Introduction

Immune evasion by microorganisms has been, and continuesto be, an area of intense scientific interest. While promisingto provide crucial insights into the fundamental features ofthe microbe–host relationship, it is hoped by many that thestudy of immune evasion by microbial pathogens will alsolead to new therapeutic and vaccine strategies to combatinfections. Interest in this area has manifested as some excel-lent review series of this subject, two of which are theImmunology and Cell Biology Special Feature on ‘Immuneevasion by pathogens’ (1996; 74: 504–574) and the morerecent series in Current Opinion in Immunology, ‘Newlessons for immunology at the host–pathogen interface’(1999; 11: 363–426).

Prominent in the study of host–pathogen relationshipshave been immune evasion strategies by viruses, which areparticularly vulnerable to the cell’s arsenal of antimicrobialdefences postinfection. Persistent infections by virusesprovide a primary example of how successful evasion oferadication by the immune system has been achieved. Ideallyfor the virus, the physiological impact of its infection will not

compromise the health of the host, so while the diminution ofimmune/inflammatory effector molecule production is bene-ficial for the virus in situ, it also provides the virus with along-term host by reducing the risk of host pathology. If aparticularly successful strategy is developed by the virus, apersistent infection could become a ‘latent’ relationship withthe host (a latent infection is also persistent, but the activityof the infection is not sufficient to evoke an ongoing immuneresponse or apparent disease). A range of effector molecules,which have previously been theoretically proposed to havepromoted the evolution of immune evasion strategies inviruses,1 are produced by human cells to repel viral infection.These factors, for example, free radicals and pro-inflamma-tory cytokines,2 are toxic for both the virus and the host andviral immune evasion strategies appear to be aimed at exert-ing control over these to primarily ensure local cellular sur-vival for the virus, as well as providing the advantage of ahealthy host in which to inhabit.

The present paper proposes, from a theoretical perspec-tive, that the relationship between virus persistence anddisease for the host can be distinguished on the basis of the‘genetic budget’, that is, the genetic capacity for a virus toevolve immune evasion strategies that are most likely to leadto a long-term relationship with the host and not result inchronic or fatal virus-induced disease symptoms. Large DNAviruses that replicate in the cell nucleus have, through theirevolution, been able to take advantage of their DNA genomeand site of replication to ‘acquire’ host genes and hence

Immunology and Cell Biology (2001) 79, 62–66

Theoretical Article

Genetic ‘budget’ of viruses and the cost to the infected host: A theory on the relationship between the genetic capacity ofviruses, immune evasion, persistence and disease

TIMOTHY B CHASTON and BRETT A LIDBURY

Gadi Research Centre, Division of Science and Design, University of Canberra, Australian Capital Territory,Australia

Summary The nature of the pathogen–host relationship is recognized as being a dynamic coevolutionary processwhere the immune system has required ongoing adaptation and improvement to combat infection. Under survivalpressure from sophisticated immune responses, adaptive processes for microbes, including viruses, have manifestedas immune evasion strategies. This paper proposes a theory that virus immune evasion can be broadly classifiedinto ‘acquisition’ or ‘erroneous replication’ strategies. Acquisition strategies are characteristic of large genomedsDNA viruses, which (i) replicate in the cell nucleus; (ii) have acquired host genes that can be used to directlymanipulate responses to infection; (iii) are often latent for the lifetime of the host; and (iv) have little or no seriousimpact on health. Alternatively, erroneous replication strategies are characteristic of small genome RNA viruses,which are recognized as being the cause of many serious diseases in humans. It is proposed that this propensity fordisease is due to the cytoplasmic site of replication and truncated temporal relationship with the host, which haslimited or removed the evolutionary opportunity for RNA viruses to have acquired host genes. This has resulted inRNA viruses relying on error-prone replication strategies which, while allowing survival and persistence, are morelikely to lead to disease due to the lack of direct viral control over potentially host-deleterious inflammatory andimmune responses to infection.

Key words: disease, evolution, genome, immune evasion, persistence, virus.

