Faculty: Isabel Novella, PhD.

Summary: My goal is to use VSV to build a framework that will be useful to understand how evolution operates, and that can be used as a basis to learn how RNA viruses evolve. As in the past, we can learn the basics with VSV, and apply them to other important human pathogens. My program has two broad (and not necessarily independent) areas of research: • Understanding fundamental principles of evolutionary genetics. We ask general questions such as what is the fate of adaptive mutations, what is the cost of specialization, what is the role of epistatic interactions, what are the consequences of genetic bottlenecks or how do asexuals recover from fitness loss during random drift. • Understanding evolutionary theory as it applies to RNA viruses and its potential use for the design of antiviral strategies. How do RNA viruses escape from antiviral molecules? How can they revert from attenuated to virulent forms? Is gene rearrangement a safe strategy to develop vaccines? Can we design environments such that viruses cannot adapt to? Can we design environments such that adaptation is limited so disease can be avoided (even if infection cannot)?

I. Theoretical models and the evolution of RNA viruses

We have collaboration with Dr. Claus Wilke (University of Texas, Austin) to use an interdisciplinary approach that allows us testing evolutionary theory. We develop biologically meaningful mathematical models, and we then test the predictions of such models. Departures from expected results can help us modify the models to reflect the biological reality, and learn about the molecular biology of VSV infection. We have two major ongoing projects:

•

Complementation. Complementation can contribute to the survival of deleterious mutants by hiding their genomes from selection, which is particularly important to understand the maintenance of reservoirs of antibody-escape and drug-resistant mutants. Even though it has been acknowledged for many decades, its effect on viral evolution has generally being ignored. We are currently assessing the effect of complementation in single mutants that differ in different functions. We are also considering periodic changes in the degree of coinfection (and thus complementation), a regime that reflects infection in the real world. Complementation can explain density-dependent selection (Figure 2) and frequency-dependent selection in viral populations, and may contribute to memory in viral quasispecies [memory is deleterious mutants that reflect the past history of a population and resist extinction]. It can also explain the delayed fixation of beneficial mutations in large populations, and effect that has been interpreted as the result of clonal interference, although complementation is an alternative explanation that was not tested.

(Click Each Image for Larger View)
Figure 2. Effect of complementation on the survival of MARM N during competition with wt.
Left panel: MARM N fitness is a function of the moi, and the theoretical prediction (solid line) was conformed experimentally (symbols).		Right panel: The speed of decay of MARM N slows down and the frequency at mutation-selection balance increases with increasing moi. The stars indicate that we halted the experiment because two passages later we observed the emergence of defective interfering particles.

•

Experimental test of theoretical evolutionary models. There are two competing theories that describe fitness increases in viral populations. Clonal interference is based on the hypothesis that, in the absence of recombination, rare beneficial mutations compete with one another and cause a deceleration in the speed of adaptation. Solitary wave is a multi-site theoretical model, so a single genome can incorporate multiple mutations sequentially. These two models are mutually exclusive and predict opposite behavior, and we are carrying out extensive and careful tests of these predictions. At this point our results support solitary wave theory (Figure 3).

(Click Each Image for Larger View)
Figure 3. Computer simulations (lines) and experimental test (symbols) of solitary wave theory for VSV populations. In both panels the gray areas represent one standard deviation around the mean value predicted by theory.
Left panel: Speed of adaptation of VSV populations passaged at different population sizes. Different symbols indicate different passage methods.		Right panel: VSV’s approach to mutation-selection balance during large population passages. The dashed line shows the prediction of the model if mutation is turned off, and shows that adaptation during the first 20 passages occurs primarily by amplification of preexisting beneficial variation.

II. Experimental testing of evolutionary and ecological theory

We are continuing the work on the effect of alternating environments in the evolution of biological systems. Ecological theory makes a number of assumptions when building models and we have been testing some of these assumptions. We have shown that viruses can adapt simultaneously to different environments even when there are fitness costs involved and that there is no correlation between the complexity of the environment and the generation of specialists or generalists. We are looking at potential correlation between phenotypic and genotypic variation and the history or selection or the degree of specialization (or generalization). We have currently a collaboration with Dr. Laura Kramer to test the effect of host switch in vivo.

III. Exploration of sequence space and fitness landscapes

Our sequencing efforts has rendered fascinating results consistent with very rugged fitness landscapes and very limited neutral sequence space. These are particularly important to understand models of viral evolution, and the latter is at the heart of the Darwinism vs. neutralism argument. I personally think that deep down they are both pretty much the same, minus details that need be considered at a more restricted level. Since this is such a basic issue of population genetics, I am very excited to be able to test experimentally these ideas. Three illustrative examples of the types of experiments we are doing are:

•	Testing potential fitness contributions of silent mutations. We have growing evidence that many silent mutations are not neutral. We are using site-directed mutagenesis to analyze silent point mutations found under selective regimes so we can quantitate the extent to which silent mutations can contribute to adaptation.
•	Testing epistatic interactions among mutations. Impaired adaptability of strains with a history of genetic bottlenecked strains suggests extensive epistatic interactions within the VSV genome. Further support came from our sequencing results, which showed that mutations typically found in non-bottlenecked strains during selection, do not appear in bottlenecked strains under similar regimes. We will use genetic backgrounds of strains with a history of bottleneck to test the adaptive value of mutations that are beneficial in non-bottlenecked strains (and vice versa). We can carry out this study in detail and map the low adaptability phenotype in some of the bottlenecked strains that showed only 2-4 nt changes compared to the progenitor wt.
•	Testing other genetic determinants of virus evolution. We have shown that relaxed selection leads to the accumulation of mutations that have an effect on the ability of the virus to adapt (adaptability) and to withstand the harmful effects of mutation (robustness). We are currently mapping mutations involved in both traits.

IV. Applying evolutionary biology to human health

We are applying the knowledge acquired during our studies on ecological theory to select oncolytic viral strains with the potential to kill cancer cells efficiently while sparing healthy cells. Our studies on viral adaptability and robustness are being applied to the generation of live-attenuated vaccine vectors that are more resistant to environmental stress and that have a lower potential for reversion (e.g. with increased safety).

V. Development of a system to study experimental evolution of Dengue virus

Dengue virus (Figure 4) is the most significant arboviral infection worldwide and it is considered an emerging virus. While we expect that work with Dengue will be more fastidious that work with VSV, its significance as a human pathogen makes it an interesting system. We are currently obtaining and developing the tools to carry out experimental evolution in the future. This work includes the isolation of genetically marked mutants and the optimization of fitness assays, which we’ll test using plaque assays, immunofluorescence and real time RT-PCR.

Isabel Novella received her Masters and Ph.D. degrees at the University of Oviedo, Spain, working on the physiology and genetics of Streptomyces and other actinomycetes. She did postdoctoral work in Foot-and-Mouth Disease virus evolution with Dr. David Andreu (University of Barcelona, Spain) and Dr. Esteban Domingo (Center for Molecular Biology, Madrid, Spain). Later she joined the group of Dr. John J. Holland at UCSD and focused on evolution of RNA virus populations. Isabel joined the Department of Microbiology and Immunology in September 1998.
Current Grants:

•	NIH (NIAID) - Determinants of RNA Virus Evolution

Connections:

•	Claus Wilke
•	Laura Kramer