Oren J. Cohen

Audrey Kinter

Anthony S. Fauci

Host factors in the pathogenesisof HIV disease

Authors' address

Orm I. Cohni, Aainy Kinter, Aiihony S. Fouci,

National Institute of Allergy and Infectious

Diseases, Laboratory of Immunoregulation,

Bcthesda. Maryland, USA.

Correspondence to:

Oren ]. Cohen

National Institute of Allergy and InfectiousDiseasesLaboratory of Immunoregulation10 Center Drive. MSC 1876Building 10 Room 1 IB 13Bethesda MD 20892-1876USAFax: 1 301 402 0070e-mail: OCohen@nih.gov

Summapy: Host factors play an important role in determining rates of dis-ease progre.ssion in human immiinodefitiency virus (HIV)-infet:ted indi-viduals. HIV is ablt lo subvert the host immune system by infecting CD4*T cells that normally orchestrate immune responses and by inducing thesecretion of proinflammatory cytokines that the virus can utilize to its ownreplicative advantage. The recognition that certain chemokine receptorsserve as necessary co-factors for HIV entry into its target cells as well as thefact that ligands for these receptors can modulate the efficiency of HIVinfection has expanded the number and scope of host factors that mayimpact the pathogenesis of HIV disease. This area of investigation will nodoubt yield novei therapeutic strategies for intervention in HIV disease;however, caution is warranted in light of the enonnous complexity of thepteiotropic cytokine and chemokine networks and the uncertainty inher-ent in manipulating these systems.

HIV-infected long-term non-progressors represent an excellentmodel to study potential host factors involved in HIV disease pathogenesis.Genetic factors certainly have a major impact on the immune responsesmounted by the host. In this regard, a polymorphism in the gene for theHIV co-receptor CC chemokine receptor 5 (CCR5), which serves as a co-receptor for macrophage (M)-tropic strains of HIV, affords a high degreeof protection against HIV infection in individuals homozygous for thegenetic defect and some degree of protection against disease progressionin HIV-infected heterozygotes. HlV-specific immtme responses, includingcytotoxic T-lymphocyte (dX) responses and neutralizing antibodyresponses, also appear to play salutary roies in protecting against diseaseprogression.

/mmuno/ogicol Rewws J997

Vol. 159 31-48Printed in Denmark. AH rights reserved

Cop>7ight © Manksgaaii 1997

immunological ReviewsISSNOIOS-2896

Introduction

The pathogenesis of human immunodeficiency virus (HIV)disease is complex and influenced by both viral and hosl factors(1). The multifactorial nature of HIV disease pathogenesis isreflected by the highly variable rates of disease progre.ssion thatare observed in individuals infected with HIY The importanceof host factors in modulating rates of disease progression is fur-ther underscored by the observation that even individuals whowere apparently infected from a common source may experi-ence widely variant clinical outcomes (2). A great deal of atten-tion has been focused recently on new insights into the viral lifecycle that have been gained by studies of the decay of plasmaviremia during potent antiretroviral therapy (3-6). These stud-

31

Cohen et al • Host factors in HIV disease pathogenesis

ies have confirmed earlier work that demonstrated high levels

of viral replication throughout the course of HIV infection

(7-9) and have greatly expanded our understanding of the

dynamics of HIV rephcation in vivo. The remarkable consistency

in quantitative estimates ofthe rate of turnover of plasma vire-

mia raises important questions regarding the observed variabil-

ity in rates of disease progression. In this regard, future studies

will need to address whether the rate of viral turnover varies

according to stage of disease or whether it is an intrinsic char-

acteristic of HIV infection. In either case, it is necessary to

invoke host factors in order to explain the great variability in

rates of clinical disease progression.

A delicate balance among a wide array of host factors likely

determines the net rate of viral replication in HIV-infected indi-

viduals. Subversion ofthe human immune system by HIV (i.e.

infection of cells that are critical components of an intact

immune system, induction of the secretion of proinflamma-

tory cytokines, and utilization of these products of immune

activation for the replicative advantage of the virus) usually tips

the balance in favor ofthe virus. The recent discovery that cer-

tain chemokine receptors (for example. CC chemokine recep-

tor (CCR)5, CXC chemokine receptor (CXCR)4, CCR3. CCR2b,

STRL33, Bonzo, and BOB) are utilized by different strains of

HIV as co-factors to gain entry into cells has greatly expanded

the number of candidate host factors that may influence the

pathogenesis of HIV disease (10-18). The ability of the

chemokine ligands of these receptors to block HIV entry into

target celts and thereby tip the balatice of immtme control over

virus replication in favor of the host is a new concept in the

field of HIV pathogenesis that has major implications for

potential therapeutic intervention.

Genetic factors may determine the outcome of interactions

between virus and host in several ways. First, the host's HIV-

specific immune responses are constrained by the individual's

major histocompatibihty complex (MHC) alleles. In addition,

the recently discovered genetic defect in the CCRS gene has a

major impact on susceptihihty to HIV infection in individuals

homozygous for the defect, and on disease progression in HIV-

infected individuals heterozygous for the defect. HlV-specific

cellular and humoral immune responses likely play an impor-

tant role in the control of viral replication, although the precise

correlates of protective immunity have not been established;

however, recent studies have shown that qualitative as well as

quantitative features of these immune responses may be impor-

tant modulators of disease progression.

Appreciation ofthe role of host factors in the pathogenesis

of HIV disease should lead to the design of novel therapeutic

strategies. The goal of tipping the balance in favor of host con-

trol over virus replication may appear to be simple; however,

the extraordinary complexity of manipulating host factors to

this end is fraught with many potential complications. This lat-

ter point is highlighted by the negative outcomes of clinical tri-

als for bacterial sepsis that targeted molecules thought to be

directly involved in the pathogenesis of sepsis (for example,

lipopolysaccharide, interleukin (IL)-1, and tumor necrosis fac-

tor (TNF)-a) (19). The need to consider therapeutic options in

the context ofa balance between pro- and anti-inflammatory

mediators and the need to consider distal interactions in a com-

plex pleiotropic cytokine network apply not only to sepsis

(20). but to HIV disease as well.

Cytokines and HIV disease: dysreguiation of cytokine

production

A highly complex network of cytokines operates to regulate the

immune system. This network is redundant and pleiotropic.

and operates in an autocrine and paracrine manner to stimulate

or suppress cellular proliferation and differentiation, and to

modulate immune function (21). Chronic immune activation

induced by HIV infection and associated opporttinistic infec-

tions results in dysregulation ofthe cytokine network. Many of

the observed alterations in cytokine production contribute to

HIV pathogenesis by further stimulating viral replication, sup-

pressing the ability of the immune system to mount a strong

antiviral response, and inducing cytokine-mediated cytopathic

effects (1, 22-24).

Similar to other chronic infections, HIV infection is associ-

ated with increased expression of proinflammatory cytokines,

especially during the later stages of disease (24). High levels of

TNF-a, IL-1 [1, and IL-6 are secreted by peripheral blood mono-

nuclear cells (PBMC) (25-30) and macrophages (3 1-34) from

HIV-infected subjects. TNF-a, IL-lp, and IL-6 are also found at

elevated levels in the serum (35-40), cerebrospinal fluid

(41-43), and tissues (44—49). High levels of expression of

these cytokines. as well as interferon (IFN)-Y (46, 47, 50) and

IL-10 (47, 50), are particularly evident in lymphoid tissue, a

major site of HIV replication throughout the course of disease

(8,9,51.52). Clironically activated and expanded CD8* T cells

(47, 53) and macrophages (22, 54) are thought to be major

contributors to the elevated cytokine levels observed in HIV-

infected subjects.

In addition to alterations in cytokine production due to

chronic immune activation. HlV-specific upregulation of cer-

tain cytokines may occur. In this regard, production of proin-

flammatory cytokines can be upregulated in PBMC, lymph

node mononuclear cells (LNMC), and macrophages after acute

32 Immunological Reviews IS9/1997

Cohen et a) • Host factors in HIV disease pathogenesis

HIV infection in vitro or after treatment with HIV proteins, such

as envelope gpI 20 and Tflt (24. 55-6Z).

Another major disruption in the cytokine pattern observed

in HIV disease is a progressive loss in the ability to produce

immunoregulatory cytokines. such as IL-2 and IL-12 (63-66).

IL-2 and IL-1 2 are critical for effective cell-mediated immune

responses, as they stimulate proliferation and lytic activity of

cytotoxie T lymphocytes (CTL) and natural killer (NK) cells.

Tliese cell-mediated immune effectors represent the primary

mechanism whereby most viral infections are cleared. In addi-

tion. IL-12 is essential for stimulating the production of

T helper (Th)l-type cytokines, including IL-2 and IFN-7. that

favor the development of cell-mediated immune responses

(67-69). While it is clear that the Th I limb of cellular immune

responses is impaired during the course of HIV infection (65.

70-73), controversy surrounds the proposed dominance of

Th2-like responses (i.e. secretion of IL-4, IL-5, and IL-10) dur-

ing progression of HIV disease. Clerici et al. showed that stim-

ulated PBMC from HIV-infected patients exhibit a preferential

Th2 pattern of cytokine secretion with disease progression (65.

70.71. 74); however, other investigators have found a skewing

of the cytokine secretion pattern of T cells from HIV infected

patients toward a ThO state (i.e. secretion of eytokines charac-

teristic of both Th I and Th2 patterns) rather than toward a Th2

state (47. 72, 73). In either case, the fmding that HIV replica-

tion is more efficient in ThO compared to Th 1 clones (72, 75)

highlights the importance of impaired Thl responses in the

pathogenesis of HIV disease (76).

Effects of cytokines on HIV replication

The effects of cytokines on HIV repUcation were recognized in

early studies wherein aetivated PBMC (77), macrophages (55.

78). and B cells (79) were shown to produce soluble factors

that could dramatically upregulate HIV expression in acutely

and chronically infected cells ofthe lymphocytic and macroph-

age lineages. These observations led to the identification of

numerous cytokines that can directly influence HIV repUcation

in infected cells (22, 24, 80) (Fig. 1).

Cytokines that have been reported to upregulate HIV repli-

cation in vitro include IL-lp. IL-2. IL-3. IL-6, IL-7 (81), IL-12

(82. 83). IL-1 5 (82. 84). TNF-a. TNF-p. and the colony-stim-

ulating factors (CSF) macrophage (M)-CSF and granulocyte-

macrophage (GM)-CSF (reviewed in (24)). IFN-a. IFN-p. and

[L-16 (85, 86) are primarily suppressors of HIV production.

whereas other cytokines. such as IL-4 (87). IL-10 (88. 89).

IL-1 3 (87). IFN-Y and TGF-|J, reduce or enhance viral replica-

tion depending on the infected cell type and the culture condi-

Fig. I. Endogenous cytokines regulate viral repUcation in CD4* T cells.Numerous cyiokines. particularly the proinflammatory cytokines TNF-a,IL- ip, and IL-6, strongly upregnlate viral replication. TGF-P and iL-10downregulatc viral replication: in the case of IL-10. this effect is at leastin part due to downregulation of proinflammatory cytokines. Tlie p che-mokines, which are secreted by a variety ofcell types including CDS' andCD8' mononuciear ceils, strongly inhibit infection by M-tropic strains ofHIV-1, whereas SDF-1 inhibits infection with T-tropic strains.[Adapted from ( i ) ]

tions (22, 24. 80). Many cytokines. such as the interferons andTNF-a, can influence HIV rephcation in both T cells and mac-rophages. while others, such as M-CSF. are cell lineage-specific.The effects of a particular cytokine are often greatly influencedby the activity of other cytokines present in the microenviron-ment. In this regard, certain cytokines have been demonstratedto act in a synergistic (88. 90. 91) or in an antagonistic (92.93) manner with other cytokines in regulating HIV replication.Finally, cytokines are pleiotropic and the overall effects of a par-ticular cytokine on HIV repUcation often reflect the balance ofboth HIV-inducing and HIV-inhibiting activities.

Proinflammatory cytokines. particularly TNF-a. are con-sidered the most potent HIV-inducing cytokines and theirmechanism of action is relatively well understood. Both TNF-aand IL-1 p activate the cellular transcription factor nuclear factor(NF)KB (94, 95). a strong inducer of HIV long terminal repeat(LTR)-mediated transcription. IL-6 alone appears to increaseHIV expression primarily by a post-transcriptional tnechanism;however. lL-6 can synergize with NFtcB-indueing cytokines toenhance HIV transcription (90). The role of endogenousproinflammatory cytokines in the regulation of HIV replicationhas been demonstrated in several cellular systems in vitro. Theproduction of HIV by tnacrophages or PBMC stimulated withphysiologic inducers of proinflammatory cytokine production,such as bacterial endotoxin or IL-2. can be partially or nearlycompletely abrogated by the addition of anti proinflammatorycytokines (96. 97). neutralizing antibodies, or receptor antag-onists (ra), such as IL-lra (98). In cultures of HIV-infectedmacrophages. the viral suppressive activity of several cytokines.

