Cellulose and surfaces

There was a time when we thought we knew everything there was to know about formation of mats (wrinkly spreader types in particular) at the air-liquid interface.  It turns out, thanks to the work of Maxime Ardré, Andrés Diaz and Tatyana Pichugina, that there is so much more to discover.   

The often told story is one that centres on cellulose production by the Wss operon.  Indeed, the cellulose-encoding Wss operon is crucial for formation of a wrinkly spreader (WS) mat — it is the major structural determinant.  However the idea that cellulose acts as a cell-cell glue is only part of the story. 

Sophie Kahn’s work in the late 1990s in which she identified genetic determinants of the adaptive wrinkly spreader mat-forming mutants resulted in discovery of the Wss (Wrinkly Spreader Structural) operon.  Hypothesising that the locus was central to success of WS type during the course of the adaptive radiation, Sophie competed ancestral (SM) SBW25 against a wss::Tn5 (cellulose defective) mutant.  In shaking broth culture fitness of the mutant was no different to the ancestral type, but as predicted, in static broth culture, the mutant was significantly less fit that the ancestor (the ancestor could generate WS types, but not the cellulose mutant).  However, as we noted in the paper, the cellulose mutant was much much less fit that we expected pointing to some phenotype associated with the cellulose operon, in the ancestral genotype, in static broth microcosms (see Spiers et al (2002)).  

In the mean time we tried to obtain some understanding of the contribution that cellulose makes to the ancestral type — in the wild: why does SBW25 (and many other bacteria including E. coli K-12) produce cellulose, what environmental signals activate its expression and what does it use cellulose for?  But with no phenotype in vitro we remained stuck.  Nonetheless, some progress was made: IVET strategies for identifying genes specifically expressed in the plant environment showed that wss is activated in both the rhizopshere and phyllosphere, where it contributes to in planta fitness.  See Gal et al (2003) and Giddens et al (2007).  But again, with no in vitro phenotype we remained stuck.  If only Rainey had not overlooked the significance of Sophie’s discovery that said that cellulose defective SBW25 must manifest a phenotype in static broth microcosms.

Fast forward some years to 2016 when Rainey was tidying up some experiments to aid publication of Peter Lind’s work on evolutionary convergence that involved quantifying diversification of SBW25 from various mutant starting points, including from SBW25 ∆wss.  Observation of 24 h static microcosms containing, separately, SBW25 ∆wss and SBW25, showed a clear phenotype associated with the mutant — a clear zone at the air-liquid interface: SBW25 ∆wss was unable to grow at the meniscus.  This pointed to the fact that ancestral (SM) SBW25 produces cellulose in static broth culture and that production occurs in response to signals encountered at the meniscus. 

Maxime Ardré took this observation further and with aid of a custom-built device obtained low resolution time-resolved data on growth of various genotypes both at the air-liquid interface and in the broth phase.  The paper can be seen here.  In short, amorphous, transparent cellulose is produced at the meniscus leading to a change in viscosity.  This somehow enables growth of tiny microcolonies up and out from the surface of the meniscus.  Just how and why remains to be determined. 

Analysis of mutants defective in Wsp, Mws and Aws c-di-GMP-producing pathways showed that all three (surprisingly) contribute to the phenotype.  We speculate that these regulatory pathways might be involved in sensing and responding to Marangoni forces.  But why involve three regulatory pathways?

Particularly intriguing were observations at the moment the thin cellulose-enabled film collapsed through the effects of gravity.  As mentioned above, cellulose when initially produced is translucent. Once microcolonies begin to emerge through the surface they soon (within a few hours) become too heavy and fall from the meniscus.  Collapse of microcolonies — and soon after the entire film — results in cellulose transforming from a translucent amorphous material into an opaque material that resembles stretched chewing gum as is typical of WS mats. 

Andrés and Tatyana are currently focussed on obtained time-resolved data at single-cell resolution using a cunning device that Andrés built combined with state-of-the-art fluorescent microscopy (including laser scanning microscopy). 

Michael Schwarz is working to improve understanding of dynamics at the population level extending the work of Maxime using low-resolution microscopy µ-beads and more.  

There remains much to be discovered and it is all very very curious.  Rainey thinks all this points to some unappreciated / undiscovered c-di-GMP-dependent behaviour at surfaces — particularly and perhaps solely at air-liquid interfaces.  If this intrigues you then contact us. 

Scroll to Top