Epistasis among Deleterious Mutations in the HIV-1 Protease

ΦX174 Attenuation by Whole-Genome Codon Deoptimization

Natural selection acting on synonymous mutations in protein-coding genes influences genome composition and evolution. In viruses, introducing synonymous mutations in genes encoding structural proteins can drastically reduce viral growth, providing a means to generate potent, live-attenuated vaccine candidates. However, an improved understanding of what compositional features are under selection and how combinations of synonymous mutations affect viral growth is needed to predictably attenuate viruses and make them resistant to reversion. We systematically recoded all nonoverlapping genes of the bacteriophage ΦX174 with codons rarely used in its Escherichia coli host. The fitness of recombinant viruses decreases as additional deoptimizing mutations are made to the genome, although not always linearly, and not consistently across genes. Combining deoptimizing mutations may reduce viral fitness more or less than expected from the effect size of the constituent mutations and we point out difficulties in untangling correlated compositional features. We test our model by optimizing the same genes and find that the relationship between codon usage and fitness does not hold for optimization, suggesting that wild-type ΦX174 is at a fitness optimum. This work highlights the need to better understand how selection acts on patterns of synonymous codon usage across the genome and provides a convenient system to investigate the genetic determinants of virulence.

Predominance of positive epistasis among drug resistance-associated mutations in HIV-1 protease

Drug-resistant mutations often have deleterious impacts on replication fitness, posing a fitness cost that can only be overcome by compensatory mutations. However, the role of fitness cost in the evolution of drug resistance has often been overlooked in clinical studies or in vitro selection experiments, as these observations only capture the outcome of drug selection. In this study, we systematically profile the fitness landscape of resistance-associated sites in HIV-1 protease using deep mutational scanning. We construct a mutant library covering combinations of mutations at 11 sites in HIV-1 protease, all of which are associated with resistance to protease inhibitors in clinic. Using deep sequencing, we quantify the fitness of thousands of HIV-1 protease mutants after multiple cycles of replication in human T cells. Although the majority of resistance-associated mutations have deleterious effects on viral replication, we find that epistasis among resistance-associated mutations is predominantly positive. Furthermore, our fitness data are consistent with genetic interactions inferred directly from HIV sequence data of patients. Fitness valleys formed by strong positive epistasis reduce the likelihood of reversal of drug resistance mutations. Overall, our results support the view that strong compensatory effects are involved in the emergence of clinically observed resistance mutations and provide insights to understanding fitness barriers in the evolution and reversion of drug resistance.

Full-Length Envelope Analyzer (FLEA): A tool for longitudinal analysis of viral amplicons

Next generation sequencing of viral populations has advanced our understanding of viral population dynamics, the development of drug resistance, and escape from host immune responses. Many applications require complete gene sequences, which can be impossible to reconstruct from short reads. HIV env, the protein of interest for HIV vaccine studies, is exceptionally challenging for long-read sequencing and analysis due to its length, high substitution rate, and extensive indel variation. While long-read sequencing is attractive in this setting, the analysis of such data is not well handled by existing methods. To address this, we introduce FLEA (Full-Length Envelope Analyzer), which performs end-to-end analysis and visualization of long-read sequencing data. FLEA consists of both a pipeline (optionally run on a high-performance cluster), and a client-side web application that provides interactive results. The pipeline transforms FASTQ reads into high-quality consensus sequences (HQCSs) and uses them to build a codon-aware multiple sequence alignment. The resulting alignment is then used to infer phylogenies, selection pressure, and evolutionary dynamics. The web application provides publication-quality plots and interactive visualizations, including an annotated viral alignment browser, time series plots of evolutionary dynamics, visualizations of gene-wide selective pressures (such as dN/dS) across time and across protein structure, and a phylogenetic tree browser. We demonstrate how FLEA may be used to process Pacific Biosciences HIV env data and describe recent examples of its use. Simulations show how FLEA dramatically reduces the error rate of this sequencing platform, providing an accurate portrait of complex and variable HIV env populations. A public instance of FLEA is hosted at http://flea.datamonkey.org. The Python source code for the FLEA pipeline can be found at https://github.com/veg/flea-pipeline. The client-side application is available at https://github.com/veg/flea-web-app. A live demo of the P018 results can be found at http://flea.murrell.group/view/P018.

Viral populations constantly evolve and diversify. In this article we introduce a method, FLEA, for reconstructing and visualizing the details of evolutionary changes. FLEA specifically processes data from sequencing platforms that generate reads that are long, but error-prone. To study the evolutionary dynamics of entire genes during viral infection, data is collected via long-read sequencing at discrete time points, allowing us to understand how the virus changes over time. However, the experimental and sequencing process is imperfect, so the resulting data contain not only real evolutionary changes, but also mutations and other genetic artifacts caused by sequencing errors. Our method corrects most of these errors by combining thousands of erroneous sequences into a much smaller number of unique consensus sequences that represent biologically meaningful variation. The resulting high-quality sequences are used for further analysis, such as building an evolutionary tree that tracks and interprets the genetic changes in the viral population over time. FLEA is open source, and is freely available online.

Involvement of a joker mutation in a polymerase-independent lethal mutagenesis escape mechanism

We previously characterized a foot-and-mouth disease virus (FMDV) with three amino acid replacements in its polymerase (3D) that conferred resistance to the mutagenic nucleoside analogue ribavirin. Here we show that passage of this mutant in the presence of high ribavirin concentrations resulted in selection of viruses with the additional replacement I248T in 2C. This 2C substitution alone (even in the absence of replacements in 3D) increased FMDV fitness mainly in the presence of ribavirin, prevented an incorporation bias in favor of A and U associated with ribavirin mutagenesis, and conferred the ATPase activity of 2C decreased sensitivity to ribavirin-triphosphate. Since in previous studies we described that 2C with I248T was selected under different selective pressures, this replacement qualifies as a joker substitution in FMDV evolution. The results have identified a role of 2C in nucleotide incorporation, and have unveiled a new polymerase-independent mechanism of virus escape to lethal mutagenesis.

•

A replacement in FMDV protein 2C confers reduced sensitivity to the mutagen ribavirin.

•

The effect of the replacement is to prevent a mutational bias evoked by ribavirin.

•

2C has an effect in nucleotide incorporation by the FMDV polymerase.

•

We describe a new molecular mechanism of escape to ribavirin-mediated extinction.

Quantitative Description of a Protein Fitness Landscape Based on Molecular Features

Understanding the driving forces behind protein evolution requires the ability to correlate the molecular impact of mutations with organismal fitness. To address this issue, we employ here metallo-β-lactamases as a model system, which are Zn(II) dependent enzymes that mediate antibiotic resistance. We present a study of all the possible evolutionary pathways leading to a metallo-β-lactamase variant optimized by directed evolution. By studying the activity, stability and Zn(II) binding capabilities of all mutants in the preferred evolutionary pathways, we show that this local fitness landscape is strongly conditioned by epistatic interactions arising from the pleiotropic effect of mutations in the different molecular features of the enzyme. Activity and stability assays in purified enzymes do not provide explanatory power. Instead, measurement of these molecular features in an environment resembling the native one provides an accurate description of the observed antibiotic resistance profile. We report that optimization of Zn(II) binding abilities of metallo-β-lactamases during evolution is more critical than stabilization of the protein to enhance fitness. A global analysis of these parameters allows us to connect genotype with fitness based on quantitative biochemical and biophysical parameters.

Epistasis as a determinant of the HIV-1 protease's robustness to mutation

The robustness of phenotypes to mutation is critical to protein evolution; robustness may be an adaptive trait if it promotes evolution. We hypothesised that native proteins subjected to natural selection in vivo should be more robust than proteins generated in vitro in the absence of natural selection. We compared the mutational robustness of two human immunodeficiency virus type 1 (HIV-1) proteases with comparable catalytic efficiencies, one isolated from an infected individual and the second generated in vitro via random mutagenesis. Single mutations in the protease (82 and 60 in the wild-type and mutant backgrounds, respectively) were randomly generated in vitro and the catalytic efficiency of each mutant was determined. No differences were observed between these two protease variants when lethal, neutral, and deleterious mutations were compared (P = 0.8025, chi-squared test). Similarly, average catalytic efficiency (−72.6% and −64.5%, respectively) did not significantly differ between protease mutant libraries (P = 0.3414, Mann Whitney test). Overall, the two parental proteins displayed similar mutational robustness. Importantly, strong and widespread epistatic interactions were observed when the effect of the same mutation was compared in both proteases, suggesting that epistasis can be a key determinant of the robustness displayed by the in vitro generated protease.

An evolutionary perspective on protein moonlighting

Moonlighting proteins serve one or more novel functions in addition to their canonical roles. Moonlighting functions arise when an adventitious interaction between a protein and a new partner improves fitness of the organism. Selective pressure for improvement in the new function can result in two alternative outcomes. The gene encoding the newly bifunctional protein may duplicate and diverge so as to encode two proteins, each of which serves only one function. Alternatively, genetic changes that minimize adaptive conflict between the two functions and/or improve control over the time and place at which each function is served can lead to a moonlighting protein. Importantly, genetic changes that enhance a moonlighting function can occur in the gene encoding the moonlighting protein itself, in a gene that affects the structure of its new partner or in a gene encoding a transcription factor that controls expression of either partner. The evolutionary history of each moonlighting protein is complex, depending on the stochastic occurrence of genetic changes such as gene duplication and point mutations, and the effects of those changes on fitness. Population effects, particularly loss of promising individuals due to random genetic drift, also play a role in the emergence of a moonlighting protein. The ultimate outcome is not necessarily the 'optimal' solution to the problem of serving two functions, but may be 'good enough' so that fitness becomes limited by some other function.

Extensive recombination-induced disruption of genetic interactions is highly deleterious but can be partially reversed by small numbers of secondary recombination events

Although homologous recombination can potentially provide viruses with vastly more evolutionary options than are available through mutation alone, there are considerable limits on the adaptive potential of this important evolutionary process. Primary among these is the disruption of favorable coevolved genetic interactions that can occur following the transfer of foreign genetic material into a genome. Although the fitness costs of such disruptions can be severe, in some cases they can be rapidly recouped by either compensatory mutations or secondary recombination events. Here, we used a maize streak virus (MSV) experimental model to explore both the extremes of recombination-induced genetic disruption and the capacity of secondary recombination to adaptively reverse almost lethal recombination events. Starting with two naturally occurring parental viruses, we synthesized two of the most extreme conceivable MSV chimeras, each effectively carrying 182 recombination breakpoints and containing thorough reciprocal mixtures of parental polymorphisms. Although both chimeras were severely defective and apparently noninfectious, neither had individual movement-, encapsidation-, or replication-associated genome regions that were on their own "lethally recombinant." Surprisingly, mixed inoculations of the chimeras yielded symptomatic infections with viruses with secondary recombination events. These recombinants had only 2 to 6 breakpoints, had predominantly inherited the least defective of the chimeric parental genome fragments, and were obviously far more fit than their synthetic parents. It is clearly evident, therefore, that even when recombinationally disrupted virus genomes have extremely low fitness and there are no easily accessible routes to full recovery, small numbers of secondary recombination events can still yield tremendous fitness gains. Importance: Recombination between viruses can generate strains with enhanced pathological properties but also runs the risk of producing hybrid genomes with decreased fitness due to the disruption of favorable genetic interactions. Using two synthetic maize streak virus genome chimeras containing alternating genome segments derived from two natural viral strains, we examined both the fitness costs of extreme degrees of recombination (both chimeras had 182 recombination breakpoints) and the capacity of secondary recombination events to recoup these costs. After the severely defective chimeras were introduced together into a suitable host, viruses with between 1 and 3 secondary recombination events arose, which had greatly increased replication and infective capacities. This indicates that even in extreme cases where recombination-induced genetic disruptions are almost lethal, and 91 consecutive secondary recombination events would be required to reconstitute either one of the parental viruses, moderate degrees of fitness recovery can be achieved through relatively small numbers of secondary recombination events.

Strong epistatic interactions within a single protein

A central question in protein molecular evolution is whether an amino acid that occurs at a given site makes an independent contribution to fitness or whether its contribution depends on other amino acid sites in the protein sequence, a phenomenon known as intragenic epistasis. In the absence of intragenic epistasis, natural selection acts on a protein mutation independent of its genetic background, and the experimentally determined fitness for a mutation should be the same across all genetic backgrounds. We tested this hypothesis by using site-directed mutagenesis to introduce a well-characterized deleterious single amino acid substitution in 56 different hepatitis C virus NS3 protease variants. The catalytic efficiency of the mutated proteases was determined and compared with the corresponding wild-type protein. Fitness effects ranged from lethality to significantly beneficial. Although primarily deleterious and lethal effects were observed (41 and 5 out of 56 tested variants, respectively), deleterious effects ranged from near neutral (-26.7% reduction of fitness) to near lethal (-98.5%). Our findings demonstrate that the introduced amino acid substitution has different fitness effects in different protein variants and provide independent support for the relevant role of intragenic epistasis in protein evolution.

Resistance mutations outside the integrase coding region have an effect on human immunodeficiency virus replicative fitness but do not affect its susceptibility to integrase strand transfer inhibitors

Most studies describing phenotypic resistance to integrase strand transfer inhibitors have analyzed viruses carrying only patient-derived HIV-1 integrase genes (INT-recombinant viruses). However, to date, many of the patients on INSTI-based treatment regimes, such as raltegravir (RAL), elvitegravir (EVG), and dolutegravir (DTG) are infected with multidrug-resistant HIV-1 strains. Here we analyzed the effect of drug resistance mutations in Gag (p2/NCp7/p1/p6), protease (PR), reverse transcriptase (RT), and integrase (IN) coding regions on susceptibility to INSTIs and viral replicative fitness using a novel HIV-1 phenotyping assay. Initial characterization based on site-directed mutant INSTI-resistant viruses confirmed the effect of a series of INSTI mutations on reduced susceptibility to EVG and RAL and viral replicative fitness (0.6% to 99% relative to the HIV-1_NL4-3 control). Two sets of recombinant viruses containing a 3,428-bp gag-p2/NCp7/p1/p6/pol-PR/RT/IN (p2-INT) or a 1,088 bp integrase (INT) patient-derived fragment were constructed from plasma samples obtained from 27 virologic failure patients participating in a 48-week dose-ranging study of elvitegravir, GS-US-183-0105. A strong correlation was observed when susceptibility to EVG and RAL was assayed using p2-INT- vs. INT-recombinant viruses (Pearson coefficient correlation 0.869 and 0.918, P<0.0001 for EVG and RAL, respectively), demonstrating that mutations in the protease and RT have limited effect on susceptibility to these INSTIs. On the other hand, the replicative fitness of viruses harboring drug resistance mutations in PR, RT, and IN was generally impaired compared to viruses carrying only INSTI-resistance mutations. Thus, in the absence of drug pressure, drug resistance mutations in the PR and RT contribute to decrease the replicative fitness of the virus already impaired by mutations in the integrase. The use of recombinant viruses containing most or all HIV-1 regions targeted by antiretroviral drugs might be essential to understand the collective effect of epistatic interactions in multidrug-resistant viruses.

Canine hepacivirus NS3 serine protease can cleave the human adaptor proteins MAVS and TRIF

Canine hepacivirus (CHV) was recently identified in domestic dogs and horses. The finding that CHV is genetically the virus most closely related to hepatitis C virus (HCV) has raised the question of whether HCV might have evolved as the result of close contact between dogs and/or horses and humans. The aim of this study was to investigate whether the NS3/4A serine protease of CHV specifically cleaves human mitochondrial antiviral signaling protein (MAVS) and Toll-IL-1 receptor domain-containing adaptor inducing interferon-beta (TRIF). The proteolytic activity of CHV NS3/4A was evaluated using a bacteriophage lambda genetic screen. Human MAVS- and TRIF-specific cleavage sites were engineered into the lambda cI repressor. Upon infection of Escherichia coli cells coexpressing these repressors and a CHV NS3/4A construct, lambda phage replicated up to 2000-fold more efficiently than in cells expressing a CHV protease variant carrying the inactivating substitution S139A. Comparable results were obtained when several HCV NS3/4A constructs of genotype 1b were assayed. This indicates that CHV can disrupt the human innate antiviral defense signaling pathway and suggests a possible evolutionary relationship between CHV and HCV.