Immunologital Reviews ] 59/1997 33

Cohen et al • Host factors in HIV disease pathogenesis

such as IL-10 and TGF-p, is attributable largely to their ability

to inhibit the secretion or activity of HlV-inducitig proinflam-

matory cytokines (92. 93. 96, 99). HIV production by infected

T cells is sensitive to both the antiproinflammatory and the

antiproliferative activity of such cyiokines (D. Goletti & A.S.

Fauci, unpublished data).

Although the role of proinflammatory and antiproinflam-

matory cytokines in the regulation of HIV replication in vivo has

not been demonstrated conclusively, several lines of evidence

suggest that these cytokines may be involved in regulating viral

production. Administration of pentoxifylline. an inhibitor of

the secretion and activity of TNF, to HIV-infected individuals

was found to reduce HIV viremia in concert with a reduction

in plasma levels of TNF-a (100. 101). The role of proinflam-

matory cytokines in maintaining steady-state levels of HIV rep-

lication is suggested by the observation that in vivo infusion of a

single bolus of IL-10 to HIV-infected subjects resulted in a rapid

and tnodest, albeit transient, decrease in plasma viremia

(D. Weissman & A.S. Fauci, unpublished data). The kinetics of

HIV suppression in vivo correlated v 'ith a dramatic reduction in

the ability of cells from these subjects to be induced in vitro to

secrete TNF-a and IL-1 p. Furthermore. IL-10 has been found to

inhibit acute HIV infection in severe combined immunodefi-

ciency (SCID) mice engrafted with human fetal thymus and

liver (102). The ability of IL-10 to suppress T-cell activation

and proliferation likely also plays a prominent role in its ability

to suppress HIV replication in vivo (103-105).

In addition to the use of immunosuppressive cytokines

which may depress HIV-inducing immune responses, cytok-

ines which stimulate T cells or anti gen-presenting cells have

been administered to HIV-infected subjects for a number of

years- The use of cytokine-based therapies aimed at immune

reconstitution in HIV disease has expanded over the past several

years, particularly with the development of potent antiretrovi-

ral therapies that limit the potential for cytokine-mediated

increases in viral replication. In this regard, administration of

IL-2 to asymptomatic HIV-infected subjects receiving concom-

mitant antiretroviral therapy results in significant and sustained

increases in CD4+ T-cell numbers with no long-term effects on

viremia (106. 107). Similar immune reconstitution therapies

are being proposed for IL-12. IL-! 3, and IL-15 (84. 108-1 15).

New studies are cotitinuing to expand the list of cytokines for

use as potential immunotherapeutic agents. A particularly

interesting cytokine-based immunotherapeutic approach is

suggested by a recent report demonstrating that transfection of

a CD4* T-cetl line with DNA encoding tbe 130 amino acid

form of IL-16 renders cells virtually resistant to HIV infection

(85). IL-! 6-mediated inhibition of HIV in this system appears

ro be due to interference with viral transcription (85, 86). This

effect may be due to the ability of IL-16 to suppress T-cell acti-

vation (116. 117). The combination of IL-2 and IL-16 is a par-

ticularly attractive option that may synergistically enhance the

expansion of CD4+ T cells (N. Parada, personal communica-

tion).

In addition to the known cytokines mentioned previously,

several unidentified soluble factors have been demonstrated to

exert dramatic HIV-modulatory activity Foremost among these

is the elusive CD8* cell-derived HIV-suppressive factor(s).

While lytic activity is an important component of CD8+ cell-

mediated HIV suppression (118). cell-free supernatants from

cultures of activated CD8- cells and cell lines are able to dratnat

ically inhibit HIV repUcation in both T cells and macrophages

(119). CD8 antiviral factor (CAF). first described by V alker et

al. (120-122), is non-cytolytic. suppresses HIV replication in

a non-MHC-restricted manner at the level of HIV LTR transcrip-

tion (123-125), and lacks identity to known cytokines (126).

A distinct group of HIV-suppressive factors secreted by

CD8+ T cells was identified by Coccbi et al. (127). These inves-

tigators attributed the HIV-suppressive actvity of CD8* cells to

the combined activities of certain cbemoattractant cytokines

(i.e. chemokines), including macrophage inflammatory pro-

tein (MlP)-la. MIP-lp, and RANTES (Regulated upon Activa-

tion, Normal T cell Expressed and Secreted). An unexplained

fmding in the study by Cocchi et al. was that although the com-

bination of the p chemokines MIP-I a. MIP-lp. and RANTES

potently suppressed the replication of several M-tropic HIV

strains, they had virtually no effect on the replication of the

T-cell line-adapted (TCLA) strain. HIV-1 IIIB. Soon after this

report, Feng et al. described the seven transmembrane orphan

receptor fusin, previously known as LESTR and HUMSTR and

currently designated CXCR4, as a co-receptor for T-cell

(T)-tropic strains of HIV (15. 16). In addition, three groups

described a new cbemokine receptor, CCR5. which bound

MIP- la. MIP-1 p. and RANTES as its natural ligands (128-130).

In light ofthe previous results of Cocchi et al. the obvious ques-

tion that arose was whether CCR5 might function as a co-

receptor for M-tropic strains of HIV A series of papers from five

different laboratories elegantly demonstrated this to be the case

(10-14). Chemokine receptors that can function as HIV co-

receptors are apparently involved in tlie fusion process that

occurs between viral and celltilar membranes (10-14. 131).

The general picture that has emerged is that M-tropic strains of

HIV-1 utilize primarily the P-chemokine receptor CCR5 (and.

to a lesser extent. CCR3) for entry and the natural CCR5 ligands

MIP-la, MIP-lp, and RANTES inhibit cellular entry of these

HIV strains. In contrast. TCLA strains, such as IIIB, exclusively

34 Immunological Reviews 159/1997

Cohen et al • Host factors in HIV disease pathogenesis

employ the a-chemokine receptor CXCR4, an interaction that

is blocked by the CXCR4 ligand stromal derived factor-1

(SDF-I). Many primary T-tropic HIV i.solaces exhibit a broad

range of CCR usage, including CXCR4 and CCR5 (132. 133).

The recent discoveries of other HIV co-receptors have already

made obsolete the simplistic notion that CCRS and CXCR4 are

the only important co-receptors for M- and T-tropic strains of

HIV, respectively (17, 18, 134).

Numerous cell types produce a variety of chemokines

(135, 136). and modulation ofthe production of these factors

may influence HIV replication in a strain-specific manner

(Fig. 1). Therefore, the overall effect of immune activation and

the secretion of proinflammatory or immunoregulatory cytok-

ines on HIV replication must now be considered in the context

1 if potential influences on chemokine production, chemokine

co-receptor expression, and the predominant viral quasispecies

that is replicating in vivo. Chemokine production, induced dur-

ing inflammation. I.s enhanced by several cytokines. including

TNF-a. IL-p, and immunoregulatory cytokines, such as IL-2

and IL-15 (135, 137-139). Thus, in HIV-infected subjects in

the early stages of disease, the ability of TNF-a to stimulate

ii chemokine production and thereby suppress M-tropic entry

may override its HIV-inducing effects; however, in individuals

harboring predominantly T-tropic quasispecies in the later

.stages of HIV disease, only the HIV-inducing activity of TNF-a

would be influential, In fact, TNF-a-mediated induction of

j3-chemokine secretion may actually enhance entry and replica-

lion of T-tropic strains of HIV (A. Kinter & A.S. Fauci, unpub-

lished data) (Fig. 2).

Similarly, cytokines that modulate the expression of

chemokine receptors would be expected to exert variable

strain-dependent effects on HIV rephcation and spread. In this

regard, IL-2 has been shown toupregulate the expression of the

M-tropic co-receptor CCR5 (140),

The puzzling bottleneck in HIV transmission that so

heavily favors emergence of M-tropic, non-syncytium-induc-

ing (NSI) strains of virus in the new host (141, 142) may in

part be due to the differential regulatory patterns of the relevant

HIV co-receptors (140. 143). In this regard, CCR5 expression is

predominantly seen in previously activated, memory T cells (i.e.

CD26'"K''CD45RA''™CD45RO+), whereas CXCR4 expression is

seen in naive, unactivated cells (i.e. CD26'™CD45RA*CD45RO-).

h is therefore plausible that the profound degree of immune

activation that occurs during acute HIV infection may result in

high expression of CCR5 and low expression of CXCR4 (M.

Ostrowski & A.S. Fauci, unpublished data). Similarly, co-infec-

lion with various other pathogens may differentially modulate

expression of HIV co-receptors and thereby exert selective pres-

Early Stage Disease

® ® M-Tropic00®® HIV

Late Stage Disease

O O T-Tropic

f iJCC

Fig. 2. Proinflammatory cytokines, such as TNF-tj;, have potentiallydichotomous effects on HIV replication. During ihe early stages ol HIVdisease (left), M-tropic strains of the virus predominate. Although TNF-acan transcriptionally upregulate HIV expressiun in infected cells, at the sametime it induces expression of tiie p chemokines (jiCC) RANTES, MIP-1 a,and MiP-1 p. These chemokines occupy the HIV co-receptor CCR5, block-ing entry of HIV into target cells. In contrast, T-tropic strains of HIV maypredominate in the late stages of HIV disease (right). In this situation, theinduction of pCC by TNF-a cannot block T-tropic HIV entry via CXCR4and in fact may enhance rephcation of T-tropic strains of HIV.

sure on HIV strains that use the co-receptors in question (H.

Moriuchi. M. Moriuchi & A. S. Fauci, unpublished data).

The observation that the natural hgands of CC chemokine

receptors that are utilized as HIV co-receptors act as potent

inhibitors of viral entry has dramatically broadened the known

.spectrum of HIV-inhibitory cytokines and the mechanisms

whereby they influence HIV rephcation. Therapeutic

approaches currently being considered include administration

of HIV-inhibitory chemokines. chemokine antagonists that

occupy HIV co-receptors without transmitting a signal (144).

and cbemokines that are engineered to hind to HIV co-recep-

tors intracellularly and remain in the endoplasmic reticulum

(i.e. a phenotypic knockout) (S-Y, Chen, personal communica-

tion). Clearly, the efficacy of such approaches would depend on

the co-receptor usage and phenotype of the predominant viral

population rephcating in vivo. In this regard, exogenous and

endogenous p chemokines suppress HIV rephcation in vitro in

Immunologiciii Ro'iews IS9/1997 35

Cohen et al • Host factors in HIV disease pathogenesis

PBMC from asymptomatic HIV-infected individuals harboring

predominantly M-tropic HIV strains (145). but not in PBMC

from individuals with more advanced disease harboring pre-

dominantly T-tropic HTV strains (146) (A. Kinter & A.S. Fauci,

unpublished data). Similarly. HIV isolates obtained longitudi-

nally from infected individuals with rapid disease progression

exhibit reduced sensitivity to inhibition by p chemokines in vitro

over time (146. 147).

A cautionary note about these potential therapies comes

from the known association of the transition from M-tropic/

NSI to T-tropic/syncytium-inducing (SI) viruses with disease

progression. The tran.sition from an NSI to SI virus may occur

by mutation of only a few amino acid residues predominantly

in the envelope V3 loop (148-1 55). The HIV envelope V3 loop

has also been shown to be a major determinant of co-receptor

usage (I 56). Given the error rate of viral reverse transcriptase

and the rapid dynamics of viral replication, mutations in the

HIV envelope gene that encode SI strains must appear very early

in disease; however, failure of such mutants to emerge until late

in the disease process indicates a change in the selective advan-

tage of such a mutation dtiring the course of disease progres-

sion. Because SI variants are able to use a broader range of entry

co-receptors (for example. CXCR4) compared with NSI

viruses, it is possible that SI variants emerge in response to high

levels of p chemokines that block cellular entry of viruses

which utiUze CCR5 (i.e. predominantly NSI viruses) (132,

146. 147). This potential effect of p chemokines should be

investigated since the emergence of T-tropic HIV strains in vivo

is associated vrith rapid CD4^ T-cell decline and disease pro-

gression (157). Further caution is warranted in hght of poten-

tial dichotomous effects of the p chemokines on HIV replica-

tion in different cell types (119, 158). The situation in vivo is no

doubt highly complex, and multiple host factors as well as reg-

ulatory aspects of co-receptor expression in different tissue

compartments likely determine the environment in which

selection for NSI or SI variants is made (1. 140, 159),

While in viiro culture systems and cell line models have

allowed investigators to identify numerous host factors that

influence HIV replication and to delineate the mechanisms

whereby these factors suppress or enhance viral replication, it

is difficult to anticipate how manipulation of these factors will

ultimately influence HIV replication in vivo. It is clear that host

factors function within the context of an interactive immuno-

regulatory cytokine network and can have pleiotropic effects on

HIV replication, some of which are viral strain-specific. Never-

theless, numerous host factors have proven or promising

immunotherapeutic potential that should be further explored

and pursued clinically for the treatment of HIV disease.

