Abstract: Summary WereviewmathematicalmodelsofHIVdynamics,disease progression,andtherapy.Westartbyintroducingabasic modelofvirusinfectionanddemonstratehowitwasused tostudyHIVdynamicsandtomeasurecrucialparameters that lead to a new understanding of the disease process. We discuss the diversity threshold model as an example of the general principle that virus evolution can drive dis- ease progression and the destruction of the immune system. Finally, we show how mathematical models can be used to understand correlates of long-term immuno- logical control of HIV, and to design therapy regimes that convert a progressing patient into a state of long-term non-progression. BioEssays 24:1178-1187, 2002. 2002 Wiley Periodicals, Inc. evolve. The infection remains asymptomatic for years before the virus load sufficiently increases and the population of CD4 T cells falls to low levels leading to the development of AIDS. Disease progression is associated with the evolution of specific viral variants that are more virulent and pathogenic (e.g. evolution of strong T cell tropism, escape from immune responses, faster viral replication, and higher degrees of cytopathicity). Anti-retroviral drug therapy has successfully been used to significantly suppress viral replication and to delay disease progression in many patients. Currently, these drugs actbytwo mechanisms:reverse transcriptaseinhibitors interfere with the process of reverse transcription and prevent the virus from infecting a cell; protease inhibitors prevent the assembly of new infectious virus by an infected cell. Because HIV integrates into the host genome, however, the infected cells remain unaffected and provide a viral reservoir. While most productively infected cells have a relatively short life- span, many cells are latently infected and are very long lived. Thus, virus eradication by drug therapy is not possible during the life time of the host. Because continued administration of drugs is associated with many problems such as side effects and the generation of drug resistance, more recent research efforts have been directed at finding therapy regimes that boost HIV-specific immune responses. In this review, we show how mathematical models can be usedtounderstandthedynamicsofHIVinfectionandtherapy. We start by describing a basic model of virus infection and continue to show how it was used to get some crucial insights into the dynamics during the asymptomatic phase of the disease. We explore howHIV evolutioncan drive disease pro-

Abstract: ▪ Abstract Influenza A viruses contain genomes composed of eight separate segments of negative-sense RNA. Circulating human strains are notorious for their tendency to accumulate mutations from one year to the next and cause recurrent epidemics. However, the segmented nature of the genome also allows for the exchange of entire genes between different viral strains. The ability to manipulate influenza gene segments in various combinations in the laboratory has contributed to its being one of the best characterized viruses, and studies on influenza have provided key contributions toward the understanding of various aspects of virology in general. However, the genetic plasticity of influenza viruses also has serious potential implications regarding vaccine design, pathogenicity, and the capacity for novel viruses to emerge from natural reservoirs and cause global pandemics.

Abstract: The resistance of cancers to conventional therapies has inspired the search for novel strategies. One such approach, namely gene therapy, is based upon the introduction of genes such as those encoding suicide proteins, tumour suppressor proteins or cytokines into tumour cells by means of a genetic vector. The efficiency with which viruses transfer their genes from one host cell to another has led to the widespread use of viruses as genetic vectors. For safety reasons, such virus vectors are generally replication-defective but, unfortunately, this has limited the efficacy of treatment by restricting the number of cells to which the therapeutic gene is delivered. For this reason, the use of replication-competent viruses has been proposed, since virus replication would be expected to lead to amplification and spread of the therapeutic genes in vivo. The replication of many viruses results in lysis of the host cells. This inherent cytotoxicity, together with the efficiency with which viruses can spread from one cell to another, has inspired the notion that replication-competent viruses could be exploited for cancer treatment. Some viruses have been shown to replicate more efficiently in transformed cells but it is unlikely that such examples will exhibit a high enough degree of tumour selectivity, and hence safety, for the treatment of patients. Our increasing knowledge of the pathogenesis of virus disease and the ability to manipulate specific regions of viral genomes have allowed the construction of viruses that are attenuated in normal cells but retain their ability to lyse tumour cells. Such manipulations have included modifying the ability of viruses to bind to, or replicate in, particular cell types, while others have involved the construction of replication-competent viruses encoding suicide proteins or cytokines. Naturally occurring or genetically engineered oncolytic viruses based upon adenovirus, herpes simplex virus, Newcastle disease virus, poliovirus, vesicular stomatitis virus, weasles virus and reovirus have been described. The results of animal studies are encouraging and a number of viruses are now being evaluated in clinical trials.

