Microbiology labs typically contain myriad flasks and stacks of petri dishes crowded with bacteria. That’s fine for someone studying their physiology or genetics. But for researchers wanting to gain insight into bacterial behavior, that laboratory setup is far from optimal.
The problem is that homogeneous environments, such as petri dishes, are quite different from the natural surroundings of microorganisms. There can be billions of bacteria on an agar plate or in an incubator’s liquid broth, but in the wild, bacteria tend to grow in dense clusters comprising a relatively small number of cells. Two collaborating groups at the University of Texas at Austin have devised a method of trapping bacteria, such as Pseudomonas aeruginosa or Escherichia coli, in microenvironments that more closely model their natural habitat—with some surprising results. (mBio, 1:e00202-10, 2010.)
Chemist and bioengineer Jason Shear’s lab invented a method of crafting tiny structures from concentrated protein solutions—microscopic cages that measure about ten to twenty micrometers across and contain a few picoliters of liquid. Microbiologist Marvin Whiteley realized that the structures, which he dubbed “lobster traps,” could ensnare small numbers of bacteria, and the two worked together to study social behavior among clumps of concentrated bacteria. They were able to trap a single Pseudomonas and let it grow and divide, in the hope that they could incite natural bacterial behavior by mimicking the stimuli experienced by aggregates of a few thousand cells within biological nooks such as alveoli in the lungs. “You don’t find a hundred billion bacteria [in a broth] in nature,” Whiteley says.
Quorum sensing involves bacteria communicating via small signaling molecules, and the behavior can result in the creation of antibiotic-resistant biofilms. But in liquid cultures it’s difficult to precisely define the population sizes and environmental conditions at which bacteria start to engage in such behavior. One model of quorum sensing posits that it’s simply a function of population density; other models propose that the rate at which molecules used to signal population size can flow through the substrate is more important.
Earlier designs for ultralow-volume bacterial traps, using conventional microfabrication materials, have foiled attempts to study quorum sensing by not allowing bacteria to organize three-dimensionally, by making it hard to precisely control fluid flow through the system, and by rendering cell growth difficult. Shear’s microstructures are made from a protein, bovine serum albumin (BSA), cross-linked at a scale that traps individual bacteria but is porous enough to allow small molecules to diffuse through the walls, so nutrients can get in and waste products can be washed away. Trapped bacteria are able to grow in these structures, and the roofs are transparent, so it’s possible to see what is going on inside.
By using different-size traps, Whiteley and Shear could control the population size they grew from a single founder bacterium. They found that with 2,600 bacteria in a trap, a reporter gene involved in quorum sensing was not turned on. However, in a bigger trap, with 8,500 bacteria at the same density, there was evidence that the microbes were signaling to each other. The flow rate in the surrounding media also had an effect: slower rates prevented information from being carried away too rapidly, enabling quorum sensing in smaller populations. Robert Palmer, who evaluated the paper for F1000, says these findings represent “a significant advance in how we think about quorum sensing.”
But the big surprise came when graduate students Jodi Connell and Aimee Wessell treated the cultures in the traps with the antibiotic gentamicin. Although bacteria at low density (~20 cells per picoliter) were mostly killed, those growing at high density (~150 to 225 cells per picoliter) survived. “Who would have thought that 150 cells could resist a concentration of antibiotic that kills 100,000 in a test tube?” Whiteley says. This finding implies that bacterial resistance can arise in very small, early bacterial colonies, which could obviously complicate treating infection with antibiotics.
Whiteley says that while the effects of substrate diffusion on quorum sensing weren’t unexpected, nobody had been able to do the experiments until now. “For my lab it’s a new way to do microbiology,” he says. Now, Whiteley and Shear want to experiment with differently shaped traps, and to look at signaling between different bacterial species in different sections within the same trap. They’re hopeful that these studies will help elucidate how bacteria live in nature, and especially in the human system. “Our blue-sky hope,” Shear says, “is to know enough to mimic certain environments in the body.”
A Hidden Jewel refers to an article evaluated in Faculty of 1000 that was published in a specialist journal. You can read the evaluation of Shear & Whiteley’s article here.