This week's paper is another recent one from my lab "Individual-level bet hedging in the bacterium Sinorhizobium meliloti", now on-line at Current Biology. Will Ratcliff did a guest post earlier, discussing a paper on the experimental evolution of bet hedging. This latest paper reports Will's own experiments.
S. meliloti is best known for its symbiosis with alfalfa. After infecting via root hairs, it reproduces inside developing root nodules.
Alfalfa nodules in our lab; copyright Inga Spence, used by permission.
When the nodules senesce, reproductive rhizobia escape into the soil, leaving the nonreproductive "workers" (bacteroids), which had been providing the plant with nitrogen, behind. A nodule may release millions of rhizobia -- we're not sure how many, actually -- each of which may have accumulated resources there, including high-energy PHB. So a single rhizobial cell that infects a root hair may end up with millions of well-endowed descendants, a few months later. But then what?
If there's another host root nearby, going back into symbiosis is probably the best way to get millions of grandchildren. But only a small fraction of the newly discharged rhizobia are likely to get a chance to re-enlist right away. The others may have to wait.
So Will took a bunch of high-PHB S. meliloti cells (a) and starved them for a month. As expected, many of the rhizobial cells apparently used up most of their PHB. (PHB per cell was measured using a flow cytometer, after staining with Nile red, which fluoresces in proportion to PHB per cell.) Low-PHB cells make up the higher of the two peaks in part (b) of the figure below.
But, to our surprise, there were still many high-PHB cells (right peak in b), even after a month of starvation.
Part (c) of the figure shows what had happened. The two green images show an S. meliloti cell dividing. Green-fluorescent protein (GFP) ended up mostly in the daughter cell. But Nile red staining shows that the mother cell kept most of the PHB.
Why not divide resources equally? The two cells are genetically identical clonemates, so there's no evolutionary reason to be selfish.
This is where bet hedging comes in. Unequal division of resources increases the chance that at least one of the cells will survive long enough to find another host plant or another source of food. Of course, it also increases the chances that one of the cells will die before finding food. In bet hedging, you always give up the chance of a big win (e.g., two survivors) in order to reduce the chance of a big loss (no survivors).
But rhizobial bet hedging is actually more sophisticated than what I've just described. The high-PHB cell further enhances its chances of long-term survival by becoming at least somewhat dormant. This is apparent, for example, in the low growth rate of the high-PHB cells in competition with a labeled competitor, relative to the growth of the low-PHB cells with the same competitor. By not growing, however, the high-PHB cells conserve their resources, so that 70% of them were still alive after 528 days of starvation!
Similar bet hedging strategies are seen in many wild plants. They make some seeds that germinate the next year and others that stay dormant for a few years. If conditions next year are very favorable (for example, if most of their competitors are killed by a disease that doesn't kill them), then many of the nondormant seeds will grow into adult plants, each producing many more seeds. But if next year is terrible (drought, fire, too many competitors, etc.), then maybe the dormant seeds will have a better chance.
Previous examples of bet hedging in bacteria, however, have involved a small fraction of the cells in a large population switching, apparently at random, to some different state. In a large population of genetically identical clonemates, there may therefore be a few individual cells able to survive catastrophe. A small chance of random switching isn't likely to occur in a small population, however, so the variability needed to ensure survival may not be there. With the bet hedging strategy Will found in S. meliloti, on the other hand, even a single cell can bet hedge, producing one faster-growing offspring and one slow-growing offspring resistant to starvation. He also showed that the dormant cells are resistant to some antibiotics, but a reviewer wanted so many additional experiments on that part of the paper that we left it out.
This material is based upon work supported by the National Science Foundation under Grant No. NSF/DEB-0918897.