Magnitude and sign epistasis among deleterious mutations in a positive-sense plant RNA virus

How epistatic interactions between mutations determine the genetic architecture of fitness is of central importance in evolution. The study of epistasis is particularly interesting for RNA viruses because of their genomic compactness, lack of genetic redundancy, and apparent low complexity. Moreover, interactions between mutations in viral genomes determine traits such as resistance to antiviral drugs, virulence and host range. In this study we generated 53 Tobacco etch potyvirus genotypes carrying pairs of single-nucleotide substitutions and measured their separated and combined deleterious fitness effects. We found that up to 38% of pairs had significant epistasis for fitness, including both positive and negative deviations from the null hypothesis of multiplicative effects. Interestingly, the sign of epistasis was correlated with viral protein-protein interactions in a model network, being predominantly positive between linked pairs of proteins and negative between unlinked ones. Furthermore, 55% of significant interactions were cases of reciprocal sign epistasis (RSE), indicating that adaptive landscapes for RNA viruses maybe highly rugged. Finally, we found that the magnitude of epistasis correlated negatively with the average effect of mutations. Overall, our results are in good agreement to those previously reported for other viruses and further consolidate the view that positive epistasis is the norm for small and compact genomes that lack genetic robustness.

Viral quasispecies evolution

Evolution of RNA viruses occurs through disequilibria of collections of closely related mutant spectra or mutant clouds termed viral quasispecies. Here we review the origin of the quasispecies concept and some biological implications of quasispecies dynamics. Two main aspects are addressed: (i) mutant clouds as reservoirs of phenotypic variants for virus adaptability and (ii) the internal interactions that are established within mutant spectra that render a virus ensemble the unit of selection. The understanding of viruses as quasispecies has led to new antiviral designs, such as lethal mutagenesis, whose aim is to drive viruses toward low fitness values with limited chances of fitness recovery. The impact of quasispecies for three salient human pathogens, human immunodeficiency virus and the hepatitis B and C viruses, is reviewed, with emphasis on antiviral treatment strategies. Finally, extensions of quasispecies to nonviral systems are briefly mentioned to emphasize the broad applicability of quasispecies theory.

Toward a systems biology perspective on enzyme evolution

Large superfamilies of enzymes derived from a common progenitor have emerged by duplication and divergence of genes encoding metabolic enzymes. Division of the functions of early generalist enzymes enhanced catalytic power and control over metabolic fluxes. Later, novel enzymes evolved from inefficient secondary activities in specialized enzymes. Enzymes operate in the context of complex metabolic and regulatory networks. The potential for evolution of a new enzyme depends upon the collection of enzymes in a microbe, the topology of the metabolic network, the environmental conditions, and the net effect of trade-offs between the original and novel activities of the enzyme.

Predictability of evolutionary trajectories in fitness landscapes

Experimental studies on enzyme evolution show that only a small fraction of all possible mutation trajectories are accessible to evolution. However, these experiments deal with individual enzymes and explore a tiny part of the fitness landscape. We report an exhaustive analysis of fitness landscapes constructed with an off-lattice model of protein folding where fitness is equated with robustness to misfolding. This model mimics the essential features of the interactions between amino acids, is consistent with the key paradigms of protein folding and reproduces the universal distribution of evolutionary rates among orthologous proteins. We introduce mean path divergence as a quantitative measure of the degree to which the starting and ending points determine the path of evolution in fitness landscapes. Global measures of landscape roughness are good predictors of path divergence in all studied landscapes: the mean path divergence is greater in smooth landscapes than in rough ones. The model-derived and experimental landscapes are significantly smoother than random landscapes and resemble additive landscapes perturbed with moderate amounts of noise; thus, these landscapes are substantially robust to mutation. The model landscapes show a deficit of suboptimal peaks even compared with noisy additive landscapes with similar overall roughness. We suggest that smoothness and the substantial deficit of peaks in the fitness landscapes of protein evolution are fundamental consequences of the physics of protein folding.

