Effects of the human CD38 glycoprotein on the early stages of the HIV-1 replication cycle

Effects of the human CD38 glycoprotein on the earlystages of the HIV-1 replication cycle

ANDREA SAVARINO,* FLAVIA BOTTAREL,†,1 LILIANA CALOSSO,* MARIA JOSE FEITO,†

THEA BENSI,† MANUELA BRAGARDO,† JOSE MARIA ROJO,§ AGOSTINO PUGLIESE,*ISABELLA ABBATE,‡ MARIA R. CAPOBIANCHI,‡ FERDINANDO DIANZANI,‡

FABIO MALAVASI,| AND UMBERTO DIANZANI†,2

*Department of Medical and Surgical Sciences, University of Turin, Italy; †Department of MedicalScience, University ‘A. Avogadro’ of Eastern Piedmont, Novara, Italy; ‡Department of ExperimentalMedicine and Pathology, University ‘La Sapienza’ of Rome, Italy; §Department of Immunology,Centro de Investigaciones Biologicas, CSIC, Madrid, Spain; and \Institute of Biology and Genetics,University of Ancona, Italy

ABSTRACT CD38 displays lateral association withthe HIV-1 receptor CD4. This association is potenti-ated by the HIV-1 envelope glycoprotein gp120. Theaim of this work was to evaluate the CD38 role in Tcell susceptibility to HIV-1 infection. Using labora-tory X4 HIV-1 strains and X4 and X4/R5 primaryisolates, we found that CD38 expression was nega-tively correlated to cell susceptibility to infection,evaluated as percentage of infected cells, release ofHIV p24 in the supernatants, and cytopathogenicity.This correlation was at first suggested by resultsobtained in a panel of human CD41 T cell linesexpressing different CD38 levels (MT-4, MT-2,C8166, CEMx174, Supt-1, and H9) and then demon-strated using CD38 transfectants of MT-4 cells (theline with the lowest CD38 expression). To addresswhether CD38 affected viral binding, we used mouseT cells that are non-permissive for productive infec-tion. Gene transfection in mouse SR.D10.CD42.F1 Tcells produced four lines expressing human CD4and/or CD38. Ability of CD41CD381 cells to bindHIV-1 or purified recombinant gp120 was signifi-cantly lower than that of CD41CD382 cells. Thesedata suggest that CD38 expression inhibits lympho-cyte susceptibility to HIV infection, probably byinhibiting gp120/CD4-dependent viral binding totarget cells.—Savarino, A., Bottarel, F., Calosso, L.,Feito, M. J., Bensi, T., Bragardo, M., Rojo, J. M.,Pugliese, A., Abbate, I., Capobianchi, M. R., Dian-zani, F., Malavasi, F., and Dianzani, U. Effects of thehuman CD38 glycoprotein on the early stages of theHIV-1 replication cycle. FASEB J. 13, 2265–2276 (1999)

Key Words: susceptibility to HIV-1 z gp120/CD4-dependentvirus binding

Cell susceptibility to infection by human immu-nodeficiency virus type-1 (HIV-1) depends on acomplex system of virus/host interactions, in whichcell-surface molecules play a pivotal role (1–3). Viral

attachment involves interaction of the HIV-1 enve-lope glycoprotein gp120 with CD4 acting as the mainviral receptor, and several members of the chemo-kine receptor family acting as co-receptors (1).Three co-receptors play a crucial role in infection ofdifferent types of human cells. CXCR4 (also knownas fusin or LESTR) is the receptor for the stromal-derived factor (SDF-1) and is mainly used by the‘lymphocyte-tropic’ X4 HIV-1 strains, which gener-ally develop in the advanced stages of infection (3,4). CCR5 is a receptor for the b-chemokines macro-phage inflammatory protein-1a (MIP-1a), MIP-1b,and RANTES, and is mainly used by the ‘macro-phage-tropic’ R5 HIV strains, which predominate inthe early stages and can infect many cell types (3, 4).CCR3 is the receptor for eotaxin and is considered tobe an important HIV-1 co-receptor on microglia (5).

Other cell-surface proteins modulate susceptibilityeither by direct interference with virus attachment orthrough their signaling pathways. For instance, ex-pression of CD26 or CD44 promotes and triggeringof CD28 either increases or decreases susceptibilityto several viral strains (6–10). Moreover, during viralbudding from infected cells, several cell-surface mol-ecules are incorporated into viral envelopes. Someare adhesion molecules and contribute to viral at-tachment by interacting with their physiological li-gands of the target cells (1, 11).

We have previously shown that CD4 binding bygp120 induces lateral associations with several leuko-cyte surface molecules (12, 13). Because HIV-1 infec-tion depends on multiple intermolecular interactionson the cell surface and precise steric interactions, wesuggested that some of these associations positively ornegatively influence cellular infection (12): CD38 was

1 The first two authors contributed equally to this work.2 Correspondence: Department of Medical Science, Univer-

sity ‘A. Avogadro’ of Eastern Piedmont, via Solaroli 17,I-28100 Novara, Italy. E-Mail: dianzani 64 med.no.unipmn.it

22650892-6638/99/0013-2265/$02.25 © FASEB

an attractive candidate, its expression being negativelycorrelated with HIV-1 infection in a panel of CD41 Tcell lines (12–14).

Human CD38, a single-chain transmembrane typeII glycoprotein, is surface-expressed by early hema-topoietic cells, lost by mature cells, and re-expressedon cell activation (15). It is detectable at high levelson mature thymocytes and activated T cells and atlow levels on resting naive cells (CD45RA1R02

cells), whereas resting memory cells (CD45RA2R01

cells) are CD382 (15, 16). CD38 is thought toexercise the following three functions on T cells: 1)as an ectoenzyme, it leads to the formation of cyclicADP-ribose, a crucial compound in regulation ofintracellular Ca21 (15, 17); 2) as an adhesion mole-cule, it mediates the interactions between leukocytesand vascular endothelial cells through CD31 (15, 16,18); 3) as a molecule involved in transmembranesignaling, its engagement results in Ca21 mobiliza-tion, costimulates cell activation, and modulates cy-tokine production (15). CD38 signaling exploitsother molecules specialized in signaling, such as CD3in T cells (19), BCR in B lymphocytes, (20), andCD16 in natural killer (NK) cells (R. Mallone, A.Funaro, and F. Malavasi, unpublished). By contrast,the relationship between its signaling and ectoen-zyme functions is not known, and CD38-inducedCa21 mobilization seems to be independent from itsenzyme activity (21).

In HIV-1 infection, CD38 has so far been studiedmainly from a prognostic perspective. Its expressionis high on peripheral blood lymphocytes in primaryinfection, decreases during transition to the asymp-tomatic phase, and then increases during progres-sion to AIDS (22). High expression on peripheralblood CD81 T cells is a negative prognostic factorand is decreased by successful HAART (highly activeantiretroviral therapy) in both adults and children(22–25). By contrast, high expression on CD41 Tcells is positively correlated with survival in children(26, 27). These findings have been attributed to theability of CD38 to mark activation of the immuneresponse (23). However, the fact that CD38 associa-tion with CD4 is increased by gp120 (12, 13) and itsexpression is correlated with HIV-1 infection, sug-gests that its role is direct (14, 28). The aim of thepresent work was to investigate this possibility byusing human and mouse cell lines transfected withthe human CD38 cDNA and assessing the effects ofCD38 expression on various steps of HIV-1 replica-tion.

