A Whiff of ...Replication – Pendulisporaceae

by Christoph
 

Just to quote myself here: When microbiologists describe a previously unknown bacterium, they are mainly interested in its lifestyle, metabolism, morphology and cell cycle. And to con­tinue with the self-quotation: For me, "getting to know" newbies always involves finding out how they ini­tiate the replication of their chromosome(s). *)
 

The four Pendulisporaceae described by Garcia et al. (2024) and introduced to STC in part 1 have unusually large genomes. Their single, circular chromosomes have lengths ranging from 12.2 to 13.6 mega-bases (Mb). One of their cousins in the taxonomic sub‑order Sorangiineae, Sorangi­um cellulosum So ce56, had for years been the bacterium with the largest known genome, 13.0 Mb, and it was found only recently that not only this branch of the myxo­bac­teria (phylum Myxococcota), but also the phylum Chloroflexota contains spe­cies with very large genomes, for example Ktedonobacter racemi­fer SO­SP1-21 (13.7 Mb). To put these genome sizes/chro­mosome lengths into perspective: the genome of the eukaryote Saccharomyces ce­revi­siae S288C has a size of 12.1 Mb, distributed over 16 chromosomes.
 

"Replicationists" assume this, but simply do not know whether bacteria with unusually large genomes ini­ti­ate the replication of their chro­mo­some like bac­teria with "normal-sized" chromosomes – think E. coli MG1655 (4.6 Mb) or B. sub­til­is 168 (4.2 Mb) – dependent on the initiator protein DnaA from an origin of replication, oriC, or whether they use unknown mechanism(s). When Schneiker et al. (2007) pub­lished the genome se­quence of Sorangium cellulosum So ce56 (a paper with 59 authors!), they had found a dnaA gene but oriC remained elusive. They state: "In the absence of the GC-skew inversion typically seen at the replication ori­gin of bacterial chromosomes, it was not possible to discern the location of oriC."
 

(click to enlarge)
Figure 2.1. Alignment of 354 bp of the oriC sequences of four Pen­du­lisporaceae and 10 related species/strains as in Figure 2.2; the se­quen­ces are aligned along the twin DnaA box R1. Color code: grey, coding region; orange/pale orange, strong/weak DUE; magenta, DnaA-trio mo­tif; red/pale red, consensus/relaxed DnaA box (5'‑TGTGNAWAA); blue/pale blue, consensus/relaxed DnaA box (5'-TTWTNCACA). Exceptionally, the genome of S. amylolyticus DSM 53668 contains a sencond dnaN gene (dnaN*). Struc­tural elements are drawn to scale; a version of this Figure that includes the base sequences is available from the author upon request. Fron­tispiece: detail of Figure 2.1

In the complete genomic sequences, single dnaA genes are annotated for P. albi­nea MSr11954, P. brun­nea MSr125­23, P. rubella MSr11368 and P. rubella MSr113­67. The pro­tein se­quences share ~85% iden­tity, indicating their close relationship on the (GTDB taxonomic) genus level. From the same (taxo­nomic) family, the DnaA protein of Sorangium cellulosum So ce56 shares ~63% iden­tity with them. In contrast, the DnaA proteins of E. coli K‑12 MG16­55 and B. subtilis 168 share only ~40% and ~50% iden­tity with DnaA of P. albi­nea MSr­11­954, respectively, which is low but in the typical range for DnaA proteins of species belonging to different phyla.
 

Given the presence of a dnaA gene, it should therefore be possible to find the ori­gin of replication, oriC, and to predict its location in the four Pendulisporaceae genomes. (I should add that molecular biolo­gists do not regard predictions as proof, but insist on experimental verification.) I started my search with the "in­formed guess" that oriC may be located in an inter­genic region adjacent to the dnaN gene as in many bacterial genomes (mostly in the dnaA·dnaN intergenic region). The dnaN-ad­jacent intergenic regions have lengths between 448 bp and 510 bp and are thus large enough to accommodate oriC. Since I had already made the experience in the past that it makes sense to include the genomes of closely related species in such an ana­ly­sis, I searched the dnaN-adjacent intergenic regions of another 10 genomes, including Sorangium cellulosum So ce56, for putative oriC sequences (these intergenic regions varied considerably in length, between 280 bp and 1211 bp).
 

I detected in all 14 sequences the conserved structural elements of bacterial oriC·s: counter-clockwise transcription of the left-flanking gene (<dnaN), a DNA-unwinding element (DUE), DnaA-trio motifs, a single or twin DnaA box in reverse orientation (R1) at a distance of ~2–3 helical turns from the DUE, and additional multiple DnaA binding sites arranged as arrays (Figure 2.1). The twin DnaA boxes R1 in the oriC·s of the Myxo­coccales are likewiese found in the oriC·s of Actinomycetota, Bacillota, and Cyano­bacteriota (yes, I'm still struggling with the new names but that is slowly wearing off). I did not detect any structural elements in a first superficial screening that would make these origins somehow "special". Also, I could not detect possible "secondary origins" with typical oriC-features in these genomes.
 

