Quantification of human immunodeficiency virus type 1 infected mononuclear cells in peripheral blood of seropositive subjects by newly developed flow cytometry analysis of the product of an in situ PCR assay

0095-1137/94/$04.00+0

Copyright X) 1994, American Society forMicrobiology

Quantification of

Human

Immunodeficiency

Virus

Type

1-Infected

Mononuclear

Cells in

Peripheral

Blood

of

Seropositive

Subjects by Newly Developed Flow Cytometry Analysis

of

the Product

of an

In

Situ PCR Assay

MARIA C.

RE,'

GIULIANO FURLINI,1 DAVIDE

GIBELLINI,'

MONICA

VIGNOLI,'

ERIC

RAMAZZOFTI,'

SIMONA

LOLLI,'

SERGIO RANIERI,2

ANDMICHELE LAPLACA1*

InstituteofMicrobiology, University of Bologna Medical School, St. Orsola General Hospital, 40138Bologna,'

and Infectious Disease Department, Santa Maria delle Croci Hospital, Ravenna, Italy

Received 26 January 1994/Returned for modification 29 March 1994/Accepted 7 June 1994

The presence of human immunodeficiency virus type 1 (HIV-1) proviral DNA in peripheral blood mononuclearcells(PBMC) of three groups (group 1, more than 500 CD4+ T cells per,ul;group 2,between 200 and 499CD4+ Tcells per gil;group 3, fewer than 200 CD4+ T cells per ulI) of HIV-1-infected patients, in different stages ofthe disease, was determined by using a newly developed flow cytometry analysis of the product ofinsituPCR assay and compared with other markers of viral replication (HIV-1 p24 antigenemia and viralisolation). Results showed varied percentages of HIV-1-infected PBMC, ranging from 0.6 to 20%. Patients with more than 500CD4+ T cells per ,ul showed the lowest percentage of HIV-1-infected PBMC (2.1 ± 1.7), compared withpatients with CD4+ T-cell counts ofbetween200and 499per ,ul (6.5% + 4.1%;P<0.001) and patients with fewerthan200CD4+ Tcellspergil (4.9%o + 4.7%;P< 0.05). Thedifferenceinthepercentage of HIV-1-infected PBMC between group 2 and group 3 patients may in part reflect the loss of CD4+ T lymphocytes in more advanced stages of the disease. However, the results clearly indicate a striking coincidence between the fall oftheCD4+T-cell countbelow400/giland thesharp increase inPBMC virus loading and p24 antigenemia. Since theprocedure isrelatively easy to perform, it could be used to monitor the evolution of HIV-1 infectionandmay prove a useful adjunct in tailoring therapeutic strategies.

A dramatic decrease in the number of circulating CD4+ helper-inducer T lymphocytes, followed by a profound de-rangementof theimmune response, is the hallmark of disease

progressioninhumanimmunodeficiencyvirus type 1

(HIV-1)-infected patients (17,21, 22,25).

The quantification of peripheral blood mononuclear cells

(PBMC)productively orlatentlyinfectedby HIV-1 is essential tounderstand variouspathogenicaspectsof thedisease,better define the stage of thedisease, and tailor therapeutic strate-gies.

Different technicalprocedures have been usedtoassessthe

frequency of HIV-1-infected cells in thebloodstream (5,28), lymphonodi (15, 20), and other regions of the body (17).

Although some studies have furnished interesting new data, conflicting results have been reported. While initial studies demonstrated that only 1 or fewer in 100 CD4+ T cells harbored HIV-1 provirus (5, 27-29), subsequent research

suggestedthat the percentage of infected PBMC may be much higher thanpreviously described

(1-4,

9, 15, 23, 26).

