...rosette is a rosette is a... (2|3)

by Christoph

I like to imag­ine that mi­cro­scopists in the late 19th cen­tury who looked at wa­ter sam­ples were hap­pily sur­prised when they saw or­ga­nized struc­tures of cell clus­ters in the whole tan­gle of pro­tists and bac­te­ria, which they then liked to call "rosettes" be­cause these were struc­tures known from botany. And on closer in­spec­tion, it turned out that some of them were ac­tu­ally bac­te­r­ial rosettes.

Of course, it could have been fun to re­search when the term "rosette" first ap­peared in the sci­en­tific lit­er­a­ture, but I found it more in­ter­est­ing to un­der­stand how bac­te­ria craft these rosette struc­tures, which bac­te­ria are known to do this, and when, that is, un­der which growth con­di­tions. Here are two (and a half) ex­am­ples, all of which go back to re­search from the early days of bac­te­ri­ol­ogy.

Planc­to­myces beke­fii rosettes

That Planc­to­myces beke­fii (phy­lum Planc­to­myce­tota) forms rosettes has long been known: ac­cord­ing to Fuerst (1995), the Hun­gar­ian re­searcher Gimesi found them in a pond in Buda­pest and ini­tially thought it was a plank­tonic fun­gus (its coni­d­io­phores, to be pre­cise). In these rosettes and those of the closely re­lated Planc­to­myces gra­cilis, few or up to ~30 ovoid cells are con­nected to their cen­ter via rigid stalks (Fig­ure 5). P. beke­fii cells mostly di­vide by bud­ding rather than bi­nary fis­sion, as is com­mon among planc­to­mycetes (see here and here in STC). You can spot a few buds in Fig­ure 2A.

Fig­ure 5. (A) Phase-con­trast mi­cro­graph of a rosette of P. beke­fii, show­ing non-cel­lu­lar stalks ra­di­at­ing to­wards a cen­tral phase-light gran­ule from spher­i­cal cells at the rosette cir­cum­fer­ence, some of which pos­sess buds. Mi­cro­graph taken dur­ing col­lab­o­ra­tive re­search of the au­thor and J. T. Sta­ley (Dept. of Mi­cro­bi­ol­ogy, Uni­ver­sity of Wash­ing­ton). Bar,10 µm. Source. (B) Pg, a rosette of P. gra­cilis with at least 11 vis­i­ble fil­a­ments ; Pb, a small, three-mem­bered rosette of P. beke­fii, the type species of the genus Planc­to­myces and morpho­type la of the Blas­to­caulis-Planc­to­myces group of bud­ding and non­pros­th­e­cately ap­pendaged bac­te­ria. Bar, 5.0 μm (Not vis­i­ble in this cutout). Source

This pe­cu­liar­ity of the planc­to­mycetes makes them eas­ily dis­tin­guish­able from the super­fi­cial­ly sim­i­lar rosettes of Caulobac­ter & com­pany, which prop­a­gate in­dis­crim­i­nately by bi­nary fis­sion (not Hy­phomi­cro­bium though, see part 3). Also, the rosettes of Caulobac­terales and Plancto­my­ce­tes can be eas­ily dis­tin­guished un­der the mi­croscope by their stalks: while the typ­i­cal stalk of the Caulobac­terales is highly flex­i­ble, that of sev­eral Planc­to­myces species is non­‑pro­sthecate, mul­ti­fib­ril­lar and pretty rigid (⌀ ~0.3 µm), but thin­ner ones are known from P. maris (⌀ ~0.1 µm). Some­times, to the de­light of the mi­cro­scopist, they are also en­crusted with iron and man­ganese ox­ide in sam­ples from na­tive fresh­wa­ter habi­tats. "Non-pros­th­e­cate" means that the stalks of the Planc­to­mycetes are not cy­to­plasm-con­tain­ing ex­tru­sions of the outer mem­brane, as in the case of the Caulo­bac­ter­a­les.

How the P. beke­fii stalks are connected/attached to the cells is not known, nor is it known how they are syn­the­sized, i.e., how they are ex­tended dur­ing growth, and how they con­nect to the cen­ter of a rosette (a mi­cro-par­ti­cle or an­other stalk tip?). It is known that many Planc­to­myce­tes can grow on sur­faces as a biofilm, but whether this also ap­plies to P. beke­fii can only be cla­rified if these or­gan­isms can be stud­ied in the lab dur­ing all phases of their life cy­cle.

Nevskia ramosa rosettes

You may never have heard of Nevskia, but they are not that rare: "Dur­ing a calm and sunny weather pe­riod 430,000 Nevskia-like bac­te­ria per mL were found in sur­face sam­ples" from a (fresh­wa­ter) ditch near their in­sti­tute in Old­en­burg, Ger­many by Plad­dies et al. (20­04). "Sur­face sam­ples" in­di­cates that Nevskia is pref­er­en­tially found in the neuston, that is, the com­pany of mi­crobes that thrive in and as the 1–50 µm-thin film at the liquid/air in­ter­face (in the lab, we would call it a "pel­li­cle"). Nevskia ra­mosa is a genus in the or­der Nevskiales and re­lated to the Xan­thomon­adales in the so far poorly re­solved "basal branches" in the (slightly out­dated) phy­lo­ge­netic tree of the Gamma­proteo­bac­teria (see here).

