218 research outputs found
The signal and the noise: characteristics of antisense RNA in complex microbial communities
No full textHigh-throughput sequencing has allowed unprecedented insight into the composition and function of complex microbial communities. With metatranscriptomics, it is possible to interrogate the transcriptomes of multiple organisms simultaneously to get an overview of the gene expression of the entire community. Studies have successfully used metatranscriptomics to identify and describe relationships between gene expression levels and community characteristics. However, metatranscriptomic data sets contain a rich suite of additional information that is just beginning to be explored. Here, we focus on antisense expression in metatranscriptomics, discuss the different computational strategies for handling it, and highlight the strengths but also potentially detrimental effects on downstream analysis and interpretation. We also analyzed the antisense transcriptomes of multiple genomes and metagenome-assembled genomes (MAGs) from five different data sets and found high variability in the levels of antisense transcription for individual species, which were consistent across samples. Importantly, we challenged the conceptual framework that antisense transcription is primarily the product of transcriptional noise and found mixed support, suggesting that the total observed antisense RNA in complex communities arises from the combined effect of unknown biological and technical factors. Antisense transcription can be highly informative, including technical details about data quality and novel insight into the biology of complex microbial communities
Étendre la diversité phénotypique et chimique d'Escherichia coli
No full textDans le passé, Nichols et al. (Cell, 2010) ont criblé une collection de mutants de déletion d’Escherichia coli (~4000 gènes) dans ~324 conditions. L’utilisation de la taille des colonies comme indicateur de croissance, leur a permis de définir des phénotypes robustes pour ~50% des mutants, conduisant ainsi à de nouvelles idées biologiques. Cependant, 50% des gènes restants n’ont pas donné de réponse significative.J’ai souhaité dépasser les limitations du précédent criblage en augmentant le crible dans 2 directions : tout d’abord, j’ai augmenté le nombre et la complexité des conditions de stress dans le but d’imiter les contraintes d'origine naturelle. Puis, j’ai adapté le test «rouge Congo» afin de tester la formation de biofilm.J’ai ainsi généré deux ensembles de données, contenant ~2 millions de phénotypes de croissance et de biofilm, dans approximativement 250 conditions de stress simples et complexes. Les données ont révélé des phénotypes pour ~80% des mutants testés, la majorité d'entre eux provenant de conditions de stress complexes et de la mesure de biofilm. L'utilisation de ces phénotypes a permis la construction d’un réseau connectant les complexes de protéines connus avec ~500 gènes de fonction inconnue.De plus, la mesure de biofilm a permis d’identifier de nouveaux acteurs impliqués dans la formation de biofilm. Les données obtenues grâce au test «rouge Congo» ont également révélé que de nombreux enzymes régulant les niveaux de ci-di-GMP servent de senseurs pour différents signaux entrants impliqués dans cette même voie. L’identité des signaux entrants a été révélée par les phénotypes spécifiques à une condition.In the past, Nichols et al. probed a knock out collection of E. coli across 324 conditions. Using colony size as a proxy for fitness allowed them to define robust phenotypes (5% FDR) for about 50% of the knockouts, leading to novel insights into E. coli’s biology. The remaining 50% of the genes did not yield a significant response even when growth was probed in such diverse conditions.I address the limitations of this previous high-throughput screen by expanding the screen design it in 2 directions: First, I increased the number and complexity of conditions to mimic naturally occurring stresses. Second, I adapted the Congo-red assay to probe biofilm formation in parallel to growth. I was able to generate two datasets that consist of more than 2 million growth and biofilm phenotypes across ~250 simple and complex stresses. The datasets revealed phenotypes for ~80% of the knockouts probed (5% FDR) with the large majority of them originating from complex stresses and the biofilm readout. Using these phenotypes, I generated a high-confident network connecting known protein complexes with >500 uncharacterized genes. In addition, the biofilm readout captured dozens of new players involved in biofilm formation. Even further, the color dataset revealed that the multiple enzymes regulating the levels of ci-di-GMP, serve as sensors for different input signals that feed into the same pathway. Identity of the input signals was revealed by the condition-specific phenotypes
