Molina-Par?s C., Lythe G. (editors) Mathematical Models and Immune Cell Biology

Подождите немного. Документ загружается.

Chapter 19

Viral Immunity and Persistence

Stephen Hickling and Rodney Phillips

Abstract Viruses survive and propagate in human populations, despite the

anti-viral immune response. Viruses have developed two principal mechanisms

in order to persist within populations. Some viruses, such as measles, are highly

infectious and quickly spread through susceptible populations. However measles is

rapidly cleared by the immune response. In contrast other less transmissible viruses

such as Epstein-Barr Virus (EBV) have developed a strategy of persisting within a

population by subverting or escaping from the host immune response.

The immune response has a number of defensive layers, which inhibit viral repli-

cation. The innate response is non-speciﬁc and mediated by cytokines that prevent

the spread of virus through tissues and promote the initiation of highly speciﬁc

adaptive immune responses. The adaptive response consists of antibodies, which

bind and neutralise free virus, and CD8

T cells that identify virally infected cells

and eradicate them.

Different immune mechanisms can by evaded by viruses. Cytokines can be in-

hibited either through preventing cytokine expression or reducing cytokine potency

through the production of viroreceptors. Additionally, immune evasion can occur

through antigenic variation. The adaptive immune response may cope with anti-

genic variation but some variability can defy the capacity of the response. Human

Immunodeﬁciency Virus (HIV) escapes adaptive immune responses by generating

antigens unrecognizable by CD8

T cell and antibody responses.

The interaction between the virus and host is a constant evolutionary struggle,

which imposes forces that drive out adapted viruses. The ability of viruses to adapt

allows them to persist and propagate within different populations.

Viruses are obligate intracellular parasites that hijack the infected cell machinery

in order to replicate and be transmitted. The balance between the pathogenicity of a

virus infection and its resolution by the immune response shapes viral survival strat-

egy. It is not to the viruses’ advantage to be highly pathogenic, as rapidly killing the

R. Phillips (



)

Peter Medawar Building for Pathogen Research, University of Oxford, South Parks Road,

Oxford OX1 3SY, UK

e-mail: rodney.phillips@ndm.ox.ac.uk

C. Molina-Par´ıs and G. Lythe (eds.), Mathematical Models and Immune Cell Biology,

DOI 10.1007/978-1-4419-7725-0

19,

 Springer Science+Business Media, LLC 2011

383

384 S. Hickling and R. Phillips

host would prevent transmission and survival within a population. However, the

virus needs to be sufﬁciently pathogenic in order to replicate and be transmitted in

a population, so a balance needs to be achieved. Consequently viruses have evolved

two principal strategies for promoting their survival within a population, which can

acquire immunity. Some viruses such as Morbilivirus (Measles) have a survival

strategy described as hit and run [1]. They are able to infect a large number of

potential hosts before being resolved by the immune response [2,3]. However, other

viruses have evolved the alternative strategy of persistence. These viruses such as

the Herpes viruses have low transmissibility but are generally poorly resolved by the

immune response. This tactic allows them to persist within a population. Countering

viral proliferation is the host immune response. The antiviral immune response con-

sists of both innate and adaptive immune responses. In terms of antiviral immunity

the innate response refers to the action of cytokines and natural killer cells (NK).

The adaptive immune response to viruses includes the production of speciﬁc antivi-

ral antibodies by B cells and the CD8

T lymphocyte response, which eliminate

virally infected cells.

Immunological Mechanisms in Antiviral Responses

The innate immune response to viral infection is mediated through the action of cy-

tokines. Cells of the innate immune response, such as macrophage and monocytes,

produce a range of antiviral cytokines. The most important being Interleukin-1, -2,

Tumour Necrosis Factor (TNF), Interferon (IFN)-˛ˇ and  [4–6]. The action of

these cytokines can promote an antiviral state in bystander cells, inducing the

expression of proteins that destroy viral genomes and inhibit replication [6–8].

Cytokines such as IFN can also upregulate the surface expression of Major His-

tocompatibility Complex molecules (MHC) – termed Human leukocyte antigens

(HLA) in humans [9].

The innate immune response also comprises NK cells. These cells are able to kill

virally infected cells. NK cells act by killing targets that lack HLA molecules which

act as the inhibitor self signal [10]. This is important for viral infections that attempt

to subvert the CD8

T cell response by downregulation of HLA class I molecules.

