotic. Based on these data I hypothesize that the gene product acts a sensor of β-lactam antibiotic. ^

Transcriptional and translational regulators of gene expression in germ cell development

Germ cell development is a highly coordinated process driven, in part, by regulatory mechanisms that control gene expression. Not only transcription, but also translation, is under regulatory control to direct proper germ cell development. In this dissertation, I have focused on two regulators of germ cell development. One is the homeobox protein RHOX10, which has the potential to be both a transcriptional and translational regulator in mouse male germ cell development. The other is the RNA-binding protein, Hermes, which functions as a translational regulator in Xenopus laevis female germ cell development. ^ Rhox10 is a member of reproductive homeobox gene X-(linked (Rhox) gene cluster, of which expression is developmentally regulated in developing mouse testes. To identify the cell types and developmental stages in which Rhox10 might function, I characterized its temporal and spatial expression pattern in mouse embryonic, neonatal, and adult tissues. Among other things, this analysis revealed that both the level and the subcellular localization of RHOX10 are regulated during germ cell development. To understand the role of Rhox10 in germ cell development, I generated transgenic mice expressing an artificial microRNA (miRNA) targeting Rhox10. While this artificial miRNA robustly downregulated RHOX10 protein expression in vitro, it did not significantly reduce RHOX10 expression in vivo. So I next elected to knockdown RHOX10 levels in spermatogonial stem cells (SSCs), which I found highly express both Rhox10 mRNA and RHOX10 protein. Using a recently developed in vitro culture system for SSCs combined with a short-hairpin RNA (shRNA) approach, I strongly depleted RHOX10 expression in SSCs. These RHOX10-depleted cells exhibited a defect in the ability to form stem cell clusters in vitro. Expression profiling analysis revealed many genes regulated by Rhox10, including many meiotic genes, which could be downstream of Rhox10 in a molecular pathway that controls SSC differentiation. ^ RNA recognition motif (RRM) containing protein, Hermes is localized in germ plasm, where dormant mRNAs are also located, of Xenopus oocytes, which implicates its role in translational regulator. To understand the function of Hermes in oocyte meiosis, I used a morpholino oligonucleotide (MO) based knockdown approach. Microinjection of Hermes MO into fully grown oocytes, which are arrested in meiotic prophase, caused acceleration of oocytes reentry into meiosis (i.e., maturation) upon progesterone induction. Using a candidate approach, I identified at least three targets of Hermes: Ringo/Spy, Xcat2, and Mos. Ringo/Spy and Mos are known to have functions in oocyte maturation, while Ringo/Spy, Xcat2 mRNA are localized in the germ plasm of oocytes, which drives germ cell specification after fertilization. This led me to propose that Hermes functions in both oocyte maturation and germ cell development through its ability to regulate 3 crucial target mRNAs. ^

Mannans and endo-beta-mannanases (MAN) in Brachypodium distachyon: expression profiling and possible role of the BdMAN genes during coleorhiza-limited seed germination

Immunolocalization of mannans in the seeds of Brachypodium distachyon reveals the presence of these polysaccharides in the root embryo and in the coleorhiza in the early stages of germination (12h), decreasing thereafter to the point of being hardly detected at 27h. Concurrently, the activity of endo-β-mannanases (MANs; EC 3.2.1.78) that catalyse the hydrolysis of β-1,4 bonds in mannan polymers, increases as germination progresses. The MAN gene family is represented by six members in the Brachypodium genome, and their expression has been explored in different organs and especially in germinating seeds. Transcripts of BdMAN2, BdMAN4 and BdMAN6 accumulate in embryos, with a maximum at 24–30h, and are detected in the coleorhiza and in the root by in situ hybridization analyses, before root protrusion (germination sensu stricto). BdMAN4 is not only present in the embryo root and coleorhiza, but is abundant in the de-embryonated (endosperm) imbibed seeds, while BdMAN2 and BdMAN6 are faintly expressed in endosperm during post-germination (36–42h). BdMAN4 and BdMAN6 transcripts are detected in the aleurone layer. These data indicate that BdMAN2, BdMAN4 and BdMAN6 are important for germination sensu stricto and that BdMAN4 and BdMAN6 may also influence reserve mobilization. Whether the coleorhiza in monocots and the micropylar endosperm in eudicots have similar functions, is discussed.

