Adaptive multiscapes: an up-to-date metaphor to visualize molecular adaptation

Abstract

Background: Wright's metaphor of the fitness landscape has shaped and conditioned our view of the adaptation of populations for almost a century. Since its inception, and including criticism raised by Wright himself, the concept has been surrounded by controversy. Among others, the debate stems from the intrinsic difficulty to capture important features of the space of genotypes, such as its high dimensionality or the existence of abundant ridges, in a visually appealing two-dimensional picture. Two additional currently widespread observations come to further constrain the applicability of the original metaphor: the very skewed distribution of phenotype sizes (which may actively prevent, due to entropic effects, the achievement of fitness maxima), and functional promiscuity (i.e. the existence of secondary functions which entail partial adaptation to environments never encountered before by the population).

Results: Here we revise some of the shortcomings of the fitness landscape metaphor and propose a new "scape" formed by interconnected layers, each layer containing the phenotypes viable in a given environment. Different phenotypes within a layer are accessible through mutations with selective value, while neutral mutations cause displacements of populations within a phenotype. A different environment is represented as a separated layer, where phenotypes may have new fitness values, other phenotypes may be viable, and the same genotype may yield a different phenotype, representing genotypic promiscuity. This scenario explicitly includes the many-to-many structure of the genotype-to-phenotype map. A number of empirical observations regarding the adaptation of populations in the light of adaptive multiscapes are reviewed.

Conclusions: Several shortcomings of Wright's visualization of fitness landscapes can be overcome through adaptive multiscapes. Relevant aspects of population adaptation, such as neutral drift, functional promiscuity or environment-dependent fitness, as well as entropic trapping and the concomitant impossibility to reach fitness peaks are visualized at once. Adaptive multiscapes should aid in the qualitative understanding of the multiple pathways involved in evolutionary dynamics.

Reviewers: This article was reviewed by Eugene Koonin and Ricard Solé.

Keywords: Adaptive landscape; Environment; Functional promiscuity; Genotype-phenotype map; Neutral networks; Phenotype size.

Fig. 1

Basic elements in the construction of the adaptive multiscape metaphor. a Schematic representation of the space of genotypes. Two genotypes are linked if they are at a distance of one mutational move. b Unfolding of the genotype space into different phenotypes in a given environment. Each group contains all genotypes (colored) that yield the same phenotype. Links joining two genotypes in the same phenotype permit neutral evolution; links joining genotypes in different phenotypes (not explicitly shown) represent mutational moves causing changes in the phenotype. Only the network structure matters to describe population dynamics. c Synthetic representation of genotype networks as circles with area proportional to phenotype size. Thickness of arrows between pairs of phenotypes represents the likelihood to attain the target phenotype conditional on being on the departure phenotype. Those links are asymmetric and weighted. In a given environment, the fitness of each phenotype is color coded, from low (blue) to high (red)

Fig. 2

Cartoon of the mapping of a genotype space into phenotypes in two different environments. Each layer (E1, E2) represents an environment. In each of them, genotypes express different phenotypes with different fitness values. Red dashed arrows indicate possible cases of functional promiscuity: if the population contains a genotype in the region of the yellow phenotype (in E1) that overlaps the blue phenotype (in E2), the function required in E2 is performed, though with a decreased proficiency. For the sake of a clearer representation, not all possible transitions between phenotypes have been depicted

Fig. 3

Population dynamics on adaptive multiscapes. The high degree of mutual accessibility of phenotypes is illustrated as an almost completely connected network where phenotypes are represented as circles with a surface proportional to the phenotype size. As depicted in the yellow phenotype, there is a complex, networked internal structure of genotypes, and molecular populations move on that network, though they occupy a tiny fraction of all possible genotypes. Therefore, there is a waiting time (stasis) before a fitter phenotype is found and fixed (punctuation). The grey curve illustrates the movement of a population within one phenotype. Different possible adaptive trajectories among phenotypes are depicted through colored links. Note that though the red phenotype can in principle be attained through fixation of an appropriate sequence of mutations, the time spent in the yellow phenotype might be, in practice, much longer than that required to find the red one

Fig. 4

Multiscape of RNA sequences of length 10 folded at two different temperatures. There are 9 non-empty phenotypes at 37 °C and 8 at 43 °C, one of them having size 1. Fitness has been chosen to be proportional to the number of unpaired nucleotides in the hairpin loop, colors indicating fitness follow the coding scale introduced in previous figures. Thick solid lines between phenotypes in an environment represent a probability of changing phenotype under a point mutation above 5%; thin solid lines stand for probabilities between 1.5 and 5%; dashed grey lines are for probabilities between 1 and 1.5%; lower probabilities exist but are not shown for clarity. The largest promiscuity occurs between the same phenotype in the two environments (see Table 1); thick dashed light-blue lines indicate likely promiscuous transitions. Phenotypes are labeled in the 37 °C environment; they occupy the same position at 43 °C

Fig. 5

Schematic representation of examples of adaptation in the framework of adaptive multiscapes. Initial environments are labeled E1 and E1’, and different subsequent environments correspond to labels E2, E2’, E3, and E4. We use stilized representations of populations at equilibrium in an environment (grey circles with black boundary) or during adaptive transients (grey circles with white boundary). Only meaningful and likely links between phenotypes in the depicted situation are shown (note the direction of arrows). Large violet arrows stand for environmental transitions. a Distribution of viral populations. Even at equilibrium, these populations maintain high levels of genotypic and phenotypic diversity in their current environments (E1). E2 represents an environment where the population is poorly adapted initially (in the blue phenotype) so it needs to search the genotype space to find and fix advantageous mutations. After the initial bottleneck, it spreads over different phenotypes again. b Drift and switch between neutral networks. Punctuation in the phenotype is enhanced by the loss of fitness of the current phenotype, which here occurs concomitantly with the exploration (search) of the neutral genotype network. A short sequence of intermediate environments is shown (E2, E3, E4). c Effects of gene duplication. The fate of a duplicated gene depends on the existence of a secondary function prior to duplication. In its absence (neofunctionalization, E1, E2) the gene can acquire random mutations and explore the surrounding genotype space. If a secondary function was present (subfunctionalization, E1’, E2’), there is a viable phenotype subjected to a selection pressure that might be rapidly optimized in E2’. In both cases, evolutionary optimization begins once a function is fulfilled. d Waddington’s canalization of an acquired character. In the language of adaptive multiscapes, a subset of genotypes in the population expresses different phenotypes in E1 or E2. Under selection (both natural and artificial) in E2, the population modifies its genomic composition, such that if E1 is restored the population will no longer express the original phenotype (there is no overlap between the equilibrium population in E2 and the orange phenotype in E1)