The different percentagesreported by different authors may bedue,inpart, totechnical differencesinthevarious studies. Acurrentlimitation of PCRonextractedDNAis thedifficulty

in correlating the presence of the viral genome with a single

cell and thus determining the exact percentage of HIV-1-carrying cells(13).Onthe otherhand,the in situhybridization

procedures suffer fromlowsensitivityandalthough in situ PCR

*Correspondingauthor.Mailingaddress: Institute ofMicrobiology,

UniversityofBolognaMedicalSchool,St. Orsola GeneralHospital,9 Via Massarenti, 40138 Bologna, Italy. Phone: 39.51.308888. Fax: 39.51.341632.

can detect a single infected cell, it requires microscopical screening of a large number of cells to determine the exact percentage ofHIV-1-harboringcells(1, 18).Thecombination of in situ PCR and flow cytometry analysis, instead, can overcome mostthese drawbacks (23, 31).

Knowing the percentage of HIV-1-infected PBMC during the different stages of the disease may provide important

information. We thereforeinvestigatedthe presence of HIV-1

proviral DNA in the PBMC of three groups of

HIV-1-seropositivepatients, using anewly developedflow cytometry

analysisof theproductof in situ PCR assay incomparisonwith other markers(HIV-1 p24antigenemiaand virusisolation)of infection.

MATERIALSANDMETHODS

Patients and controls. Peripheral blood samples were ob-tained from 53 HIV-1-seropositive subjects classified on the basis of the 1993 Centers for Disease Control revised classifi-cation system forHIVinfection(7)andfrom15healthyblood

donors, after informed consent according to the Helsinki Declaration of 1975.

Group 1 comprised 14patientswhowere asymptomaticor showed only persistent generalized lymphadenopathy with a meanCD4+T-lymphocytenumberof(696 ± 200) x

106/liter.

Group 2 included 15 symptomatic subjects, with no AIDS indicator conditions, with ameanCD4+ T-lymphocyte num-berof(347 ± 74) x

106/liter.

Group 3comprised24patients

in an advanced stage of the disease, with a mean CD4+

T-lymphocyte number of(76 ± 68) x

106/liter.

Three of the 14 asymptomatic patients, 7 of the 15 with

symptomatic disease, and 10 of the 24 with AIDS were

2152

on May 15, 2020 by guest

http://jcm.asm.org/

undergoing 3'-azido-3'-deoxythymidine (zidovudine)

treat-mentwhen the

peripheral

blood

samples

wereobtained(Table

1).

Cell

separation.

Venousbloodwascollected in EDTA(K3,

7.5%)

Vacutainers

(Becton Dickinson)

andprocessedwithin 4 h.

Low-density

PBMCwereisolated

by

collecting the floating

fractionsobtained from

leukocyte pellets

after

centrifugation

over

Ficoll-Hypaque density gradients (Sigma,

St.Louis, Mo.)

(24).

These cellswere

immediately

cultured for virusisolation

and for PCR assaysasdescribedbelow.

HIV-1

p24 antigenemia

and viral isolation.Plasma

samples

from HIV-1-infected

patients

were

immediately

frozen at

-70°C.

HIV-1

p24 antigenemia,

after immune

complex

disso-ciation

by

acid treatment, was evaluated

by

a commercial

enzyme-linked

immunosorbent assay kit for the

specific

quan-titative determinationof HIV-1 gag

p24 antigen (Du

Pontde

Nemours)

used

according

to the manufacturer's instructions.

Positive results were all confirmed

by

neutralization tests

(HIV-1 p24 confirmatory

test; Du Pont de

Nemours).

The concentration of HIV-1 gag

p24 antigen

for each

positive

sample

was

gauged by interpolation

from the standardcurve

by

quadratic regression

methods. Plasmasampleswithp24levels below 5

pg/ml

wereconsidered

negative.

For isolation of infectious

HIV-1,

106

low-density

PBMC from

patients'

bloodwere

kept

in

growth

medium

(RPMI

1640

medium; GIBCO,

Grand

Island, N.Y.) supplemented

with

10% heat-inactivated

(56°C

for 45

min)

fetal calf serum

(GIBCO),

N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic

acid

(HEPES;

5

mM),

and

gentamicin

(50 ,ug/ml)

andstimulated with 1 ,ug of

phytohemagglutinin

(Wellcome

HA

1617)

per

ml,

at

37°C.

After 2

days,

cells cocultivated with

106

phytohemag-glutinin-stimulated

PBMCobtainedfrom

healthy

donorswