Available online at www.sciencedirect.

com
R

Clinical Immunology 108 (2003) 167–174 www.elsevier.com/locate/yclim

Short Analytical Review

Protection from HIV/AIDS: the importance of innate immunity夞

Jay A. Levy,* Iain Scott, and Carl Mackewicz
Division of Hematology/Oncology, Department of Medicine, Box 1270, University of California, San Francisco, CA 94143-1270, USA

Received 15 May 2003; accepted with revision 19 June 2003

Introduction Host immune responses

Infection by HIV can occur through intimate sexual The key to resistance to HIV infection and disease pro-
activity, contact with contaminated blood or blood products, gression resides within the host immune system that con-
and transmission from mother to child [1]. Importantly, the sists of two major defense pathways: innate and adaptive
virus is infectious as a free particle or in association with immunity. Until recently, the innate immune system, the
virus-infected cells. The latter entity is found in genital first line of defense against infectious organisms, has re-
fluids at greater quantities than free infectious virions [2]. In ceived little attention in HIV infection. With the recognition
seminal fluid, for example, the amount of infectious virus of the important role of innate immune responses in helping
may be 40 –50 particles/ml, whereas infected cells may be control other infectious diseases, this rapid host immune
present at the level of 5% of all white blood cells in a response against HIV infection is becoming better appreci-
normal ejaculate (generally containing ⬃106 cells) [2]. Each ated [5,6].
of these 50,000 cells has the capability of producing 1000
particles, several of which can infect a variety of immune
cells or mucosal lining cells in the vaginal or anal canal. The Innate immune system
host defense system must, therefore, recognize both free
infectious virus and virus-infected cells to prevent and con- The innate immune system is the earliest response to
trol infection. The response must be rapid and have the microbial entry and injury, and without the innate immune
ability both to inactivate the virus and to prevent the spread system, the adaptive immune system would not have
of HIV into new cells that will become distributed through- evolved [7]. Because of the innate immune system, the
out the body. adaptive immune system has time to develop. Thus, disrup-
As with other viral infections, some people infected for tions in innate immunity can predispose burn patients to
many years show no signs of HIV infection; they remain infection and cystic fibrosis patients, lacking normal lung
clinically healthy. These “long-term survivors” or “long- surfactants, to enhanced susceptibility to bacterial infec-
term nonprogressors,” infected for ⱖ10 years, have normal tions. Individuals with mutations in the genes for comple-
CD4⫹ cell counts (⬎500 cells/␮l), and remain healthy ment or mannose-binding lectins have recurring infections
without receiving any antiviral therapy [3,4]. Some of these and individuals with disorders in macrophage and NK cell
individuals we have followed have been infected since 1978 functions often have infections [8,9] (see below).
and are, therefore, living more than 25 years without any A major function of the innate immune system involves
signs of the infection. They have the ability for long-term an intracellular signaling pathway activated by pattern rec-
control of HIV infection. ognition receptors on the surface of a variety of cells, most
of them antigen-presenting cells (e.g., dendritic cells, mac-
rophages). These toll-like receptors (TLRs), following in-
teractions with specific ligands, initiate a cascade of pro-
夞 This review was presented as part at a session on HIV infection at the cesses within the cells that leads to the activation of NF␬B
Clinical Immunology Society Meeting in San Francisco, CA, September
2002. with induction of cytokines that bring about varied immune
* Corresponding author. Fax: ⫹1-415-476-8365. responses. The TLRs, named after their counterpart first
E-mail address: [email protected] (J.A. Levy). identified in Drosophila [7], respond to a variety of different

1521-6616/$ – see front matter © 2003 Elsevier Inc. All rights reserved.
doi:10.1016/S1521-6616(03)00178-5
168 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174

Table 1 Soluble factors in innate immunity

Components of the innate and adaptive immune systems

A. Innate immune system The innate immune system has several circulatory pro-
teins (besides type 1 interferons) that can have anti-HIV
Dendritic cells B-1 cells
Macrophages Cytokines (e.g., interferon) activities (Table 1A). One in particular, the mannose-bind-
Neutrophils Chemokines ing lectin (MBL), attaches to HIV and can enhance its
NK cells Defensins engulfment and destruction by macrophages or, in the pres-
gd T cells Complement ence of complement, directly lyse the virus [9,16,17]. Low
NK-T cells Lectin-binding proteins (collectins)
mannose-binding lectin levels and polymorphisms in the
Plasmacytoid dendritic cells Fever (acute-phase reactants:
(PDCs) cathelicidins, pentraxins) MBL promoter are associated with an increased risk of HIV
CD8⫹ cellsa infection, rapid progression to AIDS, and a shorter survival
period after AIDS diagnosis [18 –20]. These observations,
B. Adaptive immune system while controversial, underline the potential effect of circu-
Dendritic cells lating innate factors in body fluids in reducing HIV infec-
Macrophages tion. Another important soluble component of plasma, com-
B lymphocytes plement, can directly interact with HIV, leading to lysis of
CD4⫹ lymphocytes virion particles [9,21]. Essentially, these soluble factors
CD8⫹ lymphocytes
attack the free virion and can be important in rapid re-
a
Noncytotoxic antiviral activity. sponses to an HIV infection. Soluble chemokines are part of
the innate immune system and have a major role in attract-
ing cells to areas of infection. HIV uses chemokine recep-
stimuli including gram-positive and gram-negative bacteria, tors on cells to enter and infect [22]. Whether high levels of
double-stranded RNA, and oligodeoxynucleotides. chemokine production play a role in vivo in preventing HIV
When comparing the innate and adaptive immune sys- spread remains controversial [22–24]. However, individuals
tems, several differences can be appreciated [5]. The innate who lack the CCR5 chemokine receptor can be less suscep-
system has a rapid response (minutes to days) whereas the tible to HIV infection [25,26].
adaptive immune system is delayed (days to weeks). The
innate immune system reacts to a “pattern” rather than to a
specific epitope that is characteristic of adaptive immunity. Cellular components of innate immunity
The innate immune system has no memory response and
will be activated to the same degree when encountering a Two cells in the innate immune system are of particular
particular infectious organism on repeated occasions. The interest to our laboratory: interferon-producing cells (IPCs),
adaptive immune system, in contrast, has memory and can now called plasmacytoid dendritic cells (PDCs), and the
respond more rapidly to a second challenge and with an CD8⫹ T lymphocytes showing a noncytotoxic anti-HIV
response (CNAR) [5,27]. The role of type 1 interferon
enhanced activity against the infectious agent.
production in HIV infection was first appreciated about 20
The innate immune system has a wide variety of cellular
years ago. Siegal and colleagues [28,29] demonstrated that
components as well as soluble products that include cyto-
the ability of peripheral blood mononuclear cells (PBMCs)
kines, chemokines, and other circulating small molecules
to produce type 1 IFN, following exposure to irradiated
(e.g., defensins, complement, mannose-binding lectins) (Ta-
herpes simplex virus, identified HIV-infected individuals
ble 1A). Each plays a unique role in rapidly preventing
who had a reduced risk of opportunistic infections and a
infection by pathogens. In contrast, the adaptive immune long survival (i.e., those with interferon-␣ production ⬎300
system consists of a limited number of cell types, primarily U/ml). In some cases, the normal production of type 1
lymphocytes, of both B- and T-cell lineage (Table 1B). interferons in HIV-infected subjects with low CD4⫹ cell
Dendritic cells (DCs) and macrophages, both antigen-pre- numbers appeared to protect individuals from disease.
senting cells (APCs), can play a prominent role in both A search for the major cellular producer of type 1 inter-
innate and adaptive immunity. ferons led to the serendipitous observation by Siegal and
Essentially, the innate immune system gives rise to cel- colleagues [30] that the plasmacytoid T lymphocyte, first
lular factors, cytokines that can have a direct antimicrobial identified in lymph nodes by Lennert et al. in 1975 [31], was
response (e.g., interferon) [10] or enhance the cellular re- the prominent producer of type 1 interferons. A similar
sponses of the innate immune system such as NK cell observation was made by Cella and colleagues [32]. Purified
activities [11]. They also induce increased adaptive immune PDCs were shown to have a CD4⫹/CD11c⫺ phenotype
responses by enhancing MHC expression on potential target and plasmacytoid morphology and to produce 200 to 1000
cells and increasing lymphocyte responses to pathogens times more interferon than other cells in response to micro-
[10,12–15]. A major function is to increase antigen presen- bial infection [30] (Table 2). The PDCs appear to be very
tation by APCs. important because of the known antiviral and antitumor
 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174 169

Table 2 Table 3
Characteristics of the plasmacytoid dendritic cell (PDC) Association between PDC number and AIDS-defining disease

CD4⫹/lin⫺/CD11c⫺: dendritic cell precursor (pDC-2) PDCs/␮l

Plasmacytoid morphology
⬍2 ⬎2
Found in lymphoid tissue; present in very low numbers in blood
Secretes interferon with exposure to herpes simplex virus and other Opportunistic infection 8 0
pathogens Kaposi’s sarcoma 3 3
Generally reduced with HIV infection No AIDS-defining disease 2 28
Loss generally mirrors CD4⫹ cell reduction