Correspondence: Dr BA Lidbury, Gadi Research Centre, Divisionof Science and Design, University of Canberra, ACT 2601, Australia.Email: lidbury@scides.canberra.edu.au

Received 3 May 2000; accepted 27 July 2000.

mimic, block and/or otherwise regulate key cell processes,which either directly threatened virus survival (e.g. cytokine-mediated antiviral responses1) or enhanced the spread ofvirus from cell to cell. RNA viruses with a cytoplasmic siteof replication have been unable to directly access host genes(although recombination with genes from coinfecting virusescan occur, e.g. paraorthomyxoviruses) and therefore have hadto rely on cruder, ‘erroneous replication’ evasion strategieswhich, while allowing persistence, are much more likely to bedetrimental to the host. While large DNA viruses can causedisease, generally the disease impact on the otherwisehealthy host is less dramatic (with the exception of pox-viruses) than that found for the smaller genome size RNAviruses (Table 1). On this basis it can be posited that the lackof genetic budget for RNA viruses, and hence a lessenedcapacity to directly influence host cell responses, has lead toRNA viruses exacting an increased ‘cost’ from the host.

A summary of virus families, including information ontheir genetic capacity, site of replication and the impact onthe health of their host, is summarized in Table 1. As with anybiological system, exceptions exist, and examples of these arealso provided in Table 1. Further to such exceptions, it is alsoacknowledged that there are many examples of RNA virusesthat persist in their natural host without causing disease (for

example, Lassa virus and lymphocytic choriomeningitisvirus (LCMV) in rodents, feline immunodeficiency virus(FIV) in non-domestic cats) and that disease outcomes areoften a function of host adaptation by the virus, whichinvolves many mechanisms additional to those discussedhere. An obvious mechanism of such adaptation is the lengthof the virus–host relationship; while this is considered in thecontext of this theory, it is discussed as a factor that haslimited the opportunity for many contemporary RNA virusesto have evolved the genetic capacity necessary to have a lessdetrimental impact on host viability and health. Therefore,the focus of this theory is the interaction of contemporaryviruses with the contemporary immune system and, subse-quently, contemporary viral disease consequences.

To illustrate our theory of viral genetic capacity, immuneevasion and persistence, two viral pathogens of humans willbe primarily discussed: human immunodeficiency virus(HIV), an RNA virus with a small genome that is frequentlyassociated with human mortality,3 and herpes simplex virus(HSV), a large genome DNA virus that persists for the lifetime of the host and causes temporary local mucosal/skin inflammation (‘cold sores’) after reactivation fromlatency in the neuron by stimuli such as UV light and stress(Table 1).4

Viral genomes, immume evasion and disease 63

Table 1 Summary of virus families, their genome size and site of replication, as correlated with the worst possible health impact of theseviruses on the infected host

Virus family (genetic material) Example Genome size (kb) Site of replication Worst impact on the infected host*

Retroviridae (RNA) HIV 9.2 Cytoplasm and nucleus AIDS and deathFiloviridae (RNA) Ebola/Marburg 19.0 Cytoplasm Haemorrhagic fever and deathParamyxoviridae (RNA) Measles virus 15.5 Cytoplasm Fever, coryza and death in virgin

populationsFlaviviridae (RNA) Dengue 12.0 Cytoplasm DHF/DSS and deathAlphaviridae (RNA) Ross River virus 12.0 Cytoplasm EPA, myalgia and long-term

incapacitationAdenoviridae (dsDNA) Ad-2/Ad-5 36.0 Nucleus Acute respiratory disease,

pharyngitisHerpesviridae (dsDNA) HSV-1 152 Nucleus Cold sores

CMV 229 Nucleus Latent infection

ExceptionsPolyomavirinae (dsDNA) JC virus Nucleus Persistent infection of ke

BK virus 5.1SV 40

Orthomyxoviridae (RNA) Influenza A ~11.5 Nucleus Respiratory/systemic symptoms and death

Poxviridae (dsDNA) Orthopoxvirus ~200 Cytoplasm Fulminant poxviral disease and death(e.g. variola virus,ectromelia)

