| REVIEW ARTICLE | |||||||
| DOI: 10.1099/vir.0.19110-0 | ||||||||
| Online 22 April 2002 | ||||||||
|
|
AIDS, caused by the retroviruses human
immunodeficiency virus type 1 and type 2 (HIV-1 and HIV-2), has
reached pandemic proportions. Therefore, it is critical to
understand how HIV causes AIDS so that appropriate therapies can be
formulated. Primarily, HIV infects and kills CD4+ T lymphocytes,
which function as regulators and amplifiers of the immune response.
In the absence of effective anti-retroviral therapy, the hallmark
decrease in CD4+ T lymphocytes during AIDS results in a weakened
immune system, impairing the body's ability to fight
infections or certain cancers such that death eventually ensues.
The major mechanism for CD4+ T cell depletion is programmed cell
death (apoptosis), which can be induced by HIV through multiple
pathways. Death of HIV-infected cells can result from the
propensity of infected lymphocytes to form short-lived syncytia or
from an increased susceptibility of the cells to death. However,
the apoptotic cells appear to be primarily uninfected bystander
cells and are eradicated by two different mechanisms: either a
Fas-mediated mechanism during activation-induced cell death (AICD),
or as a result of HIV proteins (Tat, gp120, Nef, Vpu) released from
infected cells stimulating apoptosis in uninfected bystander cells.
There is also evidence that as AIDS progresses cytokine
dysregulation occurs, and the overproduction of type-2 cytokines
(IL-4, IL-10) increases susceptibility to AICD whereas type-1
cytokines (IL-12, IFN-
) may be protective. Clearly there are
multiple causes of CD4+ T lymphocyte apoptosis in AIDS and
therapies that block or decrease that death could have significant
clinical benefit.
| INTRODUCTION |
Human immunodeficiency virus (HIV) is a retrovirus that
causes AIDS, and currently infects 42 million individuals worldwide
(WHO, 2002). One of the principal cellular targets of HIV infection
is the CD4+ T helper lymphocyte (Th). Due to their
central role in controlling immune responses, Th lymphocytes have
become a focus of study with cytomegalovirus (Zeevi et al.,
1999
), leishmaniasis (Lehmann &
Alber, 1998), cancer (Ohmi et al.,
1999), transplantation (Schirren et
al., 2000) and allergy (Robinson et
al., 1993; Macaubas et al., 1999).
The Th immune response can be grouped according to which arm of the
immune system is activated. While not absolute, type-1 helper
responses are critical in controlling intracellular infections via
cytotoxic T lymphocyte (CTL)-mediated mechanisms whereas type-2
helper responses protect against pathogens neutralized by
antibodies (Maloy et al., 2000).
CD8+ T cells represent a major arm of the cellular
immune response and their differentiation into effector cells
usually requires Th cell stimulation. The importance of CTL in
clearing virus infections is well established and has been
demonstrated for several viruses (Lukacher et al., 1984;
Murray et al., 1992; Lehmann-Grube et al.,
1993). The role of Th cells in the
generation and maintenance of functional virus-specific CTL is not
fully understood and some evidence suggests that CD4+
and CD8+ T cells can mount specific immune responses
independently (Kasaian et al., 1991).
However, studies on the impact of Th lymphocytes in CTL responses
to a chronic murine lymphocytic choriomeningitis virus (LCMV)
infection demonstrated that, when Th cells were depleted, the
LCMV-specific memory CTL responses decreased significantly,
resulting in reduced protection and a persistent virus infection
(Matloubian et al., 1994; von Herrath et al.,
1996). An HIV-1 study demonstrated that a
strong p24-specific Th proliferative response correlated with the
magnitude of the Gag-specific CTL response, and the control of
viraemia (Kalams et al., 1999).
In contrast, some in vitro data demonstrated strong CTL
responses in people with high HIV virus loads (Koenig et
al., 1995). This discrepancy suggests a
difference between immune response dynamics in vivo compared
to in vitro. In addition to stimulating CTL activity, Th
cells may also control HIV infection by producing, along with
CD8+ T cells, 
-chemokines that competitively inhibit
HIV attachment and downregulate the HIV co-receptor chemokine
receptor proteins (Kinter et al., 1998;
Saha et al., 1998).
Th cells are thought to be important in controlling HIV
infection both in the acute (Copeland & Heeney, 1996;
Rosenberg & Walker, 1998) and in the chronic phases of the
disease with a high HIV-specific CD4+ response being
noted in long-term non-progressing subjects (Rosenberg et
al., 1997). Primary HIV infection presents
with a high HIV titre that is initially controlled by a
CD8+ CTL response, along with anti-HIV antibodies (Clark
et al., 1991; Daar et al., 1991).
