A Whiff of Taxonomy – Saltatorellus ferox, again...

Part 2
 

by Christoph
 

An excursion into S. ferox chromosome replication

Wiegand et al. sequenced the Saltatorellus ferox Poly30^T genome and found it to be 7.7 Mb in size, with a comparatively high GC content of 67% (Campylobacter species ~30%, E.coli MG1655 50.8%, Strep­tomyces species ~74%). It is among the larger genomes of the Planctomycetes (~1.5–13 Mb), and organized as a single circular chromosome, which does not necessarily imply monoploidy of the small S. ferox cells. Using a webtool, I found a perfect cumulative GC-skew in the S. ferox ge­nome (see here, red curve) and the GC-skew minimum, a good indicator for the replication origin region, at position 15,437. This is reasonably close to the predicted replication origin for DnaA-de­pendent bi-directional replication, oriC­, in the 464 bp‑long intergenic region between the dnaA and dnaN genes (positions 2,008–2,432).
 

(Click to enlarge)
Figure 1. Schematic represen­tation of predicted oriC struc­tures for Salta­to­rel­lus ferox and four Planc­to­mycetes. Red half-circles: predicted DnaA-binding sites; orange el­lipses: pred­icted DNA-unwinding elements (DU­E); blue ellipses: DnaA ssDNA-binding motif (Dn­a­A-trio); gray boxes: flanking genes. The oriCs of B. subtilis and E. coli are shown for compa­ri­son. C. Weigel, unpub­lished. Front­page: S. ferox Poly30T cell budding. Source

In a prior post (Daniel's Planctomycetes saga "Tales of Mystery and Imagination"), I had predicted the oriCs of several Planctomycetes, which are also located in the dnaA-dnaN intergenic region (Figure 1). This oriC location is quite common in bacterial genomes from various phy­la but not a strict rule. Yet unlike the planctos, which have two dnaA genes, S. ferox has only one, like most bacteria. The predicted oriC of Saltatorellus ferox contains a num­ber of DnaA-binding sites, and a DNA‑unwinding element (DUE), that is, a short stretch of ~50 bp that becomes ea­sily single‑stranded when negatively supercoiled. In ad­di­tion, it contains a 'DnaA-trio' motif flanking the DUE, which has recently been shown to bind DnaA to single-stranded DNA during the initiation step of chromosome replication in B. subtilis.
 

Homologs of the chromosome segregation proteins ParAB are encoded in the S. ferox genome but not in the usual gene pair arrangement found in most bacterial genomes. Also, I could not de­tect canonical parS sites (5'‑TGTTNCACGTGAAACA), the "anchor points" for ParB, in a quick search (200 kb on both sides of oriC). Thus, while replication initiation seems to follow the known route, the exact mode of chromosome segregation in S. ferox remains to be studied. In E. coli, the XerCD recombinase resolves chromosome dimers formed during replication, and is directed to dif sites in the terminus region, where resolution takes place, by the FtsK translocase. FtsK tracks along the leading strand of chromomal DNA in oriC→ter direction, and finally "sorts" the resolved chromo­so­mes into mother and daughter cells while being anchored to the membrane at the 'di­vi­sion plane' of the septum. The S. ferox genome encodes two putative tyrosine-type recombi­na­se/inte­gra­se proteins that have low but reasonable similarity to XerC and XerD, respectively, and a full-length FtsK protein (full-lenght: including the N-terminal membrane anchor). FtsK movement along the chromosome is guided by KOPS motifs (5'‑GGGNAGGG), which usually show a skewed distri­bution, that is, more KOPS on the leading strand in a gradient from oriC to ter. Given the per­fect GC-skew of the S. ferox genome, it wasn't surprising to find a clear "KOPS skew" for 1.5 Mb to the right and to the left of oriC: on both 'chromsome arms', KOPS motifs are almost twice as frequent on the leading strand as on the lagging strand. So, I would tentatively conclude that neither chro­mo­some replication, nor chromosome segregation or division give any hints on why and how cell division occurs in such unusual ways in Saltatorellus.