Immune activation

Within 6 months to I year following primary HIV infection,

plasma viremia appears to stabilize to a steady state or set point

that is a strong prognostic indicator of the rate of disease pro-

gression (160). Underlying this deceptively stable viremia is a

high rate of virus production and clearance (approximately 10'"

virions/day). 93-99% of which is produced by newly infected

CD4* T cells (3-6). Thus, even during clinically asymptomatic

stages of HIV infection, persistent virus production serves as a

potent source of immnne activation and subsequent cytokine

secretion; diese activities, in tnrn, stimulate further viral replica-

tion.

In vitro, cellular activation is essential for productive HIV

infection of CD4+ T cells (161. 162), and agents that interfere

with T-cell activation dramatically inhibit HIV rephcation in

these cells (163, 164). The role of immnne activation in stim-

ulating HIV replication in vivo is demonstrated by increases in

viremia in HIV-infected individuals persistently or transiently

exposed to exogenous immune stimuli. In this regard. HIV-

infected natives of sub-Saharan Africa, who experience persis-

tent immune activation due to chronic exposure to parasites

and other pathogens, harbor high viral loads associated with

rapid progression of HIV disease (165. 166). Similarly, co-

infection with opportxmistic pathogens, such as active tubercu-

losis (167-170) or pneumocystis pnenmonia (171), results in

dramatic increases in levels of plasma HIV viremia that return

to baseline upon successful treatment of the opportunistic

infection (01). The source of elevated viremia during 01 was

suggested by a recent study demonstrating that lymphoid tissue

macrophages produce high levels of HIV in the setting of 01

(172).

Confirmation of the role of immune stimulation in HIV

replication has been established in studies demonstrating that

immunization of HIV-infected subjects with influenza (173) or

tetanus toxoid (1 74) antigens results in transient, but substan-

tial, increases in plasma viremia. Furthermore, PBMC from

HIV-uninfected subjects were rendered more susceptible to

HIV infection in vitro following immunization with tetanus tox-

oid (174).

Long-term non-progressors: a model to study host factors

in the pathogenesis of HIV disease

In recent years, it has become clear that in a small percentage of

HIV-infected individuals, no evidence of disease progression

can be detected over a prolonged period of time (175-183).

Definitions of long-term non-progression are somewhat arbi-

36 Inununologicd Revitws i 59/1997

Cohen et al • Host factors in HIV disease pathogenesis

irary; however, a reasonable consensus definition includes doc-

mneniation of HIV infection for more than 7 years, a CD4^

T-cell count greater than 600 cells/fiL without significant

decline over time, no symptoms of HlV-induced disease, and

no history of antiretroviral therapy (184). Although a minority

of cases of long-term non-progressive HIV infection may be

associated with attenuated strains of HIV (185-189). most data

suggest that viral attenuation is rare among long-term non-pro-

jijressors, and that host factors play a dominant role in deter-

mining the state of non-progression (180, 181, 190-192).

Genetic factors

Host genetic factors influence the rate of disease progression in

HIV infection. A number of different mechanisms may be

responsible for the observed associations between certain HLA

liaplotypes and different rates of HIV disease progression

(193-196). The ability of certain HLA molecules to efficiently

[iresent immunodominant viral epitopes in order lo generate

cell-mediated immune responses may explain an association

with slow disease progression. Conversely, other HLA mole-

cules may promote immunopathogenic responses associated

with more rapid disease progression. In a recent study

HLA-B27, B57, and B51 were most strongly associated with

slow progression of HIV disease, while HLA-A23, B37. and

B49 were associated with rapid progression (196), An HLA

profile was developed that distinguished a 6-fold difference

between rates of disease progression in rapid versus slow pro-

gressors. Other genetic factors linked to rates of HIV disease

progression include allelic forms ofthe vitamin D-binding fac-

tor Gc (197), variant alleles of mannose-binding lectin (198),

and the TNF c2 microsatellite allele (199).

CCR5 is a major co-receptor for M-tropic strains of HIV-1

(see above) (10-14). A mutant allele ofthe CCR5 gene that

contains an internal 32 base pair deletion resulting in a trun-

cated protein (200-202) has a major impact on susceptibility

to HIV infection and on rates of disease progression in HIV-

infected individuals. Homozygosity for the CCR5 mutation

results in near-total protection from HIV-1 infection (200,

202-207). Heterozygosity for the CCR5 mutation results in

decreased expression of CCR5 on the cell surface and reduced

infectability of CD4* cells with M-tropic strains of HIV-1 com-

pared to CD4+ cells from CCR5 wild-type individuals (208).

Although heterozygosity for CCR5 does not appear to afford

protection against HFV-1 infection in vivo, it may confer partial

protection against disease progression in HIV-infected individ-

uals (1S9, 202-204, 209, 210). Protection against disease pro-

gression in CCR5 heterozygotes is due in part to tbe lower viral

1.000,000

^ 100.000

IIu<z

10,000

1.000

100CCR5

Wild-TypeCCRS-A32

Heterozygotes

Fig. 3. Levels of plasma viremia are indistinguishable amongHIV-infected long-term non-progressors stratified by CCR5genotype. Dark bars represeni geometric means.

load "set point" after HIV seroconversion and a slower rate of

CD4+ T-cell depletion compared with CCR5 wild-type individ-

uals (204).

Heterozygosity for the CCR5 mutation is significantly

more common in cohorts of HIV-infected long-term non-pro-

gressors compared to HIV-infected control populations (1S9,

202, 209, 210). However, despite the fact that the frequency of

CCR5 heterozygotes is increased 2-fold among non-progres-

sors compared to HIV-infected controls, still fewer than 50% of

non-progressors are CCRS heterozygotes (202. 209). Tbe pos-

sibihty that CCR5 heterozygotes might constitute a subgroup

among non-progressors with tbe lowest viral loads and most

preserved CD4+ T-cell counts was investigated. Interestingly,

CCR5 wild-type and heterozygous long-term non-progressors

were indistinguishable with regard to multiple immunologic

and virologic parameters of disease activity (209). Mean CD4*

T-cell counts were 910 cells/|iL among CCRS wild-type non-

progressors and 885 cells/^L among CCR5 heterozygous non-

progressors. Geometric mean levels of plasma viremia were

2,104 HIV RNA copies/mL among CCR5 wild-type non-pro-

gressors and 1.995 HIV RNA copies/mL among CCR5 het-

erozygous non-progressors (209) (Fig. 3).

We have previously demonstrated that, in contrast to indi-

viduals with progressive disease, HTV-infected long-term non-

progressors maintain intact lymphoid tissue architecture (180,

211). However, a great deal of heterogeneity among non-pro-

gressors is evident in the degree of follicular hyperplasia and

viral trapping within germinal centers (180, 211). When strat-

ified according to CCR5 genotype, wild-type and heterozygous

non-progressors were again indistinguishable with regard to

these parameters (Fig. 4).

Immunological Rev'icivs 159/1997 37

Cohen et al • Host factors in HIV disease pathogenesis

CCR5 Wild-TypePatient A Paiient B

CD21

HIVIn Situ

CCR5-A32 HeterozygotesPatient C Patient D

CD21

HIVIn Situ

Fig. 4. The degree of follicularhyperplasia and the degree of virustrapping in lymph node germinalcenters are indistinguishable amongHIV-infected long-term non-progressors stratified by CCR5genotype. CD21 staining, whiclihighlights follicular dendritic cells ingerminal centers, appears red. HIVRNA detected by in siiu hybridizationappears as white grains, in darkfieldmicrographs. Representative lymphnodes from non-progressors wilhabundant and scant folhcularhyperplasia and virus trapping areshown for each CCR5 genotype.[Adapted from (209)]

Taken together, these data indicate that although CCR5 het-erozygotes have an increased chance of hecoming non-progres-sors. HIV-infected CCR5 wild-type individuals may arrive at the

same non-progressor phenotype by other mechanisms.

Host immune response

CTL responses

HlV-specific CTL play an important role in the control, alheit

incomplete, of HIV replication and spread (212, 213). High

precursor frequencies of HlV-specific CTL with broad specific-

ity have been consistently detected in long-term non-progres-

sors compared to progressors (180, 214-217). Quahtative

aspects of the HlV-specific CTL response are also important

determinants of the efficacy of the CTL response in controlling

viral replication. Maintenance of CTL responses specific for

viral core proteins is associated with a decreased risk of disease

progression (21 7, 218): this association does not appear to be

true for CTL responses against other viral proteins. Recognition

of immunodominant CTL epitopes presented hy particular

MHC class I alleles may result in potent anti-HIV activity (219)

and may in part explain the association of certain MHC class I

alleles with slower progression of HIV disease (194— 196). Fur-

thermore, the skewing of the T-cell receptor Vp repertoire in

HlV-infected patients has suggested that the ahiiity to recruit an

HlV-specific CTL response composed of a heterogeneous group

of yp families during primary infection is associated with bet-

ter control of viral rephcation and an improved prognosis com-

pared lo mobihzation and expansion of CTL from only one or

two Vp families (220). Clona! exhaustion of mobilized CTL

may he a common mechanism of viral persistence. This appears

to be a strategy for viral persistence employed by certain strains

38 Immimologicoi Reviews 159/1997

Cohen el dl • Ho i t lactor-s in HIV disease pathogenesis

of lymphocyiic choriomeningitis virus that rapidly and com-

pletely mobilize the host CTL response resulting in CTL exhaus-

tion (i.e. high-zone tolerance) (221). CTL exhaustion may

occur to some degree in HIV infection where disappearance of

certain originally expanded CTL cionotypes can he demon-

strated in the absence of viral escape mutations that might oth-

erwise explain the phenomenon (222).

Taken together, these observations argue against an immu-

nopathogenic role for CTL in HIV disease (223) and in favor of

a salutary role in the maintenance of low viral load and the state

of non-progression. This inference is further supported by the

demonstrated role of CTL in reducing levels of plasma viremia

during primary HIV infection (224. 225), and the association

of progression to AIDS with late viral escape from a long-lived

(9-12 years) immunodominant CTL response (226).

The host CTL response against HIV is constrained by the

ahihty ofthe host's MHC class I alleles to bind to various viral

t'pitopes, while the virus is constrained by the degree to which

an escape mutation impairs viral fitness. These host-virus

dynamics are extraordinarily complex given the large number

of permutations of viral epitopes and MHC class I alleles. Viral

imitations within CTL recognition epitopes (i.e. "escape

mulants") are associated with increased levels of viral rephca-

lion and progression of HIV disease (226-229), Viral escape

mutants may thrive due to the release of CTL control over their

replication and may also inhibit CTL responses against the pre-

L-scape viral epitope (230. 231), However, certain viral escape

mutations may be costly to viral fitness. In this regard, it has

lieen reported that diffuse infiltrative CD8 lymphocytosis in

MIV infection was associated with certain HLA types that appar-

ently constrain evolution of viral sequence diversity in the

envelope V3 loop (232). Other studies have highhghted the

constraints on the host CTL response imposed by MHC class I

alk'les. Il has been reported that in an HIV-infected individual,

CTL clones specific for an HLA-B14-restricted epitope of gp41

displayed very limited diversity of T-cell receptor utilization

(233). Furthermore, limited plasticity of certain CTL responses

in individuals with viral escape mutants, where the dominant

CTL response may remain largely directed at the pre-escape

viral epitope. has been demonstrated (234, 235). The possibil-

ity that increased plasticity of the CTL response may allow the

host to maintain more continuous and effective control over

viral replication was suggested by studies demonstrating

increased viral sequence diversity and generation of vigorous

L-scape mutant-specific CTL responses in slow progressors

(236. 237). A mathematical model of CTL-virus dynamics has

heen proposed that describes disease progression as a result of

viral sequence variation that escapes an immunodominant CTL

response and shifts the host response towards a weaker epitope

(238). Thus, disease progression may be the result of fitness of

viral escape mutants outpacing the plasticity of the host CTL

response, and slow progression may be the result of CTL plas-

ticity overpowering viral escape mutants with limited fitness

Soluble HlV-suppressor factors

Antiviral soluble factors elaborated by CD8* T cells may also

play a role in non-progression of HIV infection, CAF, first

described by Walker et al. (120-122), is non-cytolytic and

non-MHC-restricted, inhibits viral replication at the level of

HIV LTR transcription (1 23-125), and lacks identity to known

cytokines (126). CAF activity was found to correlate with stage

of disease (239): fewer CD8^ T cells from asymptomatic

patients without significant CD4+ T-cell depletion were

required to suppress viral replication in vitro compared to CD8+

T cells from patients with advanced stage HIV disease. Studies

of long-term non-progressors have demonstrated more potent

CD8- T'Cell-derived soluble antiviral responses compared with

progressors (181, 240).

