- Email: [email protected]
Replicative homeostasis: a mechanism of viral persistence
Medical Hypotheses (2004) 63, 515–523
http://intl.elsevierhealth.com/journals/mehy
Replicative homeostasis: a mechanism of viral persistence Richard Sallie* St John of God Hospital, Western Gastroenterology, Suite 35, 95 Monash Avenue, Nedlands, Perth, WA 6009, Australia Received 3 January 2004; accepted 21 February 2004
Summary Acute viral infection is characterised by high-level replication before prompt decline of viraemia and, commonly, viral clearance. This kinetic pattern is generally held to be due to immune control. However, infection with some viruses, notably hepatitis C (HCV), hepatitis B (HBV) and the human immunodeficiency virus (HIV), often results in chronic stable low-level spontaneously fluctuating viraemia, kinetics that are difficult to rationalize on this basis. The persistence of HCV, an RNA virus, is especially problematic and its stability, occurring despite rapid, genomic mutation is highly paradoxical. This paper outlines the hypothesis, and evidence, that viruses autoregulate replication and mutation and describes a mechanism – replicative homeostasis – explaining viral stability. Replicative homeostasis results in stable, but reactive, replicative equilibria that drive quasispecies expansion and immune escape and explain all observed viral behaviours and host responses. This paradigm implies new approaches to antiviral therapy and is broadly relevant to modulation of gene expression. c 2004 Published by Elsevier Ltd.
Background
Problem 1: mutation rate
Viruses, including hepatitis C (HCV), hepatitis B (HBV) and the human immunodeficiency viruses (HIV) cause an enormous burden of disease and premature death worldwide. These viruses replicate, at least in part, by RNA polymerases, enzymes that lack fidelity or proofreading function [1]. During replication of hepatitis C (HCV) or (HIV) each new genome will differ from the parental template by up to ten nucleotides [2] because of non-faithful template copying by RNApol (~1 105 mutations/base RNA synthesised [2]).
The infidelity of RNA polymerases poses a major problem; The probability (q) of a mutation occurring at any point on the genome during one replication cycle is q ¼ ð1 ð1 Ml Þn ), where (Ml ) is the mutation rate and (n) the genome size. For hepatitis C (~9200 bases, an RNA virus lacking DNA “hardcopy”, mutating constantly at 105 substitutions/base, q 0:0879. However, for multiple replications cycles ðhÞ; q ¼ ð1 Mnl Þh . After 20 cycles (occurring in \5 days in most patients [3,4]), the probability of any original genome remaining unmutated is 7:5 1022 , meaning effective loss of genetic information. RNA viral quasispecies stability is a paradox [5]. This “theoretical impossibility” of RNA virus stability suggests either the
*
Tel.: +61-8-93897101; fax: +61-8-93897103. E-mail address: [email protected].
0306-9877/$ - see front matter c 2004 Published by Elsevier Ltd. doi:10.1016/j.mehy.2004.02.050
516 consistently reported rates of RNA polymerase infidelity are incorrect (which, even if true, would only delay quasispecies extinction; if Ml ¼ 1010 , q < 1040 within 100 days, etc.) or that innate viral mechanism(s) control mutation during replication. On the other hand, overly faithful template copying will result in an “oligoclonal” virus susceptible to immune destruction and unresponsive to ongoing evolution of cell receptors. The necessity to retain sequence information, and the later (in terms of replicase ) RNApol evolution) requirements of viruses to access cells and evade immune assault, placed constraints on replicase (RNApol ) function that dictate replicative fidelity must be tightly controlled.
Sallie patients prior to development of antibody, in the absence of any antibody, or any demonstrable Tcell response [8,9] and without lysis of infected cells [9] it is difficult to argue, with any conviction, that immune responses are primarily responsible. Furthermore, from a viral perspective, uncontrolled, exponential growth, as might be expected from the mode of replication, would result in rapid cell lysis, host death and a reduced likelihood of stable host-to-host transmission (as would inadequate replication) a prerequisite for viral survival on an evolutionary timescale. Therefore, viral replication must also be closely regulated.
Evidence for autoregulation Problem 2: replication rate Studies of HIV, hepatitis B (HBV) and HCV, and other viruses, demonstrate broadly similar kinetics; high level, primary phase, viral replication rapidly declines and gives way to relatively constant levels of viraemia [6,7], typically 2–3 logs lower than seen acutely, for prolonged periods (Fig. 1). These kinetics pose questions that are difficult to explain on the basis of “immune control” [6] or any other host response; For example; Why does “immune control” falter after [99% of viral load and antigenic diversity is cleared? And what prevents viral clearance once \1% of viral load and antigenic diversity remain? As down-regulation of viral replication frequently occurs in
Substantial clinical and laboratory evidence suggests viral replication is auto-regulated. During successful antiviral treatment levels of virus fall sharply [3,4,6,10,11], often becoming undetectable. However, viral replication rebounds to pretreatment levels on drug withdrawal in both patients [3,4,10] and in tissue culture [12], the latter indicating replication is controlled by factors independent of either cellular or humoral immune function. Auto-regulation of HCV replication was confirmed most emphatically in patients undergoing plasmapharesis in whom 60–90% reduction in levels of virus returned to baseline, but not beyond, within 3–6 h of plasma exchange [13].
Viral replication
Figure 1 Clinical viral kinetics. Acute phase replication of wild type replication (solid line, arbitrary concentration units) rapidly subsides to basal levels for prolonged periods with minor fluctuations. Mutant forms (dashed line) predominate long term. Modified from Daar et al. [6], and Coffin [7] (for HIV), but similar patterns seen with HCV and HBV infection.
Viruses replicate by copying antigenomic intermediate templates, hence replication generally obeys exponential growth kinetics, such that ½RNAt ¼ ½RNAðt1Þ ek , where ½RNAt is the viral concentration at time (t) and k is a growth constant. However, because of RNA pol infidelity, wild type virus will accumulate at ½RNAwt t ¼ ½RNAwt ðt1Þ ð1 qÞK1 and mutant forms at ½RNAmt t ¼ ½RNAwt ðt1Þ qK1 þ ½RNAmt ðt1Þ K1 where q is the probability of mutation and K1 ¼ ek . Therefore, while wild-type virus will predominate early, replication and intracellular accumulation of mutant forms will accelerate, meaning mutant HCV RNAs will rapidly predominate (Fig. 2(a)). It is likely viral genomes would withstand significant mutation (especially in the 3rd codon position, and in hypervariable regions) without significant disruption to function. However, mutations progressively accumulate in RNAviruses [14] and ultimately severely mutated RNAs and defective
Replicative homeostasis proteins, if translated, will become dominant. It is also likely some mutant viral proteins will resist cellular export, further accelerating relative ac-
Figure 2 Intracellular viral kinetics. Schematic representation of hypothesised intracellular events during viral replication. Exponential accumulation of both wildtype (solid line) and mutant viral proteins, with mutant forms (dashed line) rapidly increasing in concentration (arbitrary units) and predominating (a). Rapidly increasing [mutant]/[wildtype] viral protein ratios (solid line) results in exponential fall in replication function (dashed line, (b)) causing subsequent fall in viral replication (a). Initial high-level replication (dashed line, (c)) falls due to inhibition of polymerase by viral products, while polymerase fidelity (solid line) rises as the inverse function.