*Worst outcome for a proportion of normally healthy, immunocompetent and non-vaccinated individuals. The fundamental information listedabove on genome size, site of replication, infection and pathology were found in Field’s Virology (1996) chapters 30, 31, 39, 40, 43, 60, 63, 64,68, 71 and 83.20 Explanation of ‘exceptions’: Polyomaviruses, like simian virus 40 (SV 40), are dsDNA viruses that replicate in the host nucleusand rarely cause disease, but contrary to the theory proposed have a small genome size. Influenza virus is a negative strand RNA virus which,unlike other RNA viruses, replicates in the nucleus rather than the cytoplasm (a very low RNA to protein ratio suggests that a nuclear site ofreplication provides essential host proteins for replication). In agreement with the theory, however, influenza A can cause serious disease in theinfected host. Poxviruses, like variola, have large dsDNA genomes with evidence of many host gene homologues acquired during evolution;however, infection, rather than being persistent or latent, can result in acute fatal disease. In addition, poxvirus replication is in the cytoplasmrather than the nucleus as found for other dsDNA viruses like adenoviruses (Ad), herpes simplex virus (HSV)-1 and cytomegalovirus (CMV).DHF/DSS, dengue haemorrhagic fever/dengue shock syndrome; EPA, epidemic polyarthritis; Ad, adenovirus; ke, kidney epithelia.

Evolution of evasion strategies: A DNA virus example

The evolution of immune evasion strategies by viruses cangenerally be said to be dependent on the specific immunecharacteristics of the host cell and the duration of viral evo-lution.4 Herpes simplex virus is an example of a virus that isthought to have evolved over a long period in parallel with thevertebrate immune system. It is regarded as having used thisevolutionary time very successfully in actively developing arange of molecular strategies to survive the host response.Immune evasion strategies documented for alpha herpes-viruses, like HSV, include the mimicking by viral glyco-proteins of host IgG Fc receptors, inhibition of class I MHCpeptide presentation and perturbation of host complementregulation.5

These mechanisms require transcription and translation ofviral genes and act to directly perturb or adjust the cellularprocesses associated with resistance to infection, such as theblocking of complement pathways via mechanisms involvingHSV gC glycoprotein binding to host complement, and HSVgE and gI glycoproteins, which together bind the Fc domainof human anti-HSV IgG (phagocytosis is not affected).6–8

Major histocompatibility complex-mediated activation ofT cells is also inhibited by the perturbation of MHC I trans-port to the cell surface via the interruption to endoplasmicreticulum (ER) protein transport by the early HSV proteinICP47,4 which is a specific regulator of transporters associ-ated with antigen presentation (TAP) and may mimic hostregulation. Another HSV-mediated control of cell responsesto infection involves early proteins that inhibit apoptotic hostDNA degradation,9 conferring a great advantage to a virusthat integrates with the host cell nuclear DNA. This regula-tion of apoptotic pathways by HSV is, however, differential;the expression of apoptosis-associated membrane phos-phatidylserine (PS), which serves as an inducer of phago-cytosis in adjacent cells, is not affected and thus facilitatescell-to-cell spread by encouraging the surrounding non-infected phagocytic cells to engulf the neighbouring HSV-positive cells.9 Another member of the herpesvirus family,Epstein–Barr virus (EBV), provides a supreme example ofherpesviruses picking up a host cell gene during its evolution,in this case illustrated by the acquisition of the human IL-10homologue, BCRFI.10 BCRFI is produced by EBV-infectedcells to deter the potent antiviral activities of TNF and IFN-γ, as well as to encourage the proliferation of B cells in which the virus replicates.

Adenoviruses (Ad) share some fundamental virologicalproperties with HSV, such as that they are both double-stranded DNA viruses and multiply in the host cell nucleus.Both HSV and Ad have a lifelong relationship with theirhosts; Ad are capable of latent infection in the tonsils, ade-noids and intestine, while HSV manifest as a recrudescentinfection of nerve ganglia. This list of characteristics incommon between Ad and HSV extends to mechanisms ofimmune evasion, which are based on direct intervention byvirus-encoded molecular mimics that have been accumulatedby the viruses over their long evolution with the host. Exam-ples for Ad include the early proteins that act in concert tocontrol TNF-sensitive pathways and class I MHC maturation,as well as activity mediated by the Ad-encoded factor E1B19K, which has anti-apoptotic effects identical to the cellular

proto-oncogene bcl-2 (for reviews see Farrell and Davis-Poynter11 and Hayder and Mullbacher12).