Adaptive Immune Responses

The adaptive immune response, unlike the innate response, is highly speciﬁc. The

production of antibody by B cells is an important effector of the adaptive im-

mune response [11, 12]. Speciﬁc antibody can bind to viral surface glycoproteins

or viral capsids [11]. Antibody binding can promote viral destruction either by an-

tibody dependent cell mediated cytotoxicity or induction of phagocytosis [11,12].

Additionally, antibody binding can inhibit viral entry by physically blocking viral

receptors binding to complementary receptors [12].

19 Viral Immunity and Persistence 385

The role of antiviral antibody is crucial for protection from a particular virus

upon repeated exposure, such as measles [2, 3]. The importance of the protective

effect of antibody is highlighted by the fact that all successful viral vaccinations

induce protective antibody responses. In addition to antibody, CD8

T lympho-

cytes are crucial to the adaptive antiviral immune response. CD8

T lymphocytes

are able to kill cells infected with viruses. CD8

T cells recognise virally infected

cells because HLA Class I molecules display viral peptide antigens on the cell

surface [13]. Class I HLA molecules are found on the surface of all nucleated

cells. The CD8

T lymphocyte recognises this epitope/HLA complex through its

unique T Cell Receptor (TCR) [14, 15]. This interaction can result in CD8

T cell

activation. Activated CD8

T lymphocytes are able to kill the target cell (express-

ing the Class I HLA with viral epitope) and are termed Cytotoxic T lymphocytes

(CTL) [13].

The CTL response is crucial for protection from viral infections, highlighted by

individuals with Severe Combined Immunodeﬁciency disease and DiGeorges syn-

drome who lack CTL. These individuals are extremely susceptible to viral infections

which are non-pathogenic in healthy people (reviewed in [16]).

HLA Class 1 Molecules and the Presentation of Viral Epitopes

to the Adaptive Immune Response

Structure of the HLA Molecules

The critical role that Class I murine MHC molecules play in the presentation of en-

dogenous, nonself peptides was ﬁrst identiﬁed in 1982 [13]. At this time it became

clear that Class I molecules present only short fragments of peptides, termed epi-

topes, to CD8

T lymphocytes. This was later conﬁrmed in 1987 when the crystal

structure of HLA Class I molecules was resolved [17].

HLA molecules are encoded within the highly polymorphic major histocompata-

bility complex loci, located in chromosome 6p. This complex spans over 3.6 Mbp

and contains approximately 118 genes. In addition to HLA Class I, the MHC also

encodes Class II HLA molecules as well as a number of cytokines and elements of

the innate immune system.

Since the initial discovery of HLA Class I molecules it became evident that there

is huge diversity in HLA Class I molecules, with 702 alleles identiﬁed to date. These

alleles can be further divided into HLA Class IA, IB and IC isotypes. Each individ-

ual has the potential to express up to two different HLA molecules of each isotype,

six in total. It is thought that most CTL are reactive against peptides presented on

HLA-A and B molecules.

The HLA Class I molecule is a non-covalently linked trimer (Fig. 19.1). The

trimer consists of the polymorphic heavy chain (encoded in the MHC), ˇ

mi-

croglobulin (ˇ

m) and the bound epitope [17]. The heavy chain consists of three

386 S. Hickling and R. Phillips

Fig. 19.1 Structure of HLA class 1 molecule. The HLA heavy chain is shown in purple (˛

, ˛

and

domains). The Heavy chain forms a binding groove, where the speciﬁc peptide binds, between

the 1 and 2 domains. The Heavy chain/peptide complex is associated with 2 microglobulin (ˇ

m),

shown in red

domains ˛

, ˛

and ˛

with a molecular weight of approximately 43 kDa. The ˛

domain spans the cell membrane [17]. The ˛

and ˛

domains form eight ˇ-pleated

sheets, upon which sit two anti-parallel ˛ helices [17]. This structure forms a plat-

form with a groove, within which the epitope sits [17, 18]. It is the ˛

and ˛

domains that form this structure which have highly polymorphic regions, allowing

a greater range of peptides to bind the groove [18].

The structure of the groove places biochemical restrictions on the peptides that

can stably bind. The groove can only accommodate peptide between 8 and 12 amino

acids in length [18]. The peptide and HLA groove interact through the formation of

hydrogen bonds at both the N- and C-terminals. The residues in the peptide that

form the hydrogen bonds are termed anchor residues.

