The agents of natural selection cause evolutionary changes in population gene pools. They include a plethora of familiar abiotic and biotic factors that affect growth, development, and reproduction in all living things. For example, the evolutionary future of a plant species’ population is shaped by physical agents such as soil moisture and fertility, and by living agents such as competitors, herbivores, pathogens, and pollinators. All of these agents are components of the external environment that surrounds every individual in nature.
But what about the physiological and biochemical conditions that prevail within an organism’s cells, tissues, and organs? The phenotypic variation available for screening by natural selection results from many developmental events occurring within the growing organism. These developmental processes and the genes that control them indirectly can influence many aspects of individual growth, morphology, and reproductive ability. Hence, natural selection can favor particular combinations of genes that optimize the growth and reproductive (Darwinian) fitness of genetically distinct organisms, mediated by the production of specific regulatory proteins, and other physiologically active metabolites. The selection pressures “derive from the internal dynamics of a functioning organism” as stated by Schwenk and Wagner (2004).
Perhaps more intriguing as hidden agents of natural selection are microscopic symbiotic organisms commonly called endosymbionts that reside within the tissues of many animals and plants. Among the best studied endosymbionts are the fungi known as endophytes that live inside the tissues of most plants. These microscopic fungi exist as tubular hyphae that grow between cells in the host plant’s leaves; many of these fungi are asexual, do not produce spores, and are completely hidden from view unless observed under the microscope (Figure 1). The hyphae grow along with the host and may infect the host’s seeds, thereby being transferred to the young offspring that will eventually emerge when the seeds germinate.
For endophytes to function as significant internal agents of natural selection, they must elicit distinct effects on the phenotypes of the host genotypes they inhabit. Several fungal endophyte species have been shown to affect growth and reproduction of their host plants in either positive or negative ways. The direction of these symbiotic effects typically depend on both environmental conditions and host genotype. For example, in the widespread perennial ryegrass Lolium perenne (Figure 2), interactive effects of host genotype with endophyte infection have been reported for a diverse set of morphological and physiological traits in the host, such as tiller production, carbohydrate storage, net photosynthesis, dry plant mass, and seed yield. For some host genotypes, endophyte-mediated effects are positive, while for others they are neutral or negative. Thus the effects of external agents of selection in the immediate environment may be modulated by endosymbionts acting as additional, internal agents of selection.
As every student of evolutionary biology knows, the result of natural selection is adaptation of the population to its local environment. This means that organisms will be better able to survive and reproduce in a particular selective environment as they become better adapted to it. But can internal agents of selection result in adaptation of host populations? For some mutualistic interactions involving endosymbiotic microbes, the answer may be “yes”. Mycorrhizal fungi, which are microbes symbiotic with the roots of many plant species, are predicted to show co-adaptation with the host populations they have evolved with. In prairies of the American Midwest, a reciprocal cross-inoculation experiment was conducted with three populations of the dominant prairie grass Andropogon gerardii (big bluestem). The roots of this large species are usually host to species of mycorrhizal fungi that improve its growth and reproduction and may be critical to its dominance in the tallgrass prairie ecosystem. Samples of soil from three prairie sites (Kansas, Minnesota, and Illinois) were reciprocally inoculated with fungi, and big bluestem plants from each population were planted into the different soil-fungi combinations. The researchers reported that root colonization of the beneficial mycorrhizal fungi was always greatest on plants in their usual local soil which contained the fungi normally living there. Generally, plants in local soil (inoculated with the local fungi) also showed greater reproduction compared to plants in soil from non-local sites. Thus, the authors maintained that local plant genotypes responded most positively to the molecular signals sent by co-adapted mycorrhizal fungal communities. This implies that the symbiotic fungi at a site had acted as internals agents of natural selection on the local host population in such a way as to maximize plant evolutionary fitness.
These examples provide evidence that the agents of natural selection are not always overt components of the external environment as is often supposed. Internal conditions within the organism interact with its genotype, affecting growth and development, and also the organism’s reproductive capacity compared to other organisms with different genotypes (i.e, its relative fitness). Symbiotic microbes within the bodies of animals and plants are an important part of this hidden side of natural selection and deserve increasing recognition by evolutionary biologists.