MATERIALS AND METHODS

Monoclonal antibodies

Fluorescein isothiocyanate (FITC)-labeled mouse monoclo-nal antibodies (mAbs) to human CD4 were purchased from

Coulter, Hialeah, FL; R-phycoerythrin (R-PE)-conjugatedmAbs to human CXCR-4 (12G5) were from PharMingen, SanDiego, CA; FITC-conjugated mouse anti-HIV-1 p24 mAbswere from Virostat, Portland, ME; unconjugated anti-CD26mAbs were a generous gift of Dr. E. C. Bosmans, Eurogenet-ics, Belgium; FITC-conjugated anti-CD38 mAbs (IB4) wereobtained as described previously (29); FITC-conjugated rab-bit anti-murine immunoglobulins were from Pierce, Rock-ford, IL. Appropriate isotype-matched mAbs were used asnegative controls.

Cells

The following human T cell lines were used: H9, Supt-1, C8166, MT-2, MT-4, and CEMx174 (14, 30–33). They weregrown in RPMI-1640 medium (Life Technologies, Inc., Gaith-ersburg, MD), supplemented with 10% (v/v) fetal calf serum(FCS; Techno-Genetics, Milan, Italy), 200 mg/ml glutamine(Merck, Darmstadt, Germany), and 40 mg/ml gentamicin(Schering-Plow, Milan, Italy).

SR.D10-CD42.F1 was a CD42 mutant cell line cloned fromthe mouse CD41 TH2 cell line D10.G4.1 (34). It was grown inClick medium with 10% FCS, 9% (v/v) b-mercaptoethanol, 5U/ml interleukin-2 (IL-2); 10 U/ml IL-4, and 25 pg/ml IL-1a.

Expression vectors

The SR.hCD4 line expressing human CD4 was produced bytransfecting the human CD4 cDNA inserted in the pNeoSRaplasmid in SR.D10.CD42.F1 cells as previously reported (35).

CD38 transfectants were produced by transfecting thehuman CD38 cDNA in SR.D10.CD42F1, SR.hCD4, and MT4cells. Full-length human CD38 cDNA (generous gift of Dr. D.Jackson, Oxford, UK) was excised from the pCDM8 plasmidwith XbaI and cloned into the corresponding site of apcDNA3.1/zeo expression vector carrying ampicillin/zeocinresistance, or pcDNA3/neo carrying ampicillin/neomycinresistance. Sequencing of the CD38 cDNA after cloning ruledout the presence of mutations in the sequence and confirmedthe correct position of the cDNA insertion in the plasmid.Plasmids were linearized with PvuI and transfected at 10mg/ml in the appropriate cell line [53106 cells in 0.8 ml ofphosphate-buffered saline (PBS)] by electroporation at 960mF and 260 V using a Gene Pulser (Bio-Rad, Hercules, CA).Transfectants carrying pcDNA3/neo were grown in culturemedium containing 0.8 mg/ml G-418 (selection medium),whereas transfectants carrying pcDNA3.1/zeo were grown inmedium containing 0.8 mg/ml zeocin.

Flow cytometry

To determine the expression of cellular and viral antigens,cells were pooled and washed in PBS with NaN3 (0.02%) andbovine serum albumin (2%; PBS A/A), and subsequentlytreated as follows.

To detect surface antigen expression, cells were suspendedin PBS A/A (2.53106 cells/ml) and incubated with saturatingconcentrations of the appropriate mAb. The negative controlsamples were incubated with isotype-matched mAbs. Whenunconjugated mAbs were used, cells were washed in PBS A/Aand subsequently incubated with FITC-labeled goat anti-mouse Ig antibodies under the same conditions. Finally, thecells were washed three times and immediately analyzed byflow cytometry (FACScan, Becton-Dickinson, Mountain View,CA).

To determine the percentage of infected cells, pellets werefixed, permeabilized, and stained with mAbs to HIV-1 p24using a commercially available kit (Caltag, Burlingame, CA),

2266 Vol. 13 December 1999 SAVARINO ET AL.The FASEB Journal

as described previously (32, 36). The fixed samples wereanalyzed by flow cytometry.

Fluorescence data were collected on a 4-decade log scaleand the relative fluorescence intensity was stated as themedian channel number (MeFI). Log values were mathemat-ically converted to linear fluorescence intensity and thecontrol antibody values for each experiment were subtracted.Fluorescence was also evaluated in terms of percentage offluorescent cells beyond the threshold value established usingcells stained with the isotype control reagents.

Infection with HIV-1

Stocks of HIV-1IIIB (a laboratory X4 strain) were obtainedfrom the supernatant of H9 IIIB cells, a persistently HIV-1-infected H9 cell line, as described elsewhere (30). Stocks ofHIV-1P1 (a laboratory-adapted X4 strain) were obtained fromacutely-infected C 8166 cells, as described previously (33).Stocks of HIV-1BAL (a laboratory-adapted R5 strain) (37) wereobtained from Dr. C. Balotta, ‘Sacco’ Hospital, Milan, Italy.Stocks of the X4 and R5/X4 primary isolates were obtainedfrom Dr. F. Piro (‘Amedeo di Savoia’ Hospital, Turin, Italy)and Dr. G. Poli (DIBIT, Milan, Italy), respectively. Bothprimary isolates were grown in phytohemagglutinin (PHA)-activated peripheral blood mononuclear cells (PBMC).Stocks of pRRL.sin.hPGK.GFP, a lentiviral vector pseudo-typed with the vesicular stomatitis virus (VSV)-G envelopeglycoprotein (38), were a generous gift of Dr. L. Naldini(IRCC Candiolo, Italy).

Viral stocks were titrated immunoenzymatically using p24antigen enzyme-linked immunosorbent assay (ELISA) kitsand biologically by the 50% end point dilution method, usingMT-2 cells (laboratory strains) or PHA-activated PBMC (pri-mary isolates). The infectious titer was expressed as cellculture infecting doses (CCID50)/ml (39).

Cells to be infected were suspended at 5 3 105 cells/ml in

Figure 1. Staining for surface CD38 in MT-4 cell transfectants:A) MT-4. M, B) MT-4.38L, C) MT-4.38I, D) MT-4.38H. Fluo-rescence using control isotype-matched mAbs (black) andfluorescence of transfectants carrying the CD38 cDNA (gray).The bold number shows the MeFI, i.e., the median fluores-cence intensity from each histogram mathematically con-verted to linear fluorescence intensity and subtracted by thecontrol antibody values.

TA

BL

E1.