activities of type 1 interferons and their abilities to modulate significantly lower than those with ⬍10 PDCs/␮l (P ⫽
the immune system [10,33]. Liu and colleagues have shown 0.03).
that PDCs can influence both innate and adaptive immunity Of importance was the rare finding of four healthy in-
[34,35]. By providing interferon-␣, they play an important fected individuals who refused therapy and had very low
part in innate immune responses. Then, following cell mat- CD4⫹ cell numbers (⬍100 cells/␮l). The absence of op-
uration with CD40 ligand (which occurs during an immune portunistic infections and cancer appeared to correlate with
response), these plasmacytoid dendritic cells mature into a normal number of PDCs present in their blood (ranging
DC-2s (dendritic cells that induce type 2 immune respons- from 0.2 to 0.5%; 3– 8 cells/␮l) [37] (our unpublished ob-
es). They produce cytokines (e.g., IL-4, IL-5 and IL-10) that servations). This potential value of PDCs in the face of low
enhance antibody production, an adaptive immune response CD4⫹ T-cell numbers is under further study. The role of
(Fig. 1). antiviral therapy in increasing the number of PDCs also
PDCs can be readily identified by flow cytometry (Fig. needs to be appreciated. Siegal and colleagues [40] have
2). These lineage-negative CD11c⫺ CD4⫹ cells constitute shown that with anti-HIV treatment, the recovery of inter-
a very small number of white cells in the peripheral blood feron production by PBMCs can be observed several
(ranging from 0.1 to 0.6% of white cells compared with up months before full reconstitution of CD4⫹ cell number.
to 40% CD4⫹ cells). This percentage of PDCs represents Further work with PDCs has demonstrated similarities
around 2–10 PDCs/␮l of blood (compared with 600 –1200 and differences in this precurser to DC-2 cells and the
CD4⫹ cells/␮l). precurser for DC-1 cells (i.e., blood-derived DCs) (Table 4).
We and others have noted that the number of PDCs is As noted above, some investigators have reported that
generally reduced with HIV infection and their loss usually blood-derived DCs or monocyte-derived DCs help direct
mirrors CD4⫹ cell reduction over time [36,37]. In support the immune response toward a type 1 pattern favoring
of these findings, studies with individuals at different clin- cellular immunity. The PDCs appear to mature into DCs
ical states demonstrated that the number of circulating that induce a type 2 response; thus, they were considered
PDCs per microliter of blood was decreased in association precursors to DC-2s [34,35]. Both DC types are lineage-
with advancement to disease (Fig. 3) [37]. Some studies negative, but the pDC-1 is CD11c⫹. Both express CD4 and
have suggested a reduction in PDCs as well as CD11c⫹ class II antigens as well as the chemokine co-receptors and
DCs in acute infection [38] but these findings require further express the IL-3 receptor (although low on pDC-1)
evaluation. A potentially important observation in our stud- [30,37,41,42]. Only the PDCs express an antigen recog-
ies was that long-term survivors appeared to have larger nized by BDCA-2 and BDCA-4 antibodies [43], and as
numbers of PDCs than healthy controls (Fig. 3) [37]. This recently reported [44] neither cell contains DC-SIGN (our
observation, which requires confirmation, may indicate a unpublished observations). This surface molecule on certain
selection of individuals who can ward off the symptoms and
CD4⫹ cell loss associated with HIV infection because of
Table 4
their high PDC number. Comparison of precursor dendritic cells (pDCs)
The association between PDC and CD4⫹ T cell numbers
indicated that at least 2 PDCs/␮l was associated with ab- Marker pDC2 (PDC) pDC1
sence of disease (Table 3) [37]. In this regard, while the Lineage ⫺ ⫺
number of subjects is limited, it is perhaps noteworthy that CD11c ⫺ ⫹
those patients with Kaposi’s sarcoma (KS) who had ⬍2 CD4 ⫹⫹ ⫹
HLA-DR ⫹⫹ ⫹⫹
PDCs/␮l continued to have cutaneous lesions, whereas
CCR5/CXCR4 ⫹ ⫹
those who had ⬎2 PDCs/␮l no longer had any new onset of CCR7 ⫹ ⫺
KS. These latter findings were not related to the number of IL-3R ⫹⫹ ⫹
CD4⫹ T cells in these individuals (⬎200 cells/␮l). The BDCA-2,4 ⫹⫹ ⫺
plasma viral load in HIV-infected subjects has also corre- DC-SIGN ⫺ ⫹a
TLR 7,9 ⫹ ⫺
lated with PDC number [37,39]. In our study, the long-term
survivors with ⬎10 PDCs/␮l had viral loads that were a
Only monocytes.
170 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174

Fig. 1. Dual role of plasmacytoid dendritic cells (PDCs) in host immunity. PDCs, on exposure to virus or other stimuli, can release type 1 interferons and
have an antimicrobial effect. After exposure to CD40 ligand, they mature into dendritic cells that play a role in adaptive immune responses.

DCs appears to bind HIV and internalize it without causing tecting reverse transcriptase activity in the culture of the
infection of the cell [45,46]. HIV can then be transferred to highly purified PDCs (⬎99% pure) co-cultured with normal
uninfected cells. Moreover, the expression of TLRs 7 and 9 CD4⫹ lymphocytes. HIV recovery is greatly enhanced if
on the PDCs has been recognized [47–50]. the PDCs are exposed to CD40 ligand, a signal for matu-
ration and a state when interferon production is not present.
A similar observation has been made by Fong et al. [51].
Interaction of PDCs with HIV When cultured cells are directly infected with HIV in
vitro (either NSI or SI type), very low level replication can
Work in our laboratory has indicated that low levels of be noted (by p24 antigen expression) in the culture super-
HIV can be detected in highly purified PDCs recovered natant. Addition of CD4⫹ T cells, however, even after
from patients, indicating that infection of this cell can take trypsinization of the PDCs (up to 6 days following the virus
place in vivo. This finding was best demonstrated by de- inoculation) to remove surface bound virus [52], readily

Fig. 3. Relationship of plasmacytoid dendritic cell number to clinical state.

Each open circle represents a value for a different study subject. Horizontal
bars indicate the median. The number of blood PDCs is increased in
Fig. 2. Identification of plasmacytoid dendritic cells by flow cytometry. long-term survivors (LTS) (P ⬍ 0.05 for all group comparisons vs LTS)
PDCs can be identified as CD4⫹, lineage⫺ cells using two-color staining and decreased in AIDS patients (P ⬍ .01 for all group comparisons vs
with anti-CD4 PE and (CD3, CD14, CD16, CD20, CD11c) FITC. About AIDS). Most of the progressors had received antiretroviral therapy for
105 peripheral blood mononuclear cells (PBMCs) were analyzed and the several months; no substantial difference in PDC number was observed
percentage of cells was determined by using the quadrant stat function between these subjects and those who were untreated. Reprinted, with
(Cell Quest). permission, from Soumelis et al. [37].
 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174 171