The virus plasma load reaches a set point that is maintained at a
plateau during the asymptomatic phase of 2–10 years
(Fig. 1). This
homeostasis becomes unbalanced, resulting in a gradual decrease in
the Th lymphocyte cell number concomitant with an increase in virus
load, signalling the onset of AIDS. A hallmark of HIV infection is
the progressive loss of Th lymphocytes as HIV disease progresses.
When CD4 levels drop from a normal of 1000 cells
mm–3 of whole blood to less than
200 mm–3, immune dysregulation
and opportunistic infections result. Evidence to date suggests that
Th responses play a major role in the control of HIV infection
(Rosenberg et al., 1997; Rosenberg & Walker, 1998;
Phenix & Badley, 2002). In addition, our studies
demonstrating HIV-specific Th responses in highly exposed
persistently seronegative individuals suggest this may play a role
in the prevention of HIV infection (Fowke et al., 2000).
In HIV disease, the mechanism for deletion of Th cells remains
unknown, although several processes probably contribute. As the
loss of Th lymphocytes is central to AIDS pathogenesis, we will
examine the possible mechanisms of their death and how their loss
contributes to disease progression. Although the pathways of
CD4+ T cell death in HIV infection are many, they mainly
result in programmed cell death (apoptosis).
Fig. 1. Kinetics of HIV disease. This is a representation of a typical HIV/AIDS disease progression in the absence of anti-retroviral
therapy. The initial spike in HIV load in the acute phase is
accompanied by an increase in HIV-specific CTLs and a decrease in
the number of CD4+ T cells. Within 1 to 2 months the
virus load is reduced and maintained at a new lower threshold by
the immune response. The lower virus load is concomitant with an
increase in CD4+ T cells, and at 2 to 3 months
seroconversion occurs as HIV-specific antibodies appear. This new
asymptomatic state is generally maintained for a number of years.
Gradually the CD4+ T cell counts decrease to below 200
mm–3 in the blood signalling the
onset of AIDS and increased susceptibility to opportunistic
infections. This is followed by an increase in the virus load,
along with a decrease in the HIV-specific CTL, and neutralizing
antibodies.
| REGULATION OF APOPTOSIS: AN OVERVIEW |
In HIV infection, disease progression correlates with
both increased virus load (Furtado et al., 1995)
and elevated levels of apoptosis (Gougeon et al., 1996).
In particular, Th cell levels inversely correlate with levels of
apoptosis (Fowke et al., 1997).
In understanding how HIV contributes to the depletion of the immune
system, one needs to be familiar with the various cell death
pathways and to determine how HIV modulates them. Apoptosis is
tightly regulated and occurs in response to either
receptor-mediated (Fas, TNFR1, DR3, DR4 and DR5) or
non-receptor-mediated (UV irradiation, DNA damage, granzymes, etc.)
signals (Evan & Littlewood, 1998;
Budihardjo et al., 1999). It has a role in many normal
physiological processes, including homeostasis of lymphocyte
populations, tissue differentiation (Golstein, 1998) and
elimination of tumorigenic, mutated or virus-infected cells
(Everett & McFadden, 1999). The response to death signals
varies depending on cell type, activation or developmental stage of
the cell, as well as the chemical or physical environment.
The apoptotic pathway involves two families of
proteins, the effectors and the regulators (Los et al.,
1999; Gross, 2001;
Zimmermann et al., 2001). Upon receiving a death signal, the
effector family of serine proteases called caspases
(cysteine-dependent aspartate-specific
proteases) are activated to catalyse a cascade of molecular
events resulting in the activation or inactivation of numerous
cellular proteins (Fig.
2). This results in the classical
apoptotic morphological and biochemical changes such as plasma
membrane blebbing, mitochondrial dysfunction and DNA fragmentation.
The regulators of the apoptotic pathway are found in a second group
of proteins, the Bcl-2 family (Strasser et al., 2000).
These proteins contain anti-apoptotic (Bcl-2, Bcl-XL) and
pro-apoptotic (Bax, Bid) members that exert their function
primarily at the mitochondrion by either preventing or inducing
mitochondrial dysfunction (Korsmeyer et al., 2000;
Alimonti et al., 2001). Anti-apoptotic molecules localize
to the outer mitochondrial membrane whereas the pro-apoptotic
molecules are sequestered in the cytoplasm by a variety of
mechanisms. Upon receiving a death signal, the pro-apoptotic
proteins translocate to the mitochondrion to interact with the
anti-apoptotic molecules, thereby promoting mitochondrial
dysfunction. The relative level of these proteins is important, as
increased amounts of anti-apoptotic proteins are able to override
even larger numbers of pro-apoptotic signals. Activation of caspase
3 is the point of no return and results in the initiation of many
downstream effector functions leading to the typical apoptotic
features such as membrane blebbing and DNA fragmentation.
Fig. 2. The
classical apoptotic pathway. Cells receive either a
receptor-mediated or a non-receptor-mediated death signal to
initiate the apoptotic pathway. Constitutive upstream caspases
(i.e. caspase 8) and pro-apoptotic Bcl-2 family proteins (i.e. Bid,
Bax) are activated, resulting in a cascade of molecular events that
act at the mitochondrion. They can induce a loss of mitochondrial
membrane potential (
), production of reactive oxygen species (ROS), permeability transition (PT) due to opening of the
permeability pore, mitochondrial swelling and ultimately release of
apoptosis-inducing factor (AIF) and cytochrome c. Release of
cytochrome c is a point of no return as cytochrome c
forms a complex with caspase 9, Apaf-1 and dATP resulting in the
autoactivation of caspase 9. Caspase 9 proceeds to cleave the
downstream effector caspases (caspase 3, 6, etc.) that in turn act
on many cellular proteins to give the typical biochemical and
morphological features such as membrane blebbing and DNA
fragmentation. Some apoptotic pathways are able to induce cell
death in a mitochondrial-independent manner that is not inhibited
by Bcl-2. In these cases, pro-apoptotic upstream molecules can
activate caspase 3 directly. However, there is a feedback loop in
which the activated caspase 3 acts on the mitochondrion to induce
dysfunction at later stages of apoptosis.
| APOPTOSIS AND THE HOMEOSTASIS OF T LYMPHOCYTES |
The Th cell's ability to become activated,
proliferate and function is critical to an effective immune
response. For CD4+ T lymphocytes to function properly
they must first become activated (Fig.
3) by antigen-presenting cells (APC) via a
two signal mechanism (Bretscher, 1999). T cell receptor (TCR)
engagement without the second co-stimulatory signal results in Th
cell death. In fact, during HIV infection the expression of several
co-stimulatory molecules is altered and may contribute to increased
levels of T cell apoptosis (Kammerer et al., 1996;
Chougnet et al., 1998; Sousa et al., 1999).
Fig. 3. Activation of T lymphocytes. Naive T cells (CD45RA) require a series of signals in order to become activated (CD45RO). The first
signal involves the recognition of a peptide in the pocket of the
MHC on the antigen-presenting cell (APC) by the T cell receptor
(TcR) on the T lymphocyte. Other signals involve the interaction of
CD28 and CD40L costimulatory molecules on the T cell with CD80/86
and CD40, respectively, on the APC.
A specific immune response must be able to expand
rapidly to process foreign antigens, but once the danger has been
removed large numbers of antigen-specific T cells are induced to
die because they are no longer needed. Receptor-mediated apoptosis
has an essential role in maintaining the homeostasis of T
lymphocyte numbers, so at the conclusion of the immune response
mature antigen-specific activated lymphocytes undergo primarily
Fas-mediated apoptosis (Lenardo et al., 1999;
Wolf & Green, 1999) (Fig.
4). Studies initially performed in
gld and lpr knockout mice have revealed that
molecules such as Fas (CD95/Apo-1) and Fas-ligand (FasL),
respectively, function to downregulate the immune response (Cohen
& Eisenberg, 1992; Nagata & Suda, 1995),
or otherwise massive and lethal lymphoproliferation occurs. Fas
expression on activated T cells does not guarantee a FasL-induced
death (Irmler et al., 1997). FLIP protects cells from
FasL-induced cell death during the initial phase of T cell
activation, but not during the later stages. This period of Fas
susceptibility is consistent with the time when the immune system
is down-regulating an immune response. In HIV infection, T cell
apoptosis is a complex process that involves both HIV-infected and
uninfected cells.
Fig. 4. Fas cell death pathway. There are five receptors (Fas, TNFR1, DR3, DR4 and DR5) in the receptor-mediated death pathway. Fas-induced
cell death requires the binding of either membrane-bound or soluble
Fas ligand (m/sFasL) to the Fas receptor on the cell surface. This
initiates the formation of the death-inducing signalling complex
(DISC), which includes Fas, FADD and caspase 8, and ultimately
results in the activation of caspase 8. The interaction of FADD
with caspase 8 can be blocked by the cellular inhibitory protein
Flip and can therefore block Fas-mediated apoptosis. There are two
types of Fas-mediated death pathways. Type 1 is
mitochondrial-independent and therefore not inhibited by the
anti-apoptotic protein Bcl-2. It involves the direct activation of
effector caspase 3 by the activated caspase 8. In contrast, type 2
proceeds via the mitochondria, resulting in mitochondrial
dysfunction and cytochrome c release, to eventually activate
caspase 3. Since Bcl-2 functions primarily at the mitochondria this
pathway can be inhibited by Bcl-2.
| CELL DEATH IN HIV-INFECTED CELLS |
The devastating effect HIV has on the immune system is
a result of infecting and killing the immune-regulating Th
lymphocytes. HIV infects and replicates primarily in activated
CD4+ T lymphocytes and to a lesser extent in macrophages
and dendritic cells. The cell specificity is due to the HIV
envelope (Env) glycoprotein binding CD4, along with chemokine
receptors CXCR4 or CCR5, for cell entry (Hogan & Hammer,
2001a, b). Once in the cytoplasm the viral
RNA is reverse-transcribed into DNA and randomly integrated into
the host cell genome. Activation of the host cell enhances the
production of viral proteins, which assemble upon budding out of
the cell, utilizing the plasma membrane as the viral envelope. HIV
plasma levels during all stages of infection range from 50 to
11x106 virions
ml–1 (Piatak
et al., 1993). The half-life of most HIV-infected
T cells in vivo is 12–36 h (Ho et al.,
1995; Wei et al., 1995;
Perelson et al., 1996), with the mechanisms of cell death
being virus- or receptor-mediated apoptosis.
