Epistasis can lead to fragmented neutral spaces and contingency in evolution

In evolution, the effects of a single deleterious mutation can sometimes be compensated for by a second mutation which recovers the original phenotype. Such epistatic interactions have implications for the structure of genome space--namely, that networks of genomes encoding the same phenotype may not be connected by single mutational moves. We use the folding of RNA sequences into secondary structures as a model genotype-phenotype map and explore the neutral spaces corresponding to networks of genotypes with the same phenotype. In most of these networks, we find that it is not possible to connect all genotypes to one another by single point mutations. Instead, a network for a phenotypic structure with n bonds typically fragments into at least 2(n) neutral components, often of similar size. While components of the same network generate the same phenotype, they show important variations in their properties, most strikingly in their evolvability and mutational robustness. This heterogeneity implies contingency in the evolutionary process.     

What is a phenotype? History and new developments of the concept

Even though the word “phenotype”, as well as the expression “genotype–phenotype relationship”, are a part of the everyday language of biologists, they remain abstract notions that are sometimes misunderstood or misused. In this article, I begin with a review of  the genesis of the concept of phenotype and of the meaning of the genotype-phenotype “relationship" from a historical perspective. I then illustrate how the development of new approaches for exploring the living world has enabled us to phenotype organisms at multiple levels, with traits that can either be measures or parameters of functions, leading to a virtually unlimited amount of phenotypic data. Thus, pleiotropy becomes a central issue in the study of the genotype–phenotype relationship. Finally, I provide a few examples showing that important genetic and evolutionary features clearly differ with the phenotypic level considered. The way genotypic variation propagates across the phenotypic levels to shape fitness variation is an essential research program in biology.

     

Why are phenotypic mutation rates much higher than genotypic mutation rates?

The evolution of genotypic mutation rates has been investigated in numerous theoretical and experimental studies. Mutations, however, occur not only when copying DNA, but also when building the phenotype, especially when translating and transcribing DNA to RNA and protein. Here we study the effect of such phenotypic mutations. We find a maximum phenotypic mutation rate, umax, that is compatible with maintaining a certain function of the organism. This may be called a phenotypic error threshold. In particular, we find a minimum phenotypic mutation rate, umin, with the property that there is (nearly) no selection pressure to reduce the rate of phenotypic mutations below this value. If there is a cost for lowering the phenotypic mutation rate, then umin is close to the optimum phenotypic mutation rate that maximizes the fitness of the organism. In our model, there is selective pressure to decrease the rate of genotypic mutations to zero, but to decrease the rate of phenotypic mutations only to a positive value. Despite its simplicity, our model can explain part of the huge difference between genotypic and phenotypic mutation rates that is observed in nature. The relevant data are summarized.     

Genotype-Phenotype Maps Maximizing Evolvability: Modularity Revisited

The mechanisms translating genetic to phenotypic variation determine the distribution of heritable phenotypic variance available to selection. Pleiotropy is an aspect of this structure that limits independent variation of characters. Modularization of pleiotropy has been suggested to promote evolvability by restricting genetic covariance among unrelated characters and reducing constraints due to correlated response. However, modularity may also reduce total genetic variation of characters. We study the properties of genotype-phenotype maps that maximize average conditional evolvability, measured as the amount of unconstrained genetic variation in random directions of phenotypic space. In general, maximal evolvability occurs by maximizing genetic variance and minimizing genetic covariance. This does not necessarily require modularity, only patterns of pleiotropy that cancel on average. The detailed structure of the most evolvable genotype-phenotype maps depends on the distribution of molecular variance. When molecular variance is determined by mutation-selection equilibrium either highly pleiotropic or highly modular genotype-phenotype maps can be optimal, depending on the mutation rate and the relative strengths of stabilizing selection on the characters.     