Abstract: Coevolution of two coupled quasispecies is studied, motivated by the competition between viral evolution and adapting immune response. In this coadaptive model, besides the classical error catastrophe for high virus mutation rates, a second "adaptation" catastrophe occurs, when virus mutation rates are too small to escape immune attack. Maximizing both regimes of viral error catastrophes is a possible strategy for an optimal immune response, reducing the range of allowed viral mutation rates to a minimum. From this requirement, one obtains constraints on B-cell mutation rates and receptor lengths, yielding an estimate of somatic hypermutation rates in the germinal center in accordance with observation.

Abstract: Viruses must mutate to survive in the face of attack by their host's immune system. A new model suggests that the viral mutation rate is optimized in an evolutionary trade-off between adaptability and genomic integrity.

Abstract: In this review we discuss the application of theoretical frameworks to the interpretation of viral gene sequence data, with particular reference to the hepatitis C virus (HCV). The increasing availability of such data means that it is now possible (and necessary) to proceed from simple qualitative models of viral evolution, to more quantitative frameworks based on statistical inference, notably population genetics and molecular phylogenetics. We argue that these approaches are invaluable tools to the virologist and are essential for understanding the dynamics of viral infection and the outcome of therapeutic strategies. We use several recent HCV data-sets to illustrate the methods.

Abstract: Based on a recent model of evolving viruses competing with an adapting immune system [1], we study the conditions under which a viral quasispecies can maximize its growth rate. The range of mutation rates that allows viruses to thrive is limited from above due to genomic information deterioration, and from below by insufficient sequence diversity, which leads to a quick eradication of the virus by the immune system. The mutation rate that optimally balances these two requirements depends to first order on the ratio of the inverse of the virus' growth rate and the time the immune system needs to develop a specific answer to an antigen. We find that a virus is most viable if it generates exactly one mutation within the time it takes for the immune system to adapt to a new viral epitope. Experimental viral mutation rates, in particular for HIV (human immunodeficiency virus), seem to suggest that many viruses have achieved their optimal mutation rate. [1] C.Kamp and S. Bornholdt, Phys. Rev. Lett., 88, 068104 (2002)

Abstract: We analyzed the evolution of the human immunodeficiency virus type 1 (HIV-1) env gene in 12 chronically infected individuals who underwent structured treatment interruptions (STIs). Analyses of length variation and of clonal sequences demonstrated highly unpredictable evolution, which may limit the strengthening of HIV-specific immune responses by STIs because of the variability in exposure to viral antigens.

Abstract: Some basic properties of RNA viruses are their high mutation rate, their enormous population sizes and their short generation time. These properties allow RNA virus populations to quickly explore fitness landscapes. A great adaptability has been amply demonstrated in experimental, as well as in natural, populations of RNA viruses. However, at least from a theoretical point of view, a limit to the extent of viral adaptation may exist as a consequence of adaptive trade-offs arising during evolution in changing environmental conditions. Here, I review previously published results searching for such fitness trade-offs. The following scenario has been explored: the cost of host-range expansion, the cost of resistance to antiviral drugs, and the adaptation to different population densities. Despite the environmental conditions tested, results show a common pattern: whenever a virus adapt to a simple environmental situation it pays a cost in terms of adaptation to alternative situations. However, in those cases where the virus has been simultaneously adapted to different environmental conditions, this cost disappears or, at least, is greatly reduced. Finally, and as another factor imposing a limit to their speed of adaptation, I review results showing that clonal interference also plays an important role during viral evolution.