Additional file 6: Table S1. of Rapid, precise quantification of bacterial cellular dimensions across a genomic-scale knockout library
No full textConditions in chemical genomics screen from [29] that exhibit negative correlation between mean cell width and S-score with p-value less than 0.000154 (Bonferroni multiple-hypothesis correction to p < 0.05 across 324 conditions; see Methods). Table S2. Conditions in chemical genomics screen from [29] that exhibit positive correlation between mean cell width and S-score with p-value less than 0.000154 (Bonferroni multiple-hypothesis correction to p < 0.05 across 324 conditions; see Methods). Table S3. Pairs of COGs and conditions in chemical genomics screen from [29] that exhibit correlations between mean cell width and S-scores with p-value less than 0.000154 (Bonferroni multiple-hypothesis correction to p < 0.05 across 324 conditions; see Methods). *: description from [29] and generously provided by Athanasios Typas. Table S4. Pairs of COGs and conditions in chemical genomics screen from [29] that exhibit correlations between mean cell length and S-scores with p-value less than 0.000154 (Bonferroni multiple-hypothesis correction to p < 0.05 across 324 conditions; see Methods). *: description from [29] and generously provided by Athanasios Typas. (DOCX 101 kb
Untersuchungen zur Promotorerkennung durch σS-haltige RNA Polymerase (EσS) in Escherichia coli
No full textTITLE AND TABLE OF CONTENTS
SUMMARIES
INTRODUCTION
GOAL SETTING
RESULTS AND DISCUSSION
LITERATURE
FIGURE APPENDIXDifferent environmental stimuli cause bacteria to exchange the sigma subunit
in the RNA polymerase (RNAP) and, thereby, tune their gene expression
according to the newly emerging needs. Sigma factors are usually thought to
recognise clearly distinguishable promoter DNA determinants. However, the case
seems to be different for the housekeeping sigma factor in Escherichia coli,
σ70 (RpoD) and σS (RpoS), the principle sigma factor in stationary phase and
in stressful conditions. These two sigma factors show a high degree of
sequence similarity and were found to bind favourably to almost identical -35
and -10 elements in-vitro. Nevertheless, the two sigma factors have clearly
different regulons in-vivo, being responsible together for the regulation of
more than 90% of the entire genome. The answer to the σS promoter selectivity
paradox has puzzled researchers for a long period and only recently has
started to being elucidated. A combination of cis-encoded and trans-acting
features allows promoter discrimination to be generated between Eσ70 and EσS.
First, by using several natural and synthetic promoters, I was able to show an
implication of the UP-element in Eσ70/EσS competition for promoters carrying
canonical 10 and 35 elements. Only EσS was shown to use efficiently the
distal UP-element sub-site, whereas full UP-elements or proximal UP-element
half-sites favoured Eσ70-mediated transcription, provided that a -35 box was
co-present. By genetic analysis this differential behaviour of the two RNAPs
was attributed to an inability of σS to interact with an adjacently placed
αCTD. In contrast, such a direct interaction between αCTD and σ70 is known to
stimulate transcription.
Furthermore, a new alignment of 80 experimentally determined σS-dependent
promoters indicated that many of them do possess a -35 element, in
contradiction to the current belief, but the majority has it misplaced by
1-2bp. Thus, EσS does not show the rigid preference of Eσ70 for 17bp spacers
between the -10 and -35 hexamers. These misplaced -35 elements were found to
be functional in many cases. In addition, the high conservation of longer or
shorter spacers than 17 bp in σS-dependent genes was due to the σS promoter
selectivity that these deviations conferred. Finally, the distinctive
structural flexibility of EσS in recognising misplaced 35 elements was
shown to be due to a different interaction of its region 4 with the β flap
subunit of RNAP and/or due to the unique way EσS binds to the extended 10
promoter region.
Apart from the intrinsic cis-acting features that generate promoter
selectivity, it is quite frequent that promoters gain their selectivity only
with the aid of additional transcriptional regulators. In the case of proP
(P2), Fis acts as a class II transcriptional regulator and co-activates the
promoter with E σS. It could be demonstrated that the selective EσS-Fis
synergy on this promoter is due to the fact that Fis binds at position -41 and
sterically forces RNAP to operate with a 16bp-spaced promoter; this
requirement can be successfully met by EσS, but not by Eσ70. In addition, the
molecular outer face of the domain 4 of σS is better suited for interacting
with Fis.