There are several mechanisms by which HIV can induce
cell death directly in the cell it infects. CXCR4-tropic HIV
isolates, generally found in the later stages of HIV infection,
preferentially infect T cells and induce membrane fusion between
cells to form a giant multinucleated cell called a syncytium.
Although syncytium formation is not necessary for the progression
to AIDS, syncytia have a short lifespan, and the emergence of
CXCR4-tropic strains correlates with an increased depletion of T
cells (Richman & Bozzette, 1994;
Sylwester et al., 1997; Kimata et al., 1999).
In addition, cell viability can be compromised because the plasma
membrane becomes disrupted or more permeable due to the continuous
budding of the virion (Fauci, 1988), or due to specific HIV proteins
such as Vpu that can induce membrane permeability (Gonzalez &
Carrasco, 2001). Virus replication in the cell also
has terminal consequences as cellular toxicity increases due to a
build up of un-integrated linear viral DNA (Shaw et al.,
1984; Levy, 1993).
Also, through cleavage, the HIV protease can inactivate
anti-apoptotic Bcl-2 while simultaneously activating pro-apoptotic
procaspase 8, making the cell more susceptible to mitochondrial
dysfunction in response to internal or external death signals
(Korant et al., 1998; Nie et al., 2002).
Both CD4+ and CD8+ T lymphocytes
are more susceptible to Fas-induced apoptosis in HIV+
individuals, and this is related to the regulation of surface
levels of CD95 (Fas) and FasL. Peripheral blood mononuclear cells
(PBMC) from HIV+ individuals express higher levels of
CD95 (Silvestris et al., 1996),
and the proportion of these T lymphocytes increases with disease
progression (Aries et al., 1995;
Baumler et al., 1996; Estaquier et al., 1996).
The HIV proteins Nef (Zauli et al., 1999),
Env (Oyaizu et al., 1994; Tateyama et al., 2000)
and Tat (Ensoli et al., 1990;
Westendorp et al., 1995) have also been implicated in
increasing CD95 and FasL levels, which presumably enhances
susceptibility to Fas-mediated killing. Nef is thought to induce
FasL by interacting specifically with the zeta chain of the TCR
complex (Xu et al., 1999). In addition, there are
pro-apoptotic influences on other aspects of the Fas death pathway.
Tat upregulates initiator caspase 8, resulting in more caspase 8
activity (Bartz & Emerman, 1999).
HIV proteins also act on the central regulators of the death
pathway, the Bcl-2 family members, that can either induce or
prevent apoptosis. The cell normally has an appropriate balance of
pro- versus anti-apoptotic proteins to ensure cell survival but HIV
can alter the balance, resulting in the disruption of mitochondrial
function. Levels of anti-apoptotic Bcl-2 are significantly lower in
HIV-infected individuals (Re et al., 1998).
In particular two HIV proteins, Tat (Sastry et al.,
1996) and HIV protease (Strack et
al., 1996), have been shown to decrease Bcl-2
levels. At the same time, Tat also has the ability to increase
pro-apoptotic Bax (Sastry et al., 1996)
and Bim (Chen et al., 2002). Other evidence demonstrates that
deletion of Vpu partially increases survival in response to CD95
cross-linking in Jurkat cells and PBMCs (Casella et al.,
1999). This could be due to the ability
of Vpu to suppress NF-
B-dependent expression of anti-apoptotic
factors like Bcl-XL (Akari et al., 2001).
CTLs are responsible for the removal of virus-infected
cells from the body, and for inducing apoptosis in HIV-infected
CD4+ cells. However, like many other viruses, HIV has
developed mechanisms to prevent, or simply delay, apoptosis of the
cells it infects. The TCR on CTLs is used to recognize infected
cells through the recognition of a non-self viral peptide in
conjunction with major histocompatibility complex class I (MHC I).