From an evolutionary perspective, it is important to notethe sequence homology between viral evasion factors andhost proteins, because this have been found for several HSVand Ad proteins;7,11,12 such molecular mimicry has beenessential to the identification of human immune and regula-tory genes that are of importance to virus survival.13 The evolution of HSV and Ad immune evasion strategies hasoccurred predominantly through the inheritance of hostgenes,4,11,12 which emphasizes the need for a long relationshipbetween the virus and the host to accumulate host factors thataid survival and hence allow these viruses to construct largegenomes. One difference between HSV and Ad is the com-paratively small size of the Ad dsDNA genome. It could beargued that this is the result of either a shorter relationshipwith its host, where it has not had the evolutionary time toaccumulate as many additional host/virus genes as HSV, oralternatively its relationship has been much longer, with evo-lutionary time available for Ad to discard superfluous genes.The other possibility is that Ad is extremely effective at targeting beneficial host genes and a large genome did notneed to be accumulated.

Evolution of evasion strategies: An RNA virus example

The retrovirus HIV offers a contrasting view of viral immuneevasion, because it mediates this by more primitive mecha-nisms that have not involved the acquisition of host genes.The HIV strategies involve the production of escape variants,inhibition of interferon-sensitive pathways and comparativelylimited MHC downregulation.11,14 The mechanisms responsi-ble for these immune avoidance activities are dependent onadaptive mutation mediated by virus-encoded reverse tran-scriptase (RT), which is prone to high error rates as well asbeing responsible for high recombination rates3, resulting in high levels of genetic variation between viral progeny. Thisgenetic variation randomly allows escape from cytotoxic T-cell recognition and subsequent evasion of immunologicalmemory.14 Rather than being deleterious to HIV, free radicalactivity has been shown to be a prominent host factor in theactivation of virus from the latent state and for increasedvirus replication.3,15 Reverse transcriptase therefore plays anintegral role in HIV replication and could not have evolvedspecifically as an evasion strategy. Instead, it is through the lack of RT error correction activities that this mode ofevasion has evolved and has indirectly, without HIV acquir-ing host genes, allowed this replicative mechanism to providean immune evasion advantage.

Interferon-induced pathways pose a substantial threat toHIV survival and replication via mechanisms involving, forexample, the activation of a latent endoribonuclease, whichdegrades both viral and host RNA; a dsRNA-dependentprotein kinase, which acts to disrupt cellular mRNA trans-lation on which the virus depends for de novo protein syn-thesis; and the induction of inducible nitric oxide synthase(iNOS).16 The mechanism of perturbation of type 1 IFN path-ways is similar for HIV and Ad, involving the transcription ofnon-coding virus-associated RNA (VA RNA) that bind to cel-lular protein kinases, thus blocking the subsequent activation(via host dsRNA) of eukaryotic protein synthesis initiation

TB Chaston and BA Lidbury64

factor (eIF-2) by leaving this key protein unphosphorylated.12

Blocking of the host activation of this IFN-associatedpathway allows the initiation of viral peptide production bythe cell to proceed unfettered.12 Acquisition of a host genedoes not provide a suitable explanation for the evolution ofthis strategy (for any virus), because it does not make sensethat the cell would carry an RNA sequence that competitivelybinds to its own protein kinases, potentially blocking the acti-vation of crucial antiviral pathways. The observation thatthese VA RNA are non-coding sequences also strongly sug-gests that this immune evasion tactic has not been acquiredfrom the host; for HIV it is therefore likely that, as for otherevasion strategies, this mechanism involves the nuances ofRT function.

Direct manipulation of host immunity by HIV appears todepend in large part on the proteins produced after mutationsin HIV-1 nef and vpu genes.11 These proteins have beenshown to reduce the levels of MHC I presentation by separatemechanisms; however, they are dependent on short-termmutation and are therefore not indicative of direct evolution-ary adjustment to the host (for review see Farrell and Davis-Poynter11). In addition, the presence of these proteins isdependent on remarkably high replication and mutationrates,3,11,17 so their effectiveness as direct evasion strategies isnot consistent.