517 cumulation of defective forms. If mutant envelope proteins are even slightly inhibitory to RNApol function, or if wild-type envelope proteins are stimulatory, but are out competed for RNApol interactions by mutant forms that will be present in vast molar excess (½RNAmt =½RNAwt > 104–5 as clinical studies confirm [15]), then progressive inhibition of viral replication will result (Fig. 2(b)). Assuming HCV can tolerate 10–30% mutation of its genome (or about 900–3000 bases) without functional disruption, then accumulation of severely mutated proteins and inhibition of RNApol will become significant after ~90–300 cycles of replication. Most cellular enzymes are under some form of kinetic control, usually by product inhibition. While simple negative feedback is sufficient to control the velocity of enzyme reactions, and the rate of product accumulation, it is inadequate to ensure the functional quality of any complex molecules synthesised. The functionality of RNApol output is determined by the protein sequences translated from the RNAs RNApol synthesizes. For viruses, and their polymerase, survival (i.e. whether the polymerase, and its viral shell can avoid immune surveillance, access cells and subsequently replicate) depends on these properties i.e. the envelope proteins topological variability. Viruses intrinsically capable of adaptation to changes within host environments, including variations in host density and cell receptor polymorphisms, will enjoy a competitive advantage over viruses lacking innate responsiveness to environmental change. This adaptability requires that viral RNApol , function be both dynamically modifiable, and controllable, and that implies a nexus between the functional output of polymerases (i.e. the functionality of envelope proteins) and both the processivity and fidelity of that polymerase. RNA polymerase function is responsive to and influenced by accessory proteins that induce conformational changes to modulate both processivity and fidelity [16], representing “proof in principle” of part of the mechanism postulated – replicative homeostasis.
Proposed mechanism of homeostasis Replicative homeostasis results from differential interactions of wild type (wt) and mutant (mt) envelope proteins on RNApol in a series of feedback epicycles that link RNApol function, RNA replication and protein synthesis. The mechanism is outlined schematically (Fig. 3(a) and (b)). Intracellular accumulation of mutant viral proteins results in
518
Figure 3 Replicative homeostasis. (a) Overview of feedback epicycles; Relatively high concentrations of wild type proteins Envwt (blue, A) out-compete mutant forms (Envmt , red) for Env:RNApol interactions, resulting in increased output of mutant RNAs, with subsequent ribosomal (green) translation and increased concentrations mutant proteins (red), relative to wild-type, returning the state to basal equilibrium. Relative excess levels of mutant forms (B, red) out-compete wild type (blue) for interactions with polymerase, favouring the formation of Envmt :RNApol , and blocking Envwt :RNApol interactions. The Envmt :RNApol interactions relatively decrease RNApol processivity but increase fidelity, increasing output of wild-type RNAs, with subsequent increase in translation of wild-type proteins restoring the equilibrium state. (b) Schematic detail of feedback epicycles; envelope proteins Envwt (blue) and Envmt (red) interact with polymerase (Pol, green) to influence rate and fidelity of RNA synthesis changing templatate, and subsequent output, of ribosomes (R). Affinity of Envwt :RNApol (Kwt:pol ) greater than Envmt :RNApol (Kmt:pol ), velocity of reaction catalysed during wild-type/polymerase (Envwt :RNApol ) (mwt:pol ) greater than mutant/ polymerase reaction Envmt :RNApol (mmt:pol ), mutation rate associated with Envwt :RNApol interactions (Muwt:pol ) greater than mutation rate of Envmt :RNApol interactions. As ½Envmt ½Envwt equilibrium results such that ½Envmt Kmt:pol ¼ ½Envwt Kwt:pol or mmt Kmt:pol ¼ mwt Kwt:pol .
progressive inhibition of RNApol by blocking the stimulatory effects of wild-type envelope: RNApol interactions that increase both replication and mutation. Progressive blockade of RNApol by mu-
Sallie tant envelope results in a less processive polymerase with increased fidelity, increasing the relative output of wild-type envelope RNAs, and subsequent preferential translation of wild type envelope proteins, hence, inexorable progression to stable equilibrium. Quasispecies stability, and other consequences (including immune escape and low-level basal replication), are inevitable outcomes once equilibrium is reached as result of these interactions. We suggest that these interactions and the resulting equilibria are important therapeutic targets, and the necessary ligand-envelope proteins or topologically homologous molecules-implicit within this hypothesis. Viruses (notably HIV) produce many accessory proteins such as nef, gag, rev and the HBeAg that effect viral replication and mutation rate. While these proteins, encoded within envelope gene sequences and thus prone to alter functionally with any envelope mutations, may interact with RNApol to reset the level of any replicative equilibrium (by changing replication rate or mutation frequency or both), stable equilibria will still result providing the sum effect of wild type proteins is to increase polymerase processivity (v) and increase mutation (Mu ) frequency relative to mutant forms i.e. MuðwtÞ > MuðmtÞ and vwt > vmt .
Mathematical modelling Intuitive analysis suggesting enzymes acting in a milieu of increasing concentrations of inhibitory molecules become progressively less active until increasing polymerase fidelity increases output of wild type proteins to cause oscillating equilibria to form may be formalised mathematically; Wild type and mutant viral RNA replication dependent on intracellular competition between mutant and wild type envelope proteins (Env) influencing polymerase properties may be described by classical Lotka–Volterra linked (because RNApol function depends on relative wildtype/mutant RNA/protein concentrations and RNAmt derives from RNAwt ) differential “predator–prey” equations of the form dNwt =dt ¼ rNwtðn1Þ ðawt jwt ½Envwtðn1Þ =amt jmt ½Envmtðn1Þ Þ and dNmt =dt ¼ rNmtðn1Þ ðawt jwt ½Envwtðn1Þ = amt jmt ½Envmtðn1Þ Þ where r is the replication rate of viral RNA, a represents the stimulation (or inhibition) coefficient for RNApol :Env interactions, j the affinity constant for RNApol :Env interactions, and ½Envwt=mt the concentration of wild type or mutant envelope at time (t ) 1). For wild type RNAðwtÞ and mutant RNAðmtÞ RNA replication by RNApol subject to differential inhibition ewt , emt by RNApol :Envwt
Replicative homeostasis
519
and RNApol :Envmt end product interactions resulting in mutation probability, p, RNA synthesis may be approximated by a series of probability functions as follows: ½RNAwt ðtÞ ¼ ½RNAwt ðt1Þ rð1 qÞ
awt jwt ½Envwt ðt1Þ amt jmt ½Envmt ðt1Þ
½RNAmt ðtÞ ¼ ½RNAwt ðt1Þ rq
½RNAwt ðt1Þ r
ð1Þ
;
awt jwt ½Envwt ðt1Þ amt jmt ½Envmt ðt1Þ
awt jwt ½Envwt ðt1Þ amt jmt ½Envmt ðt1Þ
;
þ
ð2Þ
ð3Þ
where ½Envwt=mt represents concentrations of envelope proteins. Expression (2) describes (RNAmt ) derived through unfaithful replication of wild type (RNAwt ) template, while expression (3) mutant RNA derived from replication of previously mutated RNAmt templates. Simulating viral replication using a wide range of values for awt=mt and jwt=mt produces typical acute “viral” kinetics (Fig. 4) and demonstrates the initial down regulation of viral replication and long-term stable replication to be intrinsic properties of differential product inhibition of RNApol by [Ewt ] and [Emt ] altering replicative functions that occurs independently of external events.
Figure 4 Mathematical modelling. Graphic representation of simulated kinetics demonstrating initial down regulation of viral replication and long-term stable replication is an intrinsic property of differential influences of [Envwt ] and [Envmt ] on RNApol function and occurs independently of any external influences. Replication conditions: replicative ratio ðRm Þðmmt =mwt Þ ¼ 104 , mutation rate ðMmu Þ ¼ 1 105 , but these kinetics are seen over a wide range of values mutation and replication rates. Providing Muwt > Mumt and ewt > emt , stable equilibria will form.
