You are on page 1of 11

REVIEWS

DCs and NK cells: critical effectors in


the immune response to HIV-1
Marcus Altfeld*, Lena Fadda*, Davor Frleta‡ and Nina Bhardwaj‡
Abstract | Dendritic cells (DCs) and natural killer (NK) cells have central roles in antiviral
immunity by shaping the quality of the adaptive immune response to viruses and by mediating
direct antiviral activity. HIV-1 infection is characterized by a severe dysregulation of the antiviral
immune response that starts during early infection. This Review describes recent insights into
how HIV-1 infection affects DC and NK cell function, and the roles of these innate immune cells
in HIV-1 pathogenesis. The importance of understanding DC and NK cell crosstalk during HIV
infection for the development of effective antiviral strategies is also discussed.

Despite some recent progress in HIV vaccine strategies, (TABLE 1). Both subsets specialize in detecting viruses
HIV infection remains a worldwide problem. Growing and initiating innate and adaptive immune responses
evidence indicates that the immune system is dysregu- that lead to viral elimination or control. DCs express
lated during HIV infection, and this compromises the several receptors for recognizing viruses3, including pat-
efficacy of potential immune therapies for HIV. Humoral tern recognition receptors (PRRs) such as the Toll-like
immune responses occur late in HIV-1 infection, with receptors (TLRs) and C-type lectins. DCs detect viruses
neutralizing antibodies appearing 3 months or more in peripheral tissue sites and, following activation and
after initial infection1. T cell responses are elicited around viral uptake, migrate to draining lymph nodes, where
1–2 weeks after infection, but are largely ineffective owing they trigger adaptive immune responses and promote
to the early emergence of antigen escape variants of HIV1. NK cell activation4 (TABLE 1). Activated cDCs produce
Recent reports have highlighted the dual role of the cytokines such as interleukin-12 (IL-12), IL-15 and IL-18.
innate immune system in early viral control and in con- IL-12 is crucial for cDCs to induce T helper 1 (TH1) cell
tributing to disease pathology. Dendritic cells (DCs) and responses, which subsequently promote the potent cyto-
natural killer (NK) cells are vital mediators of the innate toxic T lymphocyte (CTL) responses that are necessary
immune system and promote the development of adap- for clearing virus-infected cells5. Both IL-12 and IL-15
tive immune responses. DCs are crucial for activating and produced by cDCs can activate NK cells4 (TABLE 1). pDCs
conditioning virus-specific T cells, a process that is largely produce more type I interferons (IFNs) in response to
influenced by the preceding innate immune response. HIV than any other cell type in the body, and stimulate
NK cells impede the early spread of viruses by producing cDCs in a bystander fashion as well as directly activating
cytokines and directly killing infected cells. HIV vaccine NK cells6. In the following section, we describe recent
strategies that use DCs, through either in vitro manipula- observations that have been made concerning DC biol-
*Ragon Institute of MGH, tion of DCs isolated from patients or in vivo targeting of ogy and function during HIV-1 infection. Specifically, we
MIT and Harvard, DC subsets, are currently being investigated. The success focus on how DCs bind and recognize HIV virions, and
Massachusetts General
Hospital, Charlestown,
of these approaches will depend on a proper understand- how DCs are in turn modulated by the virus, ultimately
Massachusetts 02129, USA. ing of how DC biology is affected by HIV-1 infection. NK leading to their dysregulation in vivo. The implications of

New York University Cancer cells may be crucial for early control of HIV infection and these events underlie the need to develop vaccine strate-
Institute, NYU Langone can have important roles in editing the function of DCs, gies that enhance the ability of DCs to prime potent T cell
Medical Center, New York,
thereby affecting the ability of DCs to prime antiviral responses that can subvert the potential inhibitory effects
New York 10016, USA.
Correspondence to M.A. effector T cells. This Review focuses on the roles of these of HIV on the immune system.
and N.B. two innate cell types during HIV-1 infection.
e-mails: Binding and recognition of HIV by DCs
maltfeld@partners.org; DCs bridge innate and adaptive immunity HIV entry receptors. Binding and internalization of HIV
Nina.Bhardwaj@nyumc.org
All authors contributed
Human DCs are rare potent antigen-presenting cells by DCs is mediated by the various HIV entry recep-
equally to this work. that can be generally divided into myeloid CD11c+ ‘con- tors (which include CD4, CC-chemokine receptor 5
doi:10.1038/nri2935 ventional’ DCs (cDCs) or plasmacytoid DCs (pDCs)2 (CCR5) and CXC-chemokine receptor 4 (CXCR4)).

176 | MARCH 2011 | VOLUME 11 www.nature.com/reviews/immunol

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

Table 1 | DC subsets and their role in HIV infection


Conventional DCs (HLA-DR+CD11c+) Plasmacytoid DCs Refs
(HLA-DR+CD123+)
Langerhans cells Dermal DCs Blood DCs
CD103– CD103+ BDCA3– BDCA3+
Location Epidermis, gut Dermis Dermis Blood, Blood • Blood 2,7,
lumen secondary • Secondary lymphoid organs 9,10,
lymphoid • Peripheral tissues (for example, skin 15,27
organs and lungs)
C-type lectin Langerin DC-SIGN, Langerin, DEC205, CLEC9A BDCA2
expression DEC205 DEC205 DCIR
Role in HIV Internalize HIV Bind to HIV Unknown Unknown Unknown • Produce type I IFNs, which inhibit 9,10,
infection into degradative and transmit viral replication and induce bystander 43,45,
Birbeck granules the virus T cell death 53
to T cells • Induce TReg cells through HIV-induced
in draining IDO expression
lymph nodes • Recruit T cells to sites of HIV infection
(and thus facilitate viral spread) by
producing chemokines such as CCL5
TLR TLR2,3,5* TLR2,3,4,5* TLR2,3,4,5* TLR2,3,4,7,8 TLR2,3,8 TLR7,9 2,
expression 15,27
Cytokine IL-12, IL-15, IL-23, IL-6, TNF, IL-1β IFNα, IFNβ, IL-6, TNF 2,4–6
production
Function • Priming of antigen-specific CD4+ T cells, CD8+ T cells and B cells • Induction of TReg cells 2,4–6
• NK cell activation through IL-12 • Induction of plasma cells
• NK cell activation through type I IFNs
Pathology • Reduced frequency in peripheral blood • Reduced frequency in peripheral 25,
during HIV • Conflicting reports of functionality or dysfunction blood 26,28,
infection • Conflicting reports of functionality or 31–37
dysfunction
BDCA, blood DC antigen; CCL5, CC-chemokine ligand 5 (also known as RANTES); CLEC, C-type lectin; DC, dendritic cell; DCIR, DC immunoreceptor (also known as
CLEC4A); DC-SIGN, DC-specific ICAM3-grabbing non-integrin; IDO, indoleamine 2,3-dioxygenase; IFN, interferon; IL, interleukin; NK, natural killer; TLR, Toll-like
receptor; TNF, tumour necrosis factor; TReg, regulatory T. *There have been confounding reports regarding TLR distribution on Langerhans cells and dermal DCs.

Other chemokine receptors, including CCR3, CCR8, and specifically blocks the HIV–DC-SIGN interaction,
CCR9 and CXCR6 (REFS 7,8), have also been suggested thereby preventing HIV transmission to CD4+ T cells.
to participate in HIV-1 entry. In addition, cDCs express The different efficacies with which these two forms of
C-type lectins that bind to HIV, including DC-specific clusterin block HIV–DC-SIGN interactions may explain
ICAM3-grabbing non-integrin (DC-SIGN), langerin why sexual exposure to HIV is less likely to result in
(also known as CLEC4K or CD207) and DC immu- infection than blood-borne exposure12. DCIR was also
noreceptor (DCIR; also known as CLEC4A)7,9 (TABLE 1). recently found to bind to HIV virions and, in a similar
Viral synapse
Expression of these receptors varies based on cDC sub- way to DC-SIGN, promotes transmission of infectious
A polarized synaptic contact type, localization and activation state. Although pDCs virus to CD4+ T cells13.
point for cell to cell express C-type lectins, such as blood DC antigen 2 The C-type lectin langerin, which can bind and inter-
transmission of viruses. These (BDCA2; also known as CLEC4C), binding of HIV to nalize HIV, is expressed by epidermal Langerhans cells,
contacts can occur between
pDCs is generally mediated by interaction of the HIV and migratory Langerhans cells are believed to enhance
DCs and CD4+ T cells.
gp120 envelope protein with CD4 (REF. 10) (TABLE 1). sexually transmitted HIV infection by promoting the
Clusterin The identities of the receptors involved in virus–cDC spread of the virus to T cells in draining lymph nodes.
A ubiquitously expressed interactions can affect virus fate. Binding of HIV to In contrast to DC-SIGN-mediated internalization, HIV
glycoprotein that functions as DC-SIGN through its gp120 envelope protein leads internalized through langerin is trafficked to Birbeck
an extracellular chaperone. It is
widely expressed in various
to internalization of the virus into DC early endosomal granules, where the virus is rapidly degraded14; this sug-
types of cancer. compartments, where it is not degraded. This may gests that Langerhans cells also function as a barrier
enable DCs to deliver intact HIV virions to draining to HIV infection. However, high viral concentrations
Birbeck granule lymph nodes, where DC–T cell interactions could pro- enable langerin-independent internalization of HIV
A membrane-bound structure
mote infection of T cells through viral synapses and/ by Langerhans cells, resulting in the transport of intact
that is found in the cytoplasm
of Langerhans cells. These or cell–cell fusion. However, the contribution of this HIV to T cells by migratory Langerhans cells. In accord-
granules are rod- or tennis mode of transmission has been called into question, as ance with this, others have shown that in vitro-derived
racket-shaped with a central the transfer of HIV-1 to T cells can occur independ- Langerhans cells can transmit infectious HIV to T cells15.
linear density. Their formation ently of DC-SIGN11. Conversely, another report has Thus, the fate of HIV varies depending on the particular
is induced by langerin, an
endocytic C-type lectin
highlighted how interaction of HIV with DC-SIGN cDC receptors used to bind the virus, and this empha-
receptor that is specific to may be important for transmission of HIV. Clusterin in sizes how different DC subsets can either contribute to
Langerhans cells. semen is glycosylated differently to clusterin in blood or hinder HIV infection.

