Bacterial filamentation is an in vivo mechanism for cell-to-cell spreading

Tuan D Tran, Munira Aman Ali, Davin Lee, Marie-Anne Félix, Robert J Luallen

Preprint posted on 21 November 2021 https://www.biorxiv.org/content/10.1101/2021.06.27.450094v2

Article now published in Nature Communications at http://dx.doi.org/10.1038/s41467-022-28297-6

Skilful pathogens: a bacterium that forms filaments to colonise its nematode host

Selected by Angika Basant

Categories: cell biology, ecology, pathology

How do pathogens spread from cell to cell in a live animal? What strategies do they use to navigate complex 3D environments and defence mechanisms in their host? There are few physiological contexts where these questions can be easily addressed as in vivo imaging of internal tissues is challenging in most organisms. However, nematodes are emerging as a useful model system to study infections in this regard.

Studies in 2D cell monolayers have shown that several species of bacteria usurp the host actin cytoskeleton to propel themselves into neighbouring cells either via lateral cell membranes or via filopodia extensions. In this preprint Tran et al. describe a new mode of bacterial spread that appears to avoid the risk of extracellular exposure of the pathogen in a unique fashion. The authors discover a bacterial pathogen in the nematode Oscheius tipulae that adopts a filamentous form to spread across host intestinal cells.

[B. atropi (red) forms long filaments in wild-type nematodes. A mutant is isolated that is defective in this process. Filamentation is rescued in this mutant by expression of the gtaB gene involved in cell wall synthesis. Image from Tran et al., 2021 made available under the CC-BY-NC 4.0 license.]

Key findings:

The authors isolated a wild Oscheius tipulae nematode strain from rotting crab apples. The worms contained coccobacilli-shaped microbes in their intestinal epithelia. These microbes were originally believed to be microsporidia. However, using a panel of microsporidia- and bacteria-specific rRNA in situ hybridization probes, the authors demonstrate that the pathogen in question is actually a bacterium. It could be isolated on LB agar plates and was determined to be a new species in a clade of Bordetella which the authors name Bordetella atropi. Interestingly, staging the infection cycle in worms by pulse-chase experiments revealed the presence of short and long intracellular bacterial filaments at 16- and 24-hours post-infection (hpi) whereas by 38 and 48 hpi most worms contained only the coccobacilli form.

Filaments up to 50 μm in length were observed by confocal microscopy. More detailed analysis with transmission electron microscopy shows nucleoids that appear to be dividing and possible septum formation within these filaments. Using fluorescent dyes that mark the cytoplasm of intestinal cells and actin markers that define the apical and basolateral cell edges, the authors confirm that Bordetella atropi is indeed an intracellular pathogen. It appears to first form filaments in intestinal epithelial cells which septate into coccobacilli prior to cell exit.

The authors next isolated a mutant variant of B. atropi, LUAb7 that is incapable of making filaments in vitro. Strikingly, when infected into the nematode host, LUAb7 showed decreased anterior-posterior spreading and primarily generated coccobacilli. The infection events that were filaments were reduced to 9% compared to 95% in the wildtype strain. Furthermore, WT B. atropi filaments spread to an average of 3 cells and maximum of 8 cells at 34 hpi, whereas LUAb7 was generally seen in a single cell and occasionally in two cells.

The causative mutation in LUAb7 turned out to be a missense mutation in gtaB, a UTP–glucose-1-phosphate uridylyltransferase. GtaB (known as GalU in E. coli) catalyzes glucose-1-phosphate to UDP-glucose conversion, which is required for cell wall synthesis. The R17C mutation identified in LUAb7 modifies a predicted catalytic arginine in a conserved N-terminal motif. Importantly, complementing LUAb7 with gtaB+ from B. atropi resulted in a rescue of in vitro and in vivo filamentation and cell-to-cell spread in the host. Knocking out other members of the UDP-glucose synthesis pathway in B. atropi revealed potential positive and negative regulators of the filamentation process during infection.

What I like about this preprint:

Both the host-pathogen model system used, and the identified mechanism of cell-to-cell spread are exciting and unique. I also like that the study includes the isolation of a filamentation-defective mutant that points to a metabolic pathway and molecular players that could regulate this process of bacterial spread. This opens many new avenues of investigation.

Questions for the authors:

How do the bacteria span across membranes? What is the membrane topology at sites where the bacteria appear to “pierce” cell-cell junctions?
Is there actin accumulating on coccobacilli? Maybe to facilitate cytosolic motility or are they likely to be non-motile?
Were any events observed that point to the mechanism of cell exit by the coccobacilli? Is there a technical limitation on how many hours post-infection the infected worms can be imaged and monitored?
LuAb7 reduces host fitness to the same extent as WT bacteria. In what way is the loss in the ability of cell-to-cell spread detrimental to LuAb7? Are the total number of produced bacteria/division cycles reduced during infection?

Posted on: 30 January 2022 , updated on: 13 February 2022

doi: https://doi.org/10.1242/prelights.31322

Read preprint

(No Ratings Yet)

Author's response

Tuan Tran shared

We would like to thank Dr. Basant for the nice article highlighting the main findings of our study. You raised several interesting questions that we have been pondering and trying to address throughout the project. Due to certain technical difficulties, we only have indirect clues/evidence to answer some of these questions. Currently, we are conducting experiments to address these questions more thoroughly.