Measles virus (MV) provides another example of a smallssRNA virus that is responsible for human disease, involvingpathology of the neuronal tissue.18 Measles virus is a negative-strand RNA virus and hence does not require the activity oferror-prone RT for replication, but has developed a strategybased on accessing cells that are susceptible to infection. Thebasis of this vulnerability to MV infection is to do with neuronal dsRNA-dependent protein kinase (PKR) not beingactivated by MV RNA, resulting in the virus-mediated block-age of neuronal NF-κB binding to the IFN-β promoter, whichin turn is required for the induction of MHC I. Τhis allowsMV escape from virus-specific T-cell activation and couldcontribute to persistence in neuronal cells.18 Due to the lackof alternative NF-κB activating pathways in neuronal cells,this evasion strategy provides MV with a cellular refuge fromthe hostile immune response. This mechanism is dependenton a negative-strand RNA replicating system and is not theresult of genetic communication with the host or RT function,but is associated with cell selection-based camouflage fromthe immune system. A link between MV persistence and thedownregulation of the cell receptor CD46, which sensitizescells to complement-mediated lysis, has also been observed.19

‘Acquisition’ versus ‘erroneous replication’ evasionstrategies are a reflection of differences in geneticbudget

The production of escape variants via high replication ratesand replicative volatility in the RNA genome can be consid-ered, in the context of this theory, an erroneous replicationstrategy. Via the production of large numbers of progeny,combined with the selective pressure of a sophisticatedimmune response to viruses, an enhanced probability of pro-ducing escape mechanisms tailored to the pressures of hostantiviral activities (as described briefly earlier) has allowedsuch RNA viruses to survive and persist by exploiting many

cellular pathways. In contrast, ‘acquisition’ evasion strategiesare associated with DNA viruses that integrate with the hostnucleus and have actively targeted and acquired key hostantiviral and regulatory genes through a long evolutionaryrelationship, which can then be used via molecular mimicrymechanisms to adjust host cell responses to support the long-term needs of the virus. These opposing strategies are eitherdependent on adaptation in the short term by random muta-tion or the acquisition of host homologues over a very longperiod.4,11 In this context, this theory emphasizes that dis-tinctions between the acquisition and erroneous replicationstrategies also extend to the likelihood of a disease outcomefor the host. The infection style of small RNA viruses, withtheir reliance on crude evasion mechanisms as well as theirdiminished ability to control cellular responses, is morelikely to lead to uncontrolled inflammatory and immuneresponses which, while damaging for the virus, are also dele-terious to the general health of the infected host. In compar-ison, large ‘genetic budget’ DNA viruses are less likely tocause serious disease through their acquired abilities todirectly control host inflammatory and immune activities atthe cellular level.

Conclusion

Viruses can be broadly classified as relying on either acqui-sition or erroneous replication immune evasion strategies and both strategies can result in persistent infection of thehost. Based on the theory proposed herein, these strategiesare linked to the genetic ‘budget’ of the virus and, as anextension of this, to the subsequent ‘cost’ of infection to thehost (Table 1). Acquisition strategies are mediated by virus-encoded gene homologues of key host genes acquiredthrough lengthy evolutionary relationships of DNA virusesthat replicate in the nucleus, while erroneous replicationstrategies are mediated by extra- and intracellular processesthat are characteristic of replicative conditions specific to thevirus and are associated with RNA viruses that replicate inthe cytoplasm. Viruses such as Ad and HSV have the luxuryof genetic capacity and through the use of active immune/inflammatory evasion strategies have reduced their impact onthe infected host. Smaller RNA viruses, such as HIV and MV,have the disadvantage of comparative evolutionary youth andreplicative location and, therefore, have had an associatedtemporal and intracellular inability to actively access hostgenes. This has forced these viruses to use primitive escapemechanisms that may allow persistent infection, but signifi-cantly enhance the risk of inducing disease in their host.