The down-regulation of MHC I surface expression by Nef (Schwartz
et al., 1996) and Tat (Howcroft et al.,
1993) avert this recognition and can
prevent CTL-mediated killing of the HIV-infected cell. If the cell
were to receive a death signal, HIV has a secondary defence already
in place. Vpr protein, which is expressed at low levels in the
cell, is responsible for the increase in anti-apoptotic Bcl-2,
while simultaneously decreasing pro-apoptotic Bax (Conti et
al., 1998). In contrast, other evidence
indicates that Vpr induces a prolonged G2 cell cycle
delay followed by death in M phase (Watanabe et al.,
2000), and that apoptosis is mediated
through interaction of Vpr with the mitochondrion permeability
transition pore, which opens the pore causing mitochondrial
swelling and release of cytochrome c (Jacotot et al.,
2001; Roumier et al., 2002).
The pro-apoptotic functions of Vpr may be secondary effects
dependent upon induction of cell cycle arrest. At the same time,
Vpr transactivates the viral promoter, the long terminal repeat
(LTR), to increase virus replication (Gummuluru & Emerman,
1999). The cytoprotective effects of Vpr
may play a role during early infection, allowing productive virus
replication, but ultimately it promotes apoptosis toward the later
stages. Finally, Nef, Vpu and Env all decrease CD4 on the surface
of infected cells (Crise et al., 1990;
Willey et al., 1992; Salghetti et al., 1995).
The fact that three HIV proteins have this function implies a
critical role in the survival and replication of the virus.
Down-regulation of CD4 would prevent a super infection and
Env-induced apoptosis through the CD4 molecule. Overall, HIV has
evolved multiple mechanisms to promote survival for long enough to
ensure a productive infection and this may be supported by the fact
that infected cells do not undergo apoptosis as readily as
uninfected bystander cells (Finkel et al., 1995).
However, even infected cells have a shortened lifespan and,
therefore, partially contribute to the overall decrease in Th cells
(Ho et al., 1995; Wei et al., 1995;
Perelson et al., 1996).
| CELL DEATH IN UNINFECTED CELLS |
Apoptosis plays a major role in killing uninfected
cells. Although the Th cells infected by HIV can be directly killed
by the virus, or by HIV-specific CTL, there are generally more
apoptotic cells than infected cells (Embretson et al.,
1993). This is confirmed by direct
evidence in lymph nodes where apoptosis was seen primarily in the
uninfected bystander cells (Finkel et al., 1995).
There are two mechanisms by which uninfected cells could be killed:
either by HIV proteins released from infected cells acting on
neighbouring uninfected cells, or by activation-induced cell death
(AICD).
Effect of apoptotic HIV proteins on uninfected cells
Inactivated HIV virions (Esser et al., 2001 Soluble gp120 induces apoptosis in uninfected Th cells
through cross-linking of the CD4 molecule (Banda et al.,
1992 The HIV protein Tat, secreted from infected cells
(Chang et al., 1997 HIV Nef protein released into the extracellular matrix
induces death in neuronal cells (Trillo-Pazos et al.,
2000 Activation-induced cell death (AICD) in uninfected
cells
One interesting feature of apoptosis is that many of
the molecular steps required for apoptosis are shared by cellular
activation pathways. For example, the nuclear condensation that
occurs in programmed cell death is similar to that which occurs
prior to cell proliferation. Also, caspase activation, long known
to be associated with apoptosis, occurs when cells become activated
and proliferate (Alam et al., 1999 The binding of Env to the CD4 molecule renders the
CD4+ cell more susceptible to Fas-mediated killing
(Algeciras et al., 1998 There are other HIV proteins that do not increase CD95
expression but still sensitize the cells to Fas-induced death. The
Tat protein has the ability to increase pro-apoptotic proteins
caspase 8 and Bax, while simultaneously decreasing the
anti-apoptotic Bcl-2 (Sastry et al., 1996 It is not surprising, given the death of Th lymphocytes
in HIV infection, that there is a significant dysregulation of
cytokine responses, which likely influences Th cell apoptosis
susceptibility. It has been hypothesized that as HIV disease
progresses there is a shift in the cytokine response from a
predominantly type-1 cellular immune response (IFN- Resistance to apoptosis in vitro is associated
with a predominant type-1 response (Gougeon & Montagnier,
1993 Although some of the strongest pro-apoptotic cytokine
signals are TNF- The role of T cell growth factors IL-2 and IL-15 in
HIV-induced apoptosis is also unclear. IL-2 displays both pro- and
anti-apoptotic properties depending on the cellular
microenvironment or activation status of a particular cell.
Treatment with recombinant IL-2 or IL-15 can increase or decrease a
cell's susceptibility to apoptosis. These discrepancies may
be due to the activation status of the cells undergoing apoptosis.
IL-2 and IL-15 were shown to increase susceptibility to
CD95-mediated apoptosis (Naora & Gougeon, 1999 It is clear that cytokines play a multifactorial role
in apoptosis during HIV disease pathogenesis. The loss of
anti-apoptotic cytokines IFN- HIV causes AIDS and the loss of Th lymphocytes is a
central factor in the progression of the disease. Due to the vital
role of these cells in regulating and amplifying the immune
response, any decline in their number results in deficits in both
humoral and cell-mediated immunity. Understanding how these cells
are eliminated is critical to the development of new, effective
therapies for AIDS. However, difficulties arise as there are
multiple mechanisms involved in the death of the Th lymphocytes.