Cor

rela

tion

betw

een

base

line

expr

essi

onle

vels

ofsu

rfac

ean

tigen

san

dpr

opor

tion

ofin

fect

edce

lls(I

F%)

afte

rH

IV-1

inoc

ulat

ion

ina

pane

lof

hum

anC

D41

Tce

lllin

es

CD

4C

XC

R-4

HL

Acl

ass

IIC

D26

CD

28C

D38

CD

71H

IV-1

infe

ctio

n

MeF

Ia%

bM

eFI

%M

eFI

%M

eFI

%M

eFI

%M

eFI

%M

eFI

%IF

%c

CE

d

MT

-427

96

3292

113

667

9028

46

7210

0U

D—

UD

—37

68

,5

233

611

479

7281

MT

-223

26

4392

135

634

9111

26

5390

ND

—N

D—

496

22,

512

36

5827

5167

C81

6625

76

1910

037

26

126

9840

06

9810

025

620

,5

UD

—84

69

1019

16

9268

4255

CE

Mx1

7433

86

2410

031

56

3690

389

676

100

436

8,

56

66

,5

104

612

42N

D33

3130

Supt

-147

26

5510

061

16

5510

0U

D—

119

611

6020

96

981

262

679

8113

56

6863

147.

5H

924

16

1795

299

681

9537

36

8996

UD

—U

D—

179

652

8236

56

4595

6,

5P

valu

ese

0.58

0.61

0.19

0.36

0.96

0.68

0.63

—0.

57—

,0.

01*

,0.

05*

0.43

0.34

,0.

01*

aM

ean

1sd

ofth

em

edia

nfl

uore

scen

cein

ten

siti

es(M

eFI)

obta

ined

from

thre

eex

peri

men

ts.b

Perc

enta

geof

posi

tive

cells

(PFC

;med

ian

from

thre

eex

peri

men

ts).

cPe

rcen

tage

ofce

llspo

siti

veto

stai

nin

gfo

rin

trac

ellu

lar

HIV

-1p2

4at

4da

ysaf

ter

infe

ctio

n(m

edia

nfr

omth

ree

expe

rim

ents

).d

Perc

enta

geof

cells

show

ing

cyto

path

icef

fect

s(C

E)

at5

days

post

-infe

ctio

n.e

t-Tes

tfo

rco

rrel

atio

n;

sign

ifica

nce

thre

shol

d:P

,0.

05.

UD

,un

dete

ctab

le;

ND

,n

otdo

ne;

P,

corr

elat

ion

toIF

%.

2267CD38 AND HIV-1 INFECTION

fresh culture medium 24 h before the infection. Cell pelletswere infected with the HIV-1 stock suspensions at a titer of5–10 3 105 CCID50/ml; the number of cells was adjusted so asto have a multiplicity of infection (MOI) of ;1. The cells werethen incubated at 37°C for 2 h, washed three times with PBS,resuspended at 2.5 3 105 cells/ml in fresh culture medium,transferred to 24-well plastic microtiter trays (Nunc, Kam-strup, Denmark), and incubated at 37°C in a 5% CO2humidified atmosphere. At different times after infection, cellviability was assessed microscopically by the trypan blue-dyeexclusion method. In the case of MT-4 cells, which display atypical clustered pattern, clusters were dissociated by pipet-ting, and reclustering was examined microscopically after a3-h incubation at 37°C (14, 30).

Proviral DNA was detected by PCR 2 days after infection.Cells were lysed in 500 ml of 500 mM Tris-HCl pH 9, 2 mMEDTA, 10 mM NaCl, 1% SDS plus 80 ng of proteinase K/mlat 55°C for 24 h. Cellular DNA was prepared by phenolextraction, followed by ethanol precipitation. DNA contentwas determined spectrophotometrically at 260 nm and equalamounts of DNA for each experimental condition underwentserial dilutions that were amplified by PCR using primers fora conserved gag region sequence (sense: 59-ATAATCCAC-CTATCCCAGTAGGAGAAAT-39; anti-sense: 59-TTTGGTC-CTTGTCTTATGTCC-AGAATGC-39), detecting full-length,double-stranded viral DNA in both the integrated and unin-tegrated form. The following amplification conditions wereemployed: denaturation (5 min at 94°C) was followed by 35cycles of denaturation (1 min at 94°C), annealing (1 min at55°C), extension (1 min at 72°C), and final incubation (10min at 72°C). The amplification products of 114 bp obtainedwere electrophoresed on 3% agarose gels and visualized byethidium bromide staining (37). As a control, the humangene bax was amplified by PCR in the same DNA preparations(sense: 59-TCTCCTGCAGGATGATTGC-39; anti-sense: 59-TC-CCCAGGTCCTCACAGAT-39). A standard curve for thequantification of proviral DNA copies was prepared usingserial 10-fold dilutions of the DNA extracted from a suspen-sion of 8E5 cells, containing one proviral copy per cell (40),and amplified under the same conditions. Amplified prod-ucts of the correct size were quantified using Gel Doc(Bio-Rad, Hercules, CA). The intensity units 3 mm2

(INT3mm2) of the serially diluted standard samples wereused to construct the standard curve, using the least squaresmethod. The proviral copy numbers of the experimentalsamples were calculated by reporting the values of the lastpositive sample dilution on the standard curve.

HIV-1 binding to cells

HIV-1 binding to cells was evaluated as previously reported(41). Briefly, 106 cells were incubated with 1 ml of the viralsuspensions (titer: 10 ng of p24/ml) for 2 h at 37 or 4°C. Cellswere then washed twice in PBS, lysed in 500 ml PBS with 0.5%Triton X-100 (v/v), and p24 was quantitated immunoenzy-matically in cell lysates using an ELISA kit (NENTM LifeScience Product, Inc., Boston, MA).

Cell staining with purified HIV-1 gp120 was evaluated byincubating cells (2.53106 cells/ml) with saturating concen-trations (10 mg/ml) of FITC-conjugated gp120 (IntracellCorp, London, UK) in PBS A/A for 30 min at 4°C. Cells werethen washed and analyzed by flow cytometry. The negativecontrol samples were incubated with PBS A/A alone.

FRET assay

The OKT4 mAb (ATCC) was conjugated to Cy3 dye with theFluoroLink-Ab Cy3 Labeling Kit (Amersham, Little Chalfont,

Buckinghamshire, UK). Conjugations were judged to besuccessful by spectrophotometric and spectrofluorometricmeasurements. Cells were washed with ice-cold PBS 1 5%FCS, 0.1% NaN3, incubated on ice for 1 h simultaneously withthe FITC-conjugated mAb (the donor fluorophore) and theCy3-conjugated OKT4 (the accepting fluorophore), thenwashed, resuspended in PBS 1 0.1% NaN3, and analyzed

Figure 2. Inhibition of HIV-1IIIB replication in MT-4 cellstransfected with the CD38 cDNA. A) p24 levels in superna-tants at 2 and 5 days post-infection; B) correlation betweenbaseline expression of CD38 and relative production of p24in transfectants at day 5 post-infection; C) cell survival afterHIV-1 infection. Cells were infected with HIV-1IIIB and grownin selection medium; p24 was measured in supernatants andexpressed as means 6 sd from three experiments. In A,asterisks indicate values significantly different than thosedisplayed by MT-4. M cells (significance threshold: P,0.05,repeated-measures ANOVA1Student-Neuman-Keuls test). InB, mean p24 values at day 5 were plotted against CD38 MeFIsof clones (average from three independent experiments) andthe line that best fitted the points was estimated by the leastsquares method (solid line: r520.97, P,0.05, t test forcorrelation). The dotted lines mark the 95% confidencelimits of the regression line. In C, the results of one repre-sentative experiment are reported.


immediately. A FACScan flow cytometer was used to deter-mine energy transfer between FITC and Cy3-labeled proteinson the cell surface. Fluorescence resonance energy transfer(FRET) to Cy-3 was detected by using standard methods (42).FITC was excited at 488 nm and Cy3 emissions were collectedat .600 nm. Data from 10,000 cells/test were stored in ‘listmode’ and analyzed with LYSYS II software (Becton Dickin-son). The median linear channel of fluorescence was calcu-lated and used as the indicator for the presence (a positiveshift over background) or absence (no shift or negative shift)of energy transfer.