References

1 Lidbury BA. Was exposure to directly antiviral cytokines duringprimary infection an important selective pressure in the evolu-tion of unique immune evasion strategies by viruses? Immunol.Cell Biol. 1994; 72: 347–50.

2 Rosen GM, Pou S, Ramos CL, Cohen MS, Britigan BE. Freeradicals and phagocytic cells. FASEB J. 1995; 9: 200–9.

3 Sandstrom PA, Murray J, Folks TM, Diamond AM. Antioxidantdefenses influence HIV-1 replication and associated cytopathiceffects. Free Radic. Biol. Med. 1998; 24: 1485–91.

Viral genomes, immume evasion and disease 65

TB Chaston and BA Lidbury66

4 Johnson DC, Hill AB. Herpesvirus evasion of the immunesystem. Curr. Top. Microbiol. Immunol. 1998; 232: 149–77.

5 Davis-Poynter NJ, Farrell HE. Masters of deception: A review ofherpes virus immune evasion strategies. Immunol. Cell Biol.1996; 74: 513–22.

6 Nagashunmugam T, Lubinski J, Wang L et al. In vivo immuneevasion mediated by the herpes simplex virus type 1 immuno-globulin G Fc receptor. J. Virol. 1998; 72: 5351–9.

7 Lubinski JM, Wang L, Soulika AM et al. Herpes simplex virustype 1 glycoprotein gC mediates immune evasion in vivo. J. Virol. 1998; 72: 8257–63.

8 Wurzner R. Evasion of pathogens by avoiding recognition oreradication by complement, in part via molecular mimicry. Mol. Immunol. 1999; 36: 249–60.

9 Jerome KR, Tait JF, Koelle DM, Corey L. Herpes simplex virustype 1 renders cells resistant to cytotoxic T-lymphocyte-inducedapoptosis. J. Virol. 1998; 72: 436–41.

10 Hsu DH, de Waal Malefyt R, Fiorentino DF et al. Expression ofinterleukin-10 activity by Epstein–Barr virus protein BCRF1.Science 1990; 250: 830–2.

11 Farrell HE, Davis-Poynter NJ. From sabotage to camouflage:Viral evasion of cytotoxic T lymphocyte and natural killer cell-mediated immunity. Cell. Dev. Biol. 1998; 9: 369–78.

12 Hayder H, Mullbacher A. Molecular basis of immune evasionstrategies by adenoviruses. Immunol. Cell Biol. 1996; 74:504–12.

13 Cohen JI. The biology of Epstein–Barr virus: Lessons learnedfrom the virus and the host. Curr. Opin. Immunol. 1999; 11:365–70.

14 Wilson CC, Brown RC, Korber BT et al. Frequent detection ofescape from cytotoxic T-lymphocyte recognition in perinatalhuman immunodeficiency virus (HIV) type 1 transmission ofHIV from mother to infant. J. Virol. 1999; 73: 3975–85.

15 Elbim C, Pillet S, Prevost MH et al. Redox and activation statusof monocytes from human immunodeficiency virus-infectedpatients: Relationship with viral load. J. Virol. 1999; 73: 4561–6.

16 Karupiah G, Xie Q, Buller RML, Nathan C, Duarte C,MacMicking JD. Inhibition of viral replication by interferon-γ-induced nitric oxide synthase. Science 1993; 261: 1445–8.

17 Borrow P, Shaw GM. Cytotoxic T-lymphocyte escape viral vari-ants: How important are they in viral evasion of immune clear-ance in vivo? Immunol. Rev. 1998; 164: 37–51.

18 Dhib-Jalbut S, Xia J, Rangaviggula H, Fang Y, Lee T. Failure ofmeasles virus to activate nuclear factor κb in neuronal cells:Implications on the immune response to viral infections in thecentral nervous system. J. Immunol. 1999; 162: 4024–9.

19 Yant S, Hirano A, Wong TC. Identification of a cytoplasmic Tyr-X-X-Leu motif essential for down regulation of the humancell receptor CD46 in persistent measles virus infection. J. Virol.1997; 71: 766–70.

20 Fields BN. Field’s Virology, 3rd edn. Philadelphia: Lippincott-Raven, 1996.