HIV-infected Th lymphocytes have a shortened lifespan due to
syncytia formation, lysis by CTL and direct cytopathic effects of
HIV, but the number of apoptotic cells in infected individuals
greatly exceeds the number of HIV-infected cells. This suggests
that HIV has additional detrimental effects on the uninfected
bystander Th cell population. These bystander cells are attacked by
two differing mechanisms. Firstly, several HIV proteins, whether
attached to the virion or released by infected cells, can utilize a
number of disparate death pathways to initiate apoptosis in
uninfected cells. It is unclear in the in vivo situation
what the relative contribution that gp120, Tat or Nef makes to the
overall level of apoptosis; however, a therapy may be developed
that would require the neutralization of all of their effects. On
the other hand, there may be a point further along at which the
HIV-induced apoptotic pathways converge. If we were to identify
this point then blocking death may require intervention at a single
focal point. Clearly more understanding of the death pathways
initiated by these proteins is required. The second mechanism of
bystander killing is AICD. HIV-infected individuals have higher
levels of immune activation, which have been suggested to
contribute to increased apoptosis. The dominant death pathway in
AICD is through Fas and, therefore, inhibiting this path would be a
logical point for therapeutic intervention. Currently, it is
unknown whether the HIV proteins or AICD kill the majority of
uninfected cells. Further studies of AICD in models of HIV and
chronic immune stimulation could help determine its significance in
disease progression.
Local environmental factors can also influence cell
survival and the effectiveness of the HIV-specific immune response.
During HIV infection the switch from the production of some type-1
cytokines to other type-2 cytokines could be very important since,
respectively, they are associated with either protection from or
enhancement of AICD.
This review has highlighted the many different modes by
which HIV induces apoptotic cell death in both infected and
uninfected Th cells. One of the ultimate goals of research in this
field is to prevent or minimize death in Th cells. If this were
achieved it may be possible to convert HIV infection from a
progressively immunosuppressive and ultimately fatal disease to a
chronic manageable infection.
J. B. A. is supported by a postdoctoral fellowship from
the Canadian Institutes for Health Research (CIHR). The authors
would also like to thank the CIHR, the Manitoba Health Research
Council and CANVAC for their financial support and Dr Jody Berry
and John Rutherford for a critical review of this manuscript.
Adachi, Y., Oyaizu, N., Than, S.,
McCloskey, T. W. & Pahwa, S. (1996). IL-2 rescues in
vitro lymphocyte apoptosis in patients with HIV infection:
correlation with its ability to block culture-induced
down-modulation of Bcl-2. J Immunol 157,
4184–4193.
Evan, G. & Littlewood, T.
(1998). A matter of life and cell death. Science
281, 1317–1322.
Golstein, P. (1998). Cell death
in us and others. Science 281, 1283.
Gougeon, M. L. & Montagnier, L.
(1993). Apoptosis in AIDS. Science 260,
1269–1270.
Nagata, S. & Suda, T. (1995).
Fas and Fas ligand: lpr and gld mutations. Immunol Today
16, 39–43.
Strasser, A., O'Connor, L.
& Dixit, V. M. (2000). Apoptosis signaling. Annu Rev
Biochem 69, 217–245.
© 2003 SGM This article is now available in the July
2003 print issue of JGV (vol. 84, 1649–1661).
The complete issue of the journal may be seen in electronic form on JGV
Online.)
and HIV proteins released into the extracellular environment can
have dramatic effects on uninfected cells. HIV proteins such as
gp120, Tat, Nef and Vpu have been shown to induce cell death in
uninfected cells.
). The binding initiates part of the
T cell activation pathway; however, in the absence of the TCR being
specifically activated, this signal results in apoptosis or
inhibition of antigen-induced cell activation (Marschner et
al., 2002). Soluble and membrane-bound gp120
induce, through cell receptors such as CD4, CXCR4 and CCR5, both
Fas-dependent (upregulation of Fas/FasL, decreased FLIP) and
Fas-independent (increased Bax, decreased Bcl-2) apoptotic pathways
(Arthos et al., 2002). Recently, CCR5-tropic viruses were
shown to induce Fas and caspase 8-dependent apoptosis of uninfected
Th cells (Algeciras-Schimnich et al., 2002).
Cell surface presentation of gp120 can induce a bystander cell
death that requires close cell-to-cell contact and gp41 function
(Blanco et al., 2003). gp120 also has an inhibitory
effect when cells at the G0/G1 phase of the
cell cycle, such as naive T cells, are most sensitive to
gp120-mediated negative signalling, whereas memory T cells are less
affected.
