Teflon bacteria

Over the last 20 years or so we’ve been realizing that even bacteria don’t always live alone.  Instead, individual cells often stick to each other to make a biofilm, secreting a substance colloquially known as slime (also known to biologists as extracellular matrix).  Slime sounds wet, or sticky; but new work from Joanna Aizenberg’s lab (Epstein et al. 2011 Bacterial biofilm shows persistent resistance to liquid wetting and gas penetration.  PNAS doi: 10.1073/pnas.1011033108) now shows that biofilms have the least wettable surfaces known on earth: they’re considerably more resistant to wetting than Teflon.

How do you quantitate wettability?  Wettability is a function both of the liquid and of the surface.  What you do is take a drop of the liquid, put it on the surface, and measure the angle of contact between the drop and the surface.  If the surface is highly wettable by that liquid, the drop spreads and the angle is low — less than 90°.  If not, i.e. if it’s energetically unfavorable for that particular liquid to spread on that particular surface, the drop stays round and the angle is large.  The theoretical limit of non-wettability is a contact angle of 180°.

Teflon, which is made of polytetrafluoroethylene, is highly non-wettable by water, with a contact angle of about 110°.  But because its non-wettability mostly comes from the fact that it’s very hydrophobic, its contact angle quickly decreases when the liquid you’re using has more hydrophobic character: if you’re using a 50:50 ethanol/water mix, the contact angle is only 60°.  (You can test this at home with a non-stick pan and a bottle of 100 proof vodka.  Let me know the results.)

Biofilm surfaces are astoundingly better than Teflon.  Not only are they even less wettable by water (contact angle ~135°), they also remain non-wettable as you mix in ethanol; even at 70% ethanol, the contact angle is still about 130°.  Only at 90% ethanol do you start seeing significant wettability.  This is really rather important.  If you’re trying to sterilize a surface using ethanol, and the bacteria you’re trying to get rid of are living in a biofilm, nothing less than 90% ethanol will do you any good at all.  At lower ethanol concentrations, the ethanol doesn’t wet the surface, and that means that it can’t get into the biofilm to kill the bacteria.  Epstein et al. show this directly by imaging penetrance into a biofilm using a rhodamine dye; the dye fails to penetrate, leaving most of the bacteria in the biofilm untouched.

The authors tried some other solvents, such as acetone, methanol and isopropanol — the story’s the same, the biofilm resists them all. Reassuringly, bleach does get through — but Lysol doesn’t, and even drain opener needs to sit on the film for a significant time to have an effect.  You start to understand why biofilm infections can be so hard to get rid of.

What’s causing this extraordinary resistance?  Epstein et al. made the guess that it’s all to do with the slime — the extracellular matrix — and took a look at mutant bacteria that are deficient in important components of the matrix.  One mutant was unable to make the matrix exopolysaccharide, another was unable to make the matrix protein.  A third mutant — and this was a clever one to look at — overproduced both protein and exopolysaccharide.  All three showed different behavior from wild type.  The exopolysaccharide-deficient mutant essentially lost all resistance to wetting.  The protein-deficient mutant was slightly impaired, although still about as good as Teflon.  And the overproduction mutant was intermediate between the wild-type and the protein-deficient mutant.  Clearly, the exopolysaccharide is the main component that’s important for the non-wettability of the surface, but equally clearly it’s not as simple as more exopolysaccharide = less wettable.  [Nothing is ever simple.]

Given the extraordinary range of liquids that fail to wet the biofilm surface, the authors suspected that the non-wettability was due to more than just chemistry.  They therefore took a detailed look at the topography of the wild-type and mutant surfaces.  The wild-type has a highly crinkled appearance, whether you look at the 100 µm scale or the 10 µm scale, or even smaller.  Such multi-scaled crinkling was shown a few years ago to be important in creating oil-resistant surfaces.  The shape of the surface matters: if it curves back on itself, then the liquid will reach its desired contact angle without necessarily spreading very far.  If the curving-back is very sharp, producing an overhang for example (like a hoodoo; the surface features that provide oil resistance have been dubbed “micro-hoodoos”), then the liquid would have to change its contact angle in an undesirable way to spread past the overhang.  Instead, it stays put, and the surface remains unwetted.  The exopolysaccharide-deficient mutant is much smoother than the wild-type; and the overexpression mutant, although it’s more wrinkled on the macroscopic scale, turns out to look less crinkled at the 10-100 µm scale.

To separate the effect of topography from that of surface chemistry, Epstein et al. next replicated the crinkled shapes of the biofilm surface in epoxy resin, by first making a negative mold of the surface in PDMS (polydimethylsiloxane) and then pouring epoxy into the mold.   And indeed the shape of the surface alone gives a considerable non-wetting effect; the epoxy replica of the wild-type surface does about as well as Teflon (contact angle for water ~115°).  When this surface is made hydrophobic, by coating with perfluorodecanethiol via vapor-phase deposition, it does better than Teflon (contact angle for water ~130°).  But neither version of the epoxy surface manages to replicate the extraordinary breadth of non-wettability the live biofilm shows; like Teflon, the epoxy surfaces are much more easily wetted when you add more ethanol to the mix.  There’s still a puzzle here to solve: how does the biofilm resist liquids that have low surface tension?

Returning to the practical question of how to get rid of bacterial biofilms, Epstein et al. wondered whether delivering biocides in the vapor phase would work better than trying to get them into the film as liquids.  They used a radiation contrast agent in a test vapor to see how far this approach would get.  Alas, not very far: the gas penetrates only about 10 µm below the surface; the exopolysaccharide mutant is again much easier to get at, with penetrance of ~100 µm.  Clearly this paper raises issues that are going to be important for designing microbicides that are effective against biofilms, as well as novel antiwetting materials.

Epstein AK, Pokroy B, Seminara A, & Aizenberg J (2011). Bacterial biofilm shows persistent resistance to liquid wetting and gas penetration. Proceedings of the National Academy of Sciences of the United States of America, 108 (3), 995-1000 PMID: 21191101