RESULTS

Cell phenotype and susceptibility to HIV-1infection

The first step was to compare expression of CD38and infection by the X4 HIVIIIB strain in a panel ofhuman CD41 T cell lines selected on account oftheir permissivity to HIV-1 infection and homoge-neous expression of the main receptors for X4

strains (i.e., $90% of CD41 and CXCR-41 cells).HIV-1 infection, evaluated as proportion of infectedcells scored 4 days after infection by flow cytometry(IF%), negatively correlated to the CD38 expressionlevel (Table 1), evaluated both as MeFI and propor-tion of positive cells (r 5 20.92 and 20.87, respec-tively). A similar correlation was obtained betweenCD38 expression and the proportion of dead cells at5 days after infection (r520.95). These results werein line with our previous observations with differentapproaches (14). By contrast, IF% and cytopathiceffects did not correlate with the expression levels ofCD4 and CXCR4, nor of activation markers otherthan CD38 (CD26, CD71, HLA Class II).

HIV-1 replication in human transfectantsexpressing CD38

These data suggested that CD38 is a negative regu-lator of HIV-1 infection, even though differences insusceptibility due to other unknown features of the

Figure 3. Inhibition of primary HIV-1 isolate replication in MT-4 cells transfected with the CD38 cDNA. A) p24 levels insupernatants of cultures infected with a X4 primary isolate; B) cell survival after infection with the X4 primary isolate. C) p24levels in supernatants of cultures infected with a R5/X4 primary isolate; D) cell survival after infection with the R5/X4 primaryisolate. Cells were infected with the clinical isolates (MOI51) and grown in selection medium; p24 and counts of viable cellswere measured at different times after infection. In panels C and D, cells were split 1:2 every other day from day 4 after infection.One experiment representative of two.


cell lines, such as differences in growth kinetics ordifferential expression of molecules other thanCD38, could not be ruled out. To address this issue,we transfected the human CD38 cDNA (inserted inthe pcDNA3.1neo plasmid) into MT-4 cells andevaluated HIV-1 infection of transfected clones ex-pressing different levels of CD38. MT-4 cells werechosen because they had the lowest CD38 expressionand the highest susceptibility to infection. Threetransfectants expressing low (MT-4.38L), intermedi-ate (MT-4.38I), and high (MT-4.38H) levels of CD38,respectively, were selected. A control mock transfec-tant (MT-4.M) was produced by transfecting MT-4cells with the empty pcDNA3.1neo plasmid (Fig. 1).In all transfectants, expression of CD4 and CXCR-4,as assessed by immunofluorescence and flow cytom-etry, was similar to that displayed by parental MT-4cells (data not shown). Transfectants and parentalcells were incubated with HIV-1IIIB, washed, andcultured at 37°C. Infection was evaluated in terms ofp24 release in the supernatants. Results indicate thatHIV-1 replication was negatively correlated to CD38expression (Fig. 2). At day 2 and day 5 after infec-tion, parental MT-4 cells and mock MT-4-M cellsdisplayed the highest levels of p24, followed in orderby MT-4.38L, MT-4.38I, and MT-4.38H. In MT-4Mcells, maximal levels of p24 were reached at day 5after infection, and in MT-4.38 cells at day 7. At thelatter time, differences were no longer significant,suggesting that CD38 affects early events of infec-tion. In MT-4.38I and MT-4.38H, the low levels ofvirus replication were strictly paralleled by reducedcytopathic effects, as shown by the cell viability curves(Fig. 2). Similar results were obtained when cyto-

pathic effects were assessed as loss of ability to formcell clusters in infection driven by HIV-1IIIB (data notshown). In MT-4 cells, in fact, HIV-1IIIB acts as aslow/low syncytium-inducing strain whose cytopathiceffect consists of loss of cell ability to cluster and isstrictly correlated to the level of viral replication (14,30). When infected with HIV-1IIIB, MT-4.38I andMT-4.38H cells maintained a partial ability to formclusters, whereas MT-4.38L, MT-4.M, and parentalcells lost this ability. Similar impairment of HIV-1replication in CD38-expressing cells was also ob-served by using another laboratory-adapted X4 strain(HIV-1P1; data not shown).

To investigate whether the inhibitory effect ofCD38 expression on HIV replication was restricted tolaboratory-adapted HIV-1 strains, we used two pri-mary isolates, an X4 rapid/high syncytium-inducingand an X4/R5 slow/low syncytium-inducing. Theresults showed that both primary isolates had lowerreplication levels in CD38-expressing cells than inparental CD38-negative MT4 cells (Fig. 3).

To assess whether CD38 affected the proviral DNAformation, we evaluated HIV-1 proviral genome incellular DNA extracted from HIV-1IIIB-infected MT-4.38H and MT-4.M cells by semi-quantitative PCR onday 2. The amount of proviral DNA was lower inMT-4.38H cells than in MT-4.M cells (Fig. 4).

HIV-1 binding to mouse T cell expressing thehuman CD4 and/or CD38

These data suggested that CD38 expression couldaffect some event(s) preceding proviral DNA com-

Figure 4. Semi-quantitative PCR analysisof HIV-1 gag in DNA extracted fromMT-4.38H and MT-4.M cells on day 2after infection. A) PCR products ob-tained with primers for HIV-1 gag and thehuman gene bax from serial dilutions ofDNA extracted from MT-4.38H andMT-4. M cells. The amount of gag DNAwas lower in MT-4.38H than in MT-4. Mcells. By contrast, no difference wasfound when the exon 4 of the humangene bax was amplified. B) Quantitativeanalysis of the PCR signal. Two thousandnanograms of DNA were loaded in lanesA, 666 ng in lanes B, 222 ng in lanes C,and 74 ng in lanes D. Estimation ofproviral copies, performed by compari-son with a standard curve, indicated thatMT-4.38H cells contained 1.5 copies ofHIV-DNA per cell, whereas MT-4.M cellscontained 5.2 copies per cell.


pletion. One possibility was that the lateral associ-ation of CD38 with CD4 interferes with viral bind-ing. To investigate this possibility, we used mouseT cells that can be rendered susceptible to HIVbinding by transfection with human CD4, al-though they remain non-permissive for productiveinfection. Therefore, we transfected human CD4(inserted in the pNeoSRa plasmid) and/or CD38(inserted in the pcDNA3.1zeo plasmid) in differ-ent combinations into the mouse SR.D10.CD42.F1T cell clone, a CD42 variant of the D10 TH2 T cell

clone. Three clones expressing human CD4(SR.hCD4), human CD38 (SR.hCD38), or bothmolecules (SR.hCD4.38), respectively, were se-lected (Fig. 5). Because CD4 and CD38 display abasal level of lateral association in human T cells(12), we used FRET to evaluate whether humanCD4 and CD38 maintained lateral associations inthe murine environment. These experimentsshowed that energy transfer was obtained whenSR.hCD4.38 cells stained with the Cy3-conjugatedmAb to CD4 were co-stained with the FITC-conju-

Figure 5. Staining for human CD4 and human CD38 of mouse SR. D10. CD42. F1-derivedtransfectants: control SR. D10. CD42.F1, SR.hCD4 (human CD41), SR.hCD38 (humanCD381), SR. hCD4.38 (human CD41CD381). Black curves show the background fluores-cence obtained by staining cells with the control isotype-matched mAbs. Gray curves showthe fluorescence of cells stained with the mAbs to human CD4 and CD38, as indicated.


gated mAb to CD38. In contrast, no transfer wasdetected when they were co-stained with FITC-conjugated control mAbs to mouse CD2 or CD3(Fig. 6).