Thus, as far as it is reasonably possible to draw conclusions from predictions, it can be argued that bacteria with circular singular chromosomes of >12 Mb in length initiate replication like bacteria with "normal-sized" chromosomes. Or, to look at the other end of the length spectrum: ...like bacteria with mini genomes such as the Stam­mera capleta symbiont of Cassida rubiginosa beetles, whose chromosome is only 0.27 Mb long. It is fine to know this, and that the mechanism of DnaA-dependent initiation of chromosome replication from oriC has apparently no restrictions on chromosome length, but it is certainly not important enough to warrant a proper scientific publication.
 

(click to enlarge)
Figure 2.2. (Left) Phylogenetic tree classifying the Pendulisporaceae (bold type) new myxobacterial family constructed using MrBayes. Val­ues at node represent the posterior probability. Source. (Right) Synteny of genes in the oriC region. Red dot, position of oriC. Genes syntenic with oriC are indicated by gene names in bold. Single genes or groups of contiguous genes are separated "|". Gene order in Hali­angium ochra­ceum DSM 14365 [NC_013440.1] is shown instead of that in H. tepidum SMP-10 for which the assembled genome was not available. The genes for Aetherobacter fasciculatus [NZ_CAWUQY000000000.1] were found in the contigs of the un-assembled genome.Exceptionally, the genome of S. amylolyticus DSM 53668 contains a sencond dnaN gene (dnaN*) By the author

There's something else that's also worth a side note. As you see in Figure 2.2 (please use the "click to enlarge" option to read it), all members of the order Poly­an­giales (NCBI taxonomy) with one exception have oriC (red dot) adjacent to a syn­tenic gene triplet dnaN·recF·gyrB. I has repeatedly been argued that the strong conservation of genes in the oriC region reflects the fine‑tuning of their transcriptional regulation, as these genes have twice the copy number in repli­cat­ing cells. In fact, the arrangement of genes (synteny) in the oriC region in genomes from more than half of the known bacterial phyla is almost or comple­tely identical to that shown here for Geomonas oryzisoli and, in most cases, oriC is localized in the dnaA·dnaN intergenic region (Figure 2.2). However, just as many cases are known in which this conspicuous synteny is partially or completely mixed up by intra‑genomic recombination ("rearrangements"). A good example of this is the disruption of the syntenic oriC gene cluster in the Polyangiales, where only the oriC-adjacent gene triplet dnaN·recF·gyrB is preserved, while the mnmE·­yidCD·­rnpA·rpmH gene cluster is at a completely different genomic location, and dnaA, rsmG and gyrA are isolated genes in completely different contexts. Since the species of Polyangiales are certainly several hundred million years old, it is hard for me to imagine that they would not have used this long time to adapt the transcription of the genes close to the "new" oriC position in such a way that the double copy number during replication could be an impediment.
 

A final thought. Why could Schneiker et al. (2007) not pinpoint oriC when they sequenced Sorangium cel­lu­lo­sum So ce56 while I had no issues? For this genome, I obtained a clean inflection point in the cumulated GC‑skew curve at pos. 11,364,776, at a distance of approximately 9,2 kb from oriC at pos. 11,355,356 (c11355106..c11355605 [NC_010162.1]), which is reasonable since inflection points are not always directly at oriC but often at a distance of several kb. I suspect that they were "trapped" in the then prevailing ideas of the location of oriC in bacterial genomes: Ogasawara & Yoshikawa(1992) had convincingly argued that the (evolutionary) original localization of oriC is upstream of the dnaA gene as in B. subtilis (we now know that unwinding occurs in the dnaA·dnaN intergenic region), and Worning et al. (2006) had shown that, as a rule, oriC adjacent to dnaA is located near the inflection point of the GC skew. Since the dnaA-upstream region contains only 1 consensus and 5 degenerate DnaA boxes they probably thought it unlikely that this is the oriC, and the lack of detectable GC‑skew inflection point near dnaA convinced them in this. Additional criteria for the detection of oriC as I applied here were not available to them back in 2007.
 

Do you want to comment on this post? We would be happy about it! Please comment on Mastodon, Bluesky, or on 𝕏 (formerly Twitter).
 

*) see here in STC for the Lilliputians (formerly CPR bac­ter­ia, now Patescibacteria), here for Planctomycetota, here for Borrelia, and recently for the Atri­bacteria.