Testing the hypothesis This hypothesis is simply tested. Addition of envelope viral proteins to a stable viral should change rates of viral replication and mutation, both easily measured. Specifically, addition of mutant envelope proteins or withdrawal of wild-type envelope proteins should reduce viral replication and rates of mutation while removal of mutant envelope proteins or addition of wild-type envelope proteins should increase viral replication and mutation. In fact, observations relevant to all of these suggested perturbations to viral equilibria have already been carried out in a variety of systems and circumstances (Table 1). All outcomes are completely consistent with those predicted by replicative homeostasis.
Discussion Replicative homeostasis is a mechanism that logically explains the initial decline of viral replication and viral kinetics observed during chronic infection. It immediately resolves the paradox of why RNA viral quasispecies are intrinsically stable and explains how these viruses persist and, thereby, cause disease. Replicative homeostasis also explains many unresolved viral behaviours and implies a general approach to antiviral therapy. Viral mutation, under replicative homeostasis, is not a random, passive process. Sustained immune recognition of viral envelope (Fig. 5) would change the replicative equilibrium to favour high affinity wild-type polymerase interactions that, in turn cause rapid replication of mutant forms, reactively driving quasispecies expansion and generating the extreme antigenic diversity seen in RNA quasispecies. Viruses enjoy a massive potential numerical superiority over immune responses; theoretically, a small envelope protein of, say, 20 amino acids could assume 2020 (about 1026 ) possible conformations, greatly in excess of the ~1010 antibody or CTL receptor conformations either humoral and cellular immune responses can generate. A direct consequence of this mismatch and the stable, but reactive, equilibria achieved by viruses because of replicative homeostasis is that once viral infection is established, the clinical outcome is primarily determined by the viruses’ ability to maintain homeostatic control of the quasispecies, rather than the hosts’ response to that quasispecies. Immune escape and maximal cell tropism are inevitable consequences of this process, and the potential antigenic diversity generated.
520
Sallie
Table 1 Outcomes after mutant/wild-type envelope ratios alter during stable viral equilibria Change to equilibrium
Expected outcome
Method
Virus/system
Addition of Wild-typeEnv
Increased replication
1. Cell culture co-transfection
HBV//Cell 1.Altered replication* culture [26] HIV//Cell culture* [27] HIV/Dengue// 1. Reduced replication Human [23] HIV/Measles// 2. Reduced infectivity Human [22] HBV// Human [28]
Increased mutation Addition of MutantEnv
Decreased replication Decreased mutation
Removal of Wild-typeEnv Removal of MutantEnv
Decreased replication Decreased mutation Increased replication
1. Co-infection [23,22] 2.Therapeutic vaccination [28] 3.Cell culture [27,29] 4. Primate vaccine [30]
HBsAb post OLT for HBV Heterologous vaccination
Increased mutation
HIV//Cell culture [27,29] * HIV/SIV// macaques [30] * Human therapeutics** HIV//Influenza Human vaccination [20]
Observed outcome
Reduces replication
Increased replication and mutation [20] Replication bursts quasispecies mutation [21]
HIV//tetanus Human vaccination [21] Examples of references only, more complete list supplied. Broadly, and in many studies, in HBV, HIV, simian (SIV) and feline (FIV) immunodeficiency virus systems. (a) chimeras expressing heterologous envelope genes with common polymerase backbone demonstrated replication characteristics identical to envelopeparental clones. (b) heterologous envelope proteins inhibit wild type viral replication. ** IV Anti-HBsAg routine during liver transplantation for HBV. *
Replicative homeostasis, and its failure, explains the varied outcomes of viral infection; viral failure to down-regulate replication by polymerase inhibition may result in rapidly progressive or fulminant disease (characterised by massively polyclonal, but ineffectual, immune responses), while inadequate viral replication or generation of diversity will result in viral clearance. Stable, homeostatic equilibria will result in chronic infection with episodic fluctuations in viral replication and host response (e.g. ALT), of the sort typical of HCV or HIV. The widely varied spectrum and tempo of viral diseases is more rationally explained on the basis of a spectrum of polymerase properties that result in variable, reactive, but stable replicative equilibria than highly variable and unpredictable (yet genetically homogeneous) immune responses. Homeostatic systems functioning without external perturbations – such as viruses replicating in systems devoid of replicative or immunologic challenge – rapidly progress to stasis. In tissue culture, viruses replicating in the absence of immune challenge to their replicative equilibria, are
unable (and do not need) to generate antigenic diversity by replicative homeostasis, a phenomenon probably responsible for attenuation of viral virulence. By contrast, in dilute serial viral passage, where high affinity wild-type envelope/ polymerase interactions predominate over lower affinity mutant envelope/polymerase interactions, increased viral replication and mutation would be predicted, and has been confirmed [17]. Some viruses (or some strains of viruses), such as HBV, may use a replicative strategy that involves high level replication triggered when opportunity for specific forms of transmission (e.g. vertical mother ) child) is possible but relative inactivity between times, while others such as herpes zoster have prolonged latencies. Replicative homeostasis with prolonged replication inhibition (i.e. dormancy) mediated by RNApol :Env interactions facilitates those strategies. Changing intracellular relative wild-type and mutant envelope concentrations will disturb RNApol :envelope interactions and alter replicative equilibria. Hence once extracellular/intracellular
Replicative homeostasis
Figure 5 Intracellular/extracellular dynamics of replicative homeostasis. Mechanism of antibody dependent enhancement (ADE) [18,19] and acceleration of replication seen during heterologous vaccination for influenza [18] or tetanus [30] (a) and of inhibition of viral replication during heterologous infection (e.g. during measles [21] or Dengue [22] co-infection) or administration of mutant envelope (b).
envelope concentrations equilibrate, extracellular antibodies will effect intracellular viral replication (Fig. 5). This disturbance of viral replicative equlibria rationally explains antibody-dependent enhancement (ADE) of HIV [18], Dengue, Murray Valley encephalitis [19], and other viruses. Increased HIV replication and mutation after influenza [20] and tetanus [21] vaccination; reduced HIV replication during measles [22] and Dengue [23] co-infection; clearance of HBV without hepatocyte lysis or evidence of T cell dependent cytotoxicity [9], are similarly explained by this mechanism. Previously unexplained and problematic viral behaviours and host responses, including long-term non-progression of HIV; HBV persistence despite a robust immune response; long-term antigenic oscillations [24]; spontaneous reactivation of HBV (among many other viruses); and hypermutation of HIV, for example, all rationally resolve within this conceptual framework. There are clear and quite specific therapeutic implications of replicative homeostasis, as well as more general implications. The putative envelope/ polymerase interactions suggested herein are obvious sites for drug targeting and suggest a site of interferon action. Interferon is not effective in HIV or in all patients with HBV, and its efficacy in HCV is
521 genotype-dependent, strongly implying interferon acts in a direct, virus-specific manner unrelated to “immune enhancement”, as both in-vitro data [25] and clinical kinetic studies imply [3]. Heterologous envelope proteins from other viruses or different genotypes of the same virus, or their structural homologues, would probably effect replication in a similar fashion to interferon, an hypothesis supported by suppression of HIV replication during measles [22] and Dengue [23] co-infection. Replicative homeostasis may alter perceptions of the strategies underpinning the immune response. It is possible the primary purpose of the initial heterogeneous polymorphic humoral response to viral infection – typically pentameric IgM – is to push viral replication towards equilibria favouring production of homogeneous wild-type virus, facilitating a concerted and more focussed humoral and/or cytotoxic T cell response; Strong neutralizing IgG antibodies – HBsAb, for example – may develop as a consequence of restricted viral replication and mutation that permits effective and specific immune recognition, rather than the proximate cause of it, but once developed, high affinity wild-type antibodies ensure mutant envelope proteins remain dominant within cells maximising polymerase inhibition. Replicative homeostasis is an adaptation that facilitates stable viral replication in cells and maximises probability of cell-to-cell (and host-tohost) transmission, a prerequisite for viral survival on an evolutionary time scale. A subtle, more primordial, function of envelope/polymerase complexing may explain how replicative homeostasis evolved; Polymerase function contingent upon recognition of and response to complex three-dimensional complementarities between polymerase and envelope proteins is a sophisticated encryption technique, effectively “locking” the polymerase and minimises the likelihood any competing RNA (or DNA) molecules are replicated. In the fierce competition for finite intracellular resources, reproductive strategies that maximise proliferation of “self” genes, while thwarting propagation of “rival” genes, are strongly selected for, and are highly conserved on an evolutionary timescale. The interferons, and other cytokines, are cellular defense mechanisms that long antedate the immune system. If the interferons are functionally homologous to viral envelope proteins, its possible acquisition of these genes resulted from positive selection of beneficial virus-cell symbiosis occurring early in eukaryotic cellular evolution, a process responsible for retention of other genes. Although proposed specifically to explain RNA viral quasispecies stability, replicative homeostasis
522 is, essentially, a mechanism that regulates RNA transcription and modulates protein expression. In addition to regulating phenotypic expression, replicative homeostasis, permits, and depends upon, proteins (phenotype) inducing, and subtly influencing, the degree of genetic mutation at an RNA level. If protein expression (i.e. phenotype) controls RNApol properties in a manner contingent on that proteins’ functionality and controllably influences the degree of mutation of RNA templates RNApol synthesises, a subtle form of “quality control” is exerted over protein synthesis. This mechanism also accelerates, and directs, adaptation; While introduction of lethal mutations to most RNA genomes may not adversely influence quasispecies stability to any degree, replicative homeostasis ensures any RNA mutations that do arise, and that are translated into a beneficial phenotype, will tend to be retained. Finally, accessory proteins also influence the fidelity and processivity of DNA-dependent RNA polymerases. Control of DNA-dependent RNApol transcription by DNA viruses, cellular micro-organisms (e.g. malaria), and eukaryotic cells, mediated subtly modulating protein expression to mediate immune escape, control cell division and differentiation, or other functions, would not be surprising.