The murine clones were used to evaluate the effectsof CD38 expression on CD4-dependent virus/cell bind-ing in the absence of other human molecules. More-over, they did not express endogenous murine CD38and CD4 (35), thus avoiding interference of the hu-man molecules with their mouse homologs. Transfec-tants and parental cells were incubated with HIV-1IIIB

for 2 h at 37°C, washed, and lysed. Cell-associated p24was evaluated by ELISA in cell lysates (Fig. 7A).SR.hCD4 cells displayed higher levels of cell-associatedp24 than parental and SR.hCD38 cells. The double-transfectant SR.hCD4.38 displayed significantly lowerlevels than those displayed by SR.hCD4 cells and simi-lar to those displayed by CD42 clones. The differencecannot be attributed to discrepancies in expressionlevels of human CD4, which were similar in SR.hCD4and SR.hCD4.38 cells. Similar results were obtainedusing HIV-1P1, and performing the experiments at 4°C(data not shown). Because at 4°C the entry process issignificantly reduced, these results support the hypoth-esis that CD38 carries out its inhibitory effects at thestep of HIV-1 binding to target cells. The generality andspecificity of this effect was further investigated byincubating cells with the R5 HIV-1BAL strain and usingan HIV-1 vector pseudotyped with the VSV-G envelopeglycoprotein as a positive control. Figure 7B shows thatSR.hCD4.38 displayed higher levels of cell-associatedp24 than those displayed by SR.hCD4 cells whereHIV-1BAL was used, whereas no differences were foundwhere the pseudotyped vector was used.

To further confirm the possibility that CD38 inhib-its HIV-1 binding to target cells, cells were incubated

with FITC-labeled HIV-1IIIB gp120 for 30 min at 4°C,washed, and analyzed by flow cytometry. We foundthat gp120 staining was significantly lower inCD41CD381 cells than in their CD382 counterparts(Table 2).

DISCUSSION

This work shows that CD38 expression negativelymodulates CD41 T cell susceptibility to infection bydifferent HIV-1 strains. This effect was initially sug-gested by evaluating the correlation between CD38expression levels and susceptibility to infection inseveral human CD41 T cell lines. The results werethen confirmed by transfecting the CD38 gene in ahuman CD41CD382 T cell line and in a murine Tcell clone expressing human CD4. The resultsshowed that productive infection was significantlydelayed in cells expressing high levels of CD38. Thiseffect was not restricted to laboratory-adapted X4virus strains, since it was observed also with X4 andX4/R5 primary isolates.

The effect seemed to be mediated by early events inthe replication cycle of HIV because 1) reduction ofvirus yield was maximal at early times of infection and2) at 48 h after infection, cells expressing high CD38levels displayed lower levels of full-length proviral DNAthan cells expressing low CD38 levels. It is likely thatthe effect is mainly exerted at pre-entry stages becauseit was detectable in binding experiments with wholevirus or purified gp120 and was detectable in bothhuman (data not shown) and murine cells.

Inhibition of HIV-1 infection by the lateral associa-tion between CD38 and CD4 potentiated by gp120 was

Figure 6. FRET between human CD4 andCD38 in SR.hCD4.38 cells. CD4 displaysFRET with CD38, but not with mouse CD2or CD3. Cells were stained with Cy3-conju-gated CD4 mAb and the indicated FITC-conjugated mAb. FITC was excited at 488nm and Cy3 emissions were collected at.600 nm. The FACS profiles show onerepresentative experiment. Each quadrantshows Cy3 emissions at .600 nm in theabsence and in the presence of the indicatedFITC-conjugated mAb. A right shift of thecurve indicates FRET. The bar graph showsthe mean 6 sd of the median fluorescenceintensities, expressed as median fluorescentchannels, from three experiments.


postulated in earlier studies (12–14). The ability ofgp120 to recruit CD38 seems to be highly conserved,since it was detected with glycoproteins derived fromHIV-1IIIB, HIV-1SF2, HIV-1MN, and HIV-1451 (13). More-over, it did not presumably involve chemokine recep-tors because no lateral association was detected be-tween CD38 and CXCR4 or CCR5 in the presence orabsence of gp120 (U. Dianzani, personal observation).It could be that CD38 docking to CD4 affects the gp120interaction with CD4 and/or viral co-receptors by stericinterference. However, we cannot rule out the possibil-ity that steps following gp120/CD4 interaction may alsobe affected by CD38.

The results with CD381 transfectants minimize thepossibility that the negative correlation between

CD38 expression and HIV-1 infection is due tofactors independently controlling both variables.This possibility was raised by the observation thatretinoic acid-responsive elements (RARE) controlthe CD38 gene as well as the expression of HIV-1:endogenous RARE positively regulate the CD38gene, whereas viral RARE in the long-terminal re-peats (LTR) negatively regulate HIV-1 genes (28,43). CD38 expression, however, was not under thecontrol of RARE in transfectants.

Inhibition could also be linked to the enzymatic orco-stimulatory functions of CD38. CD38 producesnicotinamide (NAm), a compound reported to in-hibit HIV-1 replication and cytopathogenicity (14,44) probably through inhibition of poly(ADP-ribose)polymerase (45). This possibility seems to be unlikelybecause the concentrations of NAm required toinhibit HIV-1 replication are much greater thanthose produced by CD38 in cell culture (P. Deterre,personal communication). Alternatively, co-stimula-tion might enable CD38 to inhibit in the samemanner as CD28, whose triggering can decrease cellsusceptibility to macrophage-tropic HIV-1 strains byinhibiting expression of CCR5 (46). However, thispossibility appears equally unlikely because CD38-mediated inhibition was detectable in the mousemodel in the absence of viral co-receptors andproductive infection. It was also detectable in short-term binding experiments at 4°C, which minimizedthe effect of putative signals triggered by gp120through CD38.