), can be endocytosed by neighbouring
cells (Ensoli et al., 1990; Mann & Frankel, 1991;
Zagury et al., 1998). Tat upregulates caspase 8 (Bartz
& Emerman, 1999) and FasL (Li-Weber et al.,
2000), and induces apoptosis in neurons
(New et al., 1997) and Th cells (Li et al.,
1995). A Fas-independent Tat-mediated
mechanism of bystander T cell death has also been suggested (Zhang
et al., 2001). Tat can upregulate TNF-related
apoptosis-inducing ligand (TRAIL) on monocytes which may then
interact with uninfected T cells to induce apoptosis (Zhang et
al., 2001). However, Tat also protects T cells
from TRAIL-induced apoptosis (Gibellini et al., 2001).
Other anti-apoptotic effects of exogenous Tat include upregulation
of Bcl-2, which was observed in the Jurkat cell line and PBMCs
(Zauli et al., 1995).
) and a wide range of blood cells by
a Fas-independent mechanism (Okada et al., 1997,
1998). Much of the toxicity of Nef is
likely due to its myristylated N terminus, which can insert into
the plasma membrane and induce cell death in uninfected
CD4+ and CD4– T cells (Azad,
2000). It has been suggested that Nef
plays a role in allowing HIV-infected cells to evade the immune
response by inducing cytotoxic activity in uninfected
CD8+ T cells (Silvestris et al., 1999)
and down-regulating CD4 expression in neighbouring CD4+
T cells (Pugliese et al., 1999).
Also, the extracellular addition of Vpu, or its C terminus, can
cause membrane disruption and induce cell death in CD4+
and CD4– cells (Azad, 2000).
Clearly a number of extracellular HIV products are capable of
inducing apoptosis in uninfected cells.
;
Kennedy et al., 1999). With these shared similarities it
may not be surprising that activation and cell death are closely
linked. AICD is a normal multi-step regulatory mechanism that
primes a cell for death to limit an activated immune response
(Hanabuchi et al., 1994). Priming can be achieved either by
repeated stimulation through CD3/TCR (Kabelitz et al.,
1995), sole stimulation through the CD4
receptor (Banda et al., 1992)
or activation without co-stimulation (Borthwick et al.,
2000). Although not normally expressed at
high levels on resting T cells, Fas and FasL can be induced (Suda
et al., 1995). In fact, allo-stimulation induces
the expression of Fas and FasL on the surface of T cells
(O'Flaherty et al., 2000).
In HIV infection excessive immune activation has been suggested to
induce apoptosis through Fas/FasL (Badley et al., 1998,
1999; Dockrell et al., 1999).
) and also causes an increase in
effector caspase 3, 6 and 8 activity, which can be blocked with
soluble CD4 (Cicala et al., 1999,
2000; Algeciras-Schimnich et al.,
2002).
This suggests a CD4-dependent signalling mechanism, which is also
known to block antigen-specific activation of primary lymphocytes
(Marschner et al., 2002). T cells acquire an AICD-resistant
phenotype when correctly activated by antigen. Engagement of CD4 by
Env before TCR signalling prevents the upregulation of the Fas
death pathway regulator FLIP, changing T cells from Fas-resistant
to Fas-sensitive (Somma et al., 2000).
;
Bartz & Emerman, 1999). Recent data have suggested that
elevated levels of AICD in HIV infection may also be the result of
altered FasL levels. Tat upregulates FasL expression through Egr
transactivation of the FasL promoter resulting in increased AICD
(Yang et al., 2002).
CYTOKINE RESPONSES IN HIV DISEASE AND THEIR EFFECT ON APOPTOSIS 
, TNF-
,
IL-12), to a type-2 humoral response (IL-4, IL-5, IL-10, IL-13)
(Clerici & Shearer, 1994). However, additional studies of
cytokine profiles and HIV disease progression fail to completely
corroborate these findings (Galli et al., 2001).
Type-1 cytokine responses decrease as HIV disease progresses. While
there is no clear evidence that type-2 responses increase, there is
no concurrent decrease in type-2 responses. One possible
explanation for the decrease in type-1 responses may be the
increased susceptibility of cells expressing IFN- or TNF- to
AICD, presumably due to decreased Bcl-2 expression in these type-1
Th cells (Ledru et al., 1998).
Thus, during HIV disease progression, dysregulation of cytokine
responses results in a diminished ability to mount effective type-1
cytokine responses.
). Therefore, deficiencies in type-1
responses could have drastic effects on apoptotic events regulating
normal T cell homeostasis, and on HIV-induced AICD. IL-12 can
protect against Fas-mediated apoptosis and AICD in HIV-infected
patients (Estaquier et al., 1995).
In addition, AICD in lymphocytes isolated from HIV-infected
patients was blocked by the addition of recombinant IL-12 and
IFN-, whereas type-2 cytokines IL-4 and IL-10 were shown to
increase susceptibility to AICD (Clerici et al., 1996).