RANTES. MlP-ltt. and MIP-1|J are also important antiviral

soluble factors secreted by CD8* T cells as well as other cell

types (127, 145). These chemokines are natural ligands for the

chemokine receptor and M-tropic HIV co-receptor CCR5 and

inhibit viral replication primarily at the level of cell entry. Con-

flicting data have been obtained regarding a relationship

between levels of these chemokines and progression of HIV

disease (182, 240-245). These conflicting data are not surpris-

ing since they came from studies of sera or stimulated PBMC. A

recent report does, however, support a possible role for the CC

chemokines in the protection of some exposed uninfected indi-

viduals against HIV infection. Upon stimulation with HIV anti-

gens, CD4* T cells from these individuals secreted high levels

of CC chemokines that were capable of inhibiting the replica-

tion of M-tropic strains of HIV in vilro (246). Ultimately, levels

of expression ofthese chemokines in lymphoid tissue, the pri-

mary site of HIV replication in vivo, may be the most relevant

measurement with regard to an association with rates of disease

progression.

The differential chemoattractant properties of RANTES, a

ligand for CCR5 which attracts memory T cells (247). and SDF-

1. the hgand for CXCR4 which attracts naive T cells (140). may

influence the migration of populations of anti-HIV effector

cells as well as potential HIV target cells in lymphoid tissue. It

is interesting to speculate that a number of scenarios involving

HIV co-receptors and their Ugands in lymphoid tissue may

impact viral rephcation (Fig. 6), limited availability of HIV co-

Immunoiogicul Rfvicvvs 159 /1997 39

Cohen et al • Host factors in HIV disease pathogenesis

CTL-X RBCImtmoni ofCTL-X'

\Efficimt CTL-X'

response

Clearance olepitope X

PervstHKtofnwmorji CTL-X

X' niutanon doei notraintneit

Genetic Factors

CXCR4]CCR2bCCR3 ISTRL33

CCR5 +/+ CCR5 +/- CCR5 - / -

Immunoregulatory Factors

I RANTES, MIP-1a.MIP-1p

MIP-ip

RANTES

CCR5

SDF-1

CCRS

CCRS CXCR4

CXCR4

CXCR4

Fig. S. The possible outcomes following CTl recognition of an HIVepitope are complex. An efliciciu CTL response against epitope X mayresult in clearance of the epitope and persistence of memory CTL (loprighl). Persistence of the viral epiiope may be the result of a variety offactors, including poor recognition of ihe epitope by CTL, defectiveclonogenic potential ofthe CTL, downregulaiion of HLA class I moleculesby viral proteins, and/or exhaustion of the CTL clone (middle right).Alternatively, a CTL escape mutation may occur in the epitope (bottomright). The outcome in this situation depends both on the relativeefficiency of the CTL response directed against the escape mutant as wellas on the relative cost of the mutaiion to viral fitness.

Fig. 6. Genetic as well as immunoregulatory factors govern theavaikbility of functional HIV co-receptors. The CCRS-A3 2 alleleencodes a molecule that does not function as an HIV co-receptor:individuals who are homozygous for this mutant allele are affordediiear-coniplete protection against HIV Infection, whereas heterozygotesare partially protected against disease progression. Polymorphisms inother co-recepior genes will likely also be found to correlate with rales ofdisease progression. Upregulation ofthe co-receptor ligands (RANTES,MlP-la, andMlP-lpinthecaseofCCR5,andSDF-l in the case ofCXCR4) and/or downregulation ofthe co-receptors themselves may alsolimit the availability of functional HIV LO-receptors.

receptors may occur by virtue of genetic polymorphisms {i.e.

CCR5-A32) or by downregulation of their messenger RNAs or

protein products (248). Alternatively, upregulacion ofthe nat-

ural ligands ofthe HIV co-receptors may prevent HIV access to

functional co-receptors and thereby limit infection of target

cells. However, it is important to appreciate other consequences

of occupancy of HIV co-receptors by their natural ligands. As

noted above, high concentrations of RANTES. MIP-la. and

MIP-1 p may inhibit entry and replication of M-tropic strains of

HIV; however, they may also represent a selective pressure by

the host immune system that may favor the emergence of T-

tropic strains of HIV Ligation of CCR5 may also result in intra-

cellular signaling events that may actually enhance rephcation

of T-tropic strains of HIV (A. Kinter & A.S. Fauci, unpublished

data), These possibilities serve as a cautionary note to therapeu-tic strategies that employ chemokines or their antagonists(144). Finally, IL-16 has been reported to be a soluble antiviralfactor (249); however, a relationship between IL-I 6 prodtic-tion and disease progression remains to be established (240.245).

Humoral responses

The relationship between humora l i m m u n e responses to HIV

and disease progression remains uncertain ( 2 5 0 ) . N o n - p r o -

gressive HIV infection has been associated wi th lack of an anti-

body response to an epi tope (residues 5 0 3 - 5 2 8 ) wi th in the

carboxy-terminus of g p l 2 a ( 2 5 1 ) , wi th maintenance of high

levels of p24-specific ant ibodies ( 2 5 2 ) . and wi th maintenance

40 Immunoiogica! Reviews 159/J 997

Cohen et al - Host factors in HIV disease pathogenesis

of neutralizing antibodies (180. 181. 253). It has been

reported that the presence of HIV IIIB-neutralizing antibodies

correlated with a more favorable prognosis (254). Subsequent

studies demonstrated that the presence of neutralizing antibod-

ies to primary HW isolates and to autologous virus was associ-

ated with non-progression (255, 256). Furthermore, viral

oscape from neutralizing antibody responses is associated with

emergence of the SI phenotype of HIV and with disease pro-

gression (255, 257). HIV-infected long-term non-progressors

tend to maintain antibody responses that can neutralize a broad

[lanel of primary isolates and also maintain neutralizing anti-

bodies against autologous virus isolates; however, non-progres-

sors are a heterogeneous group with regard to these neutraliz-

ing antibody responses (253, 256). Whether the maintenance

of neutralizing antibodies in non-progressors is simply a

marker for a relatively intact immune system or whether these

antibodies play an active role in determining the state on non-

progression remains ttnclear

Lymphoid tissue: substrate for immune competence

The morphologic abnormalities of lymphoid tissue a.ssociated

with HIV disease progression are important determinants of

immunodeficiency (8, 258-264). Despite the long period of

HIV infection in long-term non-progressors, histopathologic

examination of lymph node biopsies from these individuals

revealed only mild HlV-related abnormalities, such as follicular

hyperplasia (180, 211). Follicular involution, fibrosis, and

lymphocyte depletion, associated with progressive HIV disease,

were found to be lacking in lymph nodes from non-progres-

sors. The degree of follicular hyperplasia seen in non-progres-

sors is significantly lower compared to that seen in progressors

and qualitatively distinct as well, without evidence oflarge geo-

graphic germinal centers extending into the nodal medulla

(180, 21 1). It is likely that preservation of lymphoid architec-

ture in non-progressors is a reflection ofthe lower levels of viral

replication over time in these individuals. Regardless of the

mechanisms responsible for lower levels of viral replication in

non-progressors, preservation of lymphoid tissue architecture

i.s a critical component of the immunocompetence observed.

This further highlights the need to understand the mechanisms

responsible for the destruction of lymphoid architecttire dur-

ing progression of HIV disease. If inimunorestorative strategies

in advanced HIV infection are to be successftJ, substrate for the

generation of immune responses (i.e. intact lymphoid tissue)

must be present, necessitating the prevention or reversal ofthe

histopathologic abnormalities of lymphoid tissue associated

with HIV disease progression.

ReferencesFaiK-i A. Host factors and the pathogenesis ofHlV-induced disease.Naiure 1996;384:5Z9-534.Dii S-L. et al. Divergeni patterns ofprogression to AIDS after infection from ihesame source: Human immunodeficiencyvirus type 1 evolution and antiviral responses.J Virol 1997;7l;428+-l-295.Ho DD, Neumann AU Perelson AS. Chen W.Leonard JM. Markowit^ M. Rapid turnover ofplasma virions and CD4 lymphocytes in HIV-1 infection.Nattire 1995:373:123-126.Wei X. et al. Viral dynamics in humanimmunodeficiency virus type I infection.Nature 199Si373:1 i7-l 22.Perelson AS, Neumann AU, Markowitz M,Leonard IM. Ho DD. HIV-1 dynamics in vivo:virion clearance rate, infected cell life-span,and viral generation time.Science l996;271:iS82-lS86.Perelson AS. et al. Decay characteristics ofHIV- i infected compartments duringcombinaiion iherapyNature 1997:387:188-191.

10.

12.

PiatakM. eial. High levels of HIV-1 in plasma 13,during all stages of infection determined hycompetitive PCR.Science 1993;259:l 749-1754.PantaieoG. ei al. HIV infection is active and 14.progressive in lymphoid tissue during theclinically latent stage of disease.Nature l993;362r3S5-3S8.Embretson ], et al. Massive covert infection ofhelper T lymphocytes and macriiphages hy 15.HIV during the incubation period of AIDS.Nature 1993:362:359-362.Deng H. et al. Identification of a major co-receptor for primary isolates of HIV-1.Nature I996;38t :661-666. 16,

Dragic T. et ai. HiV-1 entry into CD4+ cells ismediated by the chemokine receptor CC-CKR-S,

Nature 1996;38l:667-673.Alkhatib G. et at. CC CKR5: A RANTES,MIP- i a, MIP-11) receptor as a fusion cofactorfor macrophage-tropic HIV-1.Science 1996:272:1955-1958.

Choe H. et al. The p-chemokine receptorsCCR3 and CCRS faciliiaie infection hyprimary HIV-i isolates.Cell 1996:85:1 135-1 138.E)oranz B, el al. A dual tropic primary HIV-1isolate that uses fusin and the |i-chemokinereceptors CKR-5, CKR-3, andCKR-2basfusion CO factors.Cell 1996;8S:I149-|IS8.Feng Y. Broder C, Kennedy R Berger E. HIV-1entry cofactor: functional cDNA cloning of aseven-transmembrane domain, G-proteincoupled receptor.Science 1996:272:872-877.Berson I, Long D. Doranz B. Rucker J. Jtrik F,Doms R, A seven-transmembrane domainreceptor involved in fusion and entry of T-cell-tropic human immunodeficiency virustype I strains.J Virol 1996;7O:6288-629S,

Immunological Reviews 159 / ! 997 41

Cohen et al • Host factors in HIV disease pathogenesis

17. Liao F, Alkhatib G. Peden K. Sharma G.

Berger E, Farber J. STRL3 3, a novelchemokine receptor-Uke protein, functions asa fusion cofactor for both macrophage-cropicand T ceil line-tropic HIV-1.

JExpMed 1997;18S:2O15-2O23.18. Deng H. Unutmaz D, KewaiRamani V.

Littman D. Expression cloning of newreceptors used by simian and humanimmunodeficiency viruses;.

Nature 1997:388:296-300.19. Glauser M. The inflammatory cylokines. New

developments in the pathophysiology andtreatment of septic shock.

DTUgs 1996;52 (Suppl 2):9-17.

20. Bone R. Immunologic dissonance; acontinuing evolution in our understanding ofthe systemic inflammatory responsesyndrome (SIRS) and the multiple organdysfimction syndrome (MODS).

Ann Intern Med l996;t25:690-691.21. Paul W, Seder R. Lymphocyte responses and

cytokines.Cell !994;76:241-251.

22. Alonso K, Poiitiggia P. Medenica R, Riz20 S.Cytokine patterns in adults with AIDS.Immunol Invest 1997;26:34!-350.

23. Miedema F. Immunologica] abnormalities inthe natural history of HIV infection:mechanisms and chiiical relevance.Immunndefic Rev 1 992;3:1 73-193,

24. Poll G, Fauci A, Role of cytokines in thepatliogenesis of human immunodeficiencyvirus infection. In: Aggarwal B. Puri R, eds.Human cytokines; their role in disease andtherapy. Cambridge, MA: Blackwell Science;199S.p, 421-^49,

25. Lau A, Der S, Read S. Williams B- Regulationof tumor necrosis factor receptor expressionby add-labile interferon-a from AIDS ,5era,AIDS Res Hum Retroviruses 199I;7:54S-SS2,

26. Roux-Lombard R Modoux C, Cruchaud A.Dayer I. Purified blood monocytes from HIV-I infected patients produce high levels ofTNF-a and IL-1.

Clin Immunol Immunopathol 1989;50:374-384,

27. Hober D. Haque A. Wattre P Beaucaire G,Mouton Y, Gapron A. Production of tumornecrosis factor-alpha (TNF-a!pha) andinterleukin-! (IL-1) in patients with AIDS.Enhanced level of TNF-alpha is related to ahigher cytotoxic activity.

Gin Hxp Immunol 1989;78:329-333.28. Voth R, et al. Differential gene expression of

IFN-alpha and tumor necrosis factor-alpha inperipheraJ blood mononuclear cells frompatients with AIDS related complex and AIDS.I Immunol 1990:144 970-975,

29, Vyakarnam A, et al. Altered production oftumour necrosis factors alpha and beta andinterferon gamma by HIV-infectedindividuals.