CD8ⴙ cell antiviral responses

The CD8⫹ cell exhibiting an anti-HIV response that

does not involve cell killing appears to be another important
cellular component of the innate immune system [5] (see
also Table 1A):
1. It is not retrovirus-specific, but broadly reactive
against a wide variety of HIV-1, HIV-2, and SIV
strains [27]. This antiviral response has also been
shown to reduce LTR transcription of murine and
avian retroviruses [59].
Fig. 4. HIV-infected cells stimulate interferon-␣ production by plasmacy- 2. It is not HLA-restricted [60].
toid dendritic cells. HIV-infected CD4⫹ cells, but not HIV alone [HIV 3. The response is found very early in primary infection,
supernatant (sup.)], induced substantial type 1 interferon production. Con- before anti-HIV antibodies are present [61].
trol uninfected CD4⫹ cells and supernatants containing no virus show no 4. It is found in uninfected individuals who are exposed
effect. A representative experiment of at least six separate studies is shown.
to HIV but not in those without contact within the last
6 –12 months [62]. Thus, a memory response does not
appear to be involved.
demonstrates evidence of this virus infection. Since PDCs
5. It is mediated by a secreted CD8⫹ cell anti-HIV
lack DC-SIGN expression, whether HIV can attach to the
factor, CAF, that affects viral transcription [5,27,63].
cell surface of PDCs by a mechanism other than DC-SIGN
or DC-SIGNR requires further study. Our relative inability This CD8⫹ cell anti-HIV response (CNAR) contrasts
to infect PDCs in vitro contrasts with previous studies with the classic adaptive MHC-restricted cytotoxic immune
showing susceptibility of these cells to HIV infection response of CD8⫹ cells that involves antigen-specific rec-
[44,51,53,54]. These latter studies either indicated low-level ognition leading to lysis of the infected cell. CNAR can be
replication of HIV, as we have noted, or, in some cases, did demonstrated with a very low input of CD8⫹ cells (e.g., 1
not evaluate highly purified PDCs. CD8⫹ cell:20 CD4⫹ cells) compared with the larger num-
Our laboratory has also demonstrated that inoculation of ber of CD8⫹ cells required for anti-HIV cytotoxic activity.
normal PDCs with HIV-1 does not kill these cells, nor does CNAR is observed to a greater extent with CD8⫹ cells from
it induce high levels of interferon-␣ production. Substantial long-term survivors versus progressors, indicating its clini-
type 1 interferon production occurs only after exposure of cal relevance [64 – 67]. It is associated with a CD8⫹
PDCs to HIV-infected cells (with X4 or R5 isolates) (Fig. 4, CD28⫹ cell phenotype [65,68]; whereas cytotoxic activity
Table 5). Cell killing does not seem to be involved since is observed primarily with CD8⫹ CD28⫺ cells [69].
HIV-2UC1, a noncytopathic isolate [55], can induce type 1 CNAR does not affect activation or proliferation of CD4⫹
interferon release. Importantly, induction of interferon pro- cells [27]. As noted above, CNAR is active against a wide
duction by HIV-infected cells leads to a reduction in HIV variety of HIV isolates, thus demonstrating its lack of fine
production by the CD4⫹ cells ([51] our unpublished re- specificity at the effector phase. A similar type of process
sults), most likely resulting from the release of interferon-␣ has been observed as well in nonhuman primates and in cats
that has known anti-HIV activity [56]. Others have reported
IFN induction by HIV alone [44,51,57] but only low-level
production of interferon was noted, and, in some studies, Table 5
highly purified PDCs were not used. Induction of type 1 interferon (IFN) production by plasmacytoid
In summary, purified PDCs can be infected by NSI and dendritic cells
SI HIV-1 isolates, but only low-level virus production takes Stimulus IFN production
place. The presence of CD4⫹ T cells in culture would
HIV-1(R-5) a
⫾
increase the sensitivity to infection. HIV-infected cells best HIV-1(X4)a ⫾
stimulate the production of interferon by PDCs that then HIV-1gp120 ⫺
leads to reduction in virus replication in the infected cells. CD4⫹ cells(R-5) ⫹⫹
This process, we believe, takes place primarily in the lymph CD4⫹ cells(X4) ⫹⫹
CD4⫹ cellsb ⫹⫹
node where there can be a large number of infected cells HSV(irrad.) ⫹⫹
[58]. The release of interferon would block virus replication
a
in the tissue. The subsequent maturation of the PDCs in Results shown with R5 and X4 high-titered virus, CD4⫹ cells infected
with R5 or X4 HIV-1 isolates, and HIV-1 recombinant gp120. Irradiated
association with CD40 ligand (on CD4⫹ T cells) then leads herpes simplex virus (HSV) is the positive control.
to the emergence of mature dendritic cells that can activate b
Noncytopathic isolate that does not decrease CD4 expression on
adaptive immune responses against HIV. CD4⫹ cells [55].
172 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174

Table 6 known cytokines, CAF, as well as mannose-binding lectins

Cytokines lacking identity to the CD8⫹ cell antiviral factor (CAF)a and complement.
IL-1 to IL12, IL-16 Leukemia inhibitory factor The importance of the innate immune system is apparent
Interferon-␣, -␤, -␥ Lymphotactin in many infectious diseases and in HIV infection. Its im-
TNF-␣, -␤ IP-10 mediate response can be appreciated as well as its function
TGF-␤ Granzymes A,B
in continually activating and helping the adaptive immune
GM-CSF Granulysin
RANTES Protegrins, histatins, defensins system maintain its defense against HIV. Approaches to
MIP-1␣, ␤ enhance its activity can be valuable in therapy for HIV
MCP-1,3 TNF-1 receptors, I, II infection, Moreover, this rapid recognition of HIV and HIV-
GRO-␣, -␤ sFas infected cells will be very helpful in vaccines. Furthermore,
a
Studies conducted by the authors [1,24,75,80] and unpublished obser- understanding how innate immune responses can best stim-
vations. ulate the adaptive immune system would be useful for
vaccine development.

