Schematic representation and sequence characterization of the different <i>relA</i> mutations identified in the Δ<i>spoT</i> clones studied here.

<p>A. The designation of each Δ<i>spoT</i> clone characterized in this work is indicated below the <i>E. coli relA</i> gene representation. The RelA region encoding the N-terminal 455 amino acid residues bearing ppGpp synthetase activity <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Cashel1" target="_blank">[9]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Gropp1" target="_blank">[45]</a> is shadowed. B. Amino acid sequence comparisons of the N-terminal regions (∼400 amino acid residues) of <i>S. equisimilis</i> Rsh, <i>E. coli</i> RelA, and SpoT. The alignments of the corresponding regions were constructed by using ClustalW (<a href="http://www.genome.jp/tools/clustalw" target="_blank">http://www.genome.jp/tools/clustalw</a>) using default parameters, and subsequently refined to maximize secondary structure similarities between <i>S. equisimilis</i> Rsh <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Hogg1" target="_blank">[46]</a> and RelA (as predicted by Jpred3; <a href="http://www.compbio.dundee.ac.uk" target="_blank">http://www.compbio.dundee.ac.uk</a>). The corresponding amino acid sequence positions are indicated at the right. Identical (*) or conserved (:) amino acids at a given position among the three sequences are indicated below the alignments, and deletions/insertions by hyphens (−). The sequence regions spanning the ppGpp hydrolase (⌜ ⌝) and synthetase (⌜ ⌝) domains of <i>S. equisimilis</i> Rsh, as well as the central-3 helix bundle region joining these domains defined by the ⌝ and boundaries <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Hogg1" target="_blank">[46]</a> are indicated above the alignments. The different amino acid substitutions and the Trp39 deletion (Δ) identified in <i>E. coli</i> RelA in Δ<i>spoT</i> clones 8–19 (this work) are indicated below the alignments. Conservation of equivalent residues in <i>S. equisimilis</i> Rsh and <i>E. coli</i> SpoT is indicated by highlighting the corresponding amino acids in bold. Conserved residues whose site-directed mutagenesis has been found by other authors to promote loss (G251E) or reductions (H354Y) in <i>E. coli</i> RelA ppGpp synthesis capability <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Gropp1" target="_blank">[45]</a> are indicated by closed arrowheads below the alignments. The HDXXED motif typical of metal-dependent pyrophosphohydrolases located in the hydrolase domain of bifunctional enzymes <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Hogg1" target="_blank">[46]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Sajish1" target="_blank">[55]</a> is underlined. In turn, the synthetase domain motif distinguishing monofunctional (EXDD) from bifunctional (RFKD) enzymes <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Hogg1" target="_blank">[46]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Sajish1" target="_blank">[55]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0106938#pone.0106938-Sajish2" target="_blank">[56]</a> is double underlined. The sources of the sequences were: <i>S. equisimilis</i> Rsh, ref. 46; <i>E. coli</i> K-12 MG1655 SpoT, GenBank accession NP_418107.1; <i>E. coli</i> RelA BW25113: this work. The latter is identical to <i>E. coli</i> K-12 MG1655 RelA, GenBank accession NP_417264.1.