NATURE REVIEWS | IMMUNOLOGY VOLUME 11 | MARCH 2011 | 177

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

Innate recognition of HIV by DCs. Whereas binding Candida albicans or Mycobacterium tuberculosis also
and internalization of HIV is mediated by a variety of triggers RAF1-dependent phosphorylation, suggesting
DC-expressed surface receptors, activation of DCs by that co-infection could enhance the productive tran-
HIV is mainly thought to involve intracellular mem- scription of HIV24. Although this in vitro work argues for
bers of the TLR family. Human pDCs express TLR7 a role of TLR8 in promoting HIV replication in cDCs,
and TLR9 (REF. 16), and recognition of HIV-1 single- it is unclear whether this is a physiological mechanism
stranded RNA by TLR7 leads to pDC production of for productive infection in cDCs, as replication within
type I IFNs and other inflammatory cytokines. The ini- cDCs is very low. However, it raises the issue of whether
tial gp120–CD4 interaction promotes endocytosis of the use of TLR8 agonists as immunotherapeutic adju-
HIV-1 by pDCs and the subsequent activation of TLR7 vants might increase productive replication within cDCs
by viral RNA17. Genomic RNA from HIV-1 contains containing integrated HIV-1.
several immunostimulatory GU-rich sequences, which
activate the TLR7 pathway 18. Following the detection Autophagy and HIV recognition. Autophagy leads to the
of HIV by this route, pDCs rapidly secrete high levels of delivery of cytosolic components (such as signalling mol-
IFNα owing to their constitutive expression of IFN ecules and damaged organelles) to lysosomal compart-
regulatory factor 7 (IRF7)19. ments, which contain PRRs25. Following HIV-1 infection,
However, HIV does not induce the maturation of fusion between endosomes and autophagosomes is
cDCs by activating TLRs17. This is surprising consider- inhibited, thereby preventing autophagy-mediated viral
ing that blood cDCs express TLR7 and TLR8, both of degradation26. This inhibition occurs through activation
which can be activated by HIV-derived single-stranded of mammalian target of rapamycin (mTOR), a serine/
RNA. One possible explanation is low viral replica- threonine kinase that is a negative regulator of autophagy.
tion within cDCs; only 1–3% of cultured DCs become In accordance with inhibition of autophagy, HIV-1 rep-
infected in vitro, and circulating blood DCs isolated from lication in DCs leads to decreased cathepsin activity in
patients with HIV do not appear to be infected with the these cells, possibly owing to a blockade of lysosomal
virus20,21. A recent study using a pseudotyped virus that fusion, which is necessary for cathepsin activation. This
lacked HIV envelope protein showed that when this results in enhanced survival of HIV-1 in phagosomes and
strain of HIV-1 carries the vpx gene, infection of DCs decreased presentation of viral antigens27–29.
promotes their maturation and production of type I IFNs, In contrast to the aforementioned study, which
and facilitates antiviral T cell immunity. This response is showed that HIV can activate TLR8, HIV-mediated inhi-
mediated by interaction of newly synthesized HIV-1 cap- bition of autophagy was shown by others to inhibit DC
sid with cellular cyclophilin A (also known as PPIase A) responsiveness to the classical TLR4 and TLR8 agonists
and activation of the type I IFN-inducing transcription lipopolysaccharide (LPS) and single-stranded RNA26.
factor IRF3 through an unknown cytoplasmic sen- One explanation for this discrepancy may be that acti-
sor 22. Thus, it is possible that there are too few virions vation of TLR8 by HIV precedes autophagy inhibition.
within HIV-1-exposed cDCs to trigger TLR signalling, Alternatively, selective blockade of autophagosomes by
or that vpx is essential for this interaction. However, HIV may depend on the particular HIV receptor used
although viral replication in pDCs is also low, pDCs for viral entry. Clearly, additional studies are necessary to
rapidly respond to HIV-1 through TLR7. Therefore, understand how HIV-1 can modulate cDC responsive-
another explanation is that HIV interaction with C-type ness to TLR agonists. It will also be important to deter-
lectins may abrogate subsequent TLR responsiveness in mine how HIV affects different blood cDC subtypes, such
cDCs. HIV, in a similar way to Mycobacterium spp., has as the newly described BDCA3+ (also known as CD141+)
been implicated in inhibiting TLR stimulation of cDCs blood DC subset30.
through its interactions with DC-SIGN23 and this is
discussed in more detail below. Dysregulation of DCs in HIV-1 infection
Decreased DC frequency. HIV-1 infection leads to a
Can TLR signalling promote HIV replication in DCs? progressive reduction in blood DC numbers (both cDCs
De novo replication of integrated HIV-1 in imma- and pDCs), which correlates with increasing plasma
ture cDCs can be initiated by TLR8- and DC-SIGN- virus load and disease progression31. This decrease in
mediated signal transduction events. HIV is targeted DC frequency occurs early during acute HIV infection
to TLR8-containing endosomal compartments by and is sustained in the later stages of infection20. Patients
DC-SIGN, and this initiates the transcription of inte- with chronic HIV infection also have fewer blood DCs
grated HIV-1 DNA through TLR8 activation and sub- compared with uninfected controls20,31.
sequent nuclear factor-κB (NF-κB) signalling 24. The Direct infection of DCs has been suggested to be
Autophagy
binding of gp120 to DC-SIGN leads to RAF1-mediated responsible for decreased DC frequency during HIV
An evolutionarily conserved
process in which acidic phosphorylation of the p65 subunit of NF-κB, thereby infection31. Several studies have evaluated whether DCs
double-membrane vacuoles permitting transcription elongation of nascent HIV-1 from the blood of patients with chronic HIV-1 infection
sequester intracellular transcripts and productive transcription within infected are infected with HIV, with some showing evidence of
contents (such as damaged DCs24. Thus, ligation of both TLR8 and DC-SIGN is infection but others suggesting that no infection occurs21,32.
organelles and
macromolecules) and target
required for the induction of signal transduction path- Given the low level of HIV-infected DCs, it is unlikely
them for degradation, through ways that promote the synthesis of complete viral tran- that direct infection can adequately explain the decreased
fusion to secondary lysosomes. scripts from integrated proviral DNA. Co-infection with DC frequency in the blood of patients with HIV.