infected with the feline immune deficiency virus (FIV)

[70 –73]. Acknowledgments
CAF is not found in granules [74] and lacks identity to
other known cytokines and chemokines (Table 6) [27,75] This research was supported by NIH Grant RO1A130350
(unpublished observation). It appears to be a protein of and Universitywide AIDS Research Program Grant
about 10 –50 kDa, produced only by CD8⫹ cells, and at CC02-SF-002.
extremely low levels (4 U/ml of culture fluid where 1 unit ⫽
50% inhibition of HIV replication). Thus, approaches to its
identification are requiring novel molecular and biochemi- References
cal procedures [76]. CAF (as is characteristic of CNAR) [1] J.A. Levy, HIV and the Pathogenesis of AIDS (American Society of
blocks virus transcription; it does not affect any of the Microbiology, Washington, DC, ed. 2nd Edition, 1998).
processes leading to infection including interaction with cell [2] J.A. Levy, Human immunodeficiency viruses and the pathogenesis of
surface receptors, reverse transcription, and integration AIDS, J. Am. Med. Assoc. 261 (1989) 2997–3006 .
[3] J.A. Levy, HIV pathogenesis and long-term survival, AIDS 7 (1993)
[77,78]. Most recently we have noted this anti-HIV re-
1401–1410.
sponse may involve a protease [79]. [4] G. Pantaleo, S. Menzo, M. Vaccarazza, C. Graziosi, O.J. Cohen, J.F.
Demarest, et al., Studies in subjects with long-term nonprogressive
human immunodeficiency virus infection, N. Engl. J. Med. 332
(1995) 209 –216.
Characteristics of long-term survivors [5] J.A. Levy, The importance of the innate immune system in control-
ling HIV infection and disease, Trends Immunol. 22 (2001) 312–316.
Observations on long-term survivors indicate that they [6] F.P. Siegal, G.T. Spear, Innate immunity and HIV, AIDS 15 (2001)
S127–137.
have virologic and immunologic features that can be related
[7] R. Medzhitov, C. Janeway Jr., Innate immunity, N. Engl. J. Med. 343
to at least two components of innate immunity: PDC func- (2000) 338 –344.
tion and CD8⫹ cell noncytotoxic antiviral activity [5]. The [8] C.A. Janeway Jr., How the immune system works to protect the host
PDCs can ward off opportunistic infections and cancer even from infection: a personal view, Proc. Natl. Acad. Sci. USA 98 (2001)
in the face of low CD4⫹ cell numbers. CD8⫹ T cells can 7461–7468.
[9] G.T. Spear, Interaction of non-antibody factors with HIV in plasma,
inhibit HIV replication and provide clinical benefit by a
AIDS 7 (1993) 1149 –1157.
noncytotoxic mechanism. This response may be beneficial [10] C. Bogdan, The function of type I interferons in antimicrobial im-
since the infected cells are blocked in their expression of munity, Curr. Opin. Immunol. 12 (2000) 419 – 424.
virus and can potentially maintain normal function. [11] C.A. Biron, K.B. Nguyen, G.C. Pien, L.P. Cousens, T.P. Salazar-
Mather, Natural killer cells in antiviral defense: function and regula-
tion by innate cytokines, Annu. Rev. Immunol. 17 (1999) 189 –220.
[12] V. Brinkmann, T. Geiger, S. Alkan, C.H. Heusser, Interferon alpha
Summary increases the frequency of interferon gamma-producing human
CD4⫹ T cells, J. Exp. Med. 178 (1993) 1655–1663.
[13] L. Rogge, D. D’Ambrosio, M. Biffi, G. Penna, L.J. Minetti, D.H.
The importance of the innate immune system is becom- Presky, et al., The role of Stat4 in species-specific regulation of Th
ing evident in HIV infection. It is the first line of immune cell development by type I IFNs, J. Immunol. 161 (1998) 6567– 6574.
defense, is found at mucosal surfaces, and is active against [14] L.P. Cousens, R. Peterson, S. Hsu, A. Dorner, J.D. Altman, R.
many pathogens, and “pattern” recognition can permit a Ahmed, et al., Two roads diverged: interferon alpha/beta- and inter-
leukin 12-mediated pathways in promoting T cell interferon gamma
rapid response. The key innate immune cellular participants
responses during viral infection, J. Exp. Med. 189 (1999) 1315–1328.
include dendritic cells and macrophages and involve two [15] K. Hiroishi, T. Tuting, M.T. Lotze, IFN-alpha-expressing tumor cells
major cellular components discussed in this article: PDCs enhance generation and promote survival of tumor-specific CTLs,
and CD8⫹ cells. The humoral components involve many J. Immunol. 164 (2000) 567–572.
 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174 173