Acknowledgements I thank Professors WD Reed, MG McCall, RA Joske, Bill Musk, Roger Williams, AE Jones, Steve Feinstone and Jay Hoofnagle for critical clinical and scientific guidance, Sophie J Coleman, Matt and Tim for all else.
References [1] Steinhauer DA, Domingo E, Holland JJ. Lack of evidence for proofreading mechanisms associated with an RNA virus polymerase. Gene 1992;122(2):281–8. [2] Preston BD, Poiesz BJ, Loeb LA. Fidelity of HIV-1 reverse transcriptase. Science 1988;242(4882):1168–71. [3] Neumann AU, Lam NP, Dahari H, et al. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferonalpha therapy. Science 1998;282(5386):103–7. [4] Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD. HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 1996;271(5255):1582–6. [5] Eigen M. Viral quasispecies. Sci Am 1993;269(1):42–9. [6] Daar ES, Moudgil T, Meyer RD, Ho DD. Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N Engl J Med 1991;324(14): 961–4.
Sallie [7] Coffin JM. HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 1995;267(5197):483–9. [8] Lok AS, Lai CL, Wu PC, Leung EK, Lam TS. Spontaneous hepatitis B e antigen to antibody seroconversion and reversion in Chinese patients with chronic hepatitis B virus infection. Gastroenterology 1987;92(6):1839–43. [9] Guidotti LG, Rochford R, Chung J, Shapiro M, Purcell R, Chisari FV. Viral clearance without destruction of infected cells during acute HBV infection. Science 1999;284(5415): 825–9. [10] Nevens F, Main J, Honkoop P, et al. Lamivudine therapy for chronic hepatitis B: a six-month randomized dose-ranging study. Gastroenterology 1997;113(4):1258–63. [11] Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995;373(6510): 123–6. [12] Abdelhamed AM, Kelley CM, Miller TG, Furman PA, Isom HC. Rebound of hepatitis B virus replication in HepG2 cells after cessation of antiviral treatment. J Virol 2002;76(16): 8148–60. [13] Manzin A, Candela M, Solforosi L, Gabrielli A, Clementi M. Dynamics of hepatitis C viremia after plasma exchange. J Hepatol 1999;31(3):389–93. [14] Duarte E, Clarke D, Moya A, Domingo E, Holland J. Rapid fitness losses in mammalian RNA virus clones due to Muller’s ratchet. Proc Natl Acad Sci USA 1992;89(13): 6015–9. [15] Piatak Jr M, Saag MS, Yang LC, et al. High levels of HIV-1 in plasma during all stages of infection determined by competitive PCR. Science 1993;259(5102):1749–54. [16] Erie DA, Hajiseyedjavadi O, Young MC, von Hippel PH. Multiple RNA polymerase conformations and GreA: control of the fidelity of transcription. Science 1993;262(5135): 867–73. [17] Sanchez-Palomino S, Rojas JM, Martinez MA, et al. Dilute passage promotes expression of genetic and phenotypic variants of human immunodeficiency virus type 1 in cell culture. J Virol 1993;67(5):2938–43. [18] Fust G. Enhancing antibodies in HIV infection. Parasitology 1997;115 Suppl:S127–140. [19] Wallace MJ, Smith DW, Broom AK, et al. Antibodydependent enhancement of Murray Valley encephalitis virus virulence in mice. J Gen Virol 2003;84(Pt 7):1723–8. [20] Kolber MA, Gabr AH, De La Rosa A, et al. Genotypic analysis of plasma HIV-1 RNA after influenza vaccination of patients with previously undetectable viral loads. Aids 2002;16(4): 537–42. [21] Ostrowski MA, Krakauer DC, Li Y, et al. Effect of immune activation on the dynamics of human immunodeficiency virus replication and on the distribution of viral quasispecies. J Virol 1998;72(10):7772–84. [22] Moss WJ, Ryon JJ, Monze M, Cutts F, Quinn TC, Griffin DE. Suppression of human immunodeficiency virus replication during acute measles. J Infect Dis 2002;185(8):1035–42. [23] Watt G, Kantipong P, Jongsakul K. Decrease in human immunodeficiency virus type 1 load during acute dengue fever. Clin Infect Dis 2003;36(8):1067–9. [24] Nowak MA, May RM, Phillips RE, et al. Antigenic oscillations and shifting immunodominance in HIV-1 infections. Nature 1995;375(6532):606–11. [25] Chung RT, He W, Saquib A, et al. Hepatitis C virus replication is directly inhibited by IFN-alpha in a full-length binary expression system. Proc Natl Acad Sci USA 2001;98(17):9847–52.
Replicative homeostasis [26] Torresi J, Earnest-Silveira L, Civitico G, et al. Restoration of replication phenotype of lamivudine-resistant hepatitis B virus mutants by compensatory changes in the fingers subdomain of the viral polymerase selected as a consequence of mutations in the overlapping S gene. Virology 2002;299(1):88–99. [27] Steffy KR, Wong-Staal F. Transdominant inhibition of wild-type human immunodeficiency virus type 2 replication by an envelope deletion mutant. J Virol 1993;67(4): 1854–9.