178 | MARCH 2011 | VOLUME 11 www.nature.com/reviews/immunol

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

Instead, the decline in blood DC frequency during factors have been implicated in promoting the chronic
viral infection may be due to indirect mechanisms. immune activation that is observed during HIV infec-
Aberrant IFNα production during HIV infection may tion, either by direct activation of DCs or indirectly
impede the differentiation of cDCs from monocytes33 through stimulation of cells such as macrophages 42.
or other precursors. Also, DCs from patients with HIV Such stimuli may lead to partial maturation of cDCs
show increased apoptosis compared with DCs from in vivo, making these cells tolerant to subsequent stimuli
uninfected individuals34. ex vivo. Accordingly, several reports have indicated that
A third possibility is that decreased DC frequen- cDCs from chronically infected patients are less effi-
cies in the blood may be due to the redistribution of cient at stimulating T cell activation than cells from
DCs to secondary lymphoid organs. Indeed, several uninfected individuals31,40.
groups have reported an increase in pDCs in the lymph Understanding exactly how cDCs are functionally
nodes35,36 and spleens37 of patients with HIV. The over- impaired during HIV infection is crucial for effective
all functional capacity of DCs found in the secondary HIV vaccine design. This is especially true if functional
lymphoid organs of patients with HIV remains to be impairment of DCs results from bystander mechanisms
adequately evaluated. during HIV infection, as these mechanisms are likely to
Finke et al.38 reported that cDC numbers returned impede most DC-targeting approaches, including the
to normal levels following antiretroviral therapy (ART). reintroduction of ex vivo-manipulated DCs into patients.
Another recent study showed no effect of primary HIV-1
infection on cDC levels, and although pDC and CD4+ Altered pDC function. As with cDC function, it is
T cell numbers were reduced, they were restored following unclear whether pDC function is impaired during HIV
ART39. The differential restoration of cDC and pDC num- infection. Several groups have argued that type I IFN
bers may be due to multiple components of the immune production by blood pDCs is attenuated during both
system going awry beyond repair in the chronic phase of acute and chronic HIV infection43,44, whereas others
infection. Altogether, these data warrant a closer scrutiny have indicated that circulating pDCs in patients with
of the functional status of DC subsets during the immune HIV viraemia show a normal type I IFN response45.
recovery associated with ART. In the former case, it is possible that circulating pDCs
were activated in vivo by HIV before isolation and were,
Altered cDC function. Although there is an overall therefore, hyporesponsive to a secondary stimulus
consensus that blood DC numbers decline during HIV delivered ex vivo. Consistent with this, activation mark-
infection, it is still unclear whether DCs from patients ers are upregulated by blood pDCs during acute HIV
with HIV are functionally impaired. As indicated previ- infection, suggesting that these pDCs have responded to
ously, HIV interaction with DC-SIGN might negatively the virus in vivo20. Furthermore, IFN-producing pDCs
affect TLR-induced activation of cDCs. Accordingly, accumulate in the draining lymph nodes of patients with
HIV-1 infection interferes with the maturation of blood HIV35,36, although these pDCs are prone to apoptosis34,36.
cDCs and monocyte-derived DCs in response to in vitro Nonetheless, evidence is emerging that pDCs have a cru-
activation stimuli28,40. cial role in the immune dysregulation that leads to the
Other studies have reported that cDCs are not func- development of AIDS.
tionally defective following HIV exposure20,21, showing pDCs that are exposed to HIV upregulate the
that ex vivo isolated cDCs from patients with acute HIV enzyme indoleamine 2,3-dioxygenase (IDO), which
infection may even be hyperresponsive to TLR stimula- metabolizes tryptophan to kynurenine, and this upreg-
tion20 and produce increased levels of IL-12, IL-6, TNF ulation is dependent on the gp120–CD4 interaction.
and CCL3 (also known as MIP1α) compared with Through expression of IDO, pDCs promote the differ-
cDCs from uninfected controls. By contrast, compared entiation of naive T cells into regulatory T (TReg) cells,
to pDCs isolated from uninfected controls, pDCs iso- which suppress effector T cell activation46. It remains
lated from patients with acute HIV infection can show to be elucidated whether increased frequencies of TReg
decreased IFNα production during the early stages of cells prevent effective HIV-specific adaptive immune
infection, but increased IFNα production at later stages responses and exacerbate HIV-associated pathology, or
of infection20. whether TReg cells increase as a result of chronic immune
These discordant results may result from the limita- activation in an effort to limit associated immuno-
tions of analysing isolated DCs ex vivo. Modulation of pathology. By promoting TReg cell responses, pDCs may
DC function in vivo may not necessarily be a result be responsible for the decrease in TH17 cells that occurs
of direct HIV exposure, but could instead be due to during HIV infection47; this decrease is thought to lead
factors that are produced by the host in response to the loss of gut integrity and microbial translocation
to infection. Moreover, depending on the stage of HIV in patients with HIV42.
infection, DCs may be affected by the microenviron-
ment. For example, during chronic HIV infection, Response to HIV-1 in humans: an overkill?
monocytes have been shown to upregulate programmed Uncontrolled and persistent inflammation contributes
cell death protein 1 (PD1) and produce IL-10, possibly to the pathology that is associated with HIV infection.
in response to the high levels of gut-derived micro- IFNα-producing pDCs can contribute to the T cell loss
bial products that are present in the plasma of these that is characteristic of patients with AIDS by upreg-
patients41. Circulating LPS and other pathogen-derived ulating CD4+ T cell expression of the pro-apoptotic

NATURE REVIEWS | IMMUNOLOGY VOLUME 11 | MARCH 2011 | 179

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

Figure 1 | cDCs and pDCs during HIV infection. Conventional dendritic cells (cDCs) are responsible for initiating
antiviral adaptive immunity. Evidence indicates that cDC function is impaired during HIV-1 infection. This is either due to
direct viral interactions with cDCs (for example, interaction with DC-SIGN (DC-specific ICAM3-grabbing non-integrin) or
blockade of autophagy) or a result of indirect mechanisms, such as the production of interleukin-10 (IL-10) by monocytes
during infection. This cDC dysfunction could contribute to a lack of effective antiviral adaptive immunity. By contrast,
plasmacytoid DCs (pDCs) are mediators of innate immunity. pDCs activated by HIV-1 produce type I interferons (IFNs),
which, in addition to inhibiting viral replication, may contribute to bystander CD4+ T cell death. Furthermore, evidence
shows that pDCs produce T cell-attracting chemokines, which may facilitate viral spread by providing a source of new
T cells for HIV to infect. Finally, HIV-exposed pDCs prime regulatory T (TReg) cells, which could impair cDC function and
block effector T cell activity, further blunting adaptive immunity. Thus, pDCs can promote deleterious immunopathology
during HIV-1 infection. IDO, indoleamine 2,3-dioxygenase; TH17, T helper 17; TLR7, Toll-like receptor 7.

molecules TNF-related apoptosis-inducing ligand Additional data from non-human primate mod-
(TRAIL; also known as TNFSF10), death receptor 5 els suggests that pDCs can facilitate HIV infection. In
(DR5; also known as TNFRSF10B), CD95 (also known the vaginal mucosa of macaques, an initial founder
as FAS) and CD95 ligand (CD95L; also known as population of T cells becomes infected with SIV but the
FASL)48,49. This may lead to the apoptosis of uninfected recruitment of additional target T cells is required for
CD4+ T cells and promote generalized CD4+ T cell loss. viral spread. It was shown that an early influx of acti-
Interestingly, HIV-infected women are almost two times vated pDCs occurs at the initial site of SIV infection
more likely to develop AIDS than HIV-infected men and these cells subsequently recruit T cells by secreting
with similar viral loads, and this may be because acti- chemokines such as CCL5 (also known as RANTES)56.
vated pDCs from HIV-infected women produce more As mentioned above, HIV-exposed pDCs can pro-
type I IFNs than pDCs from HIV-infected men50. mote the development of TReg cells, and these may pre-
Studies of non-human primate models of simian vent the induction of effector T cells by inhibiting cDC
immunodeficiency virus (SIV) infection support the activation. Thus pDCs may promote the widespread
idea that pDCs have pathological roles. Comparing dysregulation of DC function that occurs during HIV
SIV infection in African green monkeys (Cercopithecus infection (FIG. 1). Furthermore, as DCs can promote NK
aethiops), sooty mangabeys (Cercocebus atys) and cell activation, DC dysfunction may contribute to dys-
macaques (Macaca sp.), it was shown that SIV infection regulated NK cell activity during HIV infection. This is
in macaques progresses to AIDS at a rate similar to discussed in more detail below.
that in HIV-1-infected humans51. African green monkeys
and sooty mangabeys, both of which are natural hosts NK cells in HIV-1 infection
for SIV, do not develop AIDS despite showing high levels NK cells and their receptors. NK cells promote anti-
of virus replication52. Attenuated type I IFN responses viral and antitumour immunity 57 by producing pro-
are believed to contribute to the lack of SIV-induced inflammatory cytokines and by lysing infected or trans-
immune pathology in sooty mangabeys53. Gene pro- formed cells. In addition, NK cells interact with T cells
filing has shown that in both non-pathogenic models and DCs to shape the magnitude and quality of adaptive
(African green monkeys and sooty mangabeys) and immune responses58,59. To date, no specific NK cell recep-
pathogenic models (macaques) of SIV infection there is tors that directly recognize HIV-1-infected cells have been
a comparable IFN-inducible response to acute SIV infec- identified, and the NK cell response to HIV-1-infected
tion, but only in pathogenic models is this IFN response cells appears to be regulated by the balance of inhibi-
maintained during chronic infection54,55. tory and activating signals delivered to NK cells by these

180 | MARCH 2011 | VOLUME 11 www.nature.com/reviews/immunol

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

Perforin infected cells. Below, we discuss the mechanisms by which CD56hi NK cells express high levels of C-type lectins
A component of cytolytic NK cell receptors determine NK cell function in response and NCRs, but have low-level or no expression of KIRs.
granules that participates in to viral infections, with a particular focus on HIV-1. NK Although originally thought to be distinct cell lineages, it
the permeabilization of plasma cell responses are induced against cells that lack expression is now believed that these two NK cell subsets represent
membranes, allowing
granzymes and other cytotoxic
of the MHC class I molecules that bind NK cell inhibitory different stages of NK cell maturation, with CD56hi NK
components to enter target receptors60 (known as ‘missing self ’ recognition) or by cells being less mature than CD56low NK cells62,63. In addi-
cells. cells that overexpress ligands for NK cell activating recep- tion, a third population of NK cells has been described,
tors. In addition to their role in controlling NK cell effec- consisting of CD56– cells that accumulate during chronic
Granzyme A
tor functions, cognate interactions between inhibitory viral infections64–66. These CD56– NK cells express a simi-
A member of a family of serine
proteinases that are found,
receptors and their ligands during NK cell development lar receptor profile to CD56low NK cells, but are poorly
primarily, in the cytoplasmic confer both self-tolerance and functionality to NK cells, cytotoxic and do not secrete cytokines. However, cord
granules of cytotoxic T a process termed ‘licensing’ (BOX 1). blood CD3–CD56– NK cells appear to be functionally
lymphocytes and natural killer The most frequently studied inhibitory receptors competent67, suggesting differences in the differentiation
cells. Granzymes enter target
cells through perforin pores,
include the highly polymorphic killer cell immuno- status of these phenotypically similar NK cell subsets.
then cleave and activate globulin-like receptors (KIRs), which are specific for clas-
intracellular caspases to sical MHC class I molecules, and the non-polymorphic NK cells and the antiviral response. Several mechanisms
initiate target cell apoptosis. CD94–NKG2A receptor that recognizes the non-classical have been proposed for NK cell-mediated recognition
MHC molecule HLA-E. NK cell activating receptors of HIV-1-infected cells (FIG. 2). NK cells may be able to
include NKG2D, activating members of the KIR family, detect HIV-infected cells either directly through receptor-
the natural cytotoxicity receptors (NCRs) and the Fcγ mediated interactions that have not yet been identified
receptor CD16. A comprehensive list of inhibitory and or indirectly following antibody-mediated cross-linking
activating receptors expressed on NK cells is beyond the of CD16 (an Fc receptor for IgG). NK cell responses to
scope of this Review and is summarized elsewhere61. HIV-1-derived peptides have also been observed, although
Most NK cells in the peripheral circulation (approxi- whether these responses are mediated through CD16 or
mately 90%) are CD56low, constitutively produce high other NK cell receptors remains to be elucidated68,69. Thus,
numbers of cytolytic granules and have the capacity to identification of receptors that mediate NK cell recogni-
spontaneously lyse target cells in the absence of prior tion of HIV-1 remains an important research topic.
sensitization. CD56low NK cells express abundant levels A number of studies have suggested that HIV-1 uses
of KIRs, C-type lectins and NCRs, and are thought to be specific strategies to evade NK cells. The HIV protein
functionally and phenotypically mature cells. The remain- Nef (negative factor) is known to selectively downregu-
ing 10% of circulating NK cells are CD56hi and are often late expression of HLA-A and HLA-B, but not HLA-C or
referred to as the ‘immunoregulatory’ NK cell subset as HLA-E, in infected cells. This allows HIV to evade CTL
they are poorly cytotoxic, but following stimulation they responses, which are largely directed against HLA-A-
secrete large amounts of pro-inflammatory cytokines. and HLA-B-restricted epitopes, and prevent the killing
of HIV-infected cells by NK cells, which express inhibi-
tory receptors that bind to HLA-C and/or HLA-E70–72.
Box 1 | NK cell licensing Furthermore, Nef can impair NK cell activity by down-
regulating the expression of NKG2D ligands (specifically
Natural killer (NK) cell inhibitory receptors for self MHC class I molecules have important MICA (MHC class I polypeptide-related sequence A),
roles in controlling the NK cell response to potential target cells. More recently, it has ULBP1 (also known as N2DL1) and ULBP2 (also known
been demonstrated that these receptors ensure NK cell tolerance towards self. Studies in
as N2DL2)) on infected cells73.
humans and mice have shown that during development NK cells undergo an education
process, in which MHC class I-specific inhibitory receptors are involved in the calibration
of NK cell effector functions97,99. Engagement of these receptors with MHC class I Altered NK cell phenotype and function during HIV-1
molecules provides a positive signal to NK cells that leads to licensing of fully competent infection. HIV-1 infection is associated with a functional
mature peripheral NK cells. These mature NK cells can sense and lyse autologous cells impairment of NK cells that is evident early after infec-
that have downregulated their expression of MHC class I molecules97. NK cells that fail to tion and continues during disease progression. There is
undergo this education process, owing to a lack of MHC-specific inhibitory receptor an inverse correlation between viraemia and NK cell-
engagement, are unlicensed and have a reduced (but not absent) functional activity mediated suppression of HIV-1 replication, and this may
compared with licensed NK cells97–99. Thus, two major types of self-tolerant NK cells exist be attributed to several possible mechanisms74. First,
with regard to MHC class I: licensed NK cells, which are self-tolerant because they peripheral NK cells in HIV-1-infected individuals have
express inhibitory receptors for self; and unlicensed NK cells, which are also self-tolerant
decreased intracellular stores of perforin and granzyme A,
as they are not functionally competent. NK cell licensing does not seem to be an ‘on–off’
switch, but rather a dynamic and quantitative process125,126. The strength of the
and this may account for the decreased cytotoxic capac-
interaction between inhibitory MHC class I-specific receptors and their cognate MHC ity of NK cells75. Persistent HIV-1 viraemia has also been
class I molecules appears to be proportional to the level of functional responsiveness. NK shown to result in aberrant expression of several inhibitory
cells licensed on stronger inhibitory receptor–MHC interactions respond with increased and activating NK cell surface receptors. Lower expression
strength and frequency than NK cells licensed on weaker inhibitory signals125. Indeed, a of NCRs is associated with the decreased in vitro cytotox-
hierarchy of NK cell responses to MHC class I-negative cells has been observed that is icity of NK cells from HIV-infected individuals76, and this
proportional to the number of different killer cell immunoglobulin-like receptors (KIRs) may affect NK cell-mediated clearance of virus-infected
for self-HLA molecules that are expressed by the NK cell. Certain KIR and HLA alleles cells in vivo. However, in other studies, NCR-mediated
are also associated with more responsive NK cells100,127, suggesting that KIR–HLA NK cell activation has been suggested to contribute to
combinations can differentially affect NK cell effector capacities.
pathology by promoting the loss of uninfected CD4+

NATURE REVIEWS | IMMUNOLOGY VOLUME 11 | MARCH 2011 | 181

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

how NKG2C functions in HIV infection in the context of


diminished NCR-mediated responses. An increase in cir-
culating NKG2C+ NK cells in healthy individuals corre-
lates with cytomegalovirus (CMV)-seropositive status79,
and it is possible that the expansion of NKG2C+ NK cell
populations during HIV-1 infection may be a result of
CMV reactivation in HIV-1-infected individuals. Indeed,
perturbation of the NK cell compartment during HIV-1
infection may have serious consequences for protection
from opportunistic infections.
Finally, NK cells that express CD4 and CXCR4 can
be infected with HIV-1 in vitro, resulting in altered func-
tion81. This suggests that CD4+ NK cells may be a reservoir
for HIV-1 in vivo, and further investigation is required to
explore this possibility.

KIRs and HLA molecules in HIV infection


Early studies showed that the control of HIV replication
is associated with the expression of certain HLA-B alle-
les that have the Bw4 epitope82, including HLA-B*57 and
HLA-B*27. These molecules serve as ligands for the NK
cell inhibitory receptor KIR3DL1 and as putative ligands
for the NK cell activating receptor KIR3DS1 (FIG. 2).

KIR3DS1 and HLA-Bw4-80I. Martin et al.83 first dem-


onstrated the impact of KIRs and HLA molecules on
the outcome of HIV infection. The first study to dem-
onstrate a link between expression of KIR and HLA
Figure 2 | NK cell-mediated recognition of HIV-1-infected cells. To date, it is not molecules and the outcome of HIV infection showed
fully understood how natural killer (NK) cells recognize HIV-1-infected cells, and different that individuals expressing KIR3DS1 in conjunction
mechanisms have been proposed. The expression on infected cells of NK cell activating
with HLA-Bw4-80I alleles (HLA-Bw4 alleles that have
receptor ligands, such as NKG2D ligands, results in the direct activation of NKG2D+ NK
cells and target cell lysis (a). Changes in the epitopes presented by HLA class I molecules
an isoleucine residue at position 80) have a slower pro-
might allow the engagement of activating killer cell immunoglobulin-like receptors gression to AIDS than individuals that express either one
(KIRs), and promote NK cell activation (b). Similarly, changes in HLA class I-presented or neither of these molecules. These findings were sup-
epitopes on HIV-1-infected cells can result in the disruption of the binding of inhibitory ported by studies demonstrating that KIR3DS1+ NK cells
KIRs, leading to NK cell activation (c). Finally, antibodies binding to HIV-1-infected cells preferentially expand during primary HIV-1 infection in
can cross-link the Fcγ receptor CD16 and activate CD16+ NK cells (d). HLA-Bw4-80I+ subjects84 and efficiently suppress HIV-1
replication in HLA-Bw4-80I+ target cells85. Furthermore,
KIR3DS1 expression has been associated with enhanced
T cells during HIV infection77. Therefore, the role of NK cell function during primary HIV-1 infection86, and
NCRs in controlling or contributing to HIV-1 pathology the proportion of individuals homozygous for KIR3DS1
is currently not understood and further investigation is was found to be higher in a cohort of HIV-1-exposed but
required. Second, a significant proportion (15–25%) of persistently uninfected individuals than in a group with
NK cells in the peripheral circulation of HIV-1-infected primary HIV-1 infection87.
individuals express some markers of activated NK cells However, studies aimed at demonstrating a direct
(for example, HLA-DR and CD69) but not others (for interaction between KIR3DS1 and HLA-Bw4-80I have
example, CD25 and NKp44 (also known as NCR2))78. not shown such interactions88,89. Two hypotheses are
This suggests that there is an incomplete activation of proposed in the literature. First, KIR3DS1 may recog-
NK cells in these individuals, and this may be due to nize HLA-Bw4-80I in the presence of HLA-presented
chronic stimulation resulting in NK cell exhaustion and viral peptides or stress-induced self peptides90. Indeed
anergy. In addition, although overall NK cell numbers are HIV-1 infection has been shown to uniquely alter the
unchanged during HIV infection, infected individuals presentation of host-encoded peptides by MHC class
have fewer CD3–CD56+ NK cells and show expansion of I molecules91. Second, recognition of HLA-Bw4-80I
the functionally anergic CD3–CD56– NK cell subset64,65. by KIR3DS1 might require an additional cellular pro-
HIV-1 infection has also been associated with an tein that is expressed during HIV-1 infection. This
expansion of NK cell populations that express the acti- hypothesis is supported by evidence from murine cyto-
vating NKG2C receptor and decreased expression of megalovirus (MCMV) infection, in which the NK cell
the inhibitory NKG2A receptor (predominantly on activating receptor Ly49P was shown to interact with the
CD56low NK cells)79. This may alter the balance of NK mouse MHC molecule H2-Dk and mediated recogni-
cell responses towards activation. NKG2C is functional tion of MCMV-infected cells only in the presence of an
in HIV-positive individuals80; however, it is not known additional, as-yet-undefined protein92,93.