Abstract: In order to better appreciate issues regarding the design and the utility of lentiviral vectors, the lentiviral life cycle and, in particular, how the host cell cycle influences lentiviral replication, will be discussed. Since most of the events in lentiviral replication have been best characterized for the primate lentiviruses, including human immunodeficiency virus-1 (HIV-1), HIV-2 and simian immunodeficiency virus (SIV), the discussion will focus on these viruses. The primate lentiviruses contain ten open reading frames (Fig. 1). The gag open reading frame directs the synthesis of structural virion proteins and proteins which direct the encapsidation of genomic viral RNA. The pol open reading frame encodes the viral enzymes which are involved in synthesis of viral cDNA and which direct the integration of viral into cellular DNA. The env open reading frame encodes the structural envelope proteins which mediate attachment of the virion to the cell surface and fusion of viral with target cell membranes. Sequences within the long terminal repeat (LTR) regulate viral gene expression both at the transcriptional and post-transcriptional levels. The LTR contains cis acting regulatory sequences and sequences which mediate the binding of trans-acting viral regulatory proteins. Gag, pol, and env open reading frames are a basic characteristic of retroviral genomes including primate and non-primate lentiviruses as well as simple animal onco-retroviruses such as murine leukemia virus (Fig. 1). A number of additional small open reading frames distinguish the primate and non-primate lentiviruses from simple animal onco-retroviruses. The Tat and Rev proteins regulate lentiviral gene expression at the transcriptional and post-transcriptional levels respectively (CULLEN 1998; JEANG et al. 1999). Tat protein binds to a cis-acting element (TAR), located within the LTR, to up-regulate the activity of the promoter. Rev recognizes a cis-acting element (RRE) located in the central portion of the viral envelope gene to posttranscriptionally regulate viral gene expression. The remaining open reading frames encode what are referred to as the accessory or auxiliary proteins. These terms are somewhat of a misnomer since they imply that these proteins facilitate, but are not essential for, viral replication. However, the Vif protein, which is common to all lentiviruses, with the exception of EIAV, is essential for the replication of the primate lentiviruses (BORMAN et al. 1995; COURCOUL et al. 1995; HARMACHE et al. 1995; REDDY et al. 1995). Activities of these accessory proteins have been comprehensively reviewed elsewhere (TRONO 1995; EMERMAN and MALIM 1998). Although, with the exception of Vif, the accessory proteins appear dispensable for viral replication and pathogenicity in vivo (DESROSIERS et al. 1998) they are likely to contribute to some unique aspects of primate lentiviral biology. For example, and as will be discussed later, the Vpr/Vpx proteins may facilitate the entry of the primate lentiviruses into non-dividing cells by promoting nuclear uptake of the viral genome. Collectively, these unique proteins contribute to viral fitness in that they allow the virus to adapt, or to function within, inhospitable environments. As such, loss of any of these functions would impair the ability of the virus to compete with its wild-type counterpart, ultimately leading to loss of the viral variant from the virus population. This point will become more apparent as the overlapping functions of some of the viral proteins in promoting viral entry into non-dividing cells are discussed. What may be considered redundant features of the viral genome may, in a competitive setting, confer distinct fitness advantages which allow viruses bearing these apparent redundancies to predominate. Ultimately, viral evolution preserves the fittest viruses and the fact that primate lentiviruses have several determinants which may promote nuclear uptake of the viral genome underscores the essential contribution of these proteins to viral replication and persistence in the host.

Abstract: Each year a new flu vaccine is produced, and judging which strains to target is a tricky business. A new study evaluating viral evolution suggests a more systematic approach to predicting next year's virus.

Abstract: A survey of virus infection in garlic plants cultivated in Korea was conducted for three years. Most virus-infected garlic plants (Allium sativum) showed typical symptoms on the leaves such as yellow mosaic, stripes, and distortion. Through immunosorbent electron micro-scopy and RT-PCR analysis, the complex mixtures of viruses including garlic viruses of the genus Allerivirus, gaylic strain of Leek yellow stripe virus of the genus Potyvirus, and Garlic latent virus of the genus Carlavirus were identified in the virus-infected garlic plants. Among these viruses, Allexivirus was the most frequently detect-ed in the regions surveyed. Using sets of differential primers for Allexivirus genomes, two members of the genus were amplified and sequenced from the purified viruses. The deduced amino acid sequences for the coat proteins and the nucleic acid binding proteins of two viruses showed high homologies to Garlic virus A (CarV-A) and Garlic virus D (GarV-D) of Allekivirus. This is the first report of GarV-A and GarV-D in Korea. This suggests that Allexivirus in gavlic plants in Korea was mixed and varied. Phylogenetic analyses showed that the genus Allexivirus was diversi(ied by the processes of accumulation and evolution of viruses in garlic plants due to the long period of repeated vegetative propagation.