Glin Exp Immunol I991;84:1O9-1! 5,30, Weiss L, Haeffner-Cavaillon N. Laude M.

Gilquin J, Kazatchkine MD, HIV infection isassociated with the spontaneous productionof interleukin-1 (IL-1) in vivo and with anabnormal release of IL-1 alpha in vitro.AIDS 1989:3:695-699.

31, Millai A. Miller R, Foley N, Meager A,Semple S, Rook G, Production of tumornecrosis facror-alpha by blood and lungmononuclear phagocytes from patients withhuman immunodeficiency virus-related lurigdisease.

Am] Respir Gell Mol Biol 1991 ;S: 144-148,32, Krishnan V, Meager A, Mitchell D, Pinching A,

Alveolar macrophages in AIDS patients:increased spontaneous tumor necrosis factoralpha production in Pneumocystis cariniipneumonia.

Ciin Exp Immunol 1990:80:156-160,33, Antinori A. et al. Alveolar macrophages from

AIDS patients spontaneously produce elevatedlevels of TNF-aipha in vitro,Allergol Immunopathol 1992;20:249-254.

34, Israel-Biet D. Cadranel ], Beldjord K, AndrieuJ. Jeffrey A, Even P Tumor necrosis factorproduction in HIV-seroposiiivt subjects:relationship with lung opportunisticinfections and HIV expression in alveolarmacrophages,

JImmunol 1991 ;147:490-494,35, Reddy M, Sorrell S. Unge M. Grieco M,

Tumor necrosis factor and HIV p24 antigenlevels in serum of HIV-infected populations.J Acquir Immune Defic Syndr 1988:1:

36, Lahdevirta J, Maury C. Teppo A. Repo H.Elevated levels of circulating eachectin/tumornecrosis factor in patients with acquiredimmunodeficiency syndrome.AmJMed 1988:85:289-291.

37, Arditi M. Kabat W. YogevR. Serum tumornecrosis factor alpha, interleukin I -beta, p24antigen concentrations and CD4+ cells atvarious stages of human immunodeficiencyvirus 1 infection in children.

Pediatr Infect Dis 1991:10:450-455,38, Scott-Algara D, Vuillier F, Marascescu M.

deSaintMartin J, Dighiero G. Sermn levels ofIL-2, IL-1 alpha, TNF-alpha. and solublereceptor of IL-2 in HIV-1 -infected patients.AIDS Res Hum Retroviruses 1 99 1:7:381-386,

39, Breen EC, et al. Infection with HIV isassociated with elevated IL-6 levels andproduction,JImmunol 1990:144:480-484,

40, Rautonen J, Rautonen N, Martin NL. Philip R,Wara DW Serum interleukin-6 concentrationsare elevated and associated with elevatedtumor necrosis factor-alpha andimmunoglobulin G and A concentrations inchildren with HIV infection.

AIDS 1991;S:13I9-1325.41, Grimaldi LM, et al. Elevated alpha-tumor

necrosis factor levels in spinal fluid fromHIV-1 -infected patients with centralnervous system involvement,AnnNeuroI 1991:29:21-25,

42, Gallo R Frei K, Rordorf G, Uzdins J,Tavolato B, Fontana A. Human immuno-deficiency virus type I (HIV-1) infection ofthe central nervous system: an evaluation ofcytoldnes in cerebrospinal fluid,

J Neuroimmunol 1989:23:109-116.43, Lauren2i MA. Siden A, Per.sson MA,

Norkrans G. Hagberg L. Chiodi F.Cerebrospinal fluid interIeukin-6 activity inHIV infection and inflammatory and non-inflammatory di.seasc ofthe nervous system,Clin Immunol Immunopathol 1990:57:233-241,

44, Kotler Dp Reka S, Clayton F, Intestinalmucosal inflammation associated withhuman immunodeficiency infection.Dig Dis Sci 1993:38:1 119-1127,

45, Emilie D, et al. Production of imerieukins inhuman immunodeficiency virus-1-rephcating lymph nodes,JClin Invest 1990:86:148-159,

46, Boyle MJ. ei al. Increased expression ofJnterferon-gamiTia in hyperplastic lymphnodes from HIV-infected patienls.

Glin Exp Immunol I993:92:!00-105,47, Graziosi G, et al. Lack of evidence for the

dichotomy of TH 1 and TH2 predominance inHIV-infected individuals.

Science 1994:265:248-252.

48, Wesselingh S, et al, Intracerebral cytokinemessenger RNA expression in acquiredimmunodeficiency .syndrome dementia,AnnNeuroI 1993:33:576-582,

49, Dreno B, Milpied B, Dutartre H, Litoux P.Epidermal interieukiii I in normal skin ofpatients with HTV infection,

BrJ Dermatol 1990:123:487-*92-50, Graziosi C. et al. Kinetics of cytokine

expression during primary humanimmunodeficiency virus type I infection.Proc Natl Acad Sci USA 1996:93:4386--J391.

51, Fox C, Tenner-Ricz K. Racz P, Firpo A, Pizzo PFauci A, Lymphoid germinal centers arereservoirs of human immunodeficiency virustype I RNA,) Infeci Dis 199 1:164:1051-1057.

52, Pantaleo G, et al, Lymphoid organs functionas major reservoirs for humanimmunodeficiency virus.Proc Nad Acad Sci USA 1991:88:9838-9842,

42 Immunological Rmcm 159/1997

Cohen et al • Host factors in HtV disease pathogenesis

S i. Jassoy C, el al. Human immunodeficiencyvirus type I -specific cytotoxic T lymphocytesrelease gamma interferon. tumor necrosislactor alpha (TNF-alplia), and TNF-beiawhen they encounter their target antigens,JVirol 1993;67:Z844-2852.

54. Esser R, et al. Secretory repertoire of HIV-infected human monocytes/maerophages.Pathohiology i991;S9:2t9-222-

55. Cbiisf KA, Robbins PB. Fernie B. OsiroveJM.Fauci AS. Viral antigen stimulation oftheproduction of human monokines capable ofrfgiilating HIV-I expression.

JImmunol l989;143:470-^75.56. Molina J, Scadden D, Byrn R. Dinarello C,

Groopman J. Pri:>diictioii of tumor necrosisfactor alpha and interleukin 1 beta bymonocytic cells infected with humanimminiodeficiency virus.

JCtinlnvesi 1989;84:733-737,

57. Rieckmann P. Poli G. Fox C, Kehrl J. Fauci A,Recombinant gpl 20 specifically enhancesmmor necrosis facior-alpha production andIg sccreti(.)n in B lymphocytes from HIV-infecied individuals but nol fromseronegaiive donors.

JImmunol 1991:147:2922-2927.58. Merrill ], Koyanagi Y, Chen I. Interleukin-1

and tumor necrosis factor alpha can beiiidiicfd from mononutlear phagocytes byhuman immunodeficiency virus type 1binding to (he CD4 receptor.

JVirol i989;63;4404-4408.59. Brady H. Abraham D. Pennington D, Miles C.

Jenkins S. Dzier^ak E. Altered cytokineexpression in T lymphocytes from humaninimunodefitiency virus Tat transgenic mice.J Virtii 1995;69:7622-7629.

60. Hotman F. Wright A. Etohadwala M.Wong-Staal F, Walker S. Exogenous tatproiein activates human endothelial cells.Blood 1993:82:2774-2780.

(i 1, Blaz.evic V, Heino M. Lagerstedt A, Ranki A,Krohn K. Interleukin-10 gene expressioninduced by HIV-1 Tat and Rev in the cells ofHIV-1 infected individuals.I Atquir Immune Defic Syndr Hum Retrovirol1996:13:208-214,

62, T -ring S, et al. Synthetic peptidescorresponding to sequences in HIV envelopegp41 and gpi 20 enhance in vitro productionof inierleukin-1 and tumor necrosis factorbut dcprtvss production of lntorferon-alpha,interferon-gamma and interleukin-2.Viral Immunol 1991:4:33-42.

<> 3, Lane HC. Depper JM, Greene WC, Whalen G.Waldmann TA, Fauci AS. Qualitative analysisof immime function in patients with theacquired immunodeficiency syndrome.Evidence for a selective defect in solubleantigen recognition.N EnglJMed 198S:3I3:79-84.

64- Schulick RD. Clerici M, Dolan MJ.

Shearer GM. limiting dilution analysis ofinterleukin-2-producing T cells responsive torecall and alloantigens in humaninununodeficiency virus-infected anduninfected individuals.EurJImmunol 1993:23:412-417.

65. Clerici M, et al. Restoration of HlV-specificcell-mediated immune responses byinterleiikin-11 in vitro.Science 1993:262:1721-1724,

66. Meyaard L, Schuitemaker H, Miedema F.T-cell dysfunction in HIV infection: anergydue to defective antigw-presenttng cellfunction?

Immunol Today 1993;14:I61-164.67. Romagnani S, Human TH1 and TH2 subsets:

regulation of differentiation and role inprotection and immtinopathology,

Int Arch Allergy Inimunoi 1992:98:279-285.

68. Hsieh CS. Macatonia SE. TVipp CS. Wolf SF.O'Garra A. Murphy KM, Developtnent ofTH ICD4-I- T cells throngh IL-12 produced byListeria-induced maerophages.

Science 1993:Z6O:S47-S49.69. Trinchieri G. Interleukin-12 and its role in the

generation of THl celts,tmmunol Today 1993:14:335-338.

70. Cierici M. et al. Changes in interleukin-Z andinterleukin-4 production in asymptomatic,buman immunodeficiency virus-seropositiveindividuals.

JClin Invest 1993;91:759-765,71. Clerici M. et al. Role of interleukin-10 in

T helper cell dysfunction in asymptomaticindividuals infected with the liumanimmunodeficiency virus,

JCiin Invest 1994:93:768-775.72. Maggi E. et al. Ability of HIV to promote a

TH 1 to THO shift and co replicatepreferentially in TH2 and THO cells.Science 1994;265:24-l-248.

73. Meyaard L. Otto S. Keet I. vanljer R,Miedema F. Changes in cytokine secretionpatterns of CD4-t- T-cell clones in humanimmunodeficiency virus infection.Blood 1994:84:4262-4268.

74. Clerici M, Shearer G. A TH 1-TH2 switch is acritical step in the etiology of HIV infection.Immunol Today 1993:14: !07-l 11,

7 5, Vyakarnam A, Matear R Martin S, Wagstaff M,Th 1 cells specific for HIV-1 gag p24 are lessefficient than ThO ceils in supporting HIVreplication, and inhibit virus replication inThO cells.Immunology 1995:86:85-96.

76, Clerici M. et al. Type 1 cytokine productionand low prevalence of viral isolation correlatewith long term non-progression in HIVinfection,AIDS Res Hum Retroviruses 1996:12:1053-1061,

77, FolksTM.JustementJ. Kinter A. Dinarello CA.Fauci AS, Cytokine-induced expression ofHIV-1 in a chronically infected promonocytecell line.

Science 1987:238:800-802.

78, Clouse KA. et al, Monokine regulation ofhuman immunodeftciency virus-1expression in a chronically infected humanT cell clone,JImmunol 1989:142:431-438,

79, Rieckmann P. Poli G. Kehrl ]H. Fauci AS.Activated B lymphocytes from humanimmunodeficiency virus-infected individualsinduce virus expression in infected T cellsand a promonocytic cell line, U1,JExpMed 1991:173:1-5,

80, Poli G. Fauci AS, Cytokine modulation ofHIV expression,

Semin Immunol 1993;5:I65-173,81, Smithgall M, Wong JG, Critchett K, HafFar O.

IL-7 up-regulates HIV-1 replication innaturally infected peripheral btoodmononuclear cells,

JImmunol 1996:156:2324-2330-82, Bayard-McNeeley M. Doo H, He S. Hafner A,

Johnson W, Ho J. Differential eflects ofinterleukin-12, interleukin-15, andinterleukin-2 on human immunodeficiencyvirus type 1 replication in vitro,Clin Diagn Lab Immunol 1996:3:547-553,

83, Kiiiter AL. Bende SM, Hardy EC. Jackson R.Fauci AS, Interleukin 2 induces CD8+ T cell-mediated suppression of humanimmunodeficiency virus replication in CD4+T cells and this effect overrides its ability tostimulate virus expression,

Proc Nail Acad Sci USA 1995:92:I0985-I0989,

84, Lucey D, et al. In vitro immunologic andvirologic effects of interleukin 1S onperipheral blood mononuclear cells fromnormal donors and humanimmunodeficiency virus type 1 -infectedpatients.

Ciin Diagn Lab Immunol 1997;4:43-48.85, Zhou P. Goldstein S, Devadas K. Tewari D.

