A Whiff of Taxonomy – Pendulisporaceae

by Christoph
 

(click to enlarge)
Figure 1.1. Legend see Figure 1.2. Frontispiece: Micro­sco­p­ic image of a Pendulispora rubella colony. © HIPS/Gar­cia

Microbiologists have always been fascinated by myxobacteria, that is for over a hundred years, because of their lifestyle, coordinated swarm hunting, and ability to form myxospore con­tain­ing fruiting bodies, an easily visible but unique differentiation feat among bacteria. And for just as long, microbio­logists have been tearing their hair out over the difficulties of cultivating these hunter‑gath­er­ers in the lab. How­ever, their eyes lit up when they realized over the last two decades that many of the 101 known myxobacterial species from seven taxonomic fami­lies of this phylum (Myxococcota, formerly lumped into the messy drawer Deltaproteobacteria) are skilful producers of pharmacologically interesting molecules, for example antibiotics and cy­tostatics, to an extent that comes close to that of Actinomy­cet­ota like Stretomyces sp. or Nocardia sp..
 

(click to enlarge)
Figure 1.2. Colony morphology of Pendulispora rubella MSr11367T (A), P. rubella MSr11368 (B), P. brunnea MSr­12523T (C), and P. albinea MSr11954T (D) shows their powdery spores under a stereomicroscope (top to bottom). Magnification of P. albinea MSr11954T: stereo photomicrograph highlighting aerial white spores and the approximate height from the surface (top) and aerial spores viewed under scanning electron micro­scopy showing the cell arrangement, intercellular con­nections, and a layer of wrinkled coat enclosing the cells (bottom). Source

The interest in such pharmacologically interesting natural products (NPs) that they synthesize was the major incentive for Garcia et al. (2024) to hunt for unknown myxobacteria: "The agar isolation setup was baited with both filter paper and E. coli DSM 1116^T on the same Petri dish plate. The myxobacteria were recognized by their swarming on agar and by lysis of the bacteri­al bait. We performed isolation and purification by cutting the swarm colony edge and trans­fer­ring it to standard buffered baker’s yeast agar (bufVY/2)." (see footnote for E. coli DSM 1116^T.)
 

From their successful hunt, they selected four isolates which, according to 16S rRNA analysis, are close "cousins" of Labilithrix luteola, Polyangium fumosum, and Sorangium cellulosum in the Sorangiineae taxonomic sub‑order of the Myxococcota. Within the Sorangiineae, the newly baptized species Penduli­spora rubella, P. albinea, and P. brunnea belong to the newly estab­lished family Pendulisporaceae, with isolates MSr11367 and MSr11368 representing virtually identical strains of P. rubella.
 

The growth stages of these four isolates are characterized by an advancing swarm colony pat­tern with distinctive flare-like edges (Figures 1.1+1.2), slender rod-shaped vegetative cells (L 4.0–6.0 μm, blunted ends), and dormant, rounded spores (see here). They are differentiated into three aerobe species that share tolerance to acidic as well as alkaline growth conditions (range tolerance pH 4–12), and resistance to all tested antibiotics. As growth was occurred at 18, 22, 30, 37 °C, they are mesophilic. The Pendulisporaceae appear to be well adapted to a wide range of environmental conditions, and tested positive for preying on E. coli DSM 1116T, Micrococcus luteus DSM 1790 (Actinomycetota), Yarrowia lipolytica Po1h (yeast) and S. cerevisiae (yeast).
 

(click to enlarge)
Figure 1.3. Positons of presumptive mobile genetic ele­ments ('genomic islands' identified with islandviewer 4 soft­ware). Mo­dified from Source

The Pendulisporaceae do not exhibit a multicellular fruiting-body stage; the name "Penduli…" reflects the hanging or free spores observed, in contrast to the fruiting-body-encapsulated spores found in other myxobacteria. Instead, they possess a unique powdery appearance in their colonies, com­parable to the morphology observed in many Streptomyces strains that form aerial hyphae. Among the four strains, P. albinea MSr11954 stands out as the most ge­neti­cally distant and exhibits the most distinct morphology. Its colony appears whitish and consists of interconnected cells (Figure 1.2D, inlet). The spores are reminiscent of the com­partmentalized Streptomyces spores on aerial hyphae – but do not show the "ladder" pheno­type seen so beautifully in Pictures Considered #52 – and differ from typical myxospores in lacking a sporangiole coat or slime envelope encapsulating the spores, which typically differ­entiate from inner cells in the fruiting body.

And what about specialized metabolites in the Pendulisporaceae, for whose synthesis the myxobacteria are known?
 

(click to enlarge)
Figure 1.4. Biological activities of P. rubella MSr11367, P. ru­bella MSr11368, P. albinea MSr11954, and P. brun­nea MSr12523 crude extracts against a panel of Gram-positive and Gram-negative bacteria, as well as fungi. Ac­tivity score from 0 (no inhibitory activity) to 8 (stron­gest inhibitory activity) determined by 1:1 serial dilu­tions of the crude extracts. Source

Garcia et al. (2024) found by computational genome analyses that the genomes of the Penduli­spora­ceae are spiked with 6–44 genomic islands that consist mostly of transposons and inser­tion sequences, which are recognizable by their deviant DNA sequence composition and the presence of genes usually associated with mobile DNA elements like IS elements (Figure 1.3; doesn't the distribution of genomic island in P. albinea MSr11954 look much like a barcode?). In a cursory look at the P. brunnea MSr12523 genome alone, I found ~200 annotated transpos­ase genes, including pseudogenes and truncated remnants. The latter is a clear indication of permanently ongoing genome reorganizations. The genomic islands often overlap with bio­synthetic gene clusters (BGCs), suggesting that they may facilitate the mobility of biosynthetic traits within myxobacteria and to more distant relatives.
 