Abstract: Emergence of new viruses is dependent on the intrinsic and extrinsic constraints exerting on viral evolution. Intrinsic constraints are semantic and grammatical in nature. They are analysed here in reference to Hamming's spaces, driving to a new interpretation of the evolution of the quasispecies of Manfred Eigen. Extrinsic constraints are relevant to the fact that viral evolution is always a co-evolution story, with two or three partners implicated (the viruses, their hosts and eventually their vectors). They imply that viral phylogenies are disconnected, and viruses constitute a polyphyletic system. A possible consequence is that potential viral families are already known, or at least are present in nature, in such a manner that the frames for future viral evolution are already determined and that the probability for the emergence of a new frame is nil. Nevertheless, the emergence of new pathogens in the existing frames remain possible.

Abstract: Infection of rhesus macaques with chimeric simian-human immunodeficiency viruses (SHIV) is an established model to study acquired immunodeficiency syndrome (AIDS) pathogenesis Such a controlled system allows for detailedanalysis of the molecular determinants of viral pathogenesis in addition to studying host-specific immune responses that modulate disease progression Furthermore, the use of a pathogenic molecular clone affords the opportunity to study both viral evolution within a host and to examine the generation of tissue specific variants In this report we describe viral diversification within tissues of two rhesus macaques infected intravenously with the CXCR4-specific molecular clone SHIV S F 3 3 A 2 Heteroduplex tracking analysis (HTA) was used to determine the complexity of viral DNA within distinct lymphoid tissues Not surprising, heterogeneity of the proviral quasispecies in tissues obtained during the acute infection was limited However, tissues obtained at necropsy harbored a more diverse and often different population of env variants As the inoculating virus is a molecular clone, the variants generated are likely due to the presence of tissue specific selective forces rather than a founder's effect

Abstract: This chapter focuses on the variation in plant viruses and their origins and evolution. Like other living entities, viruses substantially resemble the parent during their replication but can change to give rise to new types or strains. This inherent variation enables viruses to adapt to new and changing situations and over longer periods of time results in new viruses. Strains can vary in the severity of disease they cause in the field and can mutate to break crop-plant resistance to a virus. A range of procedures is available for isolating virus variants either from nature or following some form of mutagenesis or other manipulation outside the plant. The molecular mechanisms by which variation within a virus population is produced are similar to those found in cellular organisms. These mechanisms include mutations involving single nucleotide changes or the addition or deletion of one or a few nucleotides; recombination, deletions, or additions of blocks of nucleotide sequences; rearrangement of nucleotide sequences; and re-assortment among multipartite genomes. There are numerous suggestions for the origins of viruses; however, three sources are usually considered. One is that viruses have descended from primitive precellular life forms, whereas the other one is that they arose from some cell constituent that escaped the normal control mechanism and became self-replicating entities. It is also believed that viruses are derived from degenerate cells that eventually parasitized normal cells. There is ample experimental evidence from the study of existing viruses to demonstrate that viruses continue to evolve, but the rates of evolution for most viruses and virus genes are very difficult to establish with any degree of certainty.

Abstract: Norwalk-like viruses(NLVs) that is one genus of the family Caliciviridae are major causative agents of nonbacterial acute gastroenteritis in human. NLVs have not yet been propagated in cell cultures and model animals, which restricts the fundamental studies. However, cDNA from several NLVs can be expressed in the cell-free system, bacterial cells, insect cells and mammalian cells. Studies on polyprotein processing by virus-encoded protease, enzymatic properties of RNA helicase, protease and RNA polymerase, capsid assembly, interaction between viral proteins or genomes and cellular proteins, the molecular mechanism of translation and transcription, and the crystal structure of the capsid protein and other viral proteins are in progress. Results will be useful for development of drugs for diarrheal therapy.