These anchor residues are determined by the amino acids found both in the ˇ-

pleated sheets and ˛ helices of the HLA molecule. The amino acid sequence in the

rest of the bound peptide interact with the groove, however there is much greater

ﬂexibility in these amino acid sequences [17,18].

HLA Class I Antigen Presentation

Presentation of peptides on HLA Class I molecules is crucial for an effective im-

mune response. The cellular process for the production and presentation of optimal

peptides is a highly orchestrated process, which can be broadly divided into peptide

cleavage and HLA loading and surface expression (Fig. 19.2).

19 Viral Immunity and Persistence 387

Fig. 19.2 Mechanisms of CD8

T cell killing. (a) Schematic showing the granzyme exocytosis

model of killing. Granzyme B enters the target cell by perforin. Granzyme triggers the pro-

apoptotic caspase cascade in the target cell. (b) Schematic depicts receptor mediated target cell

death. Engagement of receptors containing death domains (such as Fas) with the appropriate lig-

and on the CTL causes trimerisation of death domains, which triggers the pro-apoptotic caspase

cascade

The majority of peptides loaded into HLA Class I molecules are derived from

the cytoplasm including self-peptides [20]. In non-infected cells, proteins are con-

stitutively cleaved into peptide fragments by the cylindrical proteasome, also called

the 20S or core particle proteasome [21]. The 20S proteasome cleaves proteins into

oligo-peptides between 3 and 15 amino acids in length [22]. The proteasome’s cylin-

drical shape is formed from two ˛ rings, between which are two ˇ rings [21, 23].

Each ring is composed of either seven ˛ or seven ˇ subunits [23]. The ˇ1,ˇ2 and ˇ5

subunits of the 20S proteasome have peptidylglutamyl, trypsin and chymotrypsin

activities, which primarily act at the C terminus of the peptide [24, 25]. The re-

maining ˇ subunits are N-terminal acting threonine proteases [25]. The ˛ rings

are thought to control the entry of peptides into the catalytic ˇ rings, with poly-

ubiquitinated peptides being preferentially cleaved [21].

Under the inﬂuence of IFN, the 20S proteasome is modulated by the addition of

19S subunits, which can bind to either one, or both ends of the 20S proteasome [21].

IFN also upregulates the expression of other catalytic subunits, LMP2, LMP7 and

MECL-1 [26, 27]. These subunits are incorporated into the ˇ rings, replacing the

ˇ1, ˇ2 and ˇ5 subunits. This IFN induced complex forms the 26S immunopro-

teasomes. The modiﬁed ˇ subunits alter the peptide cleavage speciﬁcity of the

immunoproteasome [26, 27]. The trypsin and chymotrpsin activity is increased at

388 S. Hickling and R. Phillips

the carboxy side of basic and hydrophobic residues while caspase activity at acidic

residues is reduced. This consequently causes the immunoproteasome to produce

longer peptides, which have more hydrophobic and basic residues at the Carboxy

termini [26,27]. Residues with these biochemical characteristics tend to have higher

binding afﬁnity for HLA Class I molecules [17, 28]. Since basic and hydrophobic

residues are more likely to form anchor residues.

Peptides produced by the immunoproteasome are transported into the Endoplas-

mic Reticulum (ER) [29]. The ER is where newly synthesised HLA Class I heavy

chains are located. Peptides are actively transported into the ER via the Trans-

porter Associated with Antigen presentation (TAP) [29]. TAP is a transmembrane

heterodimer, encoded in the MHC [30] which preferentially transports basic and

hydrophobicpeptides into the ER lumen [30,31]. In the ER peptides are further opti-

mized at the N-terminus by ER amino-peptidase (ERAAP) [32,33]. ERAAP cleaves

peptides of different lengths at different rates. 10mers are cleaved at higher rate then

9mers and little activity is directed against 8mers. This hierarchy of ERAAP activity

increases the production of optimal peptides for HLA Class I loading [34,35].

TAP is also a crucial component in the peptide loading complex (PLC) [36]. The

PLC allows HLA Class I heavy chains to maintain structural conformation before

peptide loading [36]. The complex comprises TAP, unloaded Class I Heavy chain

associated with ˇ

m and the molecular chaperone tapasin [37]. This complex is al-

ways found with the same stoichiometry of 1 TAP: 4 tapasin: 4 heavy chains [38].

Binding of the optimal peptides to the PLC confers conformational change, re-

leasing the Class I trimer complex from the PLC. This release consequently allow

trafﬁcking of peptide loaded Class I HLA molecules to the cell surface, to be sam-

pled by patrolling CD8

T lymphocytes [39].