182 | MARCH 2011 | VOLUME 11 www.nature.com/reviews/immunol

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

KIR3DL1 and HLA-Bw4-80I. Certain alleles of the inhibi- harness these important innate effector cells in vaccine-
tory receptor KIR3DL1 have been shown to provide pro- induced responses, particularly in the context of the
tection against HIV-1 disease progression94. KIR3DL1 is recent description of NK cell-mediated recall responses
highly polymorphic, and different alleles result in different in mice103,104.
levels of KIR3DL1 protein expression on NK cells95,96. A
genetic study on a large cohort of HIV-1-infected indi- DC–NK cell crosstalk in HIV infection
viduals revealed that individuals co-expressing high levels In addition to their own antiviral functions, NK cells can
of KIR3DL1 and HLA-Bw4-80I had slower HIV-1 disease regulate antiviral immunity by modulating DC func-
progression than individuals expressing low levels of both tion58,105–110. Crosstalk between NK cells and DCs results
KIR3DL1 and HLA-Bw4-80I94. Furthermore, KIR3DL1+ in activation of both cell types, with DCs upregulating NK
NK cell populations are expanded in HIV-positive cell effector functions and NK cells inducing further mat-
individuals who express HLA-Bw4-80I, highlighting uration of DCs (FIG. 3). Fernandez et al.58 initially showed
the importance of this receptor–ligand interaction in that DCs can promote NK cell activity against tumours
HIV-1 infection84. in vivo, and several subsequent studies have explored
These data showing a protective effect for an inhibi- DC-mediated ‘priming’ of NK cells4,111. Both cytokine
tory KIR in HIV infection seem to contradict previous production and cell–cell interactions have been found to
reports, which suggested that NK cell activation is pro- be involved in this process. Cytokines produced by cDCs,
tective during HIV infection. However, this discrepancy such as IL-12 and IL-18, can promote NK cell production
could be explained by the interactions of KIRs with HLA of IFNγ in vitro4,106,108,111, and pDC production of type I
molecules during NK cell development. As described in IFNs, as well as cell–cell contact, is required to promote
BOX 1, developing NK cells undergo a licensing proc- NK cell proliferation and cytotoxicity 58,105,107,108 (FIG. 3).
ess, in which engagement of inhibitory KIRs by HLA Activated NK cells can boost ongoing adaptive
molecules arms NK cells with functional capacity 97–99. responses by producing IFNγ, which promotes TH1
The strength of interaction between inhibitory KIRs cell polarization105,109. Reciprocally, NK cell-mediated
and HLA molecules at this licensing step proportion- activation of DCs has been shown to promote the differen-
ately enhances later functional capacity. Indeed, studies tiation of DCs that are more capable of inducing efficient
have shown that KIR3DL1+ NK cells from individuals CTL responses110. Furthermore, NK cell-mediated lysis
homozygous for HLA-Bw4 respond more potently to of virus-infected cells can provide a source of apoptotic
MHC class I-negative cells than NK cells from individu- bodies for uptake by DCs; this promotes DC maturation
als expressing only one copy or no copies of HLA-Bw4 and the presentation of viral antigens to T cells.
(REF. 100). KIR3DL1hi NK cells licensed on strong inhibi- NK cells can also kill immature DCs in a process
tory signals would therefore respond more strongly referred to as ‘DC editing’111. In vitro studies have dem-
to Nef-mediated downregulation of HLA-Bw4 than onstrated that a low ratio of NK cells to immature DCs
KIR3DL1low NK cells. Interestingly, protection against promotes DC maturation, whereas a higher NK cell to
HIV-1 disease progression was also observed in indi- immature DC ratio can result in NK cell-mediated kill-
viduals with the KIR3DL1*004 allotype of KIR3DL1 that ing of DCs110. This editing process is dependent on NK
is not expressed on the cell surface, but detected intra- cell expression of NKp30 (also known as NCR3)111, and
cellularly 94. The mechanism behind this protection is seems to be mediated by NK cell subsets that express
unknown but suggests a role for intracellular KIR–HLA CD94–NKG2A but lack inhibitory KIRs112, although other
interactions in modulating antiviral immunity. receptors and mechanisms may be involved.
NK cell licensing might also explain the protective Recent evidence suggests that the crosstalk between
effect of a single nucleotide polymorphism (SNP) that NK cells and DCs is disrupted during HIV-1 infection.
was recently identified in a genome-wide association cDCs from HIV-1-infected individuals show reduced
study 101. In this study, a dimorphism 35 kb upstream of secretion of IL-12, IL-15 and IL-18, resulting in decreased
the HLA-C locus that is associated with higher levels of NK cell activation. In addition, activation of NK cells by
HLA-C mRNA and surface expression was shown to be pDCs is impaired during HIV infection because NK
associated with better control of HIV-1 viraemia101,102. cells are less responsive to type I IFNs113. Furthermore,
It is possible that during NK cell development, the inter- the anergic CD56– NK cells that accumulate during pro-
action of NK cells expressing inhibitory KIR2D receptors gressive HIV-1 infection do not produce IFNγ and TNF
with cells expressing high levels of HLA-C (the ligand for following stimulation with MHC-devoid target cells,
KIR2D receptors) may promote the functional capacity decreasing their ability to promote DC activation64,65.
of these NK cells. NK cell-mediated DC editing is also severely compro-
In summary, recent genetic and functional stud- mised during progressive HIV-1 infection, and NK cells
ies have shown a protective effect for certain KIRs from individuals with chronic HIV-1 infection show
and HLA molecules during HIV-1 disease progres- a decreased ability to kill immature DCs114. The defect
sion. The mechanisms underlying protection are appears to be largely due to an increase in the proportion
thought to involve increased NK cell activity against of CD56– NK cells with impaired NKp30 function115.
HIV-1-infected cells; however, no data currently exist The precise mechanisms by which HIV-1 impairs NK
regarding the role of NK cells during HIV-1 infection cell and DC crosstalk remain to be fully elucidated. Effects
in vivo. A better understanding of the functions of NK of HIV-expressed Nef and Tat (transactivator of transcrip-
cells during HIV-1 infection will be required in order to tion) on DCs and NK cells have been described116,117.

NATURE REVIEWS | IMMUNOLOGY VOLUME 11 | MARCH 2011 | 183

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

Figure 3 | DC–NK cell crosstalk. Dendritic cells (DCs) are activated by HIV-1 and secrete pro-inflammatory cytokines,
including interleukin-12 (IL-12), IL-15 and type I interferons (IFNs), that stimulate natural killer (NK) cells. Activated NK cells
secrete IFNγ, which promotes DC maturation and T helper 1 (TH1)-type immune responses. Furthermore, NK cells can eliminate
immature DCs, and this editing process promotes the induction of adaptive T cell immunity. HIV-1-induced functional
impairment of NK cells, as well as of DCs, interferes with this crosstalk. cDC, conventional DC; pDC, plasmacytoid DC.

HIV-1-infected DCs become resistant to NK cell- effective T and B cell-mediated protective immunity.
mediated lysis, and this is associated with the upregulation Understanding how DCs and NK cells are affected during
of cell death inhibitors in infected DCs. These inhibitors HIV infection may provide new targets for vaccine design
are upregulated by the high-mobility group box 1 pro- or even therapeutic modulation of disease. Administration
tein (HMGB1) and protect HIV-1-infected DCs from of myeloid DCs that have been pre-treated with inactivated
TRAIL-dependent apoptosis118. Furthermore, crosstalk HIV enhances immune control of HIV in patients121, sug-
with NK cells was found to promote viral replication gesting that functionally intact antigen-presenting cells are
in HIV-1-infected DCs, but this could be prevented by required to limit viral replication. Adjuvants and vaccine
blocking HMGB1 activity 119. In addition, increased pro- vectors that target cDCs and pDCs simultaneously could
duction of IL-10 during HIV-1 infection can protect promote adaptive immunity and limit TReg cell induction
immature DCs from NK cell-mediated lysis, resulting in in order to control virus entry at mucosal sites, as well
accumulation of partially mature, poorly immunogenic as systemically 122,123. For example, the approach used in
DCs in the lymph nodes of infected individuals120. a recent vaccine trial in Thailand (which used an HIV
Overall, the dysregulated crosstalk that occurs between envelope immunogen and a canarypox virus–HIV vector)
NK cells and DCs during progressive HIV-1 infection could be improved by using DC-targeted adjuvants and
appears to be a consequence of the impairment of both superior vectors in a ‘prime–boost strategy’.
DC and NK cell functions (FIG. 3). Given the crucial role Given the speed with which HIV gains entry into cells,
of DCs in determining the quality of the adaptive immune innate defences need to be rapidly mobilized. DCs could
response to infections, this compromised editing of DC be targeted to activate specific NK cells and promote
function by NK cells during HIV-1 infection might have their cytolytic functions. Epidemiological data have indi-
significant consequences for the antiviral T and B cell cated that certain NK receptors, such as KIR3DS1, are
responses. In order to develop interventions aimed important for controlling HIV disease83, and in mouse
at enhancing immunity to HIV-1, further research is models, NK cells can develop into protective virus-spe-
required to better understand the molecular mechanisms cific memory cells103,124. Therefore, the development of
involved in NK cell–DC crosstalk and how this crosstalk DC-based vaccine strategies that elicit HIV-specific NK
is disrupted during infection with HIV-1. cell responses and stimulate the production of memory
cells may be crucial for the success of future vaccines.
Harnessing DCs and NK cells for vaccination Finally, pDCs are emerging as a population with impor-
Although HIV-1 infection is considered to be a chronic tant roles in contributing to HIV-induced pathology. As
condition, immune dysregulation can occur early in infec- such, vaccine strategies that aim to promote cDC and
tion and affect subsequent disease progression. Given the NK cell responses during HIV infection would have to
complexity of HIV infection, it remains unclear how be balanced to prevent any deleterious consequences of
the innate immune system can be harnessed to induce immune activation.

