Climb Every Mountain?

See allHide authors and affiliations

Science  19 Dec 2003:
Vol. 302, Issue 5653, pp. 2074-2075
DOI: 10.1126/science.1093165

Environments are highly variable, providing myriad ecological niches that different species can exploit to ensure their survival. A population of organisms may respond to this variation by becoming niche specialists or by evolving into generalists that can exploit a variety of niches rather than just one. Evolutionary biologists have long recognized that an organism's performance (or fitness) in one specific environment is accompanied by the organism's decreased performance (or fitness) in other environments. On page 2107 of this issue, Buckling et al. (1) explore whether adaptation to specific niches constrains an organism's ability to subsequently diversify into alternative niches. They observe that the ability of the bacterium Pseudomonas fluorescens to diversify into other niches decreases as the organism becomes a specialist and exhibits increased fitness in its selected niche. The investigators attribute this negative correlation to the bacterial population “climbing different adaptive peaks in a rugged fitness landscape.”

We expect specialists—individuals tolerating a narrow range of environmental conditions—to evolve when conditions remain constant because selection favors genotypes with the highest fitness in that environment. By contrast, generalists, which tolerate a broad range of environmental conditions, are expected to emerge when conditions are variable because selection favors the genotype with the highest fitness across all environments (2). But why can't one species successfully reproduce in all possible environments? To explain evidence favoring the selection of specialist species, evolutionary biologists postulate the existence of fitness trade-offs between environments: Traits improving fitness in one environment are presumed to be detrimental in other environments. In principle, several mechanisms can produce such trade-offs. The simplest mechanism is antagonistic pleiotropy in which mutations or traits that are beneficial in one environment are harmful in others (3). A second plausible mechanism is mutation accumulation in which mutations accumulate by genetic drift in genes that may not be necessary in one environment but that may be useful in another. These mutations are neutral in the first environment but deleterious in the second (4). A third argument favoring the selection of specialists, and one that does not necessarily invoke trade-offs, relies on the fact that specialists evolve faster, because the probability of “fixing” environment-specific beneficial alleles is much higher, and their deleterious genetic load is much lower (5).

Conquering a rugged landscape.

(A) In this model of adaptive radiation, an ancestral population (yellow) faces a complex environment with three ecological niches (peaks). Peaks represent optimal genotypes for each niche. Arrows represent different alternative mutations that create potentially useful genotypes. Different mutations place the population near different peaks, allowing for greater diversity. (B) Once the population begins to climb a particular peak, only beneficial mutations that move it uphill can be fixed; alternative mutations are eliminated by selection. In the final stage, the population located at the top of the adaptive peak is under stabilizing selection; thus, every new mutation has a deleterious fitness effect and must be washed out from the population. The result is a niche specialist that exhibits trade-offs in fitness and a reduced ability to genetically diversify into different niches.

In their study, Buckling et al. propagated six replicate populations of P. fluorescens, initiated from the same clone, in a weekly batch-transfer protocol. At each transfer, they isolated a single cell of the numerically dominant phenotype from each one of the six populations and used them to initiate the next set of six populations. In laboratory flasks containing broth media that represent a spatially heterogeneous environment, adaptive radiation of P. fluorescens takes place readily (6). Starting with a single clone, three distinct phenotypes very quickly arise that are adapted to different environmental conditions (6). The researchers measured competitive fitness of each population to assess the degree of adaptation to the environment. As expected, the fitness of the numerically dominant phenotype in a particular environment significantly increased with time, suggesting that this population was climbing a specific adaptive peak (see the figure). The originality of the approach used by Buckling et al. is their enumeration of the different phenotypes (morphs) present at each transfer. Because strong bottlenecks are imposed at each transfer, populations are initiated by single cells. Hence, all the variability observed at the end of the week of propagation has been created de novo. They observed that the number of morphs produced declined significantly with time, with a negative correlation between fitness and the ability to diversify. To explain this correlation, the authors invoke the concept of “rugged fitness landscapes”: In such landscapes, each niche is matched by a different genomic configuration that occupies a given peak in the landscape. When an ancestral population is confronted with many empty ecological niches, the acquisition of mutations moves the population toward any of the nearby peaks (see the figure). Thus, at this initial stage of the adaptive radiation process, a big chunk of all potential mutations creates genotypes close to peaks and, hence, the ability of the population to diversify is boosted. However, once a population moves into the vicinity of a peak, only those mutations that move the population uphill are beneficial. All other mutations are deleterious and are removed by selection. This phenomenon becomes increasingly intense as the population climbs the peak. Eventually, once the population settles at the top of the peak, only a single morph is expected to be maintained in the population through a process of stabilizing selection.

The Buckling et al. findings depend strongly on the association between phenotype and genotype. It is well established that the so-called smooth (SM) and wrinkly-spreader (WS) phenotypes of P. fluorescens are inherited traits (6, 7). However, there is no evidence that the other phenotypes observed by Buckling and co-workers are associated with particular genotypes; they may simply reflect random asymmetries during the process of colony growth on a solid surface.

Evolution experiments with microbes provide overwhelming evidence that specialist genotypes evolve under constant environmental conditions, whereas generalists evolve under fluctuating conditions (8, 9). Indeed, experiments with microorganisms offer a rare opportunity for disentangling the mechanisms underlying fitness trade-offs (antagonistic pleiotropy being the most common mechanism) (9). In contrast to the results of Buckling et al., other evolution experiments have yielded generalist microbial genotypes (9). For example, generalist phenotypes of Chlamydomonas evolved under conditions of fluctuating periods of light and darkness. The generalist phenotypes were fitter than their ancestors under both conditions but not as fit as specialists that evolved under either light or dark conditions (10). These genotypes pay the predicted “jack-of-all-trades and master of none” fitness cost when challenged by constant environmental conditions. By contrast, in experiments with RNA viruses, populations evolving in alternating hosts improved their fitness as much in each host as did populations that evolved in a single host. This finding demonstrates that generalists can also evolve without paying a significant fitness cost (11, 12).

Whether analyzing the evolution of generalists or specialists, the Buckling et al. study demonstrates that adaptation to a single niche limits the ability of a bacterial population to diversify in a heterogeneous environment. Ability to diversify is a particularly interesting fitness component because it enables organisms to occupy a greater number of ecological niches and thus to increase their chances for survival (7). Specialists pay a fitness cost in their reduced ability to diversify and occupy other niches.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
View Abstract

Stay Connected to Science

Navigate This Article