How Does That Grab You?

Biologists are discovering that bacteria can cling to your cells much the way a "finger trap" grasps your finger.

Story by Adam Summers   ~  Illustrations by Shawn Gould

Cranberry juice, and lots of it: that's all most people know about urinary tract infections. But I always find my thoughts drifting from juice to the troubling question of invasion. Some strains of Escherichia coli, the same bacterium that lives in your gut, invade the bladder through the urethra, despite the fact that sterile urine, flowing at nearly five feet a second, scours it several times a day. Why aren't the bacteria simply washed away, like the itsy-bitsy spider climbing up the water spout? The answer lies in the ability of bacteria to stick to cells-a biomechanical feat on a molecular scale.

Covering the surface of E. coli are hairlike projections called fimbriae. Each fimbria bears a protein tip that can bind to sugar (or to sugar-coated or sugar-containing) molecules on the surfaces of cells. But getting the stickiness just right is tricky. If the fimbriae are too sticky, they'll adhere to anything that happens to be floating around-which is just about as useful as unrolled strips of masking tape collecting dust bunnies. But if the fimbriae grip too loosely, the bacteria will detach from the surface of a cell at the slightest joggle.

It turns out that not all E. coli strains are equally gummy. Some glom on tightly, sacrificing mobility for a nice stable home. Sticky strains stirred into a suspension of red blood cells "glue" the cells together, eventually resulting in cell globs that are big enough to be seen with the naked eye; the globs remain even after the stirring stops. Other strains hardly stick at all, like vagrants drifting wherever the surrounding fluid takes them.

But there's more to stickiness than bacterial strains. Evgeni Sokurenko, Viola Vogel, and their colleagues at the University of Washington in Seattle have found that even for some invasive bacteria, adhesion is not a fixed trait. Working with strains of E. coli that don't form permanent globs in suspensions of cells, they discovered that the strength of the bond between the proteins on the bacterial fimbriae and the molecules on the outer membranes of other cells can vary, depending on the strength of the force threatening to remove them. Specifically, when the suspensions were stirred, the blood cells immediately clumped together, but when the stirring stopped, the globs dissipated and the cells went back into suspension.

What seemed to be happening was that the bacteria clung tightly to the cells in response to the large shear force exerted on them by the fast-moving fluid. But as the fluid came to rest, the fimbriae's grip on the red blood cells loosened. In other words, the fimbriae seemed to act like a "finger trap," the children's toy made of woven wicker or plastic in the form of a tube. When a child inserts a finger into each end of the tube, the weaving bunches together and both fingers slip in easily. When the child tries to pull them out, though, the tube lengthens and so tightens around the fingers: the harder the pull, the tighter the hold.

Of course, a number of mechanisms have the same effect as a finger trap. And until recently investigators could not analyze how such an effect might be operating at the scale of individual fimbriae. After all, a bacterium is so small that thirty or more of them, laid end to end, would barely span the width of a human hair. And the width of a fimbria compared with the size of a bacterium is roughly as the width of our hair is to us. That makes fimbriae too small for light microscopy, high speed video, and most of the other tools familiar to readers of this column.

So Sokurenko and his colleagues turned to computers to test how some bacteria might vary the strength of their grip on other cells. The biologists built computer models of the complex protein that forms the fimbriae. As the biologists "pulled" on the computerized chemical models, the protein unfolded, bringing more of its sticky tip into contact with its point of attachment on the other cell. The harder the pull, the greater the contact area for the fimbria's tip, and so the harder it grabbed-just like the finger trap [see illustration below].

Two different schematic mechanisms have been proposed to explain how shear force could make the bacterium Escherichia coli (shown covered with blue "hairs," or fimbriae) bind tighter with host cells. In what biologists call the "catch-bond" mechanism (a, left), the force pulls a string connecting a hinged vise (in orange, representing the adhesive tip of a fimbria) to the bacterium. (The string represents the pilin domain, the part of the fimbrial protein that connects the tip to E. coli.) This action closes the hinges, strengthening the bond between the protein and a cell-receptor molecule (a sugar called mannose, in green) on the surface of the cell (red). In the "cryptic-bond" mechanism (b, right), the force on the pilin causes a previously unrecognized binding site to swivel toward the cell, creating a stronger bond between the cell's sugar molecule and the fimbria's adhesive tip. Note that an actual bacterium's fimbriae are all roughly the same length; for clarity, some are shown longer here.

The biologists also manipulated the chemical structure of the protein in the model. When they replaced just one amino acid in the fimbrial protein (to simulate the effect of a simple mutation at one site in the bacterium's DNA), the structure of the fimbria's tip became more rigid, and so the shape of the tip changed less when the protein was pulled. A different substitution in the amino acid chain made the fimbria more flexible, enabling the protein to unfold more readily, even under low flow.

The investigators then tested the effect of rigidity on the grip strength of actual E. coli bacteria with various kinds of fimbriae. Sure enough, the more rigid structure could not adapt as well to changes in shear force exerted by fluid flow, but the more flexible version unfolded fully, even in slow flows, giving it a strong grip. That corollary finding has important implications for the evolution of bacterial strains: even small genetic changes can spell the difference between a floppy, mobile strain and a rigid, stationary one.

So picture this battle plan for invading the urinary tract. In just twenty-four hours E. coli can run through more than sixty generations-enough to take advantage of what natural selection can do for its foot soldiers. The infection is launched by the variably adhesive bacteria, which can move the fastest during intervals between high flows, and can hang on the most tenaciously when the high flows come. Once the bacteria have "captured" an area, a small genetic change can turn them into highly adhesive, always-sticky colonizers. Urination won't dislodge either group; the only hope is large doses of cranberry juice (whose tannins make E. coli less sticky)-or a good antibiotic.

Adam Summers is an assistant professor of ecology and evolutionary biology at the University of California, Irvine (asummers@uci.edu).