Novel germline mutations in the APC gene and their phenotypic spectrum in familial adenomatous polyposis kindreds

Among 23 germline mutations identified in the APC screening of 45 familial adenomatous polyposis (FAP) patients, we have found 10 different novel frameshift mutations in 11 apparently unrelated patients. In two cases, an additional missense mutation was detected. One previously described as a causative germline mutation (S2621C), associated with a 1-bp insertion (4684insA) on the opposite allele, did not segregate with the FAP phenotype in the family and was therefore considered as being non-pathogenic. The other (Z1625H) was located 2 codons before a 1-bp deletion (4897delC). Both mutations were transmitted together from an FAP father to his affected son. The FAP phenotype of these 10 novel truncating mutations was clinically documented within their kindreds. Important variability was observed in the phenotype. Interestingly, we noted that a mutation (487insT) localized at the boundary of the 5’ attenuated APC phenotype region in two unrelated families resulted in classical polyposis. A clear-cut genotype-phenotype correlation could be drawn in only two instances. In one family, a 4684insA mutation led to a mild polyposis associated with early inherited osteomas and, in the family bearing the double mutation (Z1625H+4897delC), the phenotype was obviously a 3′ attenuated type. Our data illustrate the wide genetic and phenotypic heterogeneity of this condition between and within the families, making the establishment of correlations complex and any prediction in this disease difficult, although targeting the mutation site may be helpful in some specific cases. Received: 11 February 1997 / Accepted: 11 April 1997     

Ontogenetic plasticity of anatomical and ecophysiological traits and their correlations in Iris pumila plants grown in contrasting light conditions

To better understand what directs and limits the evolution of phenotype, constraints in the realization of the optimal phenotype need to be addressed. That includes estimations of variability of adaptively important traits as well as their correlation structures, but also evaluation of how they are affected by relevant environmental conditions and development phases. The aims of this study were to analyze phenotypic plasticity, genetic variability and correlation structures of important Iris pumila leaf traits in different light environments and ontogenetic phases, and estimate its evolutionary potential. Stomatal density, specific leaf area, total chlorophyll concentration and chlorophyll a/b ratio were analyzed on I. pumila full‐sib families in the seedling phase and on the same plants after 3 years of growth in contrasting light conditions typical for ontogenetic stage in question. There was a significant phenotypic plasticity in both ontogenetic stages, but significant genetic variability was detected only for chlorophyll concentrations. Correlations of the same trait between different stages were weak due to changes in environmental conditions and difference in ontogenetic reaction norms of different genotypes. Ontogenetic variability of correlation structures was detected, where correlations and integration were higher in seedlings compared with adult plants 3 years later. Correlations were affected by environmental conditions, with integration being higher in the lower light conditions, but correlations between phases being stronger in the higher light treatment. These findings demonstrated that the analyzed traits can be selected and can mostly evolve independently in different environments and ontogenetic stages, with low genetic variability as a potentially main constraint.     

What is wrong with absolute individual fitness?

One of the most basic facts about evolution is that fitness is a relative concept. It does not matter how well an organism survives and reproduces, only that it does so better than other organisms bearing alternative traits. Nevertheless, many evolutionary arguments are framed in terms of absolute individual fitness. The absolute fitness criterion (AFC) can be justified in terms of relative fitness only given certain assumptions that are frequently violated in nature. In particular, interactions must occur in groups that are randomly formed and phenotypic variation among groups must be tightly coupled to genetic variation. Complicating the genotype-phenotype relationship can cause phenotypic variation among groups to become nonrandom, even when the groups are randomly formed, favoring traits that do not maximize absolute individual fitness. Complex genotype-phenotype relationships and complex population structures require explicit models of evolutionary change based on relative fitness differences within and among groups.     