184 | MARCH 2011 | VOLUME 11 www.nature.com/reviews/immunol

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

1. McMichael, A. J., Borrow, P., Tomaras, G. D., 23. Geijtenbeek, T. B. et al. Mycobacteria target DC-SIGN 44. Sachdeva, N., Asthana, V., Brewer, T. H., Garcia, D. &
Goonetilleke, N. & Haynes, B. F. The immune to suppress dendritic cell function. J. Exp. Med. 197, Asthana, D. Impaired restoration of plasmacytoid
response during acute HIV-1 infection: clues for 7–17 (2003). dendritic cells in HIV-1-infected patients with poor
vaccine development. Nature Rev. Immunol. 10, 11–23 24. Gringhuis, S. I. et al. HIV-1 exploits innate signaling CD4 T cell reconstitution is associated with decrease
(2010). by TLR8 and DC-SIGN for productive infection of in capacity to produce IFN-α but not proinflammatory
2. Liu, Y. J. Dendritic cell subsets and lineages, and their dendritic cells. Nature Immunol. 11, 419–426 cytokines. J. Immunol. 181, 2887–2897 (2008).
functions in innate and adaptive immunity. Cell 106, (2010). 45. Lehmann, C. et al. Increased interferon alpha
259–262 (2001). This study showed that productive HIV infection in expression in circulating plasmacytoid dendritic cells
3. Rescigno, M. & Borrow, P. The host-pathogen cDCs is induced by TLR8 signalling, which initiates of HIV-1-infected patients. J. Acquir. Immune Defic.
interaction: new themes from dendritic cell biology. viral RNA transcription, and DC-SIGN signalling, Syndr. 48, 522–530 (2008).
Cell 106, 267–270 (2001). which promotes transcript elongation. 46. Manches, O. et al. HIV-activated human plasmacytoid
4. Ferlazzo, G. et al. Distinct roles of IL-12 and IL-15 in 25. Deretic, V. Multiple regulatory and effector roles of DCs induce Tregs through an indoleamine
human natural killer cell activation by dendritic cells autophagy in immunity. Curr. Opin. Immunol. 21, 2,3-dioxygenase-dependent mechanism.
from secondary lymphoid organs. Proc. Natl Acad. Sci. 53–62 (2009). J. Clin. Invest. 118, 3431–3439 (2008).
USA 101, 16606–16611 (2004). 26. Blanchet, F. P. et al. Human immunodeficiency virus-1 This paper showed that HIV-stimulated pDCs
5. Lanzavecchia, A. & Sallusto, F. Regulation of T cell inhibition of immunoamphisomes in dendritic cells induce TReg cell differentiation through
immunity by dendritic cells. Cell 106, 263–266 impairs early innate and adaptive immune responses. upregulation of IDO.
(2001). Immunity 32, 654–669 (2010). 47. Favre, D. et al. Tryptophan catabolism by indoleamine
6. Romagnani, C. et al. Activation of human NK cells by 27. Granelli-Piperno, A., Shimeliovich, I., Pack, M., 2,3-dioxygenase 1 alters the balance of TH17 to
plasmacytoid dendritic cells and its modulation by Trumpfheller, C. & Steinman, R. M. HIV-1 selectively regulatory T cells in HIV disease. Sci. Transl. Med. 2,
CD4+ T helper cells and CD4+CD25hi T regulatory infects a subset of nonmaturing BDCA1-positive 32ra36 (2010).
cells. Eur. J. Immunol. 35, 2452–2458 (2005). dendritic cells in human blood. J. Immunol. 176, 48. Herbeuval, J. P. et al. Regulation of TNF-related
7. Wu, L. & KewalRamani, V. N. Dendritic-cell 991–998 (2006). apoptosis-inducing ligand on primary CD4+ T cells by
interactions with HIV: infection and viral 28. Granelli-Piperno, A., Golebiowska, A., Trumpfheller, C., HIV-1: role of type I IFN-producing plasmacytoid
dissemination. Nature Rev. Immunol. 6, 859–868 Siegal, F. P. & Steinman, R. M. HIV-1-infected dendritic cells. Proc. Natl Acad. Sci. USA 102,
(2006). monocyte-derived dendritic cells do not undergo 13974–13979 (2005).
8. Ignatius, R. et al. The immunodeficiency virus maturation but can elicit IL-10 production and T cell 49. Herbeuval, J. P. et al. Differential expression of IFN-α
coreceptor, Bonzo/STRL33/TYMSTR, is expressed by regulation. Proc. Natl Acad. Sci. USA 101, and TRAIL/DR5 in lymphoid tissue of progressor
macaque and human skin- and blood-derived dendritic 7669–7674 (2004). versus nonprogressor HIV-1-infected patients.
cells. AIDS Res. Hum. Retroviruses 16, 1055–1059 29. Harman, A. N. et al. HIV-1-infected dendritic cells Proc. Natl Acad. Sci. USA 103, 7000–7005 (2006).
(2000). show 2 phases of gene expression changes, with 50. Meier, A. et al. Sex differences in the Toll-like receptor-
9. Geijtenbeek, T. B. et al. DC-SIGN, a dendritic cell- lysosomal enzyme activity decreased during the mediated response of plasmacytoid dendritic cells to
specific HIV-1-binding protein that enhances trans- second phase. Blood 114, 85–94 (2009). HIV-1. Nature Med. 15, 955–959 (2009).
infection of T cells. Cell 100, 587–597 (2000). 30. Poulin, L. F. et al. Characterization of human DNGR-1+ 51. McMichael, A. J. HIV vaccines. Annu. Rev. Immunol.
10. Schmidt, B., Ashlock, B. M., Foster, H., Fujimura, S. H. BDCA3+ leukocytes as putative equivalents of mouse 24, 227–255 (2006).
& Levy, J. A. HIV-infected cells are major inducers of CD8α+ dendritic cells. J. Exp. Med. 207, 1261–1271 52. Manches, O. & Bhardwaj, N. Resolution of immune
plasmacytoid dendritic cell interferon production, (2010). activation defines nonpathogenic SIV infection.
maturation, and migration. Virology 343, 256–266 31. Donaghy, H., Gazzard, B., Gotch, F. & Patterson, S. J. Clin. Invest. 119, 3512–3515 (2009).
(2005). Dysfunction and infection of freshly isolated blood 53. Mandl, J. N. et al. Divergent TLR7 and TLR9 signaling
11. Boggiano, C., Manel, N. & Littman, D. R. myeloid and plasmacytoid dendritic cells in patients and type I interferon production distinguish pathogenic
Dendritic cell-mediated trans-enhancement of infected with HIV-1. Blood 101, 4505–4511 (2003). and nonpathogenic AIDS virus infections. Nature Med.
human immunodeficiency virus type 1 infectivity is 32. Cameron, P. U. et al. Preferential infection of dendritic 14, 1077–1087 (2008).
independent of DC-SIGN. J. Virol. 81, 2519–2523 cells during human immunodeficiency virus type 1 54. Bosinger, S. E. et al. Global genomic analysis reveals
(2007). infection of blood leukocytes. J. Virol. 81, rapid control of a robust innate response in SIV-
12. Sabatte, J. et al. Human seminal plasma abrogates 2297–2306 (2007). infected sooty mangabeys. J. Clin. Invest. 119,