Fig­ure 6. Phase-con­trast pho­tomi­cro­graphs of Nevskia ra­mosa. (E) An en­rich­ment of flat rosettes of Nevskia-like cells from stag­nant wa­ter in a wa­ter­ing can. (F) Geo­met­ri­cal pat­terns of­ten formed on agar plates. (G) Safra­nine-stained flat rosettes of N. ramosa Soe1 (de­tail in the Fron­tispiece). (H) Pure cul­ture of N. ramosa strain OL1. Bars, 25 µm (A to D and F to H) and 10 µm (E). Source

Baben­zien (1967) wrote: "Young motile cells de­velop sub­mersed, then ad­sorb to the wa­ter sur­face, lose the po­lar fla­gel­lum, and form a hya­line slime stalk on the con­cave side of the cell. When a cell mul­ti­plies by bi­nary fis­sion, branch­ing of the stalk oc­curs. The re­sult­ing flat rosette can reach a size of 70 µm in di­am­e­ter" (see im­ages here from H. Cypionka's Mi­kro­biologischer Garten (eng.)). In Fig­ure 6, you see that the for­ma­tion of rosettes by Nevskia – first de­scribed, mind you, in 1892 (ref. 7) – is very ef­fi­cient. But in con­trast to the for­ma­tion of rosettes by Phae­obacter (see part 1) or Planc­to­myces (this part) or Caulobac­ter (part 3) there is 1. no ap­proximate fi­nal size of the ro­settes (num­ber of in­di­vid­ual cells per ro­sette), and 2. the indi­vi­dual cells are not con­nected to each other through "fib­rils," but share an ex­opolysac­cha­ride (EPS) sheath. The EPS con­sists mainly of rham­nose, with small amounts of glu­cose and man­nose, but it is not known how the nicely vis­i­ble branches come about, whether this is a phy­sico-chem­i­cal process or whether the Nevskia ac­tively in­flu­ence it.

No, 'Tu­ber­cle bacilli' do not make rosettes!

In my (lap­top) folder of mi­cro­bial rosette im­ages, I stum­bled across this one in Fig­ure 7. Stu­pid­ly, I had for­got­ten to note the source. At a cur­sory glance, I thought it was an ob­scure older draw­ing of a choanofla­gel­late rosette and had added it to this folder. It seemed sus­picious to me, though, that, un­like in other im­ages of choan­fla­gel­lates, the fla­gella did not ap­pear to be arranged pe­riph­er­ally on the out­side (see here for an ex­am­ple from Kay­ley Hake), but rather on the in­side. A Google re­verse im­age search was sur­pris­ingly un­suc­cess­ful. But a perk of com­mu­ni­cat­ing via "so­cial me­dia" is that you can up­load a pic­ture and ask: "Does any­one know of this pic­ture and where it comes from?" (on mastodon, or on Blue­sky).. Less than 24 hours later, a pos­i­tive re­sponse from mastodon user be­net­wict popped-up in my time­line (thanks a lot, again!). The im­age is re­pro­duced on plate 48 as num­ber VI in the mono­graph (stand-alone schol­arly book in li­brar­ian lingo):

K. B. Lehmann, and R. O. Neu­mann. At­las and es­sen­tials of bac­te­ri­ol­ogy. William Wood and Com­pany, New York, 1897

A dig­i­tal ver­sion of the book is stored in the In­ter­net Archive that you can read here or down­load as PDF.
 

Fig­ure 7. See text for de­tails. Source

The leg­end to the im­age says that it shows 'Tu­ber­cle bacilli' (My­cobac­terium tu­ber­cu­lo­sis) lin­ing the cells of a cav­ity of a "cheesy" (pus‑filled?) bronchial gland of a tu­ber­cu­lo­sis pa­tient. So, what I mis­took as fla­gella are ac­tu­ally bac­te­ria, and the sus­pi­cious rosette are lung ep­ithe­lial cells. Case solved. The au­thors fur­ther men­tion that the im­age is a copy of Pl. II., 9. from a fa­mous pub­lication by Robert Koch, Die Äti­olo­gie der Tu­berku­lose, which I found as chap­ter in a di­gi­tized ver­sion of the Mit­theilun­gen aus dem kaiser­lichen Gesundheits­amte (1884). The compar­ison of Fig­ure 7 with the orig­i­nal Taf. II., 9. re­minded me that at the end of the 19th cen­tury, "copy­ing" quite lit­er­ally meant "re-draw­ing" (with a whiff of artis­tic li­cense), for tech­nical rea­sons. To­day, we tend to think of hi‑res/low-loss pho­to­copies, or at worst, some AI‑gen­erated non­sense.

Stay tuned for part 3, in which I will turn to rosette for­ma­tion in Caulobac­ter, in Es­cherichia coli (in­evitably), and in mul­ti­cel­lu­lar mag­ne­to­tac­tic bac­te­ria (MMB) that were first men­tioned in Elio's piece Could We Have Started Out as Magneto­tactic Bac­te­ria? back in 2007.

 

I gladly men­tion that Herib­ert Cy­p­i­onka, who led most of the re­search on Nevskia men­tioned here, showed me how in­cred­i­bly fast the huge Achro­ma­tium oxa­liferum bac­te­ria (up to 30×125 µm in size) can move dur­ing a lab visit of the Lake Stech­lin branch of the IGB in 2017.

 

Do you want to com­ment on this post? We would be happy about it! Please send us an email, or com­ment on Mastodon, or on Bluesky.