Activation of CD8

T Lymphocytes and Killing of Virally

Infected Cells

Interaction of MHC Class 1 and the TCR

Loaded HLA Class I molecules expressed on the cell surface are sampled by CD8

T lymphocytes. The CD8

T lymphocyte TCR transiently binds the surface of HLA

Class I molecules, contacting both the ˛

and ˛

domains and the epitope [14,15].

When a particular TCR encounters its complementary HLA Class I-epitope com-

plex, CD8

T lymphocyte activation can occur (Fig.19.3).

Despite TCR engagement, TCRs lack the ability to initiate intracellular sig-

nalling. To overcome this TCRs are non-covalently associated with the trans-

membrane CD3 complex [40]. The CD3 complex is composed of single  and ı

sub-units in addition to two  and two  subunits [40, 41](Fig.19.4). Each CD3

associates with two TCR dimers [40]. Consequently the overall structure of the com-

19 Viral Immunity and Persistence 389

Fig. 19.3 The CD8

T cell interacts with the HLA class I molecule through its T cell receptor

(TCR). The TCR binding groove, formed by the V˛ and Vˇ chains of the TCR, contact the surface

of the HLA/peptide molecule. The TCR contacts the HLA heavy chain ˛

, ˛

and the presented

peptide. (Picture taken and modiﬁed from: Atlas of Genetics and cytogenetics Atlasgeneticsoncol-

ogy.org/Deep/mapping Author Gerald. J. Mizejewski)

plex is TCR

CD3ı



[42]. The cytoplasmic domains of CD3 subunits contain

immuno-receptor tyrosine based motifs (ITAMS). These ITAMS can phosphorylate

intracellular kinases, which facilitate intracellular signalling [43,44].

The TCR–HLA interaction alone however is not sufﬁcient to activate CD8

T lymphocytes. A substantial number of other receptors, termed co-receptors, are

also involved; CD8 itself is involved in amplifying the signal [45]. The CD8 co-

stimulatory molecule localizes to the engaged TCR and ampliﬁes CD8

T cell acti-

vation [46]. This occurs through CD8 binding in an epitope independent manner to

the ˛

domain of the HLA Class I molecule so prolonging cell engagement [45,46].

The degree that the CD8 interaction inﬂuences activation is thought to be a func-

tion of the afﬁnity of the TCR for the HLA-epitope complex [47, 48]. TCR/HLA

390 S. Hickling and R. Phillips

ζζ

δγ

αβ

TCR

CD3 CD3

ITAMs

Fig. 19.4 The TCR complexes with CD3, that comprises of single and sub-units in addition to two

 and two  subunits. The overall structure of the complex is TCR

CD3ı



. The cytoplasmic

domains of CD3 subunits contain immuno-receptor tyrosine based motifs (ITAMS) which facilitate

intracellular signalling

and co-receptor engagement leads to surface protein rearrangement [49]. HLA

molecules and co-receptors tend to be polarized to the point of initial HLA/TCR

engagement [49, 50]. This leads to multiple HLA/TCR engagements. These multi-

ple interactions form an immunological synapse, bringing the CD8

T lymphocyte

and the target cell membranes into close contact [49,50]. Multiple molecular inter-

actions promote CD8

T cell activation [51].

Mechanisms of CTL Killing

Once activated, CD8

T lymphocytes have cytotoxic capability and are conse-

quently termed cytotoxic T Lymphocytes (CTL). Both CTL and NK cells employ

a number of different mechanisms in order to destroy virally infected cells. Cy-

totoxicity is mediated by direct cellular interactions such as granule exocytosis or

engagement of death domains [52,53]. All these mechanisms direct target killing by

inducing cellular apoptosis.

Apoptosis is an energy dependent, highly ordered process of cell death. Apopto-

sis causes cells to detach from the local environment, condense and lose membrane

integrity. Apoptosis is orchestrated by a subset of proteases called caspases [54].

19 Viral Immunity and Persistence 391

Caspases cleave substrates at C-terminal aspartate residues, and disrupt intracellu-

lar structure and DNA integrity [54]. This highly controlled process limits cell death

to the target. This is in contrast to necrosis, which results in uncontrolled cellular

damage resulting in inﬂammation (reviewed in [55]).