Notkins A, Human CD4-K cells transfectedwith IL-16 cDNA are resistant to HIVinfection: Inhibition of mRNA expression.Nat Med 1997;3:659-664,

86, Maciaszek J, Parada N, Cruikshank W, CenterD, Kornfeld H, Viglianti G, IL-16 repressesHIV-1 promoter activityJImmunol 1997:158:5-8,

87, NaifH. Li S. Ho-Shon M, MathijsJ.Williamson R Cunningham A, The state ofmattu-ation of monocytes into macrophagesdetermines the effects of IL-4 and IL-13 onHIV replication,

JImmunol 1997;lS8:501-5 11.

Immunological Rrviem 159/1997 43

Cohen et al • Host factors in HIV disease pathogenesis

88, Weissmaii D, Poli G, Fauci AS. IL-10synergizes widi mutUpie cyiokines inenhancing HIV production in cells ofmonocytic lineage.

] Acqiiir Immune Defic Syndr Hum Retrovirol1995:9:442-449.

89. Finnegan A. et al, IL-10 cooperaies with TNF-aipha to activate HIV-1 from latently andacutely infected cells ofmoiiocyie/macrophage lineage.Jlmmunol i996;156:84l-8Sl,

90, Poli G, ei al. Interieukin 6 induces humanimmunodeficiency virus expression ininfected monocytic cells alone and in synergywith tumor necrosis factor alpha bytranscripcional and post-transcripnonalmechanisms.

JExpMed l990;i72;151-158,91. Poli G, Kinter AL, Fauci AS. Ituerleukin 1

induces expression ofthe humanimmunodeficiency virus alone and in syneigywith interleukin 6 in chronically infected Ulcells: inhibition of inductive effects by tbeinterleukin 1 receptor antagonisl,

Proc Natl Acad Sci USA I994;9t: 108-112,92, Scfiuitemaker H, et aJ. Prohferation-

dependent HIV-1 infection of monocytesoccurs during differentiation intomacrophages,

JClin invest 1992;89;1 154-1 160,93, Poli G, Kinter AL. Justement JS, Bressler ?.

Kehrl JH. Fauci AS. Transforming growthfactor beta suppresses humanimmunodeficiency virus expression andreplication in infected cells ofthemonocyte/macrophage lineage,JExpMed l991;173:S89-597,

94. Osbom L, Kunkel S. Nabel GJ. Tumor necrosisfactor alpha and interleukin I stimulate thehuman immunodeficiency virus enhancer byactivation of the nuclear factor kappa B,Proc Natl Acad Sci USA 1989;86;3336-2340,

95. Duh EJ. Maury WJ, Folks TM. Fauci AS.Rabson Afl, Tumor necrosis factor alphaactivates human immunodeficiency virustype 1 through induction of nuclear factorbinding to the NF-kappa fl sites in the longterminal repeat,Proc Nat! Acad Sci USA 1989;86:5974-5978,

96. Goletti D. Kinier AL, Hardy FC, Poll G.Fauci AS, Modulation of endogenous IL-lheta and IL-1 receptor antagonist results inopposing effects on HIV expression inchronically infected monocytic cells.Jlmmunol 1996;lS6:3501-3508,

97, Weissman D. Poli G. Fauci AS. Interleukin 10blocks HIV replication in macrophages byinhibiting the autocrine loop of ttmiornecrosis factor alpha and interleukin 6induction of virus.AIDS Res Hum Retroviruses 1994;10:1199-1206,

98, Kinter AL. Poli G, Fox I. Hardy E, Fauci AS,HIV replication in IL-2-stimulated peripheralblood mononuclear cells is driven in anautocrine/paracrine manner hy endogenouscytokines.

JItiimunol 1995:154:2448-2459,99, Schuitemaker H, IL4 and ILl 0 as potent

inhibitors of HIVI replication inmacrophages in vitro: a role for cytokines inthe in vivo virus host range?

Res Immunol 1994:145:588-592,100, Wallis RS, et al, Pentoxifylline therapy in

htnnan immun odeficienc)- virus-seropositivepersons with tuberculosis: a randomizedcontrolled trial,

J Infect Dis 1996:174:727-733.101, Clerici M. et al. Penioxifylline improves cell-

mediated immunity and reduces humanimmunodeficiency virus (HIV) plasmaviremia in asymptomatic HIV-seropositivepersons.

JlnfectDis l997;i75:l2IO-l215,102, KoilmannTR. etal. Inhibition ofacute in vivo

human immunodeficiency virus infection byhuman interieukin 10 treatment of SCID miceimplanted with human fetal thymus and liver.Proc Natl Acad Sci USA 1996;93:3126-3 131,

103, MasoodR, etal. E.-10inlubltsHIV-!replication and is induced by tat,Biochem Biophys Res Comniun 1994:202:374-383,

104, Moore K. O'Garra A, de Waal Malefyt R,Vieira P. Mosmann T Jnterleukin-10,Antiu Rev Immunol 1993:11:165-190,

105, Ho A, Moore K. Interleukin-10 and itsreceptor,

Ther Immunol 1994:1:173-185,106, Kovacs JA. et aJ. Controlled trial of

inEerleukin-2 infusions in patients infectedwith the human immunodeficiency virus.NFnglJMed 1996:335:1350-1356.

107, Kinter A. Fauci AS. Interleukin-2 and humanimmunodeftciency virus infection:padiogenic mechanisms and potential forimmunologicenhancement,Immunol Res 1996:15:1-15,

108, Seder R, Grabstehi K. Berzofsky J, McDyerJ,Cytokine interactions in humanimmunodeficiency virus-infectedindividuals: roles of interleukin (IL)-2.IL-12.and IL-i5.JExpMed 1995;182:1067-1077.

109, Kanai T. Thomas E, Yasutomi Y, Letvin N.IL-15 stimulates the expansion of AIDSvirus-specific CTL.Jlmmunol 1996:157:3681-3687,

I 10, Lin S. Roberts R. Ank B, Nguyen Q.Thomas E. Stiehm E, Htunanimmunodeficiency virus (HIV) type-1GPI 20-specific cell-mediated cytotoxieity(CMC) and natural killer (NK) activity inHIV-infected (HIV-I-) subjects: enhancementwithimerIeukin-2(IL-2), IL-12. and IL-15.CUn Immunol Immunopathol 1997:82:163-173,

111, Chehimi J. et al, IL-15 enhances immunefunctions ditring HIV infection,Jlmmunol 1997:158:5978-5987.

112, McKenzie A, Zuraw,ski G. Interleukin-13:characterization and biologic properties.Cancer Treat Res 1995:80:367-378,

113, Montaner L, Gordon S, Th2-mediated HIV Ivirostatic state: macrophage-specificregulaiion in vitro.

Res Immuno] 1994:145:583-587,114, Gately M, Mulqueen M. lnterleukin-12:

potential clinical applications in the treatmentand prevention of infectious diseases.Drugs 1996:52 (Suppl 2):18-25.

I I 5, Trinchieri G, Scott P The role of inicrleukin1 2 in the immune response, disease and

therapy

Immunol Today 1994:15:460-463.1 16, Cruikshaiik W. et al. IL-16 inhibition of CD3-

dependent lymphocyte activation andproliferation,

Jlmmunol 1996:157:5240-5248.117, Theodore A, Center D. Nicoll). Fine G,

Kornfeld H, Cruiksliank W CD4 ligatid IL-16inhibits the mixed lymphocyte reaction.Jlmmunol I996;157:1958-1964,

) 18 Yang OO, ei al. Suppression of human

immunodeficiency virus type I replication byCD8+ cells: evidence for HLA class !-restricted triggering of cytnlytic andnoncytolytic mechanisms.

) Virol 1997:71:3120-3128,119, Moriuchi H. Moriuchi M, Combadiere C,

Murphy P, Fatici A, CD8+ T-cell-derived.soluble fact()r(s). but not p-cheniokinesRANTES. MIP-1 a. and MIP-1 p. suppressHIV-1 replication inmonocyte/ macroph ages.Proc Nad Acad Sci USA 1996:93:15341-15345.

120. Walker CM, Moody DJ. Stites DR levy JA.CD84 Lymphocytes can control HIV infectionin vitro by suppressing virus replication.Science 1986:234:1563-1566.

121, Walker CM. Levy JA, A diffusible lymphokineproduced by CD8 + T lymphocytes suppressesHIV replication.Immunology 1989:66:628-^30.

122. Walker CM, Erickson AL. Hsueh FC. Levy JA,Inhibition of human immunodeficiency virusreplication in acutely infected CD4-t- cells byCD8+ cells involves a noncytotoxicmechanism,J Virol 1991:65:5921-5927.

44 Imnitmological Reviews 159/1997

Cohen Gt al • Host factors in HIV disease pathogenesis

I 23, Clien CH. Wt-inhold KJ. Bartlett JA.Bolognesi DP. Greetiberg ML, CD8+ Tlymphocyte-mediated inhibition of HIV-1long terminal repeat transcription: a novelantiviral mechanism.AIDS Res Hum Retroviruses 1993:9:1079-1086,

I 24. Copeland KF. McKay PJ. Rosenthal KL,Suppression of activation ofthe luimanimmunodeficiency virus long terminal repeathy CD8-»- T cells is not lentivirus specific.AIDS Res Hum Retroviruses 1995:11:1321-1326.

125, MackewiczC, Balcklxnirn DJ. LevyjA. CD8-I-T cells suppress human immunodeficiencyvirus replicaton by inhibiting viraltranscription,Proc Nail Acad Sci USA 1995:92:2308-2312,

12b. Mackewii:/ CE, Oretega H, Levy JA, Effect ofcytokines on HIV replication in CD4+-lymplioiyics: lack of identity with the CD8-I-cell antiviral factor.Ceil Immunol 1994:153:329-343,

127, Cocchi F. DeVico A, Gar7,ino-Demo A. Arya S,Gallo R. LtLsso P Identification of RANTES,MIP-1 a. and MIP-1 p as the major HIVsuppressive factors produced by CD8+T cells,Science 1995:270:181 1-1815.

I 28, Samson M. Labbe O. Mollereau C. Vassart G.Parmentier M. Molecular cloning andfunctional expression of a new CC-chomokine receptor gene. CC-CKR5.Biochemistry 1996:35:3362-3367,

129, Combadiere C. AhujaS. Tiffany H. Murphy RCloning and functional expression of CCCKR5, a human monocyle CC chemokinereieptor selective for MIP-1 (alpha). MIP-l(beia). and RANTES,JLeukocBiol 1996:60:147-152,

I iO, Raporc C. Gosling ]. Schweickart V. Gray P.Charo I. Molecular cloning and functionalcharacterization ofa novel human CCchemokine receptor (CCR5) for RANTES,MIP-1 bota. and MIP-1 alpha,J Biol Clieni i 996:271:1 7 16 1-1 7 166,

III , Oravecz T. Pall M. Norcross M, P-chemokineinhibition of monocytotropic HIV-1infection. Interference with a postbindingfusion step,Jlmmunol 1996:157:1329-1332.

I il. Simmons G, et al. Primary, syncylium-inducing human immunodeficiency virustype 1 isolates are dual-tropic and most canuse either Lestr or CCR5 as coreceptors forvirus entry,I Virol 1996.70,8355-8360.

H 3, Zliang L. Huang Y. He T. Cao Y, Ho D, HIV-1subtype and second receptor use.Nature 1996;383:768,

1 )4, He J. et al. CCR3 and CCR5 are co-receptorsfor HIV-1 infection of niicroglia.Nature 1997:385:645-649,

135, Schalt X Biology ofthe RANTES/sis cytokinefamily

Cyiokine 1991:3:165-183,136, Baggiolini M, Dewald B, Moser B, Human

chemokines: an update,

Annu Rev Immtinol 1997:15:675-705.137, Nelson P. Kim H. Manning W. Goralski T,

Krensky A, Genomic organization andiranscriptional regulation ofthe RANTESchemokine gene,

Jlmmunol 1993:151:2601-2612.138, Rathanaswami R Hachicha M. Sadick M.

Schall T. McCoU S. Expression ofthe cytokineRANTES in luiman rlieutnaioid synovialfibroblasts, Diflerent iai regulation of RANTESand interleukin-8 genes by inflammatorycytokines.

JBiolChem 1993:268:5834-5839,

139, Schall T, Bacon K. Chemokines. leukocytetrafficking, and inflanimatiim,Curr Opin Immunol 1994:6:865-873.

i 40, Bleul C, Wu L, Hoxie J. Springer T. MacKay C,The HIV coreceptors CXCR4 and CCRS aredifferentially expressed and regulated onhuman T lymphocytes,Proc Nail Acad Sci USA 1997:94:192 5-1930.