Abstract: Sir, Reverse transcriptase (RT) and protease (PRO) are the major target of antiretroviral compounds currently employed in HIV-1-infected individuals. Multiple-drug combinations with agents active against both enzymes show more benefits than monotherapy, as witnessed by a decrease in plasma HIV-1 RNA levels and an increase in CD4 cell counts. However, even following combination therapies with several drugs (nucleoside RT inhibitors, NRTI; non-nucleoside RT inhibitors, NNRTI; and protease inhibitors, PI), an incomplete viral suppression not infrequently arises with viruses showing a reduced susceptibility to more than one inhibitor in different classes. Deeks et al. recently reported the advantage of maintaining antiretrovirals even in the presence of resistance. A suggested scenario in patients who fail different therapeutic regimens requires phenotypic and genotypic monitoring of drug resistance in order to tailor antiretroviral therapy (www.hivatis.org). This approach could be successful or conversely, if a suboptimal regimen is chosen, select for viral strains with an improved replicative capacity in the presence of drugs, rendering them less susceptible to different regimens directed against both enzymes. We investigated nine HIV-1 isolates, at three different time-points for three patients (ZU, SA and CB). These viruses were isolated after patients presented a virological failure using antiretroviral regimens including a PI. First and second time-points were obtained in 1997 and 1999, and two out of three patients were subjected to a further treatment shift from an NNRTI to lamivudine in 2001, whereas the third patient was shifted to a PI-sparing regimen. Drug susceptibilities and pol gene sequences were determined as described previously. Viral isolates remained drug-susceptible to those compounds not included in their current regimen and exhibited an intermediate level of cross-resistance among PI. In each patient we detected a difference between the three time-points in RT and PRO genes (Table). GenBank accession numbers for RT: AY065954–AY065962; for PRO: AY154955 and AY065946–AY065953. The resistanceassociated mutations and drug pressure were critical variates for HIV-1 replication. In all patients, the replicative capacity of the isolate at the first and third time-points was higher in the presence of lamivudine and lower in the presence of an NNRTI. The opposite effect was detected at the second time-point. A dose–response profile was maintained with those drugs that were not experienced in vivo by our three patients, including RT and PIs, contrary to previously experienced compounds. We have shown a viral evolution in three heavily drugexperienced patients. The genotypic and phenotypic patterns of their resistant virus mirrored the therapeutic regimen used over time. In all three patients, the viral fitness as measured in their viral isolate was higher in the presence of resistant drugs and lower when the isolate was challenged in the presence of unexperienced compounds. Endorsing the observations by Deeks et al. we have underlined the risk of resistance accumulation in patients with suboptimal viral suppression who experience therapy changes over time. We believe this phenomenon should be considered by updated HIV-1 treatment guidelines.

Abstract: The selection in vivo of mutations in HIV-1 that allow viral escape from host HLA class I restricted cytotoxic T lymphocyte (CTL) responses has been documented in individuals. Recently we showed that the many HLA-A and -B allele-specific polymorphisms in HIV-1 reverse transcriptase (RT) suggested extensive CTLescape mutation at a population level. HLA class II restricted CD4 T helper responses have a central role in HIV-1 immunity and associations between HLA class II alleles and HIV-1 disease progression have been reported. However, CD4 T cell escape mutation in HIV-1 has not been proven. We sought to determine whether, as for CTLescape, CD4 T cell escape in HIV-1 was evident at a population level as HLA-DRB1 associated polymorphisms. We analysed the diversity of HIV-1 RT sequences in a large, well characterised cohort of HIV-1 infected patients over 2210 person-years of observation. We examined the relationship between the HLA-A, -B and -DRB1 alleles present in the cohort (as covariates) and polymorphism in HIV-1 RT (as the outcome) in multivariate logistic regression models. Each single residue in HIV-1 RT (positions 20-227) was examined in separate models sequentially. Power calculations and model selection steps were carried out to limit the number of comparisons, and final Bonferroni correction was made to mark out the most significant associations. Polymorphism in HIV-1 RT occurred at sites with least functional constraint against mutation. At these sites, we identified 64 characteristic polymorphisms associated with specific HLA-A or -B alleles (OR > 1, p 1, p < 0.05). Four of the 5 known T helper cell epitopes in HIV-1 RT encompassed sites of HLA-DRB1 allele-specific polymorphism found in our study. There were also 'negative associations' between polymorphism and common HLA alleles, suggesting that the HIV-1 consensus sequence could have been selected by the host population's dominant HLA. This novel, population-based approach shows that polymorphisms in HIV-1 sequence are characteristic for specific HLA class I and II alleles, in keeping with viral adaptation to both the CTLand CD4 T helper cell responses of a human host population. These findings suggest that HLA has a central role in HIV-1 primordial and contemporary evolution and has implications for epitope mapping and predicting HIV-1 dynamics in individuals.