523 [28] Pol S. Immunotherapy of chronic hepatitis B by anti HBV vaccine. Biomed Pharmacother 1995;49(3):105–9. [29] Kuwata T, Shioda T, Igarashi T, et al. Chimeric viruses between SIVmac and various HIV-1 isolates have biological properties that are similar to those of the parental HIV-1. Aids 1996;10(12):1331–7. [30] Quesada-Rolander M, Makitalo B, Thorstensson R, et al. Protection against mucosal SIVsm challenge in macaques infected with a chimeric SIV that expresses HIV type 1 envelope. AIDS Res Hum Retroviruses 1996;12(11):993–9.
http://intl.elsevierhealth.com/journals/mehy
Replicative homeostasis: a mechanism of viral persistence Richard Sallie* St John of God Hospital, Western Gastroenterology, Suite 35, 95 Monash Avenue, Nedlands, Perth, WA 6009, Australia Received 3 January 2004; accepted 21 February 2004
Summary Acute viral infection is characterised by high-level replication before prompt decline of viraemia and, commonly, viral clearance. This kinetic pattern is generally held to be due to immune control. However, infection with some viruses, notably hepatitis C (HCV), hepatitis B (HBV) and the human immunodeficiency virus (HIV), often results in chronic stable low-level spontaneously fluctuating viraemia, kinetics that are difficult to rationalize on this basis. The persistence of HCV, an RNA virus, is especially problematic and its stability, occurring despite rapid, genomic mutation is highly paradoxical. This paper outlines the hypothesis, and evidence, that viruses autoregulate replication and mutation and describes a mechanism – replicative homeostasis – explaining viral stability. Replicative homeostasis results in stable, but reactive, replicative equilibria that drive quasispecies expansion and immune escape and explain all observed viral behaviours and host responses. This paradigm implies new approaches to antiviral therapy and is broadly relevant to modulation of gene expression. c 2004 Published by Elsevier Ltd.
Background
Problem 1: mutation rate
Viruses, including hepatitis C (HCV), hepatitis B (HBV) and the human immunodeficiency viruses (HIV) cause an enormous burden of disease and premature death worldwide. These viruses replicate, at least in part, by RNA polymerases, enzymes that lack fidelity or proofreading function [1]. During replication of hepatitis C (HCV) or (HIV) each new genome will differ from the parental template by up to ten nucleotides [2] because of non-faithful template copying by RNApol (~1 105 mutations/base RNA synthesised [2]).
The infidelity of RNA polymerases poses a major problem; The probability (q) of a mutation occurring at any point on the genome during one replication cycle is q ¼ ð1 ð1 Ml Þn ), where (Ml ) is the mutation rate and (n) the genome size. For hepatitis C (~9200 bases, an RNA virus lacking DNA “hardcopy”, mutating constantly at 105 substitutions/base, q 0:0879. However, for multiple replications cycles ðhÞ; q ¼ ð1 Mnl Þh . After 20 cycles (occurring in \5 days in most patients [3,4]), the probability of any original genome remaining unmutated is 7:5 1022 , meaning effective loss of genetic information. RNA viral quasispecies stability is a paradox [5]. This “theoretical impossibility” of RNA virus stability suggests either the
*
Tel.: +61-8-93897101; fax: +61-8-93897103. E-mail address: [email protected].
0306-9877/$ - see front matter c 2004 Published by Elsevier Ltd. doi:10.1016/j.mehy.2004.02.050
516 consistently reported rates of RNA polymerase infidelity are incorrect (which, even if true, would only delay quasispecies extinction; if Ml ¼ 1010 , q < 1040 within 100 days, etc.) or that innate viral mechanism(s) control mutation during replication. On the other hand, overly faithful template copying will result in an “oligoclonal” virus susceptible to immune destruction and unresponsive to ongoing evolution of cell receptors. The necessity to retain sequence information, and the later (in terms of replicase ) RNApol evolution) requirements of viruses to access cells and evade immune assault, placed constraints on replicase (RNApol ) function that dictate replicative fidelity must be tightly controlled.
Sallie patients prior to development of antibody, in the absence of any antibody, or any demonstrable Tcell response [8,9] and without lysis of infected cells [9] it is difficult to argue, with any conviction, that immune responses are primarily responsible. Furthermore, from a viral perspective, uncontrolled, exponential growth, as might be expected from the mode of replication, would result in rapid cell lysis, host death and a reduced likelihood of stable host-to-host transmission (as would inadequate replication) a prerequisite for viral survival on an evolutionary timescale. Therefore, viral replication must also be closely regulated.
Evidence for autoregulation Problem 2: replication rate Studies of HIV, hepatitis B (HBV) and HCV, and other viruses, demonstrate broadly similar kinetics; high level, primary phase, viral replication rapidly declines and gives way to relatively constant levels of viraemia [6,7], typically 2–3 logs lower than seen acutely, for prolonged periods (Fig. 1). These kinetics pose questions that are difficult to explain on the basis of “immune control” [6] or any other host response; For example; Why does “immune control” falter after [99% of viral load and antigenic diversity is cleared? And what prevents viral clearance once \1% of viral load and antigenic diversity remain? As down-regulation of viral replication frequently occurs in
Substantial clinical and laboratory evidence suggests viral replication is auto-regulated. During successful antiviral treatment levels of virus fall sharply [3,4,6,10,11], often becoming undetectable. However, viral replication rebounds to pretreatment levels on drug withdrawal in both patients [3,4,10] and in tissue culture [12], the latter indicating replication is controlled by factors independent of either cellular or humoral immune function. Auto-regulation of HCV replication was confirmed most emphatically in patients undergoing plasmapharesis in whom 60–90% reduction in levels of virus returned to baseline, but not beyond, within 3–6 h of plasma exchange [13].
Viral replication
Figure 1 Clinical viral kinetics. Acute phase replication of wild type replication (solid line, arbitrary concentration units) rapidly subsides to basal levels for prolonged periods with minor fluctuations. Mutant forms (dashed line) predominate long term. Modified from Daar et al. [6], and Coffin [7] (for HIV), but similar patterns seen with HCV and HBV infection.
Viruses replicate by copying antigenomic intermediate templates, hence replication generally obeys exponential growth kinetics, such that ½RNAt ¼ ½RNAðt1Þ ek , where ½RNAt is the viral concentration at time (t) and k is a growth constant. However, because of RNA pol infidelity, wild type virus will accumulate at ½RNAwt t ¼ ½RNAwt ðt1Þ ð1 qÞK1 and mutant forms at ½RNAmt t ¼ ½RNAwt ðt1Þ qK1 þ ½RNAmt ðt1Þ K1 where q is the probability of mutation and K1 ¼ ek . Therefore, while wild-type virus will predominate early, replication and intracellular accumulation of mutant forms will accelerate, meaning mutant HCV RNAs will rapidly predominate (Fig. 2(a)). It is likely viral genomes would withstand significant mutation (especially in the 3rd codon position, and in hypervariable regions) without significant disruption to function. However, mutations progressively accumulate in RNAviruses [14] and ultimately severely mutated RNAs and defective
Replicative homeostasis proteins, if translated, will become dominant. It is also likely some mutant viral proteins will resist cellular export, further accelerating relative ac-
Figure 2 Intracellular viral kinetics. Schematic representation of hypothesised intracellular events during viral replication. Exponential accumulation of both wildtype (solid line) and mutant viral proteins, with mutant forms (dashed line) rapidly increasing in concentration (arbitrary units) and predominating (a). Rapidly increasing [mutant]/[wildtype] viral protein ratios (solid line) results in exponential fall in replication function (dashed line, (b)) causing subsequent fall in viral replication (a). Initial high-level replication (dashed line, (c)) falls due to inhibition of polymerase by viral products, while polymerase fidelity (solid line) rises as the inverse function.
