Behind the Paper - Evolutionary adaptations of biofilms infecting cystic fibrosis lungs promote mechanical toughness by adjusting polysaccharide production

Discover the story behind our paper, "Evolutionary adaptations of biofilms infecting cystic fibrosis lungs promote mechanical toughness by adjusting polysaccharide production", which was published in npj Biofilms and Microbiomes

Like Comment

Our article entitled "Evolutionary adaptations of biofilms infecting cystic fibrosis lungs promote mechanical toughness by adjusting polysaccharide production" was published in the journal npj Biofilms and Microbiomes. The Nature Research team had a few questions for us about our article, which we have answered below.

What was the main aim of your research and why did you decide to investigate this?

We wanted to know how different extracellular polysaccharides acted to change the mechanics of biofilms. The model organism we use, Pseudomonas aeruginosa, can have up to three different polysaccharides in its biofilm matrix. There's been an ongoing question about what different functions these polysaccharides could serve, with the underlying motivation being that there's no clear rationale for having more than one matrix polysaccharide unless they serve distinct functions. We give an overview summary of prior work on the roles of different polysaccharides in the introduction to our recent paper published in npj Biofilms and Microbiomes. This earlier work has focused on more traditional biological and chemical benefits to the biofilm.

Our initial decision to focus on mechanics of biofilms was motivated by earlier work we had done (Cooley et al, 2013 Soft Matter). In that paper we used an atomic force microscope (AFM) to pull single bacteria off of surfaces and we showed that the form of the adhesive force connecting a bacterium to a surface was very different for different patterns of polysaccharide production. This was an unexpected finding that surprised us, and we wanted to dig further into how different polysaccharides impacted the mechanics of whole biofilms. So this started off as purely curiosity-driven, basic-science research.

After doing rheology for about a year using the lab strain PA01 and isogenic variants, it was clear to us that the three polysaccharides (Psl, Pel, and alginate) had very different impacts on biofilm mechanics. About this time we became aware of recent work by another group at UT Austin that had characterized how evolution of strains infecting the lungs of patients with Cystic Fibrosis (CF) was linked to increased production of Psl. It had already been known for a long time that strains infecting CF lungs tend to evolve to increase alginate production. So, we wanted to know if our "clean" experiments with well-controlled lab strains had parallels in the "dirty" real world of in vivo disease evolution. We found that it does, and that increased Psl acts to stiffen, strengthen, and toughen biofilms grown from clinical isolates. That was very exciting to us, because it suggests that the mechanical properties we characterize may play a role in protecting the infecting biofilm from clearance.

More is known about CF biofilm infections than most other types of infections, so CF clinical isolates were a good real-world system for us to study. Our hope is that what we have learned can be a stepping-stone for better treatments for CF infections, but also for the many other cases where biofilm disease harms people's lives, such as chronic obstructive pulmonary disease and chronic wounds in diabetic and non-diabetic patients.

How did you go about designing your study?

From biological collaborators, we got a set of isogenic variants of the lab strain PAO1 that greatly overproduce, slightly overproduce, or don't produce at all, the polysaccharides Psl, Pel, and alginate. We also got a set of chronological clinical isolates from four CF patients for which the evolutionary changes in Psl and alginate production had already been determined. This was awesome for us and it allowed us to stand on the shoulders of giants in moving this work forward.

We don't have a rheometer in my lab, but we were fortunate to find a kind collaborator on campus, Professor Kishore Mohanty, who does have a rheometer and was very generous in letting us use it. Because this rheometer was never intended for biofilm work, we had to make adjustments away from ideal experiments. The biggest adjustment was that we had to pool biofilm grown on 10-15 agar petri dishes to have enough sample volume to fill the rheometer tool. There are specialized rheometers that can work with very small sample volumes, corresponding to the thickness of a single biofilm, but because this project came as a bit of a surprise to me I hadn't purchased such a rheometer when setting up my lab. What this meant was that all the biofilms we studied had their structure disrupted when they were scraped off the plate and loaded onto the rheometer. I was concerned initially that this might erase any differences in biofilm mechanics, and we are glad that turned out not to be the case.

The other major piece of experimental design we had to do was to figure out how to measure shifts in biofilm mechanics. Bacterial biology and biofilm mechanics are sensitive to a host of factors, including atmospheric humidity, growth temperature, slight variations in the composition of the growth medium (in our case, nutrient agar plates), and biological stochasticity. We have only imperfect control over some of these factors, and no control over others (in the summer months in Texas I wish I could control the humidity!). The experiment-to-experiment variation in mechanical properties for biofilms grown from the same strain was often comparable to the shifts in mechanical properties that we determined were coming from different patterns of polysaccharide production. Therefore, we had to have a way to normalize out for intrinsic variability so that we could determine what changes were specifically coming from the different polysaccharides. We did this by growing biofilms from different strains simultaneously, on the same batch of nutrient agar plates, in the incubator at the same time and exposed to the same environmental humidity, and performing rheology on all these biofilms on the same day, in immediate succession. When we did this, we found that the shifts associated with changes in polysaccharide production were consistent within each day's worth of data, independent of week-to-week variation ascribed to varying environmental conditions. This was essential in allowing us to analyse our data, because it let us use a baseline biofilm (either wild-type PA01 or the originally-isolated infecting ancestor) and find the changes in biofilm mechanics caused by changes in polysaccharide production.