Abstract: The study of RNA virus evolution has blossomed over the last 20 years. Despite the emergence of this new discipline, there has been little active debate over perhaps the most fundamental question of all. Do RNA viruses evolve in a manner that is qualitatively different from other life forms? For

Abstract: RNA viruses are rapidly emerging as extraordinarily promising agents for oncolytic virotherapy. Integral to the lifecycles of all RNA viruses is the formation of double-stranded RNA, which activates a spectrum of cellular defense mechanisms including the activation of PKR and the release of interferon. Tumors are frequently defective in their PKR signaling and interferon response pathways, and therefore provide a relatively permissive substrate for the propagation of RNA viruses. For most of the oncolytic RNA viruses currently under study, tumor specificity is either a natural characteristic of the virus, or a serendipitous consequence of adapting the virus to propagate in human tumor cell lines. Further refinement and optimization of these oncolytic agents can be achieved through virus engineering. This article provides a summary of the current status of oncolytic virotherapy efforts for seven different RNA viruses, namely, mumps, Newcastle disease virus, measles virus, vesicular stomatitis virus, influenza, reovirus, and poliovirus.

Abstract: The role of neutralizing antibodies (NAbs) during virus rebound in human immunodeficiency virus type 1 (HIV-1)-infected patients undergoing highly active antiretroviral therapy is poorly understood. Three patients in this study had NAbs to preexisting autologous HIV-1 and an episode of virus rebound after a prolonged period of virus suppression. To investigate the influence of NAbs on virus evolution, envelope genotypes of preexisting and rebound viruses were examined. Phylogenetic analysis of env (V1-V5) sequences indicated that rebound viruses had evolved from or preexisted in baseline populations. By use of envelope pseudotype viruses, rebound viruses were found to be significantly resistant to neutralization by autologous antibody in all 3 patients, indicating that rebound viruses were selected by NAbs. The site responsible for conferring neutralization resistance against autologous antibody was identified in the upstream C3 region in 2 of 3 patients.

Abstract: This chapter examines key features of the three main stages in virus evolution with examples drawn from several virus groups. It then reviews specific results with picornaviruses, as well as implications of high mutation rates and quasispecies dynamics for this important and diverse group of pathogens. With the levels of heterogeneity and viral load often seen in infected individuals, RNA virus populations include potentially all possible single mutants and decreasing amounts of multiple mutants. There are probably many routes to drug resistance, including amino acid replacements in the wall of the pocket preventing accommodation of the drug into the pocket (termed "exclusion" mutants), and replacements elsewhere in the capsid that affect viral uncoating. Resistant mutants display decreased affinity for the drug or increased affinity for the receptor, and often show low fitness values relative to their parental counterparts. Studies with picornaviral RNA-dependent RNA polymerases (replicases) face limitations derived from the difficulties in obtaining purified enzymes capable of sustaining multiple rounds of template-dependent copying. Viral quasispecies show features of complex adaptive systems such as mobilization of minority components (individual genomes from the mutant spectrum) in response to external stimuli. New developments in biochemistry and structural biology, and a deeper understanding of the principles governing viral evolution can now be combined to produce practical developments following an extensive (and necessary) accumulation of results from basic research.