Recent gene expression data demonstrated that IFN- can strongly
induce a number of pro-apoptotic genes in the TNF superfamily
(Stylianou et al., 2002), suggesting that susceptibility to
apoptosis may not exactly fit the type-1/-2 paradigm. Regardless of
the assignation of these cytokines to a type-1 or type-2 phenotype,
these data strongly suggest that cytokine dysregulation plays an
important role in HIV-induced apoptosis. As HIV disease progresses,
not only is apoptosis of Th lymphocytes enhanced due to the dual
effects of pro-apoptotic HIV proteins and increased AICD, but
cytokine dysregulation is likely to amplify these effects by
providing a pro-apoptotic environment.
and TNF family members, their role in HIV-induced
apoptosis is not clear. It is apparent that regulation of TNF is
severely dysregulated in HIV-infected patients (Zangerle et
al., 1994), as both HIV proteins and HIV
infection can induce strong TNF responses in lymphocytes
(Capobianchi, 1996). In fact, HIV infection of
lymphocytes, or monocytes, induces TNF that in turn activates
NF-B, which induces high levels of HIV and TNF transcription (Han
et al., 1996). In addition, serum TNF levels have
been shown to be elevated in HIV-symptomatic but not asymptomatic
individuals (Hober et al., 1996).
TNF and TNF family members can initiate apoptosis immediately via
membrane-bound receptors, and although it is apparent that TNF
levels are indeed dysregulated in HIV infection, there is limited
evidence for TNF-mediated apoptosis of infected cells directly, or
by acting on bystander lymphocytes. Apoptosis of bystander Th cells
can be reduced by the addition of soluble TNF receptor decoys
(Srivastava et al., 1999) and can protect against HIV
protein-induced apoptosis (Cicala et al., 2000).
However, clinical trials of anti-TNF therapies have failed to show
improvement in either immunological or clinical outcome (Walker
et al., 1996), making the role of TNF in
HIV-induced apoptosis unclear. This is not surprising considering
the multifactorial role of TNF in inducing cell activation and
death. It is likely that the role of anti-apoptotic signals that
negatively regulate TNF and TNF family member signalling such as
caspase 8 may play a more important role in determining whether or
not a particular cell undergoes apoptosis. The redundancy in
cytokine function makes elucidation of a role in induction of
apoptosis difficult.
)
while protecting against spontaneous apoptosis (Adachi et
al., 1996). The latter effect may be due to
the ability of IL-2 and IL-15 to upregulate Bcl-2 expression (Akbar
et al., 1994). Clinically, IL-2 has been useful
in the treatment of HIV infection under certain conditions. IL-2
can increase Th cell survival independent of HIV replication,
presumably by decreasing the apoptosis of bystander Th cells. Also,
treatment with recombinant IL-2 can reduce overall levels of
apoptosis in HIV-infected, but not uninfected, individuals (Adachi
et al., 1996), further underscoring the
importance of cellular activation, and the cytokine environment in
the action of these apoptotic mediators. IL-2 and IL-15 share many
similarities in their action due to utilization of common receptor
elements (Giri et al., 1995). Of considerable interest is the
observation that IL-15 may be a more potent inhibitor of apoptosis,
which is likely because IL-15 (unlike IL-2) does not appear to
induce HIV replication (Kovacs et al., 2001).
Also, the relative levels of each cytokine in the apoptotic
microenviroment are crucial to their ability to induce, or
alternately suppress, apoptosis. A mouse model on senescence
suggests that Th cells incapable of producing sufficient levels of
IL-2 were unable to proliferate and were susceptible to apoptosis,
but could be rescued by the addition of exogenous IL-2 (Nishimura
et al., 2002). Thus, the role of IL-2 and IL-15
in apoptosis is likely to be complex due to the immunoregulatory
roles for these cytokines in cellular activation. Determination of
whether a cell undergoes apoptosis will depend not only on the
levels of IL-2 and IL-15, but also on the presence or absence of
other appropriate pro- or anti-apoptotic signals such as TNF, other
cytokines or Bcl-2 expression. These molecules may act directly
through mediation of cellular activation, indirectly through the
induction of other anti-apoptotic mediators such as Bcl-2 or even
through more downstream regulatory events such as the abilities of
these cytokines to induce mediators like IFN- that regulate
apoptosis in a more direct manner.
and IL-12 during disease progression
and the increase in pro-apoptotic cytokines IL-4 and IL-10 are
likely to amplify the role of apoptosis in the loss of Th cells.
TNF- and the TNF family of cytokines can induce apoptosis in a
direct manner by inducing the appropriate (or inappropriate)
apoptotic signals, while IL-2 and IL-15 appear to act in a more
indirect manner via mediation of cell activation. Cytokines play an
important role in apoptosis, and this is likely to be even more
important during HIV infection when normal cytokine responses
become dysregulated.
CONCLUSION
REFERENCES
JGV Direct table of contents