Is evolution deterministic, hence predictable, or stochastic, that is unpredictable? What would happen if one could “replay the tape of evolution”: will the outcomes of evolution be completely different or is evolution so constrained that history will be repeated? Arguably, these questions are among the most intriguing and most difficult in evolutionary biology. In other words, the predictability of evolution depends on the fraction of the trajectories on fitness landscapes that are accessible for evolutionary exploration. Because direct experimental investigation of fitness landscapes is technically challenging, the available studies only explore a minuscule portion of the landscape for individual enzymes. We therefore sought to investigate the topography of fitness landscapes within the framework of a previously developed model of protein folding and evolution where fitness is equated with robustness to misfolding. We show that model-derived and experimental landscapes are significantly smoother than random landscapes and resemble moderately perturbed additive landscapes; thus, these landscapes are substantially robust to mutation. The model landscapes show a deficit of suboptimal peaks even compared with noisy additive landscapes with similar overall roughness. Thus, the smoothness and substantial deficit of peaks in fitness landscapes of protein evolution could be fundamental consequences of the physics of protein folding.

Fitness ranking of individual mutants drives patterns of epistatic interactions in HIV-1

Fitness interactions between mutations, referred to as epistasis, can strongly impact evolution. For RNA viruses and retroviruses with their high mutation rates, epistasis may be particularly important to overcome fitness losses due to the accumulation of deleterious mutations and thus could influence the frequency of mutants in a viral population. As human immunodeficiency virus type 1 (HIV-1) resistance to azidothymidine (AZT) requires selection of sequential mutations, it is a good system to study the impact of epistasis. Here we present a thorough analysis of a classical AZT-resistance pathway (the 41–215 cluster) of HIV-1 variants by fitness measurements in single round infection assays covering physiological drug concentrations ex vivo. The sign and value of epistasis varied and did not predict the epistatic effect on the mutant frequency. This complex behavior is explained by the fitness ranking of the variants that strongly depends on environmental factors, i.e., the presence and absence of drugs and the host cells used. Although some interactions compensate fitness losses, the observed small effect on the relative mutant frequencies suggests that epistasis might be inefficient as a buffering mechanism for fitness losses in vivo. While the use of epistasis-based hypotheses to make general assumptions on the evolutionary dynamics of viral populations is appealing, our data caution their interpretation without further knowledge on the characteristics of the viral mutant spectrum under different environmental conditions.

Simple genomes, complex interactions: epistasis in RNA virus

Owed to their reduced size and low number of proteins encoded, RNA viruses and other subviral pathogens are often considered as being genetically too simple. However, this structural simplicity also creates the necessity for viral RNA sequences to encode for more than one protein and for proteins to carry out multiple functions, all together resulting in complex patterns of genetic interactions. In this work we will first review the experimental studies revealing that the architecture of viral genomes is dominated by antagonistic interactions among loci. Second, we will also review mathematical models and provide a description of computational tools for the study of RNA virus dynamics and evolution. As an application of these tools, we will finish this review article by analyzing a stochastic bit-string model of in silico virus replication. This model analyzes the interplay between epistasis and the mode of replication on determining the population load of deleterious mutations. The model suggests that, for a given mutation rate, the deleterious mutational load is always larger when epistasis is predominantly antagonistic than when synergism is the rule. However, the magnitude of this effect is larger if replication occurs geometrically than if it proceeds linearly.

Analyzing protein structure and function using ancestral gene reconstruction

Protein families with functionally diverse members can illuminate the structural determinants of protein function and the process by which protein structure and function evolve. To identify the key amino acid changes that differentiate one family member from another, most studies have taken a horizontal approach, swapping candidate residues between present-day family members. This approach has often been stymied, however, by the fact that shifts in function often require multiple interacting mutations; chimeric proteins are often nonfunctional, either because one lineage has amassed mutations that are incompatible with key residues that conferred a new function on other lineages, or because it lacks mutations required to support those key residues. These difficulties can be overcome by using a vertical strategy, which reconstructs ancestral genes and uses them as the appropriate background in which to study the effects of historical mutations on functional diversification. In this review, we discuss the advantages of the vertical strategy and highlight several exemplary studies that have used ancestral gene reconstruction to reveal the molecular underpinnings of protein structure, function, and evolution.