The inhibition of HIV-1 replication by CD38 expres-sion may account for the lower susceptibility of resting/naive (CD45RA1) than memory (CD45R01) CD41 Tcells (47–49), since resting naive cells constitutivelyexpress CD38, whereas resting memory cells areCD382 (16). Moreover, it may play a role in thefavorable prognostic value of high levels of CD38 onCD41 cells in HIV1 children (26) and in the relativeresistance to HIV-1 infection shown by children born to

Figure 7. Levels of virus/cell association in mouse SR. D10 celllines expressing human CD4 and/or human CD38 incubatedfor 2 h with HIV-1. A) HIV-1IIIB displays higher association withSR.hCD4 cells (CD4) than with SR.hCD4.38 (CD4 CD38),SR.hCD38 (CD38) or parental SR. D10 cells (control). B)Comparison between HIV-1BAL and pRRL.sin.hPGK. GFP (VSV-env) association with SR.hCD4 and SR.hCD4.38 cells. Cells wereincubated with the viral suspensions (titer: 10 ng of p24/ml),washed and immediately lysed. pRRL.sin.hPGK. GFP is an HIV-1vector pseudotyped with the VSV-G envelope glycoprotein.Levels of virus/cell association are expressed as picograms ofp24 in 2 3 106 cells (means6sd from three experiments).Asterisks indicate values significantly different than those dis-played by SR.hCD4 cells (significance threshold: P,0.05, repeat-ed-measures ANOVA1Student-Neuman-Keuls test).

TABLE 2. Staining of mouse SR.D10 cells transfected withhuman CD4 and/or human CD38 with FITC-conjugated HIV-1gp120 or mAb to human CD4

Human moleculeexpressed

gp120a mAb to CD4b

MeFIc %d MeFIc %d

CD4 119 6 23 32 335 6 89 78CD4 CD38 78 6 7f 9 408 6 68 83CD381 23 6 7f ,5 UD ,5none 39 6 12f ,5 UDe ,5

a Cells were stained with FITC-conjugated gp120. b Cells werestained with FITC-conjugated anti-CD4 mAb. c Mean 1 sd of themedian fluorescence intensities (MeFI) obtained from four experi-ments. d Percentage of positive cells (PFC; median from four experi-ments). e Undetectable. f Values significantly different than those dis-played by SR.hCD4 cells (CD4; significance threshold: P , 0.05,repeated-measures ANOVA 1 Student-Neuman-Keuls test).


HIV1 mothers (50). CD38 expression, in fact, is basi-cally higher in children than in adults, whereas thehighest levels are found in newborns, who have highcounts of recent thymic emigrants and naive T cells(51–53). By contrast, our observation is in apparentcontrast with the notion that PHA-activated T cells,which express high levels of CD38 (CD38bright), aremore susceptible to HIV infection than resting T cells.A reasonable speculation is that CD38-mediated pro-tection may be highly effective in resting CD38bright

cells, such as recent thymic emigrants and naive cells,but not in activated CD38bright cells, which express adifferent pattern of molecules involved in viral bindingand entry (such as chemokine receptors, CD26, orCD44) and are much more efficient in supporting viralreplication at post-entry levels, which may overwhelminhibition by CD38 (41, 51, 54). Therefore, CD38-mediated protection may be highly effective in chil-dren, whose CD38bright cells are predominantly restingcells, but not in adults, where they are predominantlyactivated (53).

In conclusion, our data suggest that CD38 expres-sion negatively modulates lymphocyte susceptibilityto CD4-dependent infection by HIV-1, probably byinhibiting the gp120/CD4-dependent virus/cellbinding. It is noteworthy that CD38 may also beexpressed by cells of the monocyte/macrophagelineage, another crucial HIV-1 target (28). In thisregard, we are currently developing a model to testwhether the CD38-related inhibitory effects found inthis study may be extended to macrophage infectionby R5 HIV-1 strains.

To our knowledge, this is the first report of anaturally occurring surface molecule capable of di-rectly inhibiting susceptibility to HIV-1 infection.Elucidation of the inhibitory mechanism(s) of CD38may help to clarify those of HIV-1 infection andprovide ways of preventing it. Current antiretroviraldrugs inhibit after HIV-1 attachment/entry. Strate-gies aimed at blocking such entry are being exten-sively sought in the effort to inhibit HIV-1 replicationat several steps (55).

This work was supported by the AIDS project (Istituto Supe-riore di Sanita, Rome), the Italian National Association againstAIDS (ANLAIDS), Piedmont Section, Turin; Associazione Itali-ana Ricerca sul Cancro (AIRC, Milan), and grant FIS 98/0037(Madrid). M. Bragardo was supported by ANLAIDS, Rome, M. J.Feito by the AIDS project. We thank Dr. G. Poli, Dr. C. Balotta,Milan, Italy, Dr. L. Naldini, Dr. A. Follenzi, and Dr. F. Piro,Turin, Italy, for providing materials and suggestions. A. Savarinois personally grateful to Dr. P. Gioannini (Turin, Italy) forenlightening and encouraging discussion.

REFERENCES

1. Wyatt, R., and Sodroski, J. (1998) The HIV-1 envelope glyco-proteins: fusogens, antigens, and immunogens. Science 280,1884–1888

2. Platt, E., Madani, N., Kozak, S. L., and Kabat D. (1997)Infectious properties of human immunodeficiency type 1 mu-tants with distinct affinities for the CD4 receptor. J. Virol. 71,883–890

3. Dittmar, M. T., McKnight, A., Simmons, G., Clapham, P. R.,Weiss, R. A., and Simmonds, P. (1997) HIV-1 tropism andco-receptor use. Nature (London) 385, 495–496

4. Cohen, J. (1997) Exploiting the HIV-chemokine nexus. Science275, 1261–1264

5. He, J., Chen, Y., Farzan, M., Choe, H., Ohagen, A., Gartner, S.,Busciglio, J., Yang, X., Hofmann, W., Newman, W., Mackay,C. R., Sodroski, J., and Gabuzda, D. (1997) CCR3 and CCR5 areco-receptors for HIV-1 infection of microglia. Nature (London)385, 645–649

6. Callebaut, C., Krust, B., Jacotot, E., and Hovanessian, A. G.(1993) T cell activation antigen, CD26, as a cofactor for entry ofHIV in CD41 cells. Science 262, 2045–2050

7. Oravecz, T., Roderiquez, G., Koffi, J., Wang, J. Ditto, M.,Bou-Habib, D. C., Lusso, P., and Norcross, M. A. (1995) CD26expression correlates with entry, replication and cytopathicity ofmonocytotropic HIV-1 strains in a T-cell line. Nature Med. 1,919–926

8. Dukes, C. S., Yu, Y., Rivadeneira, E. D., Sauls, D. L., Liao, H. X.,Haynes, B. F., and Weinberg, J. B. (1995) Cellular CD44S as adeterminant of human immunodeficiency virus type 1 infectionand cellular tropism. J. Virol. 69, 4000–4005

9. Levine, B. L., Mosca, J. D., Riley, J. L., Carroll, R. G., Vahey,M. T., Jagodzinski, L. L., Wagner, K. F., Mayers, D. L., Burke,D. S., Weislow, O. S., St. Louis, D. C., and June, C. H. (1996)Antiviral effect and ex vivo CD41 T cell proliferation inHIV-positive patients as a result of CD28 costimulation. Science272, 1939–1943