141, Zhu T. el al, Genotypic and phenotypiccharacterization of HIV-I in patients withprimary infection.Science 1993:261:1179-1181,

142, Zhang LQ. Mackenzie P. Cleland A.Hoimes EC. Leigh-Brown AJ. Simmonds PSelection for specific sequences in theexternal envelope protein of humanimmunodeficiency virus type 1 uponprimary infection,JVirol 1993:67:3345-3356,

143, Unutmaz D, Littman D, Expression pattern ofHIV-1 corecepiors on T cells: implications forviral transmission and lymphocyte homing,Proc Natl Acad Sci USA 1997:94:1615-1618,

144, Simmons G, et al, Potent inhibition of HIV-1infectiviiy in macrophages and lymphocytesby a novel CCR5 antagonist.Science 1997;276:276-279,

145, Kinter AL. et al, HIV replication in CD4-fT cells of HIV-infected individuals isregulated by a balance between the viralsuppressive effects of endogenous beta-chemokines and the viral inductive effects ofother endogenous cytokines.

Proc Natl Acad Sci USA 1996:93:14076-14081,

146, Jansson M. el al. Sensitivity to inhibition byP-chemokines correlates with biologicalphenotypes of primary HIV-1 isolates,Proc Natl Acad Sci USA 1996:93:15382-15387,

147, Connor RI. Sheridan KE. Ceradini D, Choe S.Landau NR, Change in coreceptor usecorrelates with disease progression in HIV-1 -infected individuals,JExpMed 1997:185:621-628,

148. York-Higgiiis D, Cheng-Mayer C. Bauer D.Levy JA. Dina D, Human immunodeficiencyvirus type 1 cellular host range, replication,cytopatliicity are linked to the enveloperegion ofthe viral genome.

J Virol 1990:64:4016-1020,149, de)ong JJ, et al. Human immunodeficiency

virus type I clones chimeric for the envelopeV3 domain differ in syncytlum formation andreplication capacity

JViroI 1992:66:757-765,I SO, Kuiken Cl. dejong JJ. Baan E, Keulen \y

Tersmette M. Goiidsmit J. Evolution of the V3envelope domain in proviral sequences andisolates of human immunodeficiency viruslype 1 during transition ofthe viral biologicalphenoiype,

JViroI l992;66:4622-t627,151, Hwang SS. Boyle TJ, Lyerly HK. Cullen BR.

Identification ofthe envdcipe V3 loop as theprunary determinant of cell tropism inHIV-1,Science 1991:253:71-74,

152, lyceuchi Y, Akutsu M. Murayama K.Sliimuzu N. Hoshino H, Host range mutant oflitunan immunodeficiency virus lype 1:modification of cell tropism by a single pointmutation at the neutralization epitope in theenv gene.JViroI 1991:65:1710-1718,

153, Cheng-Mayer C, Shioda T, Levy JA, Hostrange, replicaiive, and cytopathic propertiesof human immunodeficiency virus type I aredetermined by a very few amino acid changesin tat and gpl 20.JViroI 1991:65;6931-6941.

154, Fouchier RA. et al. Phenoiype-associatedsequence variation in the third variabledomain ofthe human immunodeficiencyvirus type 1 gpl 20 molecule.JViroI 1992;66:3183-3187,

1S 5. Bhattacharyya D. Brooks BR. Callahan L.Positioning of positively charged residues inihe V3 loop correlates with HIV type Isyncytium-iuducing phenotype,AIDS Res Hum Retroviruses 1996:12:83-90.

156- Cocchi F, DeVico A. Garzino-Demo A. Cara A,Gallo R. Lus,w P The V3 domain of HIV-1envelope glycoprotein gpl 20 is critical forchemokine-mediated blockade of infection.Nat Med 1996:2:1244-1247.

157, Koot M, et al. Prognostic value of HIV-1syncytium-inducing phenoiype for rate ofCD4-t- cell depletion and progression to AIDS.Ann Intern Med 1993:118:681-688,

158, Schmidtniaycrova H, Sherry B, Bukrinsky M.Chemokines and HIV replicition.Nature 1996:382:767,

159, Michael N. et al. The role of viral phenotypeand CCR-5 gene defects in HIV-Itransmission and disease progression,Nat Med i997:3:338-34O.

Inimunoiogical Reviews ! 59 /1997 45

Cohen et al • Host factors in HIV disease pathogenesis

160. Mellors I\\: etal. Quantitation of HIV-l RNAin plasma predicts outcome afterseroconversion.Ann Intern Med 199S;122:S73-579.

i 61. Zack ]A. Arrigo SJ, Weitsman SR, Go AS,Haisiip A, Chen IS. HIV-1 entry into quiescentprimary lymphocytes: molecular analysisreveals a labile, latent viral structure.Cell 199O;61:2]3-Z22.

162. Bukrinsky M, Stanwick T, Dempscy M.Stevenson M. Quiescent T lymphocytes as aninducible virus reservoir in HIV-l infection.Science 1991:254:423-427.

163. Finberg R, et al. Selective elimination ofHIV- i infected cells with an interleukin-2receptor-specific cytotoxin.

Science I99l;2S2:I703-]70S.164. Bell K. Ramilo O, Vitetu E. Combined use of

an immunotoxin and cyclosporine to preventboth activated and quiescent peripheral bloodT cells from producing type 1 humanimmtinodeficiency virus.

Proc Nad Acad Sci USA 1993;90:141 1-1415.165. Quinn T, ei al. Serologic and inimmiologic

siudies in patients with AIDS in NorthAmerica and Africa. The potential role ofinfectiotis agents as cofactors in humanimmunodeficiency virus infection.

I Am Med Assoc 1987:257:2617-2621.166. Bemwich Z, Kalinkovich A, Weisman Z.

Inimtine activation is a dominant factor in thepaihogenesis of African AIDS.

Immunol Today 199S;16:187-I91.167. Pape JW, Jean SS, Ho )L. Hafner A.

Johnson WD. Effect of isoniazid prophylaxison incidence of active tuberculosis andprogression of HIV infection.Lancet 1993:342:268-272.

168. Walhs R. et al. Influence of tuberculosis onhuman immunodeficiency virus (HIV-1):enhanced cytokine expression and elevatedbeta 2-microglobulin iii HIV-1-associatedtubercu io.sis.JlnfeciDis 1993:167:43-48.

169. Whalen C. Horsburgh CR. Hom D, Uhan C,Simberkoff M. Ellner J. Accelerated course ofhuman immunodeficiency virus infectionafter tuberculosis.Am J Respir Crit Care Med 1995:151:129-135.

170. Goletti D. et al. Effect of Mycobacieriumtuberculosis on HIV replication. Rote ofimmune activation.JImmunol 1996:157:1271-1278.

171. Israel-Biet D, Cadranel J, Even P Humanimmunodeficiency virus production byalveolar lymphocytes is increased duringPneumocystis carinii pneumonia.

Am Rev Respir Dis I 993; 148:1308-1312.172. Orenstein J, Eox C. Wahl S. Macrophages a a

source of HIV during opporlunisricinfections.Science 1997:276:1857-1861.

17 3. Staprans S, et al. Activation of virusreplication after vaccination of HlV-t-infected individuals.JExpMed 1995:182:]727-I737.

174. Stanley S, et al. Effect of immunization with acommon recall antigen on viral expression inpatients infected with humanimmunodeficiency virus type 1.NEnglJMed 1996:334:1222-1230.

17 5. Lifson AR, et al. Long-term humanimmunodeficiency virus infection inasymptomatic homosexual and bisexual menwith normal CD44- lymphocyte counts:immunologie and virologic characteristics.J Infect Dis 1991; 163:959-965.

176. Sheppard HW. Lang W Asther MS,Vittinghoff E, Winkelstein W Thecharacterizadon of non-progressors: long-term HTV-1 infection with stable CD4-)- T-celllevels.

AIDS I993:7:l l59-n66.177. Buchbinder SP Katz MH, Hessol NA.

O'Malley PM, Holmberg SD. Long-termHIV-! infection wiihoui immvinologicprogression.

AIDS 1994:8:1123-1128.178. Keet IP, et al. Characteristics of long-term

asymptomatic infection with humanimmunodeficiency virus type 1 in men withnormal and low CD4+ cell counts.

J Infect Dis 1994:169:1236-1243.i 79. Munoz A, et al. Long-term survivors with

HIV-l infection: incubation period andlongitudinal patterns of CD4+ lymphocytes.J .Acquir Immune Defic Syndr Hum Retrovirol1995:8:496-505.

180. Pantaleo G. et a!. Studies in subjects withlong-term nonprogressive humanimmunodeficiency virus infection.NEnglJMed 1995:332:209-216.

181. Cao Y, Qin L, Zhang L, Safrii J. Ho DD.Virologic and immunologic characterizationof long-term survivors of humanimmunodeficiency virus type 1 infection.NEnglJMed 1995:332:201-208.

182. Vicenzi E, et ai. Hemophilia andnon progressing human immunodeficiencyvirus type I infection.Blood 1997:89:191-200.

183. Balotta C, et al. Comparable biological andmolecular determinants in HIV typeI -infected long-term non progressors andrecently infected individuals.AIDS Res Hum Retroviruses 1997:13:337-341.

184. Schrager L, Young J. Fowler M. Mathieson B,Vermund S. Long-term survivors of HIV-1infection: defmitions and researchchallenges.AIDS 1994:8 (Suppl l):S95-IO8,

185. Learmont J, et al. Long-term symptomlessHlV-1 infection in recipients of bloodproducts from a single donor.Lancet 1992:340:863-867.

186. Deacon NJ, et al. Genomic strticture of anattenuated quasi species of HIV-1 from ablood transfusion donor and recipients.Science 1995:270:988-991.

187. Kirchhoif F. Grecnough TC, BrettlerDB.Sullivan JL. Desrosiers RC. Brief report:absence of intact nef sequences in a long-term survivor with nonprogressive HIV-!infection.NEnglJMed 1995:332:228-232,

! 88. Iversen AKN. et al. Persistence of attenuatedrev genes in a human immunodeficiencyvirus type 1 -infected asymptomaticindividual.J Virol 1995:69:5743-5753.

189. Michael NL. et al. Defective accessory genesin a human immunodeficiency virus type 1 -infected long-term stirvivor lackingrecoverable virus.J Virol 1995:69:4228-4236.

190. Michael NL. Chang G, D'arcy LA. Tseng CJ.Birx DL. Sheppard HW Functionalcharacterization of humanimmunodeficiency virus type 1 nef genes inpatients with divergent rates of diseaseprogression.

JViro! 1995:69:6758-6769.191. Huang Y, Zhang L, Ho DD. Characterization

of nef sequences in long-term survivors ofhuman immunodeficiency virus type 1infection.

JViroi 1995:69:93-100.192. Huang Y, Zhang L, Ho D. Biological

characterization of nef in long term survivorsof human immunodeficiency virus type 1infection.J Virol 1995:69:8142-8146.

193. KaslowRA. etal. Al, Cw7, B8, DR3 HLAantigen combination associated with rapiddecline of T-helper lymphocytes in HIV-1infection.Lancet 1990:335:927-930.

194. Klein MR. et al. Associations between HLAfrequencies and pathogenic features ofhuman immunodeficiency virus type Iinfection in seroconverters from theAmsterdam Cohort of Homosexual Men.] Infect Dis 1994:169:1244-1249.

195. Keet IP. Klein MR, Just JJ, Kaslow RA. The roleof host genetics in the natural history of HIV-1 infection: the needles in the haystack.AIDS 1996:10(Suppl A):S59-S67.

196. Kaslow RA. el al. Influence of combinationsof human major histocompatibility complexgenes on tlie course of HIV-l infection.Nat Med 1996:2:405-41 1.

i 97. Eales L-J, et al. Association of different allelicforms of group specific componeni withsusceptibility to and ciinical manifestation ofhuman immunodeficiency virus infection.Lancet 1987:i: 999-1002.

46 ImmunologicaJ Reviews 159/1997

Cohen et al • Host factors in HIV disease pathogenesis

198. Garred P. et al, Susceptibillly to HIV infectionatid progression of AIDS in relation to variinlalleles of tnannose-binding leclin,Uncet 1997:349:236-240.

199, Khoo SH, et al. The TNF cZ microsatellitealtele is associated with the rate of HIV diseaseprogression.AIDS I997;ll:423-t28,

.M)0. SamsonM.ctal,Resistance to H!V-! infectionin caticasian individtjals bearing mutantalieles ofthe CCR-S chemoldne receptorgene.Nature 1996;382;722-725.

'1)1, Liu R. et al.Homozygous defect in HIV-l

coreceptor accotints for resistance ofsomemultiply-exposed individuals to HIV-1infection.Cell 1996:86:367-377.

,'02, Zimmerman R et al. inherited resistance toHIV-1 conferred by an inactivating mutationin CC chemokine receptor S: studies inpopulations with contrasting clinicalphenotypes, defmed racial backgrounds andquantified risks,Mol Med 1997:3:23-36.

'03, Dean M. et al. Genetic restriction of HIV-linfection and progression to AIDS by adeletion allele of ihe CKRS structural gene.Science 1996;273:1856-1862,

,'.(14, Huang Y, et al. The role of a mutant CCR5aliele in HIV-1 transmission and diseaseprogression.Nat Med 1996:2:1140-1243,

.'05, Biti R, French R, Young), Bennetts B,Stewart G, HIV-1 infection in an individualhomozygous for the CCR5 deletion aliele,Nat Med 1997:3:252-253,

, 06, O'Brien T, et al, HIV- i infection in a manhomozygous for CCR5A32.Lancet 1997;349:12I9.