Granule Mediate Cytotoxicity

CTL contain granules containing effector molecules that can induce apoptosis to

the target cells. The most important of these are perforin and granzyme B, members

of the caspase family [56]. When CTL interact with target cells the lytic granules

are polarized towards the contact point with the target [57]. This led to the granule

exocytosis model of CTL-induced killing, which postulates that the CTL exocytose

perforin and granzyme B into the cleft between the T cell membrane and the tar-

get cell. It was originally hypothesised that perforin monomers polymerise in the

target cell membrane in the presence of Ca

[58–60]. This results in the forma-

tion of transmembrane pores in the target, allowing activated granzyme B to enter

the target cell cytoplasm [61]. However, in perforin deﬁcient mice, CTL can still

kill target cells via granzyme B [62]. This, along with other observations, has led

to the hypothesis that granzyme B can be internalised into target cells by recep-

tor mediated endocytosis, with perforin playing a role in granzyme B release from

vesicles [63, 64].

Granzyme B cleaves procaspase-3 and procaspase-8 [65,66]. Both these procas-

pases require cleavage for activation, and form part of the caspase cascade which

induce apoptosis. However, granzyme B can also induce apoptosis in a caspase in-

dependent manner [67]. Granzyme B signals through the mitochondria, disrupting

energy metabolism resulting in the release of cytochrome c, a pro-apoptotic pro-

tein [67].

Although granzyme B is the main effector molecule for the induction of apop-

tosis other granzymes (A–H) and serine esterases are also released into the cell

cytoplasm. These effector molecules are also members of the caspase family and

help to induce apoptosis though cleavage of their molecular targets [67, 68].

Signalling Induced Apoptosis

In addition to the granule exocytosis model of cyotoxicity, it has been shown that

CTL can kill targets independently of granzyme B, by receptor mediated cytotox-

icity [52, 53]. This mechanism of cytotoxicity is principally used in lymphocyte

selection, but has recently been associated with auto-immunity [69].

CTL kill through the CD95 (Fas) pathway [52]. Fas is a member of the tumour-

necrosis factor receptor family of death receptors [70, 71]. Fas is expressed on the

surface of many cell types; however, Fas ligand (FasL) is only expressed on the

392 S. Hickling and R. Phillips

surface of activated CTL [52]. Ligation of Fas by FasL causes trimerization of Fas

on the cell surface. Each Fas molecule has an intracellular death domain, which

recruits and activates caspase 8, inducing the caspase signalling cascade [71].

Viral Persistence and Subversion of the Immune Response

Viral Subversion of the Immune System

The coexistence of the viral pathogen and the host is an extremely ﬁne balance. The

virus needs a large enough host population in order to survive and propagate, so

rapidly killing the host is not a successful evolutionary strategy. Viruses with low

transmission rates tend to be poorly resolved by the immune response. Thee viruses

persist within hosts in order to survive. In order to do this viruses have developed

strategies, which can interfere with the immune response designed to eliminate it.

Interference occurs against both the innate and the adaptive responses.

Subversion of Cytokine Action

Cytokines are critical for immune signalling and promoting immunological re-

sponses. Consequently, viruses have evolved a varied number of strategies to either

inhibit cytokine production or prevent their action [66]. The most common cytokines

targeted are inevitably the cytokines with the most potent antiviral effects. These in-

clude Interleukin (IL)-1, -2, Tumour necrosis factor (TNF), Interferon (IFN)-˛-ˇ

and  [4,6,8].

IFN signalling causes the induction and expression of a number of genes,

termed IFN stimulated genes (ISG). This occurs through IFN regulatory factor 3

(IRF-3), that translocates to the nucleus and binds the transcription co-activator

CBP [72]. The Adenovirus E1A protein binds to and inhibits the IRF–CBP inter-

action preventing ISG expression [72]. Hepatitis B employs a similar strategy with

the terminal protein inhibiting the IRF–CBP interaction [73].

In addition to reducing the production of cytokines, a number of other viruses

prevent their action. One mechanism, employed by the Pox family of viruses, is the

secretion of viroreceptors [74]. Viroreceptors can mimic the receptor for a given

cytokine and have potent cytokine binding capability [74]. The myxoma virus vi-

roreceptor, M-T2, and cowpox virus crmB both mimic the TNF receptor [75, 76].

Viroreceptors compete with and sequester secreted TNF [75]. This consequently

reduces the inﬂuence of TNF. Other viroreceptors, such as MT-7 produced by myx-

oma are thought to be promiscuous for a number of cytokines including IFN,IL2

and IL5 [77, 78].