But even before σS competes with σ70 for promoter recognition, one of its main
problems, and that of other alternative sigmas, is to out-compete the
vegetative σ70 from core RNAP. Crl was shown here to directly help σS compete
with σ70 for limiting amounts of core RNAP, both in-vivo and in-vitro, by
supporting EσS formation. In addition, a microarray analysis could confine the
role of Crl in stationary phase to that of an auxiliary factor of σS,
especially when the latter was present in lower cellular amounts. Finally, the
pathways enabling Crl to exert a dual role in the regulation of σS levels and
activity were identified.
To conclude, this thesis aimed to characterise the function of different cis-
and trans-acting elements that support the major stress response sigma
factor, σS, to control its substantial regulon (up to 10% of the entire
genome). In order to achieve that, σS has to harshly compete with the
housekeeping sigma, both for limited amounts of core RNAP, but also for
recognition of almost identical core promoter sequences.Unterschiedliche Umweltstimuli veranlassen Bakterien dazu, die Sigma-
Untereinheit in der RNA-Polymerase (RNAP) auszutauschen und dadurch ihre
Genexpression auf die neu aufkommenden Bedürfnisse einzustellen. Normalerweise
gilt, dass unterschiedliche Sigmafaktoren klar unterscheidbare Promotor-DNA
Determinanten erkennen. Im Fall des vegetative Sigmafaktors in Escherichia
coli, σ70 (RpoD) und σS(RpoS), dem Hauptsigmafaktor in der Stationärphase und
unter Stressbedingungen verhält es sich jedoch anders. Diese beiden
Sigmafaktoren zeigen ein hohes Maß an Sequenzähnlichkeit und binden bevorzugt
an fast identische -35 und -10 Elemente in vitro. Dennoch haben die beiden
Sigmafaktoren in vivo klar unterscheidbare Regulons und sind dabei gemeinsam
für die Regulation von mehr als 90% des gesamten Genoms verantwortlich. Die
Lösung des σS\- Promotor-Selektivitäts Paradoxes beschäftigt Forscher
bereits seit langer Zeit und erst kürzlich ist es gelungen, ihr näher zu
kommen. Eine Kombination von in cis oder in trans wirkende Eigenschaften
ermöglicht die Unterscheidung von Promotoren durch Eσ70 und EσS.
Erstens konnte ich durch die Verwendung von mehreren natürlichen und
synthetischen Promotoren eine Beteiligung des UP-Elements am Wettbewerb von
Eσ70 und EσS um Promotoren mit kanonischen -10 und -35 Elementen etablieren.
Nur EσS kann die distale Halbeseite des UP-Elements effizient nutzen, während
vollständige UP-Elemente oder proximale UP-Element Halbseiten Eσ70-vermittelte
Transkription begünstigen, vorausgesetzt, dass gleichzeitig eine -35 Box
vorhanden ist. Genetische Analysen konnten dieses unterschiedliche Verhalten
der zwei Holoenzyme darauf zurückführen, dass σS unfähig ist, mit einer
benachbart platzierten αCTD zu interagieren. Im Gegensatz dazu ist bekannt,
dass eine polare Interaktion zwischen der αCTD und σ70 die Transkription
stimuliert.
Des Weiteren wies ein Alignment von 80 experimentell bestimmten σS-abhängigen
Promotoren darauf hin, dass viele dieser Promotoren, im Widerspruch zu
derzeitigen Annahmen, ein -35 Element besitzen. Die Mehrzahl der Promotoren
trägt dieses jedoch um 1-2bp versetzt . Daher zeigt EσS nicht die rigide
Präferenz, die Eσ70 für 17bp-Spacer besitzt. Diese versetzten -35 Elemente
sind in vielen Fällen funktionsfähig. Außerdem war die starke Konservierung
von Spacern, die länger oder kürzer als 17 bp sind, in σS-abhängigen Genen auf
die σS-Promotor-Selektivität zurückzuführen, die diese Abweichungen bewirkten.