[16] A.B. Ezekowitz, M. Kuhlman, J.E. Groopman, R.A. Byrn, A human [36] S. Feldman, D. Stein, S. Amrute, T. Denny, Z. Garcia, P. Kloser, et
serum mannose-binding protein inhibits in vitro infection by the al., Decreased interferon-alpha production in HIV-infected patients
human immunodeficiency virus, J. Exp. Med. 169 (1989) 185–196. correlates with numerical and functional deficiencies in circulating
[17] M. Saifuddin, M.L. Hart, H. Gewurz, Y. Zhang, G. Spear, Interaction type 2 dendritic cell precursors, Clin. Immunol. 101 (2001) 201–210.
of mannose-binding lectin with primary isolates of human immuno- [37] V. Soumelis, I. Scott, F. Gheyas, D. Bouhour, G. Cozon, L. Cotte, et
deficiency virus type 1, J. Gen. Virol. 81 (2000) 949 –955. al., Depletion of circulating natural type 1 interferon-producing cells
[18] P. Garred, H.O. Madsen, U. Balslev, B. Hofmann, C. Pedersen, J. in HIV-infected AIDS patients, Blood 98 (2001) 906 –912.
Gerstoft, et al., Susceptibility to HIV infection and progression of [38] J. Pacanowski, S. Kahi, M. Baillet, P. Lebon, C. Deveau, C. Goujard,
AIDS in relation to variant alleles of mannose-binding lectin, Lancet et al., Reduced blood CD123⫹ (lymphoid) and CD11c⫹ (myeloid)
349 (1997) 236 –240. dendritic cell numbers in primary HIV-1 infection, Blood 98 (2001)
[19] T. Pastinen, K. Liitsola, P. Niini, M. Salminen, A.-C. Syvanen, 3016 –3021.
Contribution of the CCR5 and MBL genes to susceptibility to HIV [39] H. Donaghy, A. Pozniak, B. Gazzard, N. Qazi, J. Gilmour, F. Gotch,
type 1 infection in the Finnish population, AIDS Res. Hum. Retrovir. et al., Loss of blood CD11c(⫹) myeloid and CD11c(⫺) plasmacytoid
14 (1998) 695– 698. dendritic cells in patients with HIV-1 infection correlates with HIV-1
[20] M. Boniotto, S. Crovella, D. Pirulli, G. Scarlatti, A. Spano, L. Vatta, RNA virus load, Blood 98 (2001) 2574 –2576.
et al., Polymorphisms in the MBL2 promoter correlated with risk of [40] F.P. Siegal, P. Fitzgerald-Bocarsly, B.K. Holland, M. Shodell, Inter-
HIV-1 vertical transmission and AIDS progression, Genes Immun. 1 feron-␣ generation and immune reconstitution during antiretroviral
(2000) 346 –348. therapy for human immunodeficiency virus infection, AIDS 15
[21] B.L. Sullivan, E.J. Knopoff, M. Saifuddin, D.M. Takefman, M.N. (2001) 1603–1612.
Saarloos, B.E. Sha, et al., Susceptibility of HIV-1 plasma virus to [41] G. Grouard, M.C. Rissoan, L. Filgueira, I. Durand, J. Banchereau,
complement-mediated lysis, J. Immunol. 157 (1996) 1791–1798. Y.J. Liu, The enigmatic plasmacytoid T cells develop into dendritic
[22] E.A. Berger, P.M. Murphy, J.M. Farber, Chemokine receptors as cells with interleukin (IL)-3 and CD40-ligand, J. Exp. Med. 185
HIV-1 coreceptors: roles in viral entry, tropism, and disease, Annu. (1997) 1101–1111.
Rev. Immunol. 17 (1999) 657–700. [42] G. Penna, S. Sozzani, L. Adorini, Cutting edge: selective usage of
[23] F. Cocchi, A.L. DeVico, A. Garzino-Demo, S.K. Arya, R.C. Gallo, P. chemokine receptors by plasmacytoid dendritic cells, J. Immunol. 167
Lusso, Identification of RANTES, MIP-1alpha, and MIP-1beta as the (2001) 1862–1866.
major HIV-suppressive factors produced by CD8⫹ T cells, Science [43] A. Dzionek, A. Fuchs, P. Schmidt, S. Cremer, M. Zysk, S. Miltenyi,
270 (1995) 1811–1815. et al., BDCA-2, BDCA-3, and BDCA-4: three markers for distinct
[24] C.E. Mackewicz, E. Barker, J.A. Levy, Role of ␤-chemokines in subsets of dendritic cells in human peripheral blood, J. Immunol. 165
suppressing HIV replication, Science 274 (1996) 1393–1395. (2000) 6037– 6046.
[25] M. Samson, F. Libert, B.J. Doranz, J. Rucker, C. Liesnard, C.M. [44] A. Yonezawa, R. Morita, A. Takaori-Kondo, N. Kadowaki, T.
Farber, et al., Resistance to HIV-1 infection in caucausian individuals Kitawaki, T. Hori, et al., Natural alpha interferon-producing cells
bearing mutant alleles of the CCR-5 chemokine receptor gene, Nature respond to human immunodeficiency virus type 1 with alpha inter-
382 (1996) 722–725. feron production and maturation into dendritic cells, J. Virol. 77