The topology of the possible: formal spaces underlying patterns of evolutionary change

The current implementation of the Neo-Darwinian model of evolution typically assumes that the set of possible phenotypes is organized into a highly symmetric and regular space equipped with a notion of distance, for example, a Euclidean vector space. Recent computational work on a biophysical genotype-phenotype model based on the folding of RNA sequences into secondary structures suggests a rather different picture. If phenotypes are organized according to genetic accessibility, the resulting space lacks a metric and is formalized by an unfamiliar structure, known as a pre-topology. Patterns of phenotypic evolution-such as punctuation, irreversibility, modularity--result naturally from the properties of this space. The classical framework, however, addresses these patterns by exclusively invoking natural selection on suitably imposed fitness landscapes. We propose to extend the explanatory level for phenotypic evolution from fitness considerations alone to include the topological structure of phenotype space as induced by the genotype-phenotype map. We introduce the mathematical concepts and tools necessary to formalize the notion of accessibility pre-topology relative to which we can speak of continuity in the genotype-phenotype map and in evolutionary trajectories. We connect the factorization of a pre-topology into a product space with the notion of phenotypic character and derive a condition for factorization. Based on anecdotal evidence from the RNA model, we conjecture that this condition is not globally fulfilled, but rather confined to regions where the genotype-phenotype map is continuous. Equivalently, local regions of genotype space on which the map is discontinuous are associated with the loss of character autonomy. This is consistent with the importance of these regions for phenotypic innovation. The intention of the present paper is to offer a perspective, a framework to implement this perspective, and a few results illustrating how this framework can be put to work. The RNA case is used as an example throughout the text.     

Human Intellectual Disability Genes Form Conserved Functional Modules in Drosophila

Intellectual Disability (ID) disorders, defined by an IQ below 70, are genetically and phenotypically highly heterogeneous. Identification of common molecular pathways underlying these disorders is crucial for understanding the molecular basis of cognition and for the development of therapeutic intervention strategies. To systematically establish their functional connectivity, we used transgenic RNAi to target 270 ID gene orthologs in the Drosophila eye. Assessment of neuronal function in behavioral and electrophysiological assays and multiparametric morphological analysis identified phenotypes associated with knockdown of 180 ID gene orthologs. Most of these genotype-phenotype associations were novel. For example, we uncovered 16 genes that are required for basal neurotransmission and have not previously been implicated in this process in any system or organism. ID gene orthologs with morphological eye phenotypes, in contrast to genes without phenotypes, are relatively highly expressed in the human nervous system and are enriched for neuronal functions, suggesting that eye phenotyping can distinguish different classes of ID genes. Indeed, grouping genes by Drosophila phenotype uncovered 26 connected functional modules. Novel links between ID genes successfully predicted that MYCN, PIGV and UPF3B regulate synapse development. Drosophila phenotype groups show, in addition to ID, significant phenotypic similarity also in humans, indicating that functional modules are conserved. The combined data indicate that ID disorders, despite their extreme genetic diversity, are caused by disruption of a limited number of highly connected functional modules.     

Identical mutations and phenotypic variation

The relationship between pathogenetic mutations and disease phenotype is becoming increasingly complex. Well-delineated clinical entities can be genetically heterogeneous, and mutations in a particular gene may result in fundamental clinical differences. Genetic heterogeneity includes mutations at different gene loci or allelic mutations within a single gene, resulting in a similar phenotype. However, one and the same mutation is expected to be associated with a uniform clinical picture. In the present article, evidence is presented that this is not necessarily the case, and examples of identical mutations resulting in highly variable combinations of clinical features are discussed. Although the number of examples of this puzzling phenomenon is rapidly increasing, the underlying mechanisms are as yet poorly understood. In some cases, interacting genetic alterations can be held responsible for the phenotypic heterogeneity; in others, epigenetic phenomena provide a plausible explanation. These and other mechanisms under discussion are considered here. The Mendelian concept of monofactorial disease causation appears to be increasingly untenable for a growing number of developmental errors. Received: 15 April 1997 / Accepted: 30 April 1997     

Regulatory gene mutation: a driving force behind group a Streptococcus strain‐ and serotype‐specific variation