10. Kitchen, S. G., Korin, Y. D., Roth, M. D., Landay, A., Zack, J. A.(1998) Costimulation of naive CD8(1) lymphocytes inducesCD4 expression and allows human immunodeficiency virus type1 infection. J. Virol. 72, 9054–9060

11. Frank, I., Stoiber, H., Godar, S., Stockinger, H., Steindl, F.,Katinger, H. W., and Dierich, M. P. (1996) Acquisition of hostcell-surface-derived molecules by HIV-1. AIDS 10, 1611–1620

12. Dianzani, U., Bragardo, M., Buonfiglio, D., Redoglia, V., Fu-naro, A., Portoles, P., Rojo, J., Malavasi, F., and Pileri, A. (1995)Modulation of CD4 lateral interaction with lymphocyte surfacemolecules induced by HIV-1 gp120. Eur. J. Immunol. 25, 1306–1311

13. Feito, M. J., Bragardo, M., Bonissoni, S., Bottarel, F., Malavasi,F., and Dianzani, U. (1997) gp 120s derived from four differentsyncytium-inducing strains induce different patterns of CD4association with lymphocyte surface molecules. Int. Immunol. 9,1141–1147

14. Savarino, A., Pugliese, A., Martini, C., Pich, P. G., Pescarmona,G. P., and Malavasi, F. (1996) Investigation of the potential roleof membrane CD38 in protection against cell death induced byHIV-1. J. Biol. Reg. Homeost. Agents 10, 13–18

15. Mehta, K., Shahid, U., and Malavasi, F. (1996) Human CD38, acell surface molecule with multiple functions. FASEB J. 10,1408–1417

16. Dianzani, U., Funaro, A., Di Franco, D., Garbarino, G., Bra-gardo, M., Redoglia, V., Buonfiglio, D., De Monte, L. B., Pileri,A., and Malavasi., F. (1994) Interaction between endotheliumand CD41CD45RA1 lymphocytes: role of the human CD38molecule. J. Immunol. 153, 952–959

17. Howard, M., Grimaldi, J. C., Bazan, J. F., Lund, F. E.,Santos-Argumedo, L., Parkhouse, R. M. E., Walseth, T. F., andLee, H. C. (1993) Formation and hydrolysis of cyclic ADP-ribose catalyzed by lymphocyte antigen CD38. Science 262,1056 –1059

18. Deaglio, S., Morra, M., Mallone, R., Ausiello, C. M., Prager, E.,Gambarino, G., Dianzani, U., Stockinger, H., and Malavasi, F.(1998) Human CD38 (ADP-ribosyl cyclase) is a counter-recep-tor of CD31, an Ig superfamily member. J. Immunol. 160,1106–1115

19. Morra, M., Zubiaur, M., Terhost, J., Sancho, J., and Malavasi, F.(1998) CD38 is functionally dependent of TCR/CD3 complexin human T cells. FASEB J. 330, 1129–1135

20. Lund, F. E., Yu, N., Kim, K-M., Reth, M., and Howard, M. (1996)Signaling through CD38 augments B cell antigen receptor


(BCR) responses and is dependent on BCR expression. J. Im-munol. 159, 1455–1467

21. da Silva, C. P., Schweitzer, K., Heyer, P., Malavasi, F., Mayr,G. W., and Guse, A. H. (1998) Ectocellular CD38-catalyzedsynthesis and intracellular Ca21-signalling activity of cyclic ADP-ribose in T-lymphocytes are not functionally related. Immunol.Rev. 161, 79–93

22. Liu, Z., Hultin, L. E., Cumberland, W. G., Hultin, P., Schmidt, I.,Matud, J. L., Detels, R., and Giorgi, J. V. (1996) Elevated relativefluorescence intensity of CD38 antigen expression on CD81

cells is a marker of poor prognosis in HIV infection: results of 6years of follow-up. Cytometry 26, 1–7

23. Liu, Z., Cumberland, W. G., Hultin, L. E., Kaplan, A. H., Detels,R., and Giorgi J. V. (1998) CD81 T lymphocyte activation inHIV-1 disease reflects an aspect of pathogenesis distinct fromviral burden and immunodeficiency. J. Acquir. Immune Defic.Syndr. Hum. Retroviruses 18, 332–340

24. Autran, B., Carcelain, C., Li, T. S., Blanc, C., Mathez, D.,Tubiana, R., Katlama, C., Debre, P., and Leibowitch, J. (1997)Positive effects of combined antiretroviral therapy on CD41 Tcell homeostasis and function in advanced HIV disease. Science277, 112–116

25. Vigano, A., Saresella, M., Rusconi, S., Ferrante, P., and ClericiM. (1998) Expression of CD38 on CD8 T cells predicts mainte-nance of high viraemia in HAART-treated HIV-1-infected chil-dren. Lancet 352, 1905–1906

26. De Martino, M., Rossi, M. E., Azzari, C., Gelli, M. G., Galli, L.,and Vierucci, A. (1998) Different meaning of CD38 moleculeexpression on CD41 and CD81 cells of children perinatallyinfected with human immunodeficiency virus type-1 infectionsurviving more than five years. Pediatr. Res. 43, 752–758

27. Sirera, R., Bayona, A., Carbonell, F., Perez-Tamarit, D.,Otero, M. C., Asensi, F., and Gonzalez-Molina, A. (1998) Theexpression of CD38 and DR are markers of immune activa-tion and disease progression in HIV1 children. 12th WorldAIDS Conference: Geneva, Switzerland, June 28-July 3 1998,Abstract 213*/31166, p. 528 Casparie: Meerhugowaard, TheNetherlands

28. Ferrero, E., and Malavasi, F. (1999) The metamorphosis of amolecule: from soluble enzyme to the leukocyte receptor CD38.J. Leukoc. Biol. 65, 151–161

29. Malavasi, F., Calligaris-Cappio, F., Milanese, C., Dellabona, P.,Richiardi, P., and Carbonara, A. (1984) Characterization of amurine monoclonal antibody specific for human early lympho-hemopoietic cells. Hum. Immunol. 9, 9–20

30. Pauwels, R., De Clerq, E., Desmyter, J., Balzarini, J., Goubau, P.,Herdewijn, P., Vanderhaeghe, H., and Vandeputte, M. (1987)Sensitive and rapid assay on MT-4 cells for detection of antiviralcompounds against the AIDS virus. J. Virol. Meth. 16, 171–185

31. Strizki, J. M., Turner, J. D., Collman, R. G., Hoxie, J., andGonzalez-Scarano, F. (1997) A monoclonal antibody (12G5)directed against CXCR-4 inhibits infection with the dual-tropichuman immunodeficiency virus type 1 isolate HIV-1(89.6) butnot the T-tropic isolate HIV-1(HxB). J. Virol. 71, 5678–5683

32. Savarino, A., Calosso, L., Piragino, A., Martini, C., Gennero, L.,Pescarmona, G. P., and Pugliese, A. (1999) Modulation ofsurface transferrin receptors in lymphoid cells de novo infectedwith human immunodeficiency virus type 1. Cell Biochem. Funct.17, 47–55

33. Scheglovitova, O., Capobianchi, M. R., Antonelli, G., Guanmu,D., and Dianzani, F. (1993) CD4-positive lymphoid cells rescueHIV-1 replication from abortively infected human primaryendothelial cells. Arch. Virol. 132, 267–280