207, Theodorou I, Meyer L, Magierowska M.Katlama C. Rouzioux C, HIV-1 infection in anindividual homozygous forCCRSA32,Lancet 1997:349:1219-1220,

208, Wu L, et al. CCR5 levels and expressionpattern correlate wUh infectability bymacrophage-tropic HIV-1. in vitro,JExpMed 1997;18S:1681-1692,

209, CohenO. etal, Heterozygosity fora defectivegene for CC chemokine receptor 5 is not thesole determinant for the immunologic andvirologic phenotype of HIV-infected longterm non-progressors,]Chn Invest 1997:100:1581-1589,

' 10, Eugen-Ol.sen J, et al. Heterozygo.iiity for adeletion in the CKR-S gene leads toprolonged AIDS free survival and slower CD4T cell fall in a cohort of HTV seropositiveindividuals.

AIDS i997:II:30S-310.Ml, Panialeo G, Vatcarezza M. Graziosi C, Cohen OJ,

Fauci AS, Antiviral immunity in HIV-1infected long-term non-progressors (LTNPs).Semin Virol 1996:7:131-138,

212, Gotch F. Galhmore A, McMichael A,Cytotoxic T cells — protection from diseaseprogression - protection from infecdon,Immunol Lett 1996:51:125-128.

213, Jassoy C. Walkor E, HIV-1 -specific cytotoxic Tlymphocytes and the control of HIV-1rep h cation.Springer Semin Immunopathol 1997;18:341-3S4,

214, Harrer T, et al, Cytotoxic T lymphocytes inasymptomatic long-term nonprogressingHIV-1 infection. Breadth and specificity oftheresponse and relation to in vivo viralqtiasispecies in a person with prolongedinfection and low viral load.

JImmunol 1996:156:2616-2623,215, Harrer T, et al. Strong cytotoxic T cell and

weak neutralizing antibody responses in asubset of persons with stable nonprogressingHIV lype 1 infection.

AIDS Res Hum Retroviruses 1996:12:585-592.

216, RinaldoC, et al. High levels of anti-humanimmtmodeficiency virus type 1 (HIV-1)memory cytotoxic T-lymphocyie activity andlow viral load are associated with lack ofdisease in HIV-1 -infected long-termnonprogressors.JVirol 199S:69:S838-,5842,

217, Klein MR, et al. Kinetics of gag-specificcytotoxic T lymphocyte responses during theclinical course of HIV-1 infection: alongitudinal analysis of rapid progressors andlong-term asymptomatics,JExpMed 1995:181:1365-1372.

2 1 8, Riviere Y. et al. Gag-specific cytoioxicresponses to HIV type 1 are associated with adecreased risk of progression to AIDS-relatedcomplex or AIDS.AIDS Res Htim Retroviruses 1995:11:903-907,

219. Goulder PJ, et al. Novel, cross-restricted,conserved, and immunodominant cytotoxicT lymphocyte epitopes in slow progressors inH!V type 1 infection.AIDS Res Hum Retroviruses 1996:12:1691-1698,

220. Pantaleo G, ei al.The qualitative nature oftheprimary immune response to HIV infection isa prognosiicator of disease progressionindependent of the initial level of plasmaviremia.Proc Natl Acad Sci USA 1 997;94:254-258.

221. Moskophidis D, Lechner F, Pircher H,Zinkernagel RM. Virus persistence in acutelyinfected immumKompetent mice byexhaustion of antiviral cytotoxic effectorT cells.Nature 1993:362:758-761,

222. Pantaleo G. et al. Evidence for rapiddisappearance of initially expanded HIV-specific CD8-I- T cell clones during primaryHIV infection,Proc Natl Acad Sci USA 1997 (In press).

223. Zinkernagel RM, Hengartner H, T-cell-mediated immunopathology versus directcytolysis by virus: implications for HIV andAIDS,

Immunol Today 1994:15:262-268,224. Koup RA, et al. Temporal association of

cellular immune responses with the initialcontrol of viremia in primary humanimmunodeficiency virus type 1 syndrome,JVirol 1994:68:4650^655,

225. Borrow P, Lewicki H, Halm BH, Shaw GM,Oldstone MBA. Virus-specific CDBi-cytotoxic T-lymphocyte activity associatedwith control of viremia in primary humanimmunodeficiency virus type I infection,JVirol 1994:68:6103-6110,

226. Goulder PJR, et al. Late escape from animmunodominant cytotoxic T-Iymphocyteresponse asscKiated with progression to AIDS.Nat Med 1997:3:212-217,

227. PhiUips RE. et al. Human immunodeficiencyvirus genetic variation that can escapecytotoxic T cell recognition.Nature 1991:354:433-434,

228. Koenig S, et at. Transfer of HIV-1-specificcytotoxic T lymphocytes to an AIDS patientleads to selection for mutant HIV variants andsubsequent disease progression.Nat Med 199S:l:330-336,

229. Borrow P, et al. Antiviral pressure exerted byHIV-1 -specific cytotoxic T lymphocytes(CTLs) during primary inleciiondemonstrated by rapid selection of CTLescape virus,

Nat Med 1997:3:205-211.230. Klenerman P, et al. Cytotoxic T-cetl activity

antagonized by naturally occurring HIV-1Gag variants.Nature t994;369:403-»-07.

231. Meier U-C. etal, Cytotoxic T lymphocyte lysisinhibited by viable HIV mutants.Science 1995:270:1360-1362,

232. Itescu S, Rose S, Dwyer E. Wiiiche.ster R,Certain HLA-DR5 and -DR6 majorhistocompatibility complex class II altetes areassociated with a CD8 lymphocytic hostresponse to human immunodeficiency virustype 1 characterized by low lymphocyte viralstrain heterogeneity and slow diseaseprogression,Proc Natl Acad Sci USA 1994:91:11472-11476.

233. Kalams SA. et al. Longitudinal analysis ofT cell receptor (TCR) gene usage by humanImmunodeficiency virus I envelope-specificcytotoxic T lymphocyte clone reveals alimited TCR repertoire,JExpMed I 994:1 79:1 261-1 271.

Immunological Reviews 159/1997 47

Cohen et al - Host factocs in HIV disease pathogenesis

234, KalamsSA, et al, T ceO receptor asage and finespecificity of human immunodeficiency virus1-specific cytotoxic T lymphocyte clones:analysis of quasispecLes recognition reveals adominant response directed against a minorin vivo variant,JExpMed I996;i83:l669-1679.

2 3 S, Wilson CC. et al. Overlapping epitopes inhuman immunodeficiency virus type 1gp 120 presented by HLA A, B, and Cmolecules: effects of viral variation oncytotoxit T-lymphocyte recognition,JViroI ]997;71:I2S6-1264,

236, Wolinsky SM. et al. Adaptive evolution ofhuman imniunodefidency virus-type 1during the natural course of infection.Science 1996;272:S37-542,

237, Haas G, et al. Dynamics of viral variants inHIV-1 Nef and specific cytotoxic Tlymphocytes in vivo,

jlmmunol I996;157:421 2-4221,238, Nowak MA. et ai, Antigenic oscillations and

shifting immunodominance in HIV-1infections.

Nature 199S;375:606-6i I.239, Mackewicz CE, Ortega HW. Levy JA, CD8+

cell anti-HIV activity correlates with theclinical state ofthe infected individual,JClin Invest 1991;87:1462-I466.

240, Scaia E. et al, C-C chemokines. IL-16, andsoluble antiviral factor activiiy are increasedin cloned T ceUs from subjects with long-term nonprogressive HIV infection.Jlmmunol l997;lS8:448S-4492,

241, McKenzie S. Dallalio G, North M. Frame P.Means R, Serum chemokine levels in patientswitli non-progressing HIV infection.

AIDS 1996;10:29-33.242, Zanussi S, D'Andrea M. Simonelli C. Tirelli U,

DePaoh P Serum levels of RANTES and MIP-la in HIV-positive long-term survivors andprogressor patients,AIDS 1996:10:1431-1432,

243, ClerJci M. et al. Chemokine production inHIV-seropositive long-term asymptomaticindividuals.AIDS 1996;!O:I432-1433.

244, Chen Y, Gupta P CD8+ T-ceU-mediatedsuppression of HIV-1 infection may not bedue to chemokines RANTES. MIP-la, andMIP- i ,

AIDS i996:t0:l43-l-l43S,

245, Blazevic V. Heino M. Ranki A. Jussila T,Krohii K, RANTES. MIP. and interleiikin-l 6 inHIV infection.AIDS I996;!0:1435-1436,

246, Furci L, et al. Antigen-driven C-CChemokine-mediated HIV-1 Suppression byCD4(4-} T CeUs from Exposed UninfectedIndividuals Expressing the Wild-type CCR- 5Allele,JExpMed 1997:186:455^-60.

247, Schall T, Bacon K, Toy K. Goeddel D Selectiveattraction of monocytes and T lymphocytes ofthe memory phenotype by cytokine RANTES,Nature 1990:347:669-671,

248, Amara A. et al, HIV co-recepiordownregulation as antiviral principle:SDF-la-dependent mternalization of thechemokine receptor CXCR4 contributes toinhibition of HIV replication,JExpMed 1997:186:139-146,

249, Baier M. Werner .\. Bannert N. Metzner K.Kurth R, HIV suppression by interleukin-16,Naiure 1995:378:563,

250, Zwart G, et al. Antibody responses to HIV-1envelope and gag epitopes in HIV-1seroconverters with rapid versus slow diseaseprogression.Virology 1994;201:285-293,

251, Wong MT. et al. Longitudinal analysis of thehumoral immtine response to humanimmunodeficiency virus type I (HIV-1)gp 160 epitopes in rapidly progressing andnonprogressing HIV-1 -uninfected subjects.JlnfectDis i993:168:1 S23-1 527,

252, Hogervorst E. et al. Predictors for non- andslow progression in human

imniunodeliciency virus (HIV) type 1infection: low viral RNA copy numbers inserum and maintenance of high HIV-1 p24-speciBc but not V3-specifit antibody levels,JlnfectDis I99S;171:8I 1-821,

253, Montefiori DC, el al. Neutralizing andinfection-enhancing antibody responses tohuman immunodeficiency virus type 1 inlong-term non progressors,) Infect Dis 1996;t73:60-67.

254, Sei Y. Tsang PH. RobozJP. Sarin PS. Wallace Jl,BekesijG, Neutralizing antibodies as aprognostic indicator in the progression ofacquired immune deficiency syndrome(AIDS)-related disorders: a double-blindstiid>:JClin Immunol 1988;8:464-472.

255. Tsang ML, et al. Neutrali/ing anlibodiesagainst sequential autologous humanimmunodeficiency virus type I isolates afterseroconversion,

J Infect Dis 1994;170: i 141-i 147.256. Pilgrim A, et al. Neutralizing antibody

responses ai various stages of infection withhuman immunodeficiency virus type 1.

J Infect Dis 1997 (In press),257, Arendrup M, Nielsen C. Hansen J-ES,

Pedersen C. Mathiesen L, Nielsen JO.Autologous HIV-I neutralizing antibodies:emergence of neutralization-resist ant escapevirus and subsequent development of escapevirus neutralizing antibodies.

J Acquir Immune Defic Syndr 1992:5:303-307,

258. loachim HL, Lerner CW, Tapper ML, Thelymphoid lesions associated with theacquired iinniuiiodeficiency syjicirome.Am ) Surg Pathol I983;7:543-SS3,

259. Fernandez R. Mouradian J, Metroka C.Davies J, The prognostic value ofhistopathology in persLstenl generalizedlymphadenopathy in homosexual men,N Engl J Med 1983;309:1 85-186,

260. Janossy G, et al. An immunohistologicalapproach to persistent lymphadenopathy andits relevance to AIDS.

Clin Exp Immunol 1985:59:257-266,261, Biberfeld P, Chayt K. Marselle L. BiberfeldG,

Gallo R. Harper M. HTLV-III expression ininfected lymph nodes and relevance topathogenesis of lymphadenopathy,AmJPathol I986;125;436^42,

262, Ratz P. Tenner-Racz K. Kahi C. Feller AC.Kern P. Dietrich M, Spectrum ofmorphologicchanges of lymph nodes from patienis withAIDS or AIDS-related complexes.

Prog Allergy 1986:37:81-181,263, Pallesen G, Gersioft ], Malliiesen L, Stages in

LAV/HUV-III lymphadenitis: 1, Hisiologicaland immunohistological classification.Scand J Immunol 1987:25:83-91,

264, Tbrner R, Levine A. Gill P. Parker J. Meyer fiProgressive histopathologic abnormaliUcs inthe persistent generalized lymphadenopathysyndrome.Am J Surg Pathol 1987;n:625-632.

48 Immunoiagical Reviews 159/1997