When we saw that different polysaccharides had different effects on biofilm mechanics, we knew that something more complex than a simple physical gel was happening. To try to get a better understanding of the mechanisms underlying the effects of specific polysaccharides, we used an AFM to separate pairs of bacteria with different patterns of polysaccharide expression. This let us determine what mechanical properties were still present for single bacteria and what mechanical properties were present only for the biofilm, and therefore were emergent properties of the biofilm state. These were also the first measurements anyone has done of inter-bacterial cohesion, so this is also a cool first determination of what force and energy scales operate to stick bacteria together.

What challenges did you face?

Studying biofilms and bacteria from a mechanical and physical perspective is, in and of itself, quite new. The repertoire of methods for studying biological and chemical properties of biofilms is large, but there are not well-established ways to study the mechanics of the polymers produced by biofilms. We are one of only a few groups who have done bulk rheology on biofilms. So the whole method of growing biofilms and transferring them to the rheometer was developed trial and error. We spent 6 months taking data before realizing that our biofilms were evaporating water so fast that it was affecting our results. So we had to build a solvent trap to keep our biofilms from drying out and then re-do all of our experiments. That was a bit of a blow.

It was also a challenge to figure out how to analyse our rheology data, due to the high inter-experiment variability discussed above. Developing the comparative approach described above allowed us to do this. For the biofilms grown from clinical isolates especially, those data looked all over the map and we first tried to track chronological trends in mechanical properties and got nowhere. It was when we realized we could use the first-isolated ancestor as a baseline and do comparative measurements for strains that had evolved to have altered polysaccharide that we were able to determine that there were clear mechanical trends for the biofilms grown from bacteria that had evolved in the lungs of CF patients.

What were the key findings from your research?

Bacteria in biofilms are able to protect themselves from antibiotics and the immune system, in large part because of the polymeric matrix. Based on our results, and the finding that the biofilm elasticities we measure are comparable to the stresses that others have measured as being exerted by phagocytosing neutrophils (as well as macrophages and unicellular predators), we posit that there is likely also a mechanical component to the protective properties of the polymeric matrix. We believe that by compromising the stiffness of the polymeric shield, the immune system could more easily clear an infection naturally: it is easier to remove something that is softer and more fluid than something stiff and hard.

This has an analogy in everyday life, namely bi-annual visits to the dentist. The plaque that forms on your teeth is biofilm. You can brush off the softer plaque/biofilm with your toothbrush, but your dentist has to use a metal pick to scrape off the biofilm that’s gotten too hard to remove with your toothbrush.

What next? What further research is needed in this area?

We have two main steps forward that we're working on now:

We want to find ways to compromise biofilm mechanics. From our recently-published paper, we know what polymers make the biofilm strong, and we have some good ideas about what mechanisms underlie the strengthening effects of Psl and Pel. The key insight is that the biofilm is a network of interacting polymers and proteins. If we can compromise the network by interfering with interactions between specific components, the biofilm should become softer, weaker, and easier for the immune system to clear.

We also want to determine the mechanical limitations on neutrophil phagocytosis. Much is known about the geometric limits on phagocytosis - if the target is a rigid plastic bead, neutrophils can engulf small beads but are unable to engulf large beads. However, nothing is known about how the viscoelastic mechanics of a large target impact the ability of the body's immune cells break off and engulfs large pieces. Obtaining this knowledge would tell us what mechanical properties of biofilms would be most important to disrupt to improve immunological clearance. We are developing new assays to measure exactly this. For the inspiration of this line of research we owe a debt of gratitude to Prof. Phil Stewart's 2014 paper, "Biophysics of Biofilm Infections" in the journal Pathogens and Disease. Thus we stand on the shoulders of giants again.

Read the article in full

Vernita D Gordon

Assistant Professor, The University of Texas at Austin

B.S. in Physics and Math, Vanderbilt University, 1997 Ph.D. in Physics, Harvard University, 2003 I've been part of the faculty in the Physics Department at the University of Texas at Austin since 2010. I enjoy thinking about how the structure and mechanics of multicellular bacterial systems impinge on their biology, and how this perspective could lead to new approaches to treating disease.