517 cumulation of defective forms. If mutant envelope proteins are even slightly inhibitory to RNApol function, or if wild-type envelope proteins are stimulatory, but are out competed for RNApol interactions by mutant forms that will be present in vast molar excess (½RNAmt =½RNAwt > 104–5 as clinical studies confirm [15]), then progressive inhibition of viral replication will result (Fig. 2(b)). Assuming HCV can tolerate 10–30% mutation of its genome (or about 900–3000 bases) without functional disruption, then accumulation of severely mutated proteins and inhibition of RNApol will become significant after ~90–300 cycles of replication. Most cellular enzymes are under some form of kinetic control, usually by product inhibition. While simple negative feedback is sufficient to control the velocity of enzyme reactions, and the rate of product accumulation, it is inadequate to ensure the functional quality of any complex molecules synthesised. The functionality of RNApol output is determined by the protein sequences translated from the RNAs RNApol synthesizes. For viruses, and their polymerase, survival (i.e. whether the polymerase, and its viral shell can avoid immune surveillance, access cells and subsequently replicate) depends on these properties i.e. the envelope proteins topological variability. Viruses intrinsically capable of adaptation to changes within host environments, including variations in host density and cell receptor polymorphisms, will enjoy a competitive advantage over viruses lacking innate responsiveness to environmental change. This adaptability requires that viral RNApol , function be both dynamically modifiable, and controllable, and that implies a nexus between the functional output of polymerases (i.e. the functionality of envelope proteins) and both the processivity and fidelity of that polymerase. RNA polymerase function is responsive to and influenced by accessory proteins that induce conformational changes to modulate both processivity and fidelity [16], representing “proof in principle” of part of the mechanism postulated – replicative homeostasis.
Proposed mechanism of homeostasis Replicative homeostasis results from differential interactions of wild type (wt) and mutant (mt) envelope proteins on RNApol in a series of feedback epicycles that link RNApol function, RNA replication and protein synthesis. The mechanism is outlined schematically (Fig. 3(a) and (b)). Intracellular accumulation of mutant viral proteins results in
518
Figure 3 Replicative homeostasis. (a) Overview of feedback epicycles; Relatively high concentrations of wild type proteins Envwt (blue, A) out-compete mutant forms (Envmt , red) for Env:RNApol interactions, resulting in increased output of mutant RNAs, with subsequent ribosomal (green) translation and increased concentrations mutant proteins (red), relative to wild-type, returning the state to basal equilibrium. Relative excess levels of mutant forms (B, red) out-compete wild type (blue) for interactions with polymerase, favouring the formation of Envmt :RNApol , and blocking Envwt :RNApol interactions. The Envmt :RNApol interactions relatively decrease RNApol processivity but increase fidelity, increasing output of wild-type RNAs, with subsequent increase in translation of wild-type proteins restoring the equilibrium state. (b) Schematic detail of feedback epicycles; envelope proteins Envwt (blue) and Envmt (red) interact with polymerase (Pol, green) to influence rate and fidelity of RNA synthesis changing templatate, and subsequent output, of ribosomes (R). Affinity of Envwt :RNApol (Kwt:pol ) greater than Envmt :RNApol (Kmt:pol ), velocity of reaction catalysed during wild-type/polymerase (Envwt :RNApol ) (mwt:pol ) greater than mutant/ polymerase reaction Envmt :RNApol (mmt:pol ), mutation rate associated with Envwt :RNApol interactions (Muwt:pol ) greater than mutation rate of Envmt :RNApol interactions. As ½Envmt ½Envwt equilibrium results such that ½Envmt Kmt:pol ¼ ½Envwt Kwt:pol or mmt Kmt:pol ¼ mwt Kwt:pol .
progressive inhibition of RNApol by blocking the stimulatory effects of wild-type envelope: RNApol interactions that increase both replication and mutation. Progressive blockade of RNApol by mu-
Sallie tant envelope results in a less processive polymerase with increased fidelity, increasing the relative output of wild-type envelope RNAs, and subsequent preferential translation of wild type envelope proteins, hence, inexorable progression to stable equilibrium. Quasispecies stability, and other consequences (including immune escape and low-level basal replication), are inevitable outcomes once equilibrium is reached as result of these interactions. We suggest that these interactions and the resulting equilibria are important therapeutic targets, and the necessary ligand-envelope proteins or topologically homologous molecules-implicit within this hypothesis. Viruses (notably HIV) produce many accessory proteins such as nef, gag, rev and the HBeAg that effect viral replication and mutation rate. While these proteins, encoded within envelope gene sequences and thus prone to alter functionally with any envelope mutations, may interact with RNApol to reset the level of any replicative equilibrium (by changing replication rate or mutation frequency or both), stable equilibria will still result providing the sum effect of wild type proteins is to increase polymerase processivity (v) and increase mutation (Mu ) frequency relative to mutant forms i.e. MuðwtÞ > MuðmtÞ and vwt > vmt .
Mathematical modelling Intuitive analysis suggesting enzymes acting in a milieu of increasing concentrations of inhibitory molecules become progressively less active until increasing polymerase fidelity increases output of wild type proteins to cause oscillating equilibria to form may be formalised mathematically; Wild type and mutant viral RNA replication dependent on intracellular competition between mutant and wild type envelope proteins (Env) influencing polymerase properties may be described by classical Lotka–Volterra linked (because RNApol function depends on relative wildtype/mutant RNA/protein concentrations and RNAmt derives from RNAwt ) differential “predator–prey” equations of the form dNwt =dt ¼ rNwtðn1Þ ðawt jwt ½Envwtðn1Þ =amt jmt ½Envmtðn1Þ Þ and dNmt =dt ¼ rNmtðn1Þ ðawt jwt ½Envwtðn1Þ = amt jmt ½Envmtðn1Þ Þ where r is the replication rate of viral RNA, a represents the stimulation (or inhibition) coefficient for RNApol :Env interactions, j the affinity constant for RNApol :Env interactions, and ½Envwt=mt the concentration of wild type or mutant envelope at time (t ) 1). For wild type RNAðwtÞ and mutant RNAðmtÞ RNA replication by RNApol subject to differential inhibition ewt , emt by RNApol :Envwt
Replicative homeostasis
519
and RNApol :Envmt end product interactions resulting in mutation probability, p, RNA synthesis may be approximated by a series of probability functions as follows: ½RNAwt ðtÞ ¼ ½RNAwt ðt1Þ rð1 qÞ
awt jwt ½Envwt ðt1Þ amt jmt ½Envmt ðt1Þ
½RNAmt ðtÞ ¼ ½RNAwt ðt1Þ rq
½RNAwt ðt1Þ r
ð1Þ
;
awt jwt ½Envwt ðt1Þ amt jmt ½Envmt ðt1Þ
awt jwt ½Envwt ðt1Þ amt jmt ½Envmt ðt1Þ
;
þ
ð2Þ
ð3Þ
where ½Envwt=mt represents concentrations of envelope proteins. Expression (2) describes (RNAmt ) derived through unfaithful replication of wild type (RNAwt ) template, while expression (3) mutant RNA derived from replication of previously mutated RNAmt templates. Simulating viral replication using a wide range of values for awt=mt and jwt=mt produces typical acute “viral” kinetics (Fig. 4) and demonstrates the initial down regulation of viral replication and long-term stable replication to be intrinsic properties of differential product inhibition of RNApol by [Ewt ] and [Emt ] altering replicative functions that occurs independently of external events.
Figure 4 Mathematical modelling. Graphic representation of simulated kinetics demonstrating initial down regulation of viral replication and long-term stable replication is an intrinsic property of differential influences of [Envwt ] and [Envmt ] on RNApol function and occurs independently of any external influences. Replication conditions: replicative ratio ðRm Þðmmt =mwt Þ ¼ 104 , mutation rate ðMmu Þ ¼ 1 105 , but these kinetics are seen over a wide range of values mutation and replication rates. Providing Muwt > Mumt and ewt > emt , stable equilibria will form.