the capture and transmission of human 33. Kodama, A. et al. Impairment of in vitro generation 3556–3572 (2009).
immunodeficiency virus type 1 to CD4+ T cells of monocyte-derived human dendritic cells by 55. Jacquelin, B. et al. Nonpathogenic SIV infection of
mediated by DC-SIGN. J. Virol. 81, 13723–13734 inactivated human immunodeficiency virus-1: African green monkeys induces a strong but rapidly
(2007). involvement of type I interferon produced from controlled type I IFN response. J. Clin. Invest. 119,
13. Lambert, A. A., Gilbert, C., Richard, M., Beaulieu, A. D. plasmacytoid dendritc cells. Hum. Immunol. 71, 3544–3555 (2009).
& Tremblay, M. J. The C-type lectin surface receptor 541–550 (2010). 56. Li, Q. et al. Glycerol monolaurate prevents mucosal
DCIR acts as a new attachment factor for HIV-1 in 34. Meera, S., Madhuri, T., Manisha, G. & Ramesh, P. SIV transmission. Nature 458, 1034–1038 (2009).
dendritic cells and contributes to trans- and cis- Irreversible loss of pDCs by apoptosis during early 57. Biron, C. A., Nguyen, K. B., Pien, G. C., Cousens, L. P.
infection pathways. Blood 112, 1299–1307 HIV infection may be a critical determinant of & Salazar-Mather, T. P. Natural killer cells in antiviral
(2008). immune dysfunction. Viral Immunol. 23, 241–249 defense: function and regulation by innate cytokines.
14. de Witte, L. et al. Langerin is a natural barrier to HIV-1 (2010). Annu. Rev. Immunol. 17, 189–220 (1999).
transmission by Langerhans cells. Nature Med. 13, 35. Dillon, S. M. et al. Plasmacytoid and myeloid dendritic 58. Fernandez, N. C. et al. Dendritic cells directly trigger
367–371 (2007). cells with a partial activation phenotype accumulate in NK cell functions: cross-talk relevant in innate
15. Fahrbach, K. M. et al. Activated CD34-derived lymphoid tissue during asymptomatic chronic HIV-1 anti-tumor immune responses in vivo. Nature Med.
Langerhans cells mediate transinfection with human infection. J. Acquir. Immune Defic. Syndr. 48, 1–12 5, 405–411 (1999).
immunodeficiency virus. J. Virol. 81, 6858–6868 (2008). 59. Mailliard, R. B. et al. Dendritic cells mediate NK cell
(2007). 36. Lehmann, C. et al. Plasmacytoid dendritic cells help for Th1 and CTL responses: two-signal
16. Kadowaki, N. et al. Subsets of human dendritic cell accumulate and secrete interferon alpha in lymph requirement for the induction of NK cell helper
precursors express different Toll-like receptors and nodes of HIV-1 patients. PLoS ONE 5, e11110 function. J. Immunol. 171, 2366–2373 (2003).
respond to different microbial antigens. J. Exp. Med. (2010). 60. Karre, K., Ljunggren, H. G., Piontek, G. & Kiessling, R.
194, 863–869 (2001). 37. Nascimbeni, M. et al. Plasmacytoid dendritic cells Selective rejection of H–2-deficient lymphoma
17. Beignon, A. S. et al. Endocytosis of HIV-1 activates accumulate in spleens from chronically HIV-infected variants suggests alternative immune defence
plasmacytoid dendritic cells via Toll-like receptor–viral patients but barely participate in interferon-α strategy. Nature 319, 675–678 (1986).
RNA interactions. J. Clin. Invest. 115, 3265–3275 expression. Blood 113, 6112–6119 (2009). 61. Lanier, L. L. Up on the tightrope: natural killer cell
(2005). 38. Finke, J. S., Shodell, M., Shah, K., Siegal, F. P. & activation and inhibition. Nature Immunol. 9,
This study showed how HIV stimulates pDCs for Steinman, R. M. Dendritic cell numbers in the blood of 495–502 (2008).
type I IFN production through TLR7 signalling. HIV-1 infected patients before and after changes in 62. Romagnani, C. et al. CD56brightCD16– killer Ig-like
18. Meier, A. et al. MyD88-dependent immune activation antiretroviral therapy. J. Clin. Immunol. 24, 647–652 receptor– NK cells display longer telomeres and
mediated by human immunodeficiency virus (2004). acquire features of CD56dim NK cells upon activation.
type 1-encoded Toll-like receptor ligands. J. Virol. 81, 39. Killian, M. S., Fujimura, S. H., Hecht, F. M. & Levy, J. A. J. Immunol. 178, 4947–4955 (2007).
8180–8191 (2007). Similar changes in plasmacytoid dendritic cell and 63. Yu, J. et al. CD94 surface density identifies a
19. Honda, K. et al. IRF-7 is the master regulator of CD4 T-cell counts during primary HIV-1 infection and functional intermediary between the CD56bright and
type-I interferon-dependent immune responses. treatment. AIDS 20, 1247–1252 (2006). CD56dim human NK-cell subsets. Blood 115, 274–
Nature 434, 772–777 (2005). 40. Martinson, J. A. et al. Dendritic cells from HIV-1 281 (2010).
20. Sabado, R. L. et al. Evidence of dysregulation of infected individuals are less responsive to Toll-like 64. Mavilio, D. et al. Characterization of CD56–/CD16+
dendritic cells in primary HIV infection. Blood 116, receptor (TLR) ligands. Cell. Immunol. 250, 75–84 natural killer (NK) cells: a highly dysfunctional NK
3839–3852 (2010). (2007). subset expanded in HIV-infected viremic individuals.
This study examined the changes that occur in cDC 41. Said, E. A. et al. Programmed death-1-induced Proc. Natl Acad. Sci. USA 102, 2886–2891
and pDC frequency and function over the course of interleukin-10 production by monocytes impairs CD4+ (2005).
early HIV infection. T cell activation during HIV infection. Nature Med. 16, 65. Alter, G. et al. Sequential deregulation of NK cell
21. Smed-Sorensen, A. et al. Differential susceptibility to 452–459 (2010). subset distribution and function starting in acute
human immunodeficiency virus type 1 infection of 42. Brenchley, J. M. et al. Microbial translocation is a HIV-1 infection. Blood 106, 3366–3369 (2005).
myeloid and plasmacytoid dendritic cells. J. Virol. 79, cause of systemic immune activation in chronic HIV 66. Gonzalez, V. D. et al. Expansion of functionally skewed
8861–8869 (2005). infection. Nature Med. 12, 1365–1371 (2006). CD56-negative NK cells in chronic hepatitis C virus
22. Manel, N. et al. A cryptic sensor for HIV-1 activates 43. Kamga, I. et al. Type I interferon production is infection: correlation with outcome of pegylated
antiviral innate immunity in dendritic cells. Nature profoundly and transiently impaired in primary HIV-1 IFN-α and ribavirin treatment. J. Immunol. 183,
467, 214–217 (2010). infection. J. Infect. Dis. 192, 303–310 (2005). 6612–6618 (2009).