[26] R. Liu, W.A. Paxton, S. Choe, D. Ceradini, S.R. Martin, R. Horuk, et (2003) 3777–3784.
al., Homozygous defect in HIV-1 coreceptor accounts for resistance [45] T.B. Geijtenbeek, D.S. Kwon, R. Torensma, S.J. van Vliet, G.C. van
for some multiply-exposed individuals to HIV-1 infection, Cell 86 Duijnhoven, J. Middel, et al., DC-SIGN, a dendritic cell-specific
(1996) 367–377. HIV-1-binding protein that enhances trans-infection of T cells, Cell
[27] J.A. Levy, C.E. Mackewicz, E. Barker, Controlling HIV pathogene- 100 (2000) 587–597.
sis: the role of noncytotoxic anti-HIV activity of CD8⫹ cells, Immu- [46] D.S. Kwon, G. Gregorio, N. Bitton, W.A. Hendrickson, D.R. Littman,
nol. Today 17 (1996) 217–224. DC-SIGN-mediated internalization of HIV is required for trans-en-
[28] C. Lopez, P.A. Fitzgerald, F.P. Siegal, Severe acquired immune hancement of T cell infection, Immunity 16 (2002) 135–144.
deficiency syndrome in male homosexuals: diminished capacity to [47] A. Krug, A. Towarowski, S. Britsch, S. Rothenfusser, V. Hornung, R.
make interferon-alpha in vitro associated with severe opportunistic Bals, et al., Toll-like receptor expression reveals CpG DNA as a
infections, J. Infect. Dis. 148 (1983) 962–966. unique microbial stimulus for plasmacytoid dendritic cells which
[29] F.P. Siegal, C. Lopez, P.A. Fitzgerald, K. Shah, P. Baron, I.Z. Lei- synergizes with CD40 ligand to induce high amounts of IL-12, Eur.
derman, et al., Opportunistic infections in acquired immune defi- J. Immunol. 31 (2001) 3026 –3037.
ciency syndrome result from synergistic defects of both the natural [48] N. Kadowaki, S. Antonenko, Y.J. Liu, Distinct CpG DNA and poly-
and adaptive components of cellular immunity, J. Clin. Invest. 78 inosinic–polycytidylic acid double-stranded RNA, respectively, stim-
(1986) 115–123. ulate CD11c-type 2 dendritic cell precursors and CD11c⫹ dendritic
[30] F.P. Siegal, N. Kadowaki, M. Shodell, P.A. Fitzgerald-Bocarsly, K. cells to produce type I IFN, J. Immunol. 166 (2001) 2291–2295.
Shah, S. Ho, et al., The nature of the principal type 1 interferon- [49] H. Hemmi, T. Kaisho, O. Takeuchi, S. Sato, H. Sanjo, K. Hoshino, et
producing cells in human blood, Science 284 (1999) 1835–1837. al., Small anti-viral compounds activate immune cells via the TLR7
[31] K. Lennert, E. Kaiserling, H.K. Muller-Hermelink, T-associated MyD88-dependent signaling pathway, Nat. Immunol. 22 (2002) 22.
plasma-cells, Lancet 1 (1975) 1031–1032. [50] T. Ito, R. Amakawa, T. Kaisho, H. Hemmi, K. Tajima, K. Uehira, et
[32] M. Cella, D. Jarrossay, F. Facchetti, O. Alebardi, H. Nakajima, A. al., Interferon-alpha and interleukin-12 are induced differentially by
Lanzavecchia, et al., Plasmacytoid monocytes migrate to inflamed Toll-like receptor 7 ligands in human blood dendritic cell subsets, J.
lymph nodes and produce large amounts of type I interferon, Nat. Exp. Med. 195 (2002) 1507–1512.
Med. 5 (1999) 919 –923. [51] L. Fong, M. Mengozzi, N.W. Abbey, B.G. Herdier, E.G. Engleman,
[33] C.A. Biron, Interferons alpha and beta as immune regulators—a new Productive infection of plasmacytoid dendritic cells with human im-
look, Immunity 14 (2001) 661– 664. munodeficiency virus 1 is triggered by CD40 ligation, J. Virol. 76
[34] M.C. Rissoan, V. Soumelis, N. Kadowaki, G. Grouard, F. Briere, R. (2002) 11033–11041.
de Waal Malefyt, et al., Reciprocal control of T helper cell and [52] S. Tang, J.A. Levy, Inactivation of HIV-1 by trypsin and its use in
dendritic cell differentiation, Science 283 (1999) 1183–1186. demonstrating specific virus infection of cells, J. Virol. Methods 33
[35] N. Kadowaki, S. Antonenko, J.Y. Lau, Y.J. Liu, Natural interferon (1991) 39 – 46.
alpha/beta-producing cells link innate and adaptive immunity, J. Exp. [53] S. Patterson, J. Gross, P. Bedford, S.C. Knight, Morphology and
Med. 192 (2000) 219 –226. phenotype of dendritic cells from peripheral blood and their produc-
174 J.A. Levy et al. / Clinical Immunology 108 (2003) 167–174

