Evolution of the bacterial nucleosidase PpnN and its relation to the stringent response.

In our recent publication (Zhang et al., 2019), we demonstrate an interesting mode of regulation of purine metabolism unique to Proteobacteria. In this microreview, we would like to reflect on the ideas put forward, with special focus on protein domain architecture of the enzyme involved, its orthologues in plants, and the implications of the differential effects observed between binding of the two alarmone molecules, ppGpp (guanosine 3',5'-bisdiphosphate) and pppGpp (guanosine-5'-triphosphate-3'-diphosphate). In our previous work, we showed that the Escherichia coli nucleotide 5'-monophosphate nucleosidase, PpnN, which is conserved in Proteobacteria, cleaves its preferred substrate, guanosine monophosphate (GMP), at a much higher rate in the presence of both pppGpp and ppGpp (Figure 1A). Structural analysis reveals that binding of pppGpp leads to a conformational change in the protein that exposes its active site, suggesting this is the reason for the observed increase in activity. Finally, point mutation of the alarmone-interacting residues show a defect in binding, resulting in (i) increased basal catalytic activity of PpnN and higher competitive fitness of E. coli in an environment with fluctuating nutrient levels, and (ii) increased bacterial sensitivity towards antibiotics. In contrast, complete loss of the ppnN gene has the inverse effect, i.e. reduced competitive growth and improved antibiotic tolerance. We used these observations to propose a model in which E. coli uses PpnN to balance the need of fitness (fast growth) against tolerance towards antibiotics to improve survival.

The E. coli HicB Antitoxin Contains a Structurally Stable Helix-Turn-Helix DNA Binding Domain.

The E. coli hicAB type II toxin-antitoxin locus is unusual by being controlled by two promoters and by having the toxin encoded upstream of the antitoxin. HicA toxins contain a double-stranded RNA-binding fold and cleaves both mRNA and tmRNA in vivo, while HicB antitoxins contain a partial RNase H fold and either a helix-turn-helix (HTH) or ribbon-helix-helix domain. It is not known how an HTH DNA-binding domain affects higher-order structure for the HicAB modules. Here, we present crystal structures of the isolated E. coli HicB antitoxin and full-length HicAB complex showing that HicB forms a stable DNA-binding module and interacts in a canonical way with HicA despite the presence of an HTH-type DNA-binding domain. No major structural rearrangements take place upon binding of the toxin. Both structures expose well-ordered DNA-binding motifs allowing a model for DNA binding by the antitoxin to be generated.

(p)ppGpp Regulates a Bacterial Nucleosidase by an Allosteric Two-Domain Switch.

The stringent response alarmones pppGpp and ppGpp are essential for rapid adaption of bacterial physiology to changes in the environment. In Escherichia coli, the nucleosidase PpnN (YgdH) regulates purine homeostasis by cleaving nucleoside monophosphates and specifically binds (p)ppGpp. Here, we show that (p)ppGpp stimulates the catalytic activity of PpnN both in vitro and in vivo causing accumulation of several types of nucleobases during stress. The structure of PpnN reveals a tetramer with allosteric (p)ppGpp binding sites located between subunits. pppGpp binding triggers a large conformational change that shifts the two terminal domains to expose the active site, providing a structural rationale for the stimulatory effect. We find that PpnN increases fitness and adjusts cellular tolerance to antibiotics and propose a model in which nucleotide levels can rapidly be adjusted during stress by simultaneous inhibition of biosynthesis and stimulation of degradation, thus achieving a balanced physiological response to constantly changing environments.

Toxins, Targets, and Triggers: An Overview of Toxin-Antitoxin Biology.

Bacterial toxin-antitoxin (TA) modules are abundant genetic elements that encode a toxin protein capable of inhibiting cell growth and an antitoxin that counteracts the toxin. The majority of toxins are enzymes that interfere with translation or DNA replication, but a wide variety of molecular activities and cellular targets have been described. Antitoxins are proteins or RNAs that often control their cognate toxins through direct interactions and, in conjunction with other signaling elements, through transcriptional and translational regulation of TA module expression. Three major biological functions of TA modules have been discovered, post-segregational killing ("plasmid addiction"), abortive infection (bacteriophage immunity through altruistic suicide), and persister formation (antibiotic tolerance through dormancy). In this review, we summarize the current state of the field and highlight how multiple levels of regulation shape the conditions of toxin activation to achieve the different biological functions of TA modules.

Identification of the non-ribosomal peptide synthetase responsible for biosynthesis of the potential anti-cancer drug sansalvamide in Fusarium solani.

Sansalvamide is a cyclic pentadepsipeptide produced by Fusarium solani and has shown promising results as potential anti-cancer drug. The biosynthetic pathway has until now remained unidentified, but here we used an Agrobacterium tumefaciens-mediated transformation (ATMT) approach to generate knockout mutants of two candidate non-ribosomal peptide synthetases (NRPS29 and NRPS30). Comparative studies of secondary metabolites in the two deletion mutants and wild type confirmed the absence of sansalvamide in the NRPS30 deletion mutant, implicating this synthetase in the biosynthetic pathway for sansalvamide. Sansalvamide is structurally related to the cyclic hexadepsipeptide destruxin, which both contain an α-hydroxyisocaproic acid (HICA) unit. A gene cluster responsible for destruxin production has previously been identified in Metarhizium robertsii together with a hypothetical biosynthetic pathway. Using comparative bioinformatic analyses of the catalytic domains in the destruxin and sansalvamide NRPSs, we were able to propose a model for sansalvamide biosynthesis. Orthologues of the gene clusters were also identified in species from several other genera including Acremonium chrysogenum and Trichoderma virens, which suggests that the ability to produce compounds related to destruxin and sansalvamide is widespread.