NATURE REVIEWS | IMMUNOLOGY VOLUME 11 | MARCH 2011 | 185

© 2011 Macmillan Publishers Limited. All rights reserved


REVIEWS

67. Lu, X. et al. CD16+ CD56– NK cells in the peripheral 88. Gillespie, G. M. et al. Lack of KIR3DS1 binding to dendritic cells by natural killer cells. J. Exp. Med.
blood of cord blood transplant recipients: a unique MHC class I Bw4 tetramers in complex with CD8+ 195, 335–341 (2002).
subset of NK cells possibly associated with graft-versus- T cell epitopes. AIDS Res. Hum. Retroviruses 23, 111. Ferlazzo, G. et al. Human dendritic cells activate
leukemia effect. Eur. J. Haematol. 81, 18–25 (2008). 451–455 (2007). resting natural killer (NK) cells and are recognized
68. Stratov, I., Chung, A. & Kent, S. J. Robust NK cell- 89. Carr, W. H. et al. Cutting edge: KIR3DS1, a gene via the NKp30 receptor by activated NK cells.
mediated human immunodeficiency virus (HIV)- implicated in resistance to progression to AIDS, J. Exp. Med. 195, 343–351 (2002).
specific antibody-dependent responses in HIV-infected encodes a DAP12-associated receptor expressed on 112. Della Chiesa, M. et al. The natural killer cell-mediated
subjects. J. Virol. 82, 5450–5459 (2008). NK cells that triggers NK cell activation. J. Immunol. killing of autologous dendritic cells is confined to a cell
69. Tiemessen, C. T. et al. Cutting edge: unusual NK cell 178, 647–651 (2007). subset expressing CD94/NKG2A, but lacking
responses to HIV-1 peptides are associated with 90. Altfeld, M. & Goulder, P. ‘Unleashed’ natural killers inhibitory killer Ig-like receptors. Eur. J. Immunol.
protection against maternal-infant transmission of hinder HIV. Nature Genet. 39, 708–710 (2007). 33, 1657–1666 (2003).
HIV-1. J. Immunol. 182, 5914–5918 (2009). 91. Hickman, H. D. et al. Cutting edge: class I 113. Reitano, K. N. et al. Defective plasmacytoid dendritic
70. Collins, K. L., Chen, B. K., Kalams, S. A., Walker, B. D. presentation of host peptides following HIV infection. cell–NK cell cross-talk in HIV infection. AIDS Res.
& Baltimore, D. HIV-1 Nef protein protects infected J. Immunol. 171, 22–26 (2003). Hum. Retroviruses 25, 1029–1037 (2009).
primary cells against killing by cytotoxic T 92. Lee, S. H. et al. Susceptibility to mouse 114. Tasca, S. et al. Escape of monocyte-derived
lymphocytes. Nature 391, 397–401 (1998). cytomegalovirus is associated with deletion of an dendritic cells of HIV-1 infected individuals
71. Cohen, G. B. et al. The selective downregulation of activating natural killer cell receptor of the C-type from natural killer cell-mediated lysis. AIDS 17,
class I major histocompatibility complex proteins by lectin superfamily. Nature Genet. 28, 42–45 2291–2298 (2003).
HIV-1 protects HIV-infected cells from NK cells. (2001). 115. Mavilio, D. et al. Characterization of the defective
Immunity 10, 661–671 (1999). 93. Kielczewska, A. et al. Ly49P recognition of interaction between a subset of natural killer cells and
72. Bonaparte, M. I. & Barker, E. Killing of human cytomegalovirus-infected cells expressing H2-Dk and dendritic cells in HIV-1 infection. J. Exp. Med. 203,
immunodeficiency virus-infected primary T-cell blasts CMV-encoded m04 correlates with the NK cell 2339–2350 (2006).
by autologous natural killer cells is dependent on the antiviral response. J. Exp. Med. 206, 515–523 This paper shows that NK cell-mediated DC editing
ability of the virus to alter the expression of major (2009). is severely impaired in progressive HIV-1 infection,
histocompatibility complex class I molecules. 94. Martin, M. P. et al. Innate partnership of HLA-B and and this appears to be due to an increased
Blood 104, 2087–2094 (2004). KIR3DL1 subtypes against HIV-1. Nature Genet. 39, accumulation of CD56– NK cells with impaired
73. Cerboni, C. et al. Human immunodeficiency virus 1 733–740 (2007). NKp30 function.
Nef protein downmodulates the ligands of the This large genetic study showed that individuals 116. Quaranta, M. G. et al. HIV-1 Nef impairs the dynamic
activating receptor NKG2D and inhibits natural killer with a KIR3DL1hiHLA-Bw4-80I+ phenotype had of DC/NK crosstalk: different outcome of CD56dim and
cell-mediated cytotoxicity. J. Gen. Virol. 88, 242–250 slower HIV-1 disease progression than individuals CD56bright NK cell subsets. FASEB J. 21, 2323–2334
(2007). with a KIR3DL1lowHLA-Bw4-80I+ phenotype, (2007).
74. Kottilil, S. et al. Innate immunity in human indicating that KIR–HLA interactions can enhance 117. Poggi, A. et al. NK cell activation by dendritic cells is
immunodeficiency virus infection: effect of viremia the protective effect of HLA-Bw4 alleles. dependent on LFA-1-mediated induction of calcium-
on natural killer cell function. J. Infect. Dis. 187, 95. Yawata, M. et al. Roles for HLA and KIR calmodulin kinase II: inhibition by HIV-1 Tat
1038–1045 (2003). polymorphisms in natural killer cell repertoire C-terminal domain. J. Immunol. 168, 95–101
75. Portales, P. et al. Interferon-α restores HIV-induced selection and modulation of effector function. (2002).
alteration of natural killer cell perforin expression J. Exp. Med. 203, 633–645 (2006). 118. Melki, M. T., Saidi, H., Dufour, A., Olivo-Marin, J. C. &
in vivo. AIDS 17, 495–504 (2003). 96. Thomas, R. et al. Novel KIR3DL1 alleles and their Gougeon, M. L. Escape of HIV-1-infected dendritic
76. De Maria, A. et al. The impaired NK cell cytolytic expression levels on NK cells: convergent evolution of cells from TRAIL-mediated NK cell cytotoxicity during
function in viremic HIV-1 infection is associated with KIR3DL1 phenotype variation? J. Immunol. 180, NK-DC cross-talk—a pivotal role of HMGB1.
a reduced surface expression of natural cytotoxicity 6743–6750 (2008). PLoS Pathog. 6, e1000862 (2010).
receptors (NKp46, NKp30 and NKp44). 97. Kim, S. et al. Licensing of natural killer cells by host 119. Saidi, H., Melki, M. T. & Gougeon, M. L.
Eur. J. Immunol. 33, 2410–2418 (2003). major histocompatibility complex class I molecules. HMGB1-dependent triggering of HIV-1 replication
77. Vieillard, V., Strominger, J. L. & Debre, P. Nature 436, 709–713 (2005). and persistence in dendritic cells as a consequence
NK cytotoxicity against CD4+ T cells during HIV-1 98. Fernandez, N. C. et al. A subset of natural killer cells of NK-DC cross-talk. PLoS ONE 3, e3601 (2008).
infection: a gp41 peptide induces the expression of achieves self-tolerance without expressing inhibitory 120. Alter, G. et al. IL-10 induces aberrant deletion of
an NKp44 ligand. Proc. Natl Acad. Sci. USA 102, receptors specific for self-MHC molecules. Blood 105, dendritic cells by natural killer cells in the context of
10981–10986 (2005). 4416–4423 (2005). HIV infection. J. Clin. Invest. 120, 1905–1913
78. Fogli, M. et al. Significant NK cell activation 99. Anfossi, N. et al. Human NK cell education by (2010).
associated with decreased cytolytic function in inhibitory receptors for MHC class I. Immunity 25, This work showed that IL-10 secreted in response
peripheral blood of HIV-1-infected patients. 331–342 (2006). to HIV-1 renders immature DCs resistant to NK
Eur. J. Immunol. 34, 2313–2321 (2004). 100. Kim, S. et al. HLA alleles determine differences cell-mediated killing, resulting in the accumulation
79. Mela, C. M. et al. Switch from inhibitory to activating in human natural killer cell responsiveness and of poorly immunogenic DCs in the lymph nodes of
NKG2 receptor expression in HIV-1 infection: lack of potency. Proc. Natl Acad. Sci. USA 105, 3053–3058 infected patients.
reversion with highly active antiretroviral therapy. (2008). 121. Lu, W., Arraes, L. C., Ferreira, W. T. & Andrieu, J. M.
AIDS 19, 1761–1769 (2005). 101. Fellay, J. et al. A whole-genome association study of Therapeutic dendritic-cell vaccine for chronic HIV-1
80. Guma, M. et al. Imprint of human cytomegalovirus major determinants for host control of HIV-1. infection. Nature Med. 10, 1359–1365 (2004).
infection on the NK cell receptor repertoire. Science 317, 944–947 (2007). 122. Hansen, S. G. et al. Effector memory T cell responses
Blood 104, 3664–3671 (2004). This genome-wide association study showed are associated with protection of rhesus monkeys from
81. Bernstein, H. B. et al. CD4+ NK cells can be productively that a protective SNP associated with higher HLA-C mucosal simian immunodeficiency virus challenge.
infected with HIV, leading to downregulation of CD4 expression correlates with better control of HIV-1 Nature Med. 15, 293–299 (2009).
expression and changes in function. Virology 387, viraemia. 123. Wyand, M. S. et al. Protection by live, attenuated
59–66 (2009). 102. Thomas, R. et al. HLA-C cell surface expression and simian immunodeficiency virus against heterologous
82. Flores-Villanueva, P. O. et al. Control of HIV-1 viremia control of HIV/AIDS correlate with a variant challenge. J. Virol. 73, 8356–8363 (1999).
and protection from AIDS are associated with upstream of HLA-C. Nature Genet. 41, 1290–1294 124. Paust, S. et al. Critical role for the chemokine receptor
HLA-Bw4 homozygosity. Proc. Natl Acad. Sci. USA (2009). CXCR6 in NK cell-mediated antigen-specific memory
98, 5140–5145 (2001). 103. Sun, J. C., Beilke, J. N. & Lanier, L. L. Adaptive of haptens and viruses. Nature Immunol. 11,
83. Martin, M. P. et al. Epistatic interaction between immune features of natural killer cells. Nature 457, 1127–1135 (2010).
KIR3DS1 and HLA-B delays the progression to AIDS. 557–561 (2009). 125. Brodin, P., Lakshmikanth, T., Johansson, S., Karre, K.
Nature Genet. 31, 429–434 (2002). 104. O’Leary, J. G., Goodarzi, M., Drayton, D. L. & & Hoglund, P. The strength of inhibitory input during
This study first demonstrated the protective von Andrian, U. H. T cell- and B cell-independent education quantitatively tunes the functional
effect of KIR–HLA combinations in HIV-1 disease adaptive immunity mediated by natural killer cells. responsiveness of individual natural killer cells.
progression and showed that individuals Nature Immunol. 7, 507–516 (2006). Blood 113, 2434–2441 (2009).
expressing KIR3DS1 in conjunction with its 105. Kikuchi, T. et al. Dendritic cells pulsed with live and 126. Yu, J. et al. Hierarchy of the human natural killer cell
putative ligand (HLA-Bw4-80I) have a slower dead Legionella pneumophila elicit distinct immune response is determined by class and quantity of
progression to AIDS compared with individuals responses. J. Immunol. 172, 1727–1734 (2004). inhibitory receptors for self-HLA-B and HLA-C ligands.
lacking either or both receptor–ligand pairs. 106. Gerosa, F. et al. Reciprocal activating interaction J. Immunol. 179, 5977–5989 (2007).
84. Alter, G. et al. HLA class I subtype-dependent between natural killer cells and dendritic cells. 127. Yawata, M. et al. MHC class I-specific inhibitory
expansion of KIR3DS1+ and KIR3DL1+ NK cells J. Exp. Med. 195, 327–333 (2002). receptors and their ligands structure diverse human
during acute human immunodeficiency virus type 1 107. Gerosa, F. et al. The reciprocal interaction of NK-cell repertoires toward a balance of missing self-
infection. J. Virol. 83, 6798–6805 (2009). NK cells with plasmacytoid or myeloid dendritic cells response. Blood 112, 2369–2380 (2008).
85. Alter, G. et al. Differential natural killer cell-mediated profoundly affects innate resistance functions.
inhibition of HIV-1 replication based on distinct KIR/ J. Immunol. 174, 727–734 (2005). Acknowledgements
HLA subtypes. J. Exp. Med. 204, 3027–3036 108. Borg, C. et al. NK cell activation by dendritic cells This work was supported by grants from the US National
(2007). (DCs) requires the formation of a synapse leading to Institutes of Health (to N.B. and M.A.), the Doris Duke
86. Long, B. R. et al. Conferral of enhanced natural killer IL-12 polarization in DCs. Blood 104, 3267–3275 Charitable Foundation (to M.A.), the Bill & Melinda Gates
cell function by KIR3DS1 in early human (2004). Foundation (to N.B. and M.A.) and the Phillip T. and Susan M.
immunodeficiency virus type 1 infection. J. Virol. 82, 109. Martin-Fontecha, A. et al. Induced recruitment of Ragon Foundation (to M.A. and L.F.).
4785–4792 (2008). NK cells to lymph nodes provides IFN-γ for TH1
87. Boulet, S. et al. Increased proportion of KIR3DS1 priming. Nature Immunol. 5, 1260–1265 (2004). Competing interests statement
homozygotes in HIV-exposed uninfected individuals. 110. Piccioli, D., Sbrana, S., Melandri, E. & Valiante, N. M. The authors declare competing financial interests: see Web
AIDS 22, 595–599 (2008). Contact-dependent stimulation and inhibition of version for details.

186 | MARCH 2011 | VOLUME 11 www.nature.com/reviews/immunol

© 2011 Macmillan Publishers Limited. All rights reserved