Data from multiple bacterial pathogens are consistent with regulator‐encoding genes having higher mutation frequencies than the genome average. Such mutations drive both strain‐ and type‐ (e.g., serotype, haplotype) specific phenotypic heterogeneity, and may challenge public health due to the potential of variants to circumvent established treatment and/or preventative regimes. Here, using the human bacterial pathogen the group A Streptococcus (GAS; S. pyogenes) as a model organism, we review the types and regulatory‐, phenotypic‐, and disease‐specific consequences of naturally occurring regulatory gene mutations. Strain‐specific regulator mutations that will be discussed include examples that transform isolates into hyper‐invasive forms by enhancing expression of immunomodulatory virulence factors, and examples that promote asymptomatic carriage of the organism. The discussion of serotype‐specific regulator mutations focuses on serotype M3 GAS isolates, and how the identified rewiring of regulatory networks in this serotype may be contributing to a decades old epidemiological association of M3 isolates with particularly severe invasive infections. We conclude that mutation plays an outsized role in GAS pathogenesis and has clinical relevance. Given the phenotypic variability associated with regulatory gene mutations, the rapid examination of these genes in infecting isolates may inform with respect to potential patient complications and treatment options.     

The interplay between phenotypic and ontogenetic plasticities can be assessed using reaction-diffusion models

Every morphological, behavioral, or even developmental character expression of living beings is coded in its genotype and is expressed in its phenotype. Nevertheless, the interplay between phenotypic and ontogenetic plasticities, that is, the capability to manifest trait variations, is a current field of research that needs morphometric, numerical, or even mathematical modeling investigations. In the present work, we are searching for a phenotypic index able to identify the underlying correlation among phenotypic, ontogenetic, and geographic distribution of the evolutionary development of species of the same genus. By studying the case of Pseudoplatystoma fishes, we use their skin patterns as an auxiliary trait that can be reproduced by means of a reaction diffusion (RD) model. From this model, we infer the phenotypic index in terms of one of the parameters appearing in the mathematical equations. To achieve this objective, we perform extensive numerical simulations and analysis of the model equations and link the parameter variations with different environmental and physicochemical conditions in which the individuals develop, and which may be regulated by the ontogenetic plasticity of the species. Our numerical study indicates that the patterns predicted by a set of reaction diffusion equations are not uniquely determined by the value of the parameters of the equation, but also depend on how the process is initiated and on the spatial distribution of values of these parameters. These factors are therefore significant, since they show that an individual’s growth dynamics and apparent secondary transport processes, like advection, can be determinant for the alignment of motifs in a skin pattern. Our results allow us to discern the correlation between phenotypic, ontogenetic, and geographic distribution of the different species of Pseudoplatystoma fishes, thus indicating that RD models represent a useful taxonomic tool able to quantify evolutionary indexes.     

Phenotypic plasticity: Eco-Devo and evolution

Phenotypic plasticity refers to the ability of an organism to alter its physiology/morphology/behavior in response to changes in environmental conditions. Although encompassing various phenomena spanning multi-ple levels of organization, most plastic responses seem to take place by altering gene expression and eventually altering ontogenetic trajectory in response to environmental variation. Epigenetic modifications provide a plausi-ble link between the environment and alterations in gene expression, and the alterations in phenotype based on environmentally induced epigenetic modifications can be inherited transgenerationally. Even closely related species and populations with different genotypes may exhibit differences in the patterns and the extents of plastic responses, indicating the wide existence of plasticity genes which are independent of trait means and directly respond to environmental stimuli by triggering phenotypic changes. The ability of plasticity is not only able to affect the adaptive evolution of species significantly, but is also an outcome of evolutionary processes. Therefore, phenotypic plasticity is a potentially important molder of adaptation and evolution.     

X-linked adrenomyeloneuropathy associated with 14 novel ALD-gene mutations: no correlation between type of mutation and age of onset