Schließlich konnte gezeigt werden, dass die spezifische strukturelle
Flexibilität von EσS bei der Erkennung von versetzten -35 Elementen auf
einer unterschiedlichen Interaktion der Region 4 von σS mit der β-Flap-
Untereinheit der RNAP und/oder auf der unterschiedlichen Art und Weise, wie
EσS an die erweiterte -10 Promotorregion bindet, beruht.
Abgesehen von den intrinsischen in cis wirkenden Faktoren, die Promotor-
Selektivität bewirken, erhalten Promotoren relativ häufig ihre Selektivität
nur durch die Funktion von zusätzlichen transkriptionalen Regulatoren. Im Fall
von proP (P2) agiert Fis als ein Klasse II transkriptionaler Regulator und co-
aktiviert den Promotor zusammen mit EσS. Wir konnten zeigen, dass die
selektive EσS-Fis-Synergie an diesem Promotor darauf zurückzuführen ist, dass
Fis an Position -41 bindet und RNAP sterisch dazu zwingt, mit einem 16bp-
Spacer zu interagieren. Diese Vorraussetzung kann erfolgreich von EσS, jedoch
nicht von Eσ70 erfüllt werden. Zudem ist die nach außen gerichtete Oberfläche
von σS von vornherein besser für eine Interaktion mit Fis geeignet.
Doch noch bevor σS mit σ70 um die Promotor-Erkennung konkurriert ist eins
seiner Hauptprobleme erfolgreich mit dem vegetativen σ70 um die Kern-RNAP zu
konkurrieren. Hier wurde gezeigt, dass Crl sowohl in vivo als auch in vitro σS
direkt hilft, mit σ70 um die limitierende Menge an Kern-RNAP zu konkurrieren,
indem es die Bildung von EσS unterstützt. Außerdem konnten Microarray-Analysen
die Rolle von Crl in der Stationärphase auf die eines Hilfsfaktors für σS
einschränken, besonders wenn letzterer in niedrigeren zellulären
Konzentrationen vorhanden war. Schließlich wurden die Wege identifiziert, die
es Crl ermöglichen, eine doppelte Rolle in der Regulation von σS-Spiegeln und
-Aktivität auszuüben.
Zusammenfassend kann gesagt werden, dass es das Ziel meiner Dissertation war,
die Funktion verschiedener in cis und in trans wirkender Elemente zu
charakterisieren, die den Hauptsigmafaktor der Stressantwort σS dabei
unterstützen, sein umfangreiches Regulon (bis zu 10% des gesamten Genoms) zu
aktivieren. Dafür muss σS hart mit dem vegetative Sigmafaktor konkurrieren,
sowohl um limitierte Mengen an Kern-RNAP als auch um die Erkennung von fast
identischen Kern-Promotor-Sequenzen
Escherichia coli σ70 senses sequence and conformation of the promoter spacer region
No full textIn bacteria, promoter identification by RNA polymerase
is mediated by a dissociable p factor. The
housekeeping σ70 factor of Escherichia coli recognizes
two well characterized DNA sequence
elements, known as the ‘-10’ and ‘-35’ hexamers.
These elements are separated by ‘spacer’ DNA, the
sequence of which is generally considered unimportant.
Here, we use a combination of bioinformatics,
genetics and biochemistry to show that σ70
can sense the sequence and conformation of the
promoter spacer region. Our data illustrate how
alterations in spacer region sequence can increase
promoter activity. This stimulatory effect requires
σ70 side chain R451, which is located in close proximity
to the non-template strand at promoter
position -18. Conversely, R451 is not required to
mediate transcriptional stimulation by improvement
of the -10 element. Mutation of σ70 residue
R451, which is highly conserved, results in reduced
growth rate, consistent with a central role in
promoter recognition
Bacterial protein networks: properties and functions
No full textDistinct cellular functions are executed by separate groups of proteins, organized into complexes or functional modules, which are ultimately interconnected in cell-wide protein networks. Understanding the structures and operational modes of these networks is one of the next great challenges in biology, and microorganisms are at the forefront of research in this field. In this Review, we present our current understanding of bacterial protein networks, their general properties and the tools that are used for systematically mapping and characterizing them. We then discuss two well-studied examples, the chemotaxis network and the cell cycle network in Escherichia coli, to illustrate how network architecture promotes function
- …