Computational discovery of specificity-conferring sites in non-ribosomal peptide synthetases.

MOTIVATION: By using a class of large modular enzymes known as Non-Ribosomal Peptide Synthetases (NRPS), bacteria and fungi are capable of synthesizing a large variety of secondary metabolites, many of which are bioactive and have potential, pharmaceutical applications as e.g. antibiotics. There is thus an interest in predicting the compound synthesized by an NRPS from its primary structure (amino acid sequence) alone, as this would enable an in silico search of whole genomes for NRPS enzymes capable of synthesizing potentially useful compounds. RESULTS: NRPS synthesis happens in a conveyor belt-like fashion where each individual NRPS module is responsible for incorporating a specific substrate (typically an amino acid) into the final product. Here, we present a new method for predicting substrate specificities of individual NRPS modules based on occurrences of motifs in their primary structures. We compare our classifier with existing methods and discuss possible biological explanations of how the motifs might relate to substrate specificity. AVAILABILITY AND IMPLEMENTATION: SEQL-NRPS is available as a web service implemented in Python with Flask at http://services.birc.au.dk/seql-nrps and source code available at https://bitbucket.org/dansondergaard/seql-nrps/. CONTACT: micknudsen@gmail.com or cstorm@birc.au.dk SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Structural analysis of the yeast exosome Rrp6p-Rrp47p complex by small-angle X-ray scattering.

The RNase D-type 3'-5' exonuclease Rrp6p from Saccharomyces cerevisiae is a nuclear-specific cofactor of the RNA exosome and associates in vivo with Rrp47p (Lrp1p). Here, we show using biochemistry and small-angle X-ray scattering (SAXS) that Rrp6p and Rrp47p associate into a stable, heterodimeric complex with an elongated shape consistent with binding of Rrp47p to the nuclease domain and opposite of the HRDC domain of Rrp6p. Rrp47p reduces the exonucleolytic activity of Rrp6p on both single-stranded and structured RNA substrates without significantly altering the affinity towards RNA or the ability of Rrp6p to degrade RNA secondary structure.

Mimer: an automated spreadsheet-based crystallization screening system.

In this paper, a simple low-cost alternative to large commercial systems for preparing macromolecular crystallization conditions is described. Using an intuitive spreadsheet-based approach, the system allows the rapid calculation of relevant pipetting volumes given known stock-solution concentrations and incorporates the automatic design of custom crystallization screens via the incomplete-factorial and grid-screen approaches. Automated dispensing of the resulting crystallization screens is achieved using a generic and relatively inexpensive liquid handler.

Human 2'-phosphodiesterase localizes to the mitochondrial matrix with a putative function in mitochondrial RNA turnover.

The vertebrate 2-5A system is part of the innate immune system and central to cellular antiviral defense. Upon activation by viral double-stranded RNA, 5'-triphosphorylated, 2'-5'-linked oligoadenylate polyribonucleotides (2-5As) are synthesized by one of several 2'-5'-oligoadenylate synthetases. These unusual oligonucleotides activate RNase L, an unspecific endoribonuclease that mediates viral and cellular RNA breakdown. Subsequently, the 2-5As are removed by a 2'-phosphodiesterase (2'-PDE), an enzyme that apart from breaking 2'-5' bonds also degrades regular, 3'-5'-linked oligoadenylates. Interestingly, 2'-PDE shares both functionally and structurally characteristics with the CCR4-type exonuclease-endonuclease-phosphatase family of deadenylases. Here we show that 2'-PDE locates to the mitochondrial matrix of human cells, and comprise an active 3'-5' exoribonuclease exhibiting a preference for oligo-adenosine RNA like canonical cytoplasmic deadenylases. Furthermore, we document a marked negative association between 2'-PDE and mitochondrial mRNA levels following siRNA-directed knockdown and plasmid-mediated overexpression, respectively. The results indicate that 2'-PDE, apart from playing a role in the cellular immune system, may also function in mitochondrial RNA turnover.

The activity and selectivity of fission yeast Pop2p are affected by a high affinity for Zn2+ and Mn2+ in the active site.

In eukaryotic organisms, initiation of mRNA turnover is controlled by progressive shortening of the poly-A tail, a process involving the mega-Dalton Ccr4-Not complex and its two associated 3'-5' exonucleases, Ccr4p and Pop2p (Caf1p). RNA degradation by the 3'-5' DEDDh exonuclease, Pop2p, is governed by the classical two metal ion mechanism traditionally assumed to be dependent on Mg(2+) ions bound in the active site. Here, we show biochemically and structurally that fission yeast (Schizosaccharomyces pombe) Pop2p prefers Mn(2+) and Zn(2+) over Mg(2+) at the concentrations of the ions found inside cells and that the identity of the ions in the active site affects the activity of the enzyme. Ion replacement experiments further suggest that mRNA deadenylation could be subtly regulated by local Zn(2+) levels in the cell. Finally, we use site-directed mutagenesis to propose a mechanistic model for the basis of the preference for poly-A sequences exhibited by the Pop2p-type deadenylases as well as their distributive enzymatic behavior.