Adrenomyeloneuropathy (AMN) represents a milder form of X-linked adrenoleukodystrophy (ALD), the most frequent peroxisomal disorder. The disease is characterised by an abnormal accumulation of saturated, very long chain, fatty acids, because of altered peroxisomal ?-oxidation that concomitantly leads to demyelination in the central and peripheral nervous systems. ALD shows a highly variable phenotypic expression and extensive mutation analysis in ALD patients has failed to establish a genotype-phenotype correlation, even in the presence of the same ALD-gene defect. Therefore, we have looked for a relationship between the molecular lesion and the age of onset in 19 patients with a well-classified clinical course of AMN. The nearly complete novel spectrum of ALD gene mutations identified has revealed no obvious correlation between the type of mutation and age of AMN onset in this small series. However, intrafamiliar concordance could be observed with respect to the occurrence of adrenocortical insufficiency. This supports the idea of one (or more) additional gene(s) contributing to the phenotypic expression of ALD. Electronic Publication     

The expression of andromonoecy in Solanum hirtum (Solanaceae): phenotypic plasticity and ontogenetic contingency

Sex expression (the proportions of staminate and hermaphrodite flowers produced) in andromonoecious Solarium hirtum is phenotypically plastic, and there is genetic variation for sex expression plasticity. Changes in sex expression phenotype are inherently the result of altered development. However, the underlying developmental components of sex expression plasticity and of differences in plasticity among genotypes are unknown. This study takes an explicitly genetic and developmental approach to the study of phenotypic plasticity and examines changes in sex expression of ten clonally replicated genotypes at three levels of organization: among inflorescences, within inflorescences, and at the level of developing floral meristems. Changes in sex expression of individuals and differences among individuals are the result of a predictable interplay of resource, architectural, and floral level response within the hierarchical construction of the shoot system. Phenotypic plasticity of whole plant sex expression is ultimately due to sexual lability of individual developing flowers: floral sex is not determined until a primordium size of 9–10 mm. Until that time, sex expression remains labile and developing floral primordia can respond to changes in plant resource status. This flower level developmental lability, however, is expressed within the constraints set by the architecture and ontogenetic history of the organism. Only those floral primordia produced in distal portions of each inflorescence are labile, capable of developing into either a staminate or hermaphrodite flower, whereas those primordia in basal positions invariably develop as hermaphrodite flowers. The genotypes differ with respect to the architectural components of phenotypic plasticity and it is this architectural variation that results in differences in plasticity among genotypes. The phenomenon, in which the developmental fate of a primordium depends upon where and when it is produced within the architecture of an organism and what events have preceded it during ontogeny, can be termed “ontogenetic contingency.”     

The phenotypic variance within plastic traits under migration-mutation-selection balance

How phenotypic variances of quantitative traits are influenced by the heterogeneity in environment is an important problem in evolutionary biology. In this study, both genetic and environmental variances in a plastic trait under migration-mutation-stabilizing selection are investigated. For this, a linear reaction norm is used to approximate the mapping from genotype to phenotype, and a population of clonal inheritance is assumed to live in a habitat consisting of many patches in which environmental conditions vary among patches and generations. The life cycle is assumed to be selection-reproduction-mutation-migration. Analysis shows that phenotypic plasticity is adaptive if correlations between the optimal phenotype and environment have become established in both space and/or time, and it is thus possible to maintain environmental variance (V(E)) in the plastic trait. Under the special situation of no mutation but maximum migration such that separate patches form an effective single-site habitat, the genotype that maximizes the geometric mean fitness will come to fixation and thus genetic variance (V(G)) cannot be maintained. With mutation and/or restricted migration, V(G) can be maintained and it increases with mutation rate but decreases with migration rate; whereas VE is little affected by them. Temporal variation in environmental quality increases V(G) while its spatial variance decreases V(G). Variation in environmental conditions may decrease the environmental variance in the plastic trait.     

Coming to Grips with Evolvability

To explain the evolution of complex organisms by random mutation, drift, and selection is not a trivial task. This becomes obvious if we imagine an organism in which most genes affect most traits and all mutations are immediately expressed in the phenotype. Most of the mutations will be deleterious. Computer programmers experienced a similar problem when trying to evolve computer programs by introducing random changes to a conventional computer code, realizing that almost all random changes are “lethal.” Everyone who has done any programming knows that conventional computer languages are very brittle! Real organisms are not organized in this way but rather involve mediation between the genes and the phenotypic traits, namely development, also sometimes called the genotype–phenotype map. This map of genetic effects is structured in a way that enables evolvability, that is, enhances the probability that mutations will improve the performance of the organism. Here we outline two properties of organismal development, namely modularity and robustness. Modularity refers to the situation in which genes affect a restricted number of functionally related phenotypic characters. Robustness describes a situation in which cryptic mutations can accumulate without effect on fitness but can become visible to selection in a new environment or genetic background. We discuss recent empirical evidence in support of both phenomena and their effect on evolvability and also briefly address their evolution.     