Testing the hypothesis This hypothesis is simply tested. Addition of envelope viral proteins to a stable viral should change rates of viral replication and mutation, both easily measured. Specifically, addition of mutant envelope proteins or withdrawal of wild-type envelope proteins should reduce viral replication and rates of mutation while removal of mutant envelope proteins or addition of wild-type envelope proteins should increase viral replication and mutation. In fact, observations relevant to all of these suggested perturbations to viral equilibria have already been carried out in a variety of systems and circumstances (Table 1). All outcomes are completely consistent with those predicted by replicative homeostasis.
Discussion Replicative homeostasis is a mechanism that logically explains the initial decline of viral replication and viral kinetics observed during chronic infection. It immediately resolves the paradox of why RNA viral quasispecies are intrinsically stable and explains how these viruses persist and, thereby, cause disease. Replicative homeostasis also explains many unresolved viral behaviours and implies a general approach to antiviral therapy. Viral mutation, under replicative homeostasis, is not a random, passive process. Sustained immune recognition of viral envelope (Fig. 5) would change the replicative equilibrium to favour high affinity wild-type polymerase interactions that, in turn cause rapid replication of mutant forms, reactively driving quasispecies expansion and generating the extreme antigenic diversity seen in RNA quasispecies. Viruses enjoy a massive potential numerical superiority over immune responses; theoretically, a small envelope protein of, say, 20 amino acids could assume 2020 (about 1026 ) possible conformations, greatly in excess of the ~1010 antibody or CTL receptor conformations either humoral and cellular immune responses can generate. A direct consequence of this mismatch and the stable, but reactive, equilibria achieved by viruses because of replicative homeostasis is that once viral infection is established, the clinical outcome is primarily determined by the viruses’ ability to maintain homeostatic control of the quasispecies, rather than the hosts’ response to that quasispecies. Immune escape and maximal cell tropism are inevitable consequences of this process, and the potential antigenic diversity generated.
520
Sallie
Table 1 Outcomes after mutant/wild-type envelope ratios alter during stable viral equilibria Change to equilibrium
Expected outcome
Method
Virus/system
Addition of Wild-typeEnv
Increased replication
1. Cell culture co-transfection
HBV//Cell 1.Altered replication* culture [26] HIV//Cell culture* [27] HIV/Dengue// 1. Reduced replication Human [23] HIV/Measles// 2. Reduced infectivity Human [22] HBV// Human [28]
Increased mutation Addition of MutantEnv
Decreased replication Decreased mutation
Removal of Wild-typeEnv Removal of MutantEnv
Decreased replication Decreased mutation Increased replication
1. Co-infection [23,22] 2.Therapeutic vaccination [28] 3.Cell culture [27,29] 4. Primate vaccine [30]
HBsAb post OLT for HBV Heterologous vaccination
Increased mutation
HIV//Cell culture [27,29] * HIV/SIV// macaques [30] * Human therapeutics** HIV//Influenza Human vaccination [20]
Observed outcome
Reduces replication
Increased replication and mutation [20] Replication bursts quasispecies mutation [21]
HIV//tetanus Human vaccination [21] Examples of references only, more complete list supplied. Broadly, and in many studies, in HBV, HIV, simian (SIV) and feline (FIV) immunodeficiency virus systems. (a) chimeras expressing heterologous envelope genes with common polymerase backbone demonstrated replication characteristics identical to envelopeparental clones. (b) heterologous envelope proteins inhibit wild type viral replication. ** IV Anti-HBsAg routine during liver transplantation for HBV. *
Replicative homeostasis, and its failure, explains the varied outcomes of viral infection; viral failure to down-regulate replication by polymerase inhibition may result in rapidly progressive or fulminant disease (characterised by massively polyclonal, but ineffectual, immune responses), while inadequate viral replication or generation of diversity will result in viral clearance. Stable, homeostatic equilibria will result in chronic infection with episodic fluctuations in viral replication and host response (e.g. ALT), of the sort typical of HCV or HIV. The widely varied spectrum and tempo of viral diseases is more rationally explained on the basis of a spectrum of polymerase properties that result in variable, reactive, but stable replicative equilibria than highly variable and unpredictable (yet genetically homogeneous) immune responses. Homeostatic systems functioning without external perturbations – such as viruses replicating in systems devoid of replicative or immunologic challenge – rapidly progress to stasis. In tissue culture, viruses replicating in the absence of immune challenge to their replicative equilibria, are
unable (and do not need) to generate antigenic diversity by replicative homeostasis, a phenomenon probably responsible for attenuation of viral virulence. By contrast, in dilute serial viral passage, where high affinity wild-type envelope/ polymerase interactions predominate over lower affinity mutant envelope/polymerase interactions, increased viral replication and mutation would be predicted, and has been confirmed [17]. Some viruses (or some strains of viruses), such as HBV, may use a replicative strategy that involves high level replication triggered when opportunity for specific forms of transmission (e.g. vertical mother ) child) is possible but relative inactivity between times, while others such as herpes zoster have prolonged latencies. Replicative homeostasis with prolonged replication inhibition (i.e. dormancy) mediated by RNApol :Env interactions facilitates those strategies. Changing intracellular relative wild-type and mutant envelope concentrations will disturb RNApol :envelope interactions and alter replicative equilibria. Hence once extracellular/intracellular
Replicative homeostasis
Figure 5 Intracellular/extracellular dynamics of replicative homeostasis. Mechanism of antibody dependent enhancement (ADE) [18,19] and acceleration of replication seen during heterologous vaccination for influenza [18] or tetanus [30] (a) and of inhibition of viral replication during heterologous infection (e.g. during measles [21] or Dengue [22] co-infection) or administration of mutant envelope (b).
envelope concentrations equilibrate, extracellular antibodies will effect intracellular viral replication (Fig. 5). This disturbance of viral replicative equlibria rationally explains antibody-dependent enhancement (ADE) of HIV [18], Dengue, Murray Valley encephalitis [19], and other viruses. Increased HIV replication and mutation after influenza [20] and tetanus [21] vaccination; reduced HIV replication during measles [22] and Dengue [23] co-infection; clearance of HBV without hepatocyte lysis or evidence of T cell dependent cytotoxicity [9], are similarly explained by this mechanism. Previously unexplained and problematic viral behaviours and host responses, including long-term non-progression of HIV; HBV persistence despite a robust immune response; long-term antigenic oscillations [24]; spontaneous reactivation of HBV (among many other viruses); and hypermutation of HIV, for example, all rationally resolve within this conceptual framework. There are clear and quite specific therapeutic implications of replicative homeostasis, as well as more general implications. The putative envelope/ polymerase interactions suggested herein are obvious sites for drug targeting and suggest a site of interferon action. Interferon is not effective in HIV or in all patients with HBV, and its efficacy in HCV is
521 genotype-dependent, strongly implying interferon acts in a direct, virus-specific manner unrelated to “immune enhancement”, as both in-vitro data [25] and clinical kinetic studies imply [3]. Heterologous envelope proteins from other viruses or different genotypes of the same virus, or their structural homologues, would probably effect replication in a similar fashion to interferon, an hypothesis supported by suppression of HIV replication during measles [22] and Dengue [23] co-infection. Replicative homeostasis may alter perceptions of the strategies underpinning the immune response. It is possible the primary purpose of the initial heterogeneous polymorphic humoral response to viral infection – typically pentameric IgM – is to push viral replication towards equilibria favouring production of homogeneous wild-type virus, facilitating a concerted and more focussed humoral and/or cytotoxic T cell response; Strong neutralizing IgG antibodies – HBsAb, for example – may develop as a consequence of restricted viral replication and mutation that permits effective and specific immune recognition, rather than the proximate cause of it, but once developed, high affinity wild-type antibodies ensure mutant envelope proteins remain dominant within cells maximising polymerase inhibition. Replicative homeostasis is an adaptation that facilitates stable viral replication in cells and maximises probability of cell-to-cell (and host-tohost) transmission, a prerequisite for viral survival on an evolutionary time scale. A subtle, more primordial, function of envelope/polymerase complexing may explain how replicative homeostasis evolved; Polymerase function contingent upon recognition of and response to complex three-dimensional complementarities between polymerase and envelope proteins is a sophisticated encryption technique, effectively “locking” the polymerase and minimises the likelihood any competing RNA (or DNA) molecules are replicated. In the fierce competition for finite intracellular resources, reproductive strategies that maximise proliferation of “self” genes, while thwarting propagation of “rival” genes, are strongly selected for, and are highly conserved on an evolutionary timescale. The interferons, and other cytokines, are cellular defense mechanisms that long antedate the immune system. If the interferons are functionally homologous to viral envelope proteins, its possible acquisition of these genes resulted from positive selection of beneficial virus-cell symbiosis occurring early in eukaryotic cellular evolution, a process responsible for retention of other genes. Although proposed specifically to explain RNA viral quasispecies stability, replicative homeostasis
522 is, essentially, a mechanism that regulates RNA transcription and modulates protein expression. In addition to regulating phenotypic expression, replicative homeostasis, permits, and depends upon, proteins (phenotype) inducing, and subtly influencing, the degree of genetic mutation at an RNA level. If protein expression (i.e. phenotype) controls RNApol properties in a manner contingent on that proteins’ functionality and controllably influences the degree of mutation of RNA templates RNApol synthesises, a subtle form of “quality control” is exerted over protein synthesis. This mechanism also accelerates, and directs, adaptation; While introduction of lethal mutations to most RNA genomes may not adversely influence quasispecies stability to any degree, replicative homeostasis ensures any RNA mutations that do arise, and that are translated into a beneficial phenotype, will tend to be retained. Finally, accessory proteins also influence the fidelity and processivity of DNA-dependent RNA polymerases. Control of DNA-dependent RNApol transcription by DNA viruses, cellular micro-organisms (e.g. malaria), and eukaryotic cells, mediated subtly modulating protein expression to mediate immune escape, control cell division and differentiation, or other functions, would not be surprising.