Such an arrangement of a whole group of a dozen or more genes in biosynthetic gene clusters (BGCs) for the stepwise synthesis of a metabolite is common in bacterial genomes. Think of the his and trp operons in E. coli, but also the gum operon (12 genes) for the synthe­sis of the complex polysaccharide xanthan gum by Xanthomonas species. Gene clusters for the synthe­sis of molecules formerly known as secondary metabolites but now preferably referred to as 'specialized metabolites' or 'natural products' (NPs) are usually located together in such BGCs as operons with one or two main promoters, as this ensures their joint expression. The creativity of microbiologists is then required to find cultivation conditions that can trigger their expression.
 

(click to enlarge)
Figure 1.5. (Left) Extracts with molecular features show a positive correlation with the observed activities against S. aureus. The activity of the extracts from P. brun­nea MSr12523 is correlated with that of so­ranginin P3. (Right) Structural features of soranginin P3 differing from those of sorangicin A are highlighted in yellow, and additional and elusive stereoelements are marked with asterisks. The most relevant COSY cor­relations are bold, and the most relevant HMBC cor­relations (both elaborate nuclear magnetic reson­ance (NMR) techniques) are shown as arrows. Source

In an initial attempt to understand the functions of the various BGCs, Garcia et al. (2024) em­ployed three different growth conditions to obtain crude cell extracts of the four Pendulispora­ceae for tests of their biological activity, that is, their influence on the growth of Gram-positive and Gram-negative bacteria as well as selected fungi (Figure 1.4; not that C. neofor­mans, a fungus, is grouped together with the Gram-positives for unexplained reasons). In soil, their natural ha­bitat, the Pendulispo­raceae would of course also encounter nematodes and insects, but in­cluding some of these in the test series would probably have been too time-consuming and will have to wait for follow-up studies. The authors obtained bioactive extracts from the four Pendu­lisporaceae that were to varying extents inhibitory to the growth of many of the test organisms (note in Figure 1.4 that for E. coli only the ΔacrB strain lacking the AcrB multidrug efflux pump showed a marked response).
 

The computational analyses pointed to, literally, numerous BGCs in the four Pendulisporaceae genomes, many shared between the species, and 64 BGCs in P. brunnea MSr12523 alone. It is impossible to present the truly Herculean work that Ronald Garcia and his fourteen collabora­tors carried out to elucidate the chemical structures and synthesis pathways of many of the numerous NPs of the Pendulisporaceae in a com­prehensible manner here in a blog, (the length of this sentence reflects the 170 pages (!) of Supplementary Material of their paper.)
 

(click to enlarge)
Figure 1.6. Distinct features of the P. brunnea MSr12523 srb BGC (biosynthetic gene cluster). (A) Comparison be­tween the sor BGC from Sorangium cellulosum pro­ducing sorangicin A and the P. brunnea MSr12523 srb BGC producing sorangicin P3. Blue, shared genes; red, distinct genes; yellow, resistance gene; grey, ot­her genes. (B) Biosynthetic model for sorangicin P, based on retro-biosynthesis, phylogenetic analysis of KS domains and occurrence of further domains within the respec­tive modules. Domains introducing structural differen­ces in comparison to sorangicin A colored red. Putative trans-acting domains needed to explain the observed structures are indicated above each module without bubbles. Superscripted “0” marks inactive domains. (ACP: acyl carrier protein; DH: dehydratase; ER; enoyl­reductase; KR: ketore­duc­t­ase; KS: ketosynthase; MT: methyltransferase; PS: pyran synthase; Ppant arms drawn as wavy lines). Source

I limit myself to a superficial look at the structure of sorangicin P3 and its biosynthesis by P. brunnea MSr12523, an antibiotic known as sorangicin A from S. cellulosum. Sorangicin is, like rifampicin, an antibiotic that binds to the β‑subunit of RNA polymerase and is highly toxic to MRSA strains of S. au­re­us, for example. In Figure 1.5 you can see well how much the pro­duction of certain metabolites depends on the cultivation conditions. And Figure 1.6 shows that the elucidation of their structure requires the mastery of sophisticated NMR methods to reveal how expert microbes practice the art of combinatorial chemistry (for avid readers: François Jacob wrote his wonderful essay Evolution and Tinkering (PDF) a long time ago. I have to read it again and again when the diversity of life at the molecular and cellular level comes to overwhelm me).
 

The four Pendulisporaceae presented here have unusually large genomes, all around 13 me­ga­bases (Mb). That is 3 times larger than the genome of E. coli, and larger than the genome of the yeast S. cerevisiae, a eukaryote with a comparatively small genome. In a follow-up post, I will take a look at how the replication of these genomes might work.
 

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

Footnote   E. coli strain DSM 1116^T is identical with E. coli W (ATCC 9637) that was isolated from the soil of a cemetery near Rutgers University, Newark, NJ, USA around 1943 by Selman A. Waks­man, the discov­er­er of streptomycin. The strain was termed "Waksman's strain" or "W strain" because it showed the high­est sensitivity to streptomycin compared to other isolated E. coli strains in Waksman's collection (Source).