1. How do the bacteria span across membranes? What is the membrane topology at sites where the bacteria appear to “pierce” cell-cell junctions?

We think the two parts of this question are closely related, the insight into one could inform about the other. Mechanistically, how the bacteria span across membranes is a question of great interest to us to explore in future studies. It could be simply the force generated by the growing filaments that creates an opening for the bacteria to invade the neighboring cells. Alternatively, it has been reported that Burkholderia spp. cause cell fusion with a type 6 secretion system protein called VgrG-5 that is localized at the polar end of the bacillus. B. atropi encodes several homologs of this protein in its genome and therefore we can investigate whether B. atropi also employs similar strategy to span across membranes. Depending on what mechanism(s) used by B. atropi to cross the membranes, the topology of the membranes could be intact elsewhere except for a small opening where the bacteria push through, or the whole lateral membranes break down as cells fuse.

2. Is there actin accumulating on coccobacilli? Maybe to facilitate cytosolic motility or are they likely to be non-motile?

We have not observed host actin accumulating on coccobacilli in the cytoplasm of intestinal cells, so it is unlikely that the coccobacilli are motile in the cytoplasm. This also fits into our model where the filamentous form enables the bacteria to reach new locations (i.e., either in the same cell when the filaments are still short or neighboring cells as the filaments grow longer). We do sporadically observe host actin coating of some coccobacilli near the apical side of the cells at early infection stages, which may be a remnant of invasion as an isoform of host actin (ACT-5) is highly concentrated here.

3. Were any events observed that point to the mechanism of cell exit by the coccobacilli? Is there a technical limitation on how many hours post-infection the infected worms can be imaged and monitored?

From the TEM images at 48 hours post infection, we observed coccobacilli filled both the intestinal cells and the lumen, with the microvilli delineating intracellular-extracellular environments becoming irregular, morphologically abnormal, and hardly visible. We lean toward the possibility that the host cells are disrupted/lysed as bacteria exit, as we have not seen evidence of the bacteria surrounded by host membranes. However, it remains possible that other mechanism such as exocytosis might be involved in exit at earlier time points when the number of bacteria has not overwhelmed the host.
There is no technical limit for imaging these nematodes as they remain transparent throughout development and adulthood. For our study, we normally monitor/image worms at 48 hours post infection at the latest, since we know from our lifespan data that the worm population normally starts to die off within 2-3 days after exposure to the bacteria. Unfortunately, we currently cannot conduct live imaging of O. tipulae of long periods of time. The most common method of anesthetizing worms on agarose pad only allows a few-hour window before animals show signs of cell death, but some systems have been established to perform long-term imaging in C. elegans for up to ~ 40 hours.

4. LuAb7 reduces host fitness to the same extent as WT bacteria. In what way is the loss in the ability of cell-to-cell spread detrimental to LuAb7? Are the total number of produced bacteria/division cycles reduced during infection?

We expect that the total number of bacterial progeny produced by LUAb7 in vivo would be reduced compared to WT B. atropi, as the loss in ability to spread from cell to cell should reduce it replicative volume in the host. And, as we mentioned before, it seems unlikely that the coccobacilli (the major morphological phenotype of LUAb7) are motile in cytoplasm. We have tried to answer this question using standard colony forming unit (CFU) counts from lysed nematodes. However, we have not been able to get a clear answer to this question since there are many confounding factors during infection that make it difficult to directly compare the number of bacteria produced by LUAb7 vs. WT. These include B. atropi colonization of the intestinal lumen, variability in the frequency of infection per experiment, and carry-over and growth of B. atropi after the host nematodes are removed from the pathogen and transferred to standard media plates. We are currently trying to optimize the experimental design to address this question.

Have your say Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Also in the cell biology category:

Fetal brain response to maternal inflammation requires microglia

Bridget Elaine LaMonica Ostrem, Nuria Dominguez Iturza, Jeffrey Stogsdill, et al.

Selected by Manuel Lessi

Alteration of long and short-term hematopoietic stem cell ratio causes myeloid-biased hematopoiesis

Katsuyuki Nishi, Taro Sakamaki, Akiomi Nagasaka, et al.

Selected by Manuel Vicente, Anaisa Ferreira

Clusters of lineage-specific genes are anchored by ZNF274 in repressive perinucleolar compartments

Martina Begnis, Julien Duc, Sandra Offner, et al.

Selected by Silvia Carvalho

Also in the ecology category:

How the liver contributes to stomach warming in the endothermic white shark Carcharodon carcharias

David C. Bernvi, Geremy Cliff

Selected by Sarah Young-Veenstra

Pervasive sublethal effects of agrochemicals as contributing factors to insect decline

Lautaro Gandara, Richard Jacoby, François Laurent, et al.

Selected by Roberto Amadio

Multimodal interactions in Stomoxys navigation reveals synergy between olfaction and vision

Merid N Getahun, Steve Baleba, John Ngiela, et al.

Selected by Maitri Manjunath

Also in the pathology category:

Hypoxia blunts angiogenic signaling and upregulates the antioxidant system in elephant seal endothelial cells

Kaitlin N Allen, Julia María Torres-Velarde, Juan Manuel Vazquez, et al.

Selected by Sarah Young-Veenstra

H₂O₂ sulfenylates CHE linking local infection to establishment of systemic acquired resistance

Lijun Cao, Heejin Yoo, Tianyuan Chen, et al.

Selected by Marc Somssich