34. Diez-Orejas, R., Ballester, S., Feito, M. J., Ronda, M., Ojeda, G.,Criado, G., Portoles, P., and Rojo, J. M. (1994) Genetic andimmunological evidence for CD4-dependent association ofp56lck with the ab T-cell receptor (TCR): regulation of TCR-induced activation. EMBO J. 13, 90–99

35. Bragardo, M., Buonfiglio, D., Feito M. J., Bonissoni, S., Redoglia,V., Rojo, J. M., Ballester, S., Portoles, P., Garbarino, G., Malavasi,F., and Dianzani, U. (1997) Modulation of lymphocyte interac-tion with endothelium and homing by HIV-1 gp120. J. Immunol.159, 1619–1627

36. North, M. E., Ivory, K., Funachi, M., Webster, A. D. B., Lane,A. C., and Farrant, J. (1996) Intracellular cytokine productionby human CD41 and CD81 T cells from normal and immuno-deficient donors using directly conjugated anti-cytokine anti-

bodies and three-colour flow cytometry. Clin. Exp. Immunol. 105,517–522

37. Capobianchi, M. R., Abbate, I., Antonelli, G., Turriziani, O,Dolei, A., and Dianzani, F. (1998) Inhibition of HIV type 1 BaLreplication by MIP-1a, MIP-1 223, and RANTES in macro-phages. AIDS Res. Hum. Retroviruses 14, 233–240

38. Dull, T., Zufferey, R., Kelly, M., Mandel, R. J., Nguyen, M.,Trono, D., Naldini, L. (1998) A third-generation lentivirusvector with a conditional packaging system. J. Virol. 72, 8463–8471

39. Nakashima, H., Pauwels, R., Baba, M., Schols, D., Desmyter, J.,De Clercq, E. (1989) Tetrazolium-based plaque assay for HIV-1and HIV-2, and its use in the evaluation of antiviral compounds.J. Virol. Meth. 26, 319–329

40. Greenough, T. C., Brettler, D. B., Somasundaran, M., Panicali,D. L., Sullivan, J. L. (1997) Human immunodeficiency virus type1-specific cytotoxic T lymphocytes (CTL), virus load, and CD4load, and CD4 T cell loss: evidence supporting a protective rolefor CTL in vivo. J. Infect. Dis. 176, 118–125

41. Dolei, A., Biolchini, A., Serra, C., Curreli, S., Gomes, E.,Dianzani, F. (1998) Increased replication of T-cell-tropic HIVstrains and CXC-chemokine receptor-4 induction in T cellstreated with macrophage inflammatory protein (MIP)-1al-pha, MIP-1beta and RANTES beta-chemokines. AIDS 12,183–190

42. Szollosi, J., Tron, L., Damjanovich, S., Helliewell, S. H., Arndt,D. J., and Jovin, T. M. (1994) Fluorescence energy transfermeasurements on cell surfaces: a critical comparison of steady-state fluorimetric and flow cytometric methods. Cytometry 5,210–216

43. Maciaszek, J. W., Coniglio, S. J., Talmage, D. A., and Viglianti,G. A. (1998) Retinoid-induced repression of human immuno-deficiency virus type-1 core promoter activity inhibits virusreplication. J. Virol. 72, 3394–3400

44. Murray, M. F., and Srinivasan, A. (1995) Nicotinamide inhibitsHIV-1 in both acute and chronic in vitro infection. Biochem.Biophys. Res. Commun. 210, 954–959

45. Furlini, G., Re, M. C., and La Placa, M. (1991) Increasedpoly(ADP-ribose)polymerase activity in cells infected by humanimmunodeficiency virus type 1. Microbiologica 14, 141–148

46. Carroll, R. G., Riley, J. L., Levine, B. L., Feng, Y., Kaushal, S.,Ritchey, D. W., Bernstein, W., Weislow, O. S., Brown, C. R.,Berger, E. A., June, C. H., and St. Louis, D. C. (1997) Differen-tial regulation of HIV-1 fusion cofactor expression by CD28costimulation of CD41 T cells. Science 276, 273–276

47. Schnittman, S. M., Lane, H. C., Greenhouse, J., Justement, J. S.,Baseler, M., and Fauci, A. (1990) Preferential infection of CD41

memory T cells by human immunodeficiency virus type 1:evidence for a role in the selective T-cell functional defectsobserved in infected individuals. Proc. Natl. Acad. Sci. USA 87,6058–6062

48. Cayota, A., Vuillier, F., Scott-Algara, D., Feuillie, V., and Dighi-ero, G. (1993) Differential requirements for HIV-1 replicationin naive and memory CD4 T cells from asymptomatic HIV-1seropositive carriers and AIDS patients. Clin. Exp. Immunol. 91,241–248

49. Ostrowski, M. A., Chun, T. W., Justement, S. J., Motola, I.,Spinelli, M. A., Adelsberger, J., Ehler, L. A., Mizell, S. B.,Hallahan, C. W., Fauci, A. S. (1999) Both memory andCD45RA1/CD62L1 naive CD4(1) T cells are infected in hu-man immunodeficiency virus type 1-infected individuals. J. Virol.73, 6430–6435

50. Matheson, P. B., Weedon, J., Cappelli, M., Abrams, E. J., Shaffer,N., Bamji, M., Krasinski, K., Lambert, G., Kaul, A., Grimm, K., etal. (1995) Comparison of methods of estimating the mother-to-child transmission rate of human immunodeficiency virus type 1(HIV-1). New York City Perinatal HIV Transmission Collabora-tive Study Group. Am. J. Epidemiol. 142, 714–718

51. Bofill, M., Akbar, A. N., Salmon, M., Robinson, M., Burford, G.,and Janossy, G. (1994) Immature CD45RA(low)RO(low) T cellsin the human cord blood. I. Antecedents of CD45RA1unprimed T cells. J. Immunol. 152, 5613–5623

52. Amlot, P. L., Tahami, F., Chinn, D., and Rawlings, E. (1996)Activation antigen expression on human T cells. I. Analysis bytwo-colour flow cytometry of umbilical cord blood, adult bloodand lymphoid tissue. Clin. Exp. Immunol. 105, 176–182


53. McCloskey, T. W., Cavaliere, T., Bakshi, S., Harper, R., Fagin, J.,Kohn, N., and Pahwa, S. (1997) Immunophenotyping of Tlymphocytes by three-color flow cytometry in healthy newborns,children, and adults. Clin. Immunol. Immunopathol. 84, 46–55

54. Mo, H., Monard, S., Pollack, H., Ip, J., Rochford, G., Wu, L.,Hoxie, J., Borkowsky, W., Ho, D. D., and Moore, J. P. (1998)Expression patterns of the HIV type 1 coreceptors CCR5 and

CXCR4 on CD41 T cells and monocytes from cord and adultblood. AIDS Res. Hum. Retroviruses 14, 607–617

55. Chan, D. C., and Kim, P. S. (1998) HIV entry and its inhibition.Cell 93, 681–684

Received for publication February 22, 1999.Revised for publication August 9, 1999.


Effects of the human CD38 glycoprotein on the early stages of the HIV-1 replication cycle

Documents