Importance of replication in microarray gene expression studies: Statistical methods and evidence from repetitive cDNA hybridizations

We present statistical methods for analyzing replicated cDNA microarray expression data and report the results of a controlled experiment. The study was conducted to investigate inherent variability in gene expression data and the extent to which replication in an experiment produces more consistent and reliable findings. We introduce a statistical model to describe the probability that mRNA is contained in the target sample tissue, converted to probe, and ultimately detected on the slide. We also introduce a method to analyze the combined data from all replicates. Of the 288 genes considered in this controlled experiment, 32 would be expected to produce strong hybridization signals because of the known presence of repetitive sequences within them. Results based on individual replicates, however, show that there are 55, 36, and 58 highly expressed genes in replicates 1, 2, and 3, respectively. On the other hand, an analysis by using the combined data from all 3 replicates reveals that only 2 of the 288 genes are incorrectly classified as expressed. Our experiment shows that any single microarray output is subject to substantial variability. By pooling data from replicates, we can provide a more reliable analysis of gene expression data. Therefore, we conclude that designing experiments with replications will greatly reduce misclassification rates. We recommend that at least three replicates be used in designing experiments by using cDNA microarrays, particularly when gene expression data from single specimens are being analyzed.

Expression profiling and sequence diversity of novel DREB genes from common bean (Phaseolus vulgaris L.) and their association with drought-related traits

Common bean is a major dietary component in several countries, but its productivity is negatively affected by abiotic stresses. Dissecting candidate genes involved in abiotic stress tolerance is a paramount step toward the improvement of common bean performance under such constraints. Thereby, this thesis presents a systematic analysis of the DEHYDRATION RESPONSIVE ELEMENT-BINDING (DREB) gene subfamily, which encompasses genes that regulate several processes during stress responses, but with limited information for common bean. First, a series of in silico analyses with sequences retrieved from the P. vulgaris genome on Phytozome supported the categorization of 54 putative PvDREB genes distributed within six phylogenetic subgroups (A-1 to A-6), along the 11 chromosomes. Second, we cloned four novel PvDREB genes and determined their inducibility-factors, including the dehydration-, salinity- and cold-inducible genes PvDREB1F and PvDREB5A, and the dehydration- and cold-inducible genes PvDREB2A and PvDREB6B. Afterwards, nucleotide polymorphisms were searched through Sanger sequencing along those genes, revealing a high number of single nucleotide polymorphisms within PvDREB6B by the comparison of Mesoamerican and Andean genotypes. The nomenclature of PvDREB6B is discussed in details. Furthermore, we used the BARCBean6K_3 SNP platform to identify and genotype the closest SNP to each one of the 54 PvDREB genes. We selected PvDREB6B for a broader study encompassing a collection of wild common bean accessions of Mesoamerican origin. The population structure of the wild beans was accessed using sequence polymorphisms of PvDREB6B. The genetic clusters were partially associated with variation in latitude, altitude, precipitation and temperature throughout the areas such beans are distributed. With an emphasis on drought stress, an adapted tube-screening method in greenhouse conditions enabled the phenotyping of several drought-related traits in the wild collection. Interestingly, our data revealed a correlation between root depth, plant height and biomass and the environmental data of the location of the accessions. Correlation was also observed between the population structure determined through PvDREB6B and the environmental data. An association study combining data from the SNP array and DREB polymorphisms enabled the detection of SNP associated with drought-related traits through a compressed mixed linear model (CMLM) analysis. This thesis highlighted important features of DREB genes in common bean, revealing candidates for further strategies aimed at improvement of abiotic stress tolerance, with emphasis on drought tolerance