tive and non-productive infection with human immunodeficiency to acquired immunodeficiency syndrome, Blood 92 (1998) 3105–
virus type 1, Immunology 72 (1991) 361–367. 3114.
[54] S. Patterson, A. Rae, N. Hockey, J. Gilmour, F. Gotch, Plasmacytoid [67] A.M. Gomez, F.M. Smaill, K.L. Rosenthal, Inhibition of HIV repli-
dendritic cells are highly susceptible to human immunodeficiency cation by CD8⫹ T cells correlates with CD4 counts and clinical stage
virus type 1 infection and release infectious virus, J. Virol. 75 (2001) of disease, Clin. Exp. Immunol. 97 (1994) 68 –75.
6710 – 6713. [68] E. Barker, K.N. Bossart, S.H. Fujimura, J.A. Levy, CD28-costimu-
[55] L.A. Evans, J. Moreau, K. Odehouri, H. Legg, A. Barboza, C. Cheng- lation increases CD8⫹ cell suppression of HIV replication, J. Immu-
Mayer, et al., Characterization of a noncytopathic HIV-2 strain with nol. 159 (1997) 5123–5131.
unusual effects on CD4 expression, Science 240 (1988) 1522–1525. [69] J.V. Giorgi, in: S. Gupta (Ed.), Immunology of HIV Infection, Ple-
[56] R.J. Pomerantz, M.S. Hirsch, Interferon and human immunodefi- num, New York, 1996, pp. 181–199.
ciency virus infection, Interferon 9 (1987) 113–127. [70] M. Kannagi, L.V. Chalifoux, C.I. Lord, N.L. Letvin, Suppression of
[57] J.J. Ferbas, J.F. Toso, A.L. Logar, J.S. Navratil, C.R. Rinaldo Jr., simian immunodeficiency virus replication in vitro by CD8⫹ lym-
CD4⫹ blood dendritic cells are potent producers of IFN-␣ in re- phocytes, J. Immunol. 140 (1988) 2237–2242.
sponse to in vitro HIV-1 infection, J. Immunol. 152 (1994) 4649 – [71] B.A. Castro, C.M. Walker, J.W. Eichberg, J.A. Levy, Suppression of
4653. human immunodeficiency virus replication by CD8⫹ cells from in-
[58] G. Pantaleo, C. Graziosi, A.S. Fauci, The role of lymphoid organs in fected and uninfected chimpanzees, Cell. Immunol. 132 (1991) 246 –
the pathogenesis of HIV infection, Semin. Immunol. 5 (1993) 157– 255.
163. [72] D.J. Blackbourn, C.P. Locher, B. Ramachandran, S.W. Barnett, K.K.
Murthy, K.D. Carey, et al., CD8⫹ cells from HIV-2-infected baboons
[59] K.F.T. Copeland, P.J. McKay, K.L. Rosenthal, Suppression of acti-
control HIV replication, AIDS 11 (1997) 737–746.
vation of the human immunodeficiency virus long terminal repeat by
[73] I.S. Choi, R. Hokanson, E.W. Collisson, Anti-feline immunodefi-
CD8⫹ T cells is not lentivirus specific, AIDS Res. Hum. Retrovir. 11
ciency virus (FIV) soluble factor (s) produced from antigen-stimu-
(1995) 1321–1326.
lated feline CD8(⫹) T lymphocytes suppresses FIV replication, J. Vi-
[60] C.E. Mackewicz, M.R. Garovoy, J.A. Levy, HLA compatibility re-
rol. 74 (2000) 676 – 683.
quirements for CD8⫹ T-cell-mediated suppression of human immu-
[74] C.E. Mackewicz, B. Wang, S. Metkar, M. Richey, C. Froelich, J.A.
nodeficiency virus replication, J. Virol. 72 (1998) 10165–10170.
Levy, Lack of the CD8⫹ cell anti-HIV factor in CD8⫹ cell granules,
[61] C.E. Mackewicz, L.C. Yang, J.D. Lifson, J.A. Levy, Non-cytolytic
Blood (2003) 180 –183.
CD8 T-cell anti-HIV responses in primary infection, Lancet 344
[75] C.E. Mackewicz, H. Ortega, J.A. Levy, Effect of cytokines on HIV
(1994) 1671–1673. replication in CD4⫹ lymphocytes: lack of identity with the CD8⫹
[62] S. Stranford, J. Skurnick, D. Louria, D. Osmond, S. Chang, J. Snin- cell antiviral factor, Cell. Immunol. 153 (1994) 329 –343.
sky, et al., Lack of infection in HIV-exposed individuals is associated [76] L.S. Diaz, M. Stone, C.E. Mackewicz, J.A. Levy, Differential gene
with a strong CD8⫹ cell noncytotoxic anti-HIV response, Proc. Natl. expression in CD8⫹ cells exhibiting non-cytotoxic anti-HIV activity,
Acad. Sci. USA 96 (1999) 1030 –1035. Virology (2003) 400 – 409.
[63] C.M. Walker, J.A. Levy, A diffusible lymphokine produced by [77] C.E. Mackewicz, D.J. Blackbourn, J.A. Levy, CD8⫹ cells suppress
CD8⫹ T lymphocytes suppresses HIV replication, Immunology 66 human immunodeficiency virus replication by inhibiting viral tran-
(1989) 628 – 630. scription, Proc. Natl. Acad. Sci. USA 92 (1995) 2308 –2312.
[64] C.E. Mackewicz, H.W. Ortega, J.A. Levy, CD8⫹ cell anti-HIV [78] C.E. Mackewicz, B.K. Patterson, S.A. Lee, J.A. Levy, CD8(⫹) cell
activity correlates with the clinical state of the infected individual, noncytotoxic anti-human immunodeficiency virus response inhibits
J. Clin. Invest. 87 (1991) 1462–1466. expression of viral RNA but not reverse transcription or provirus
[65] A.L. Landay, C. Mackewicz, J.A. Levy, An activated CD8⫹ T cell integration, J. Gen. Virol. 81 (2000) 1261–1264.
phenotype correlates with anti-HIV activity and asymptomatic clini- [79] C.E. Mackewicz, C.S. Craik, J.A. Levy, The CD8⫹ cell noncytotoxic
cal status, Clin. Immunol. Immunopathol. 69 (1993) 106 –116. anti-HIV response can be blocked by protease inhibitors, Proc. Natl.
[66] E. Barker, C.E. Mackewicz, G. Reyes-Teran, A. Sato, S.A. Stranford, Acad. Sci. USA 100 (2003) 3433–3438.
S.H. Fujimura, et al., Virological and immunological features of [80] C.E. Mackewicz, J.A. Levy, W.W. Cruikshank, H. Kornfeld, D. M.
long-term human immunodeficiency virus-infected individuals who Center, Role of IL-16 in HIV replication, Nature 383 (1996) 488 –
have remained asymptomatic compared to those who have progressed 489.