Morphological variability of the Asiatic cyprinid, topmouth gudgeon Pseudorasbora parva, in its introduced European range

To assess the spatial variability in external morphology of non-native populations of topmouth gudgeon Pseudorasbora parva within an ontogenetic context, triple regression analysis (distance-based measurements) was applied to data from eight European populations (two Slovak, four Romanian, one English and one French). The data from Slovakia were also subjected to geometrical analysis (co-ordinates-based measurements) to obtain a more complex picture of the species' overall morphology. Great phenotypic variability was observed, being expressed not only in the formation of different definite phenotypes but also in the manner by which the phenotypes are achieved. Thus, both the definite phenotype and the patterns of development in invasive P. parva may be highly influenced by environmental conditions. Such great morphological (phenotypic) variability is likely to be one of the attributes that make this species such a successful invader.     

Evolution of robustness to noise and mutation in gene expression dynamics

Phenotype of biological systems needs to be robust against mutation in order to sustain themselves between generations. On the other hand, phenotype of an individual also needs to be robust against fluctuations of both internal and external origins that are encountered during growth and development. Is there a relationship between these two types of robustness, one during a single generation and the other during evolution? Could stochasticity in gene expression have any relevance to the evolution of these types of robustness? Robustness can be defined by the sharpness of the distribution of phenotype; the variance of phenotype distribution due to genetic variation gives a measure of 'genetic robustness', while that of isogenic individuals gives a measure of 'developmental robustness'. Through simulations of a simple stochastic gene expression network that undergoes mutation and selection, we show that in order for the network to acquire both types of robustness, the phenotypic variance induced by mutations must be smaller than that observed in an isogenic population. As the latter originates from noise in gene expression, this signifies that the genetic robustness evolves only when the noise strength in gene expression is larger than some threshold. In such a case, the two variances decrease throughout the evolutionary time course, indicating increase in robustness. The results reveal how noise that cells encounter during growth and development shapes networks' robustness to stochasticity in gene expression, which in turn shapes networks' robustness to mutation. The necessary condition for evolution of robustness, as well as the relationship between genetic and developmental robustness, is derived quantitatively through the variance of phenotypic fluctuations, which are directly measurable experimentally.     

Changing genetic paradigms: creating next-generation genetic databases as tools to understand the emerging complexities of genotype/phenotype relationships

Understanding genotype/phenotype relationships has become more complicated as increasing amounts of inter- and intra-tissue genetic heterogeneity have been revealed through next-generation sequencing and evidence showing that factors such as epigenetic modifications, non-coding RNAs and RNA editing can play an important role in determining phenotype. Such findings have challenged a number of classic genetic assumptions including (i) analysis of genomic sequence obtained from blood is an accurate reflection of the genotype responsible for phenotype expression in an individual; (ii) that significant genetic alterations will be found only in diseased individuals, in germline tissues in inherited diseases, or in specific diseased tissues in somatic diseases such as cancer; and (iii) that mutation rates in putative disease-associated genes solely determine disease phenotypes. With the breakdown of our traditional understanding of genotype to phenotype relationships, it is becoming increasingly apparent that new analytical tools will be required to determine the relationship between genotype and phenotypic expression. To this end, we are proposing that next-generation genetic database (NGDB) platforms be created that include new bioinformatics tools based on algorithms that can evaluate genetic heterogeneity, as well as powerful systems biology analysis tools to actively process and evaluate the vast amounts of both genomic and genomic-modifying information required to reveal the true relationships between genotype and phenotype.