Acknowledgements I thank Professors WD Reed, MG McCall, RA Joske, Bill Musk, Roger Williams, AE Jones, Steve Feinstone and Jay Hoofnagle for critical clinical and scientific guidance, Sophie J Coleman, Matt and Tim for all else.
References [1] Steinhauer DA, Domingo E, Holland JJ. Lack of evidence for proofreading mechanisms associated with an RNA virus polymerase. Gene 1992;122(2):281–8. [2] Preston BD, Poiesz BJ, Loeb LA. Fidelity of HIV-1 reverse transcriptase. Science 1988;242(4882):1168–71. [3] Neumann AU, Lam NP, Dahari H, et al. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferonalpha therapy. Science 1998;282(5386):103–7. [4] Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD. HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 1996;271(5255):1582–6. [5] Eigen M. Viral quasispecies. Sci Am 1993;269(1):42–9. [6] Daar ES, Moudgil T, Meyer RD, Ho DD. Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N Engl J Med 1991;324(14): 961–4.
Sallie [7] Coffin JM. HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 1995;267(5197):483–9. [8] Lok AS, Lai CL, Wu PC, Leung EK, Lam TS. Spontaneous hepatitis B e antigen to antibody seroconversion and reversion in Chinese patients with chronic hepatitis B virus infection. Gastroenterology 1987;92(6):1839–43. [9] Guidotti LG, Rochford R, Chung J, Shapiro M, Purcell R, Chisari FV. Viral clearance without destruction of infected cells during acute HBV infection. Science 1999;284(5415): 825–9. [10] Nevens F, Main J, Honkoop P, et al. Lamivudine therapy for chronic hepatitis B: a six-month randomized dose-ranging study. Gastroenterology 1997;113(4):1258–63. [11] Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995;373(6510): 123–6. [12] Abdelhamed AM, Kelley CM, Miller TG, Furman PA, Isom HC. Rebound of hepatitis B virus replication in HepG2 cells after cessation of antiviral treatment. J Virol 2002;76(16): 8148–60. [13] Manzin A, Candela M, Solforosi L, Gabrielli A, Clementi M. Dynamics of hepatitis C viremia after plasma exchange. J Hepatol 1999;31(3):389–93. [14] Duarte E, Clarke D, Moya A, Domingo E, Holland J. Rapid fitness losses in mammalian RNA virus clones due to Muller’s ratchet. Proc Natl Acad Sci USA 1992;89(13): 6015–9. [15] Piatak Jr M, Saag MS, Yang LC, et al. High levels of HIV-1 in plasma during all stages of infection determined by competitive PCR. Science 1993;259(5102):1749–54. [16] Erie DA, Hajiseyedjavadi O, Young MC, von Hippel PH. Multiple RNA polymerase conformations and GreA: control of the fidelity of transcription. Science 1993;262(5135): 867–73. [17] Sanchez-Palomino S, Rojas JM, Martinez MA, et al. Dilute passage promotes expression of genetic and phenotypic variants of human immunodeficiency virus type 1 in cell culture. J Virol 1993;67(5):2938–43. [18] Fust G. Enhancing antibodies in HIV infection. Parasitology 1997;115 Suppl:S127–140. [19] Wallace MJ, Smith DW, Broom AK, et al. Antibodydependent enhancement of Murray Valley encephalitis virus virulence in mice. J Gen Virol 2003;84(Pt 7):1723–8. [20] Kolber MA, Gabr AH, De La Rosa A, et al. Genotypic analysis of plasma HIV-1 RNA after influenza vaccination of patients with previously undetectable viral loads. Aids 2002;16(4): 537–42. [21] Ostrowski MA, Krakauer DC, Li Y, et al. Effect of immune activation on the dynamics of human immunodeficiency virus replication and on the distribution of viral quasispecies. J Virol 1998;72(10):7772–84. [22] Moss WJ, Ryon JJ, Monze M, Cutts F, Quinn TC, Griffin DE. Suppression of human immunodeficiency virus replication during acute measles. J Infect Dis 2002;185(8):1035–42. [23] Watt G, Kantipong P, Jongsakul K. Decrease in human immunodeficiency virus type 1 load during acute dengue fever. Clin Infect Dis 2003;36(8):1067–9. [24] Nowak MA, May RM, Phillips RE, et al. Antigenic oscillations and shifting immunodominance in HIV-1 infections. Nature 1995;375(6532):606–11. [25] Chung RT, He W, Saquib A, et al. Hepatitis C virus replication is directly inhibited by IFN-alpha in a full-length binary expression system. Proc Natl Acad Sci USA 2001;98(17):9847–52.
Replicative homeostasis [26] Torresi J, Earnest-Silveira L, Civitico G, et al. Restoration of replication phenotype of lamivudine-resistant hepatitis B virus mutants by compensatory changes in the fingers subdomain of the viral polymerase selected as a consequence of mutations in the overlapping S gene. Virology 2002;299(1):88–99. [27] Steffy KR, Wong-Staal F. Transdominant inhibition of wild-type human immunodeficiency virus type 2 replication by an envelope deletion mutant. J Virol 1993;67(4): 1854–9.
523 [28] Pol S. Immunotherapy of chronic hepatitis B by anti HBV vaccine. Biomed Pharmacother 1995;49(3):105–9. [29] Kuwata T, Shioda T, Igarashi T, et al. Chimeric viruses between SIVmac and various HIV-1 isolates have biological properties that are similar to those of the parental HIV-1. Aids 1996;10(12):1331–7. [30] Quesada-Rolander M, Makitalo B, Thorstensson R, et al. Protection against mucosal SIVsm challenge in macaques infected with a chimeric SIV that expresses HIV type 1 envelope. AIDS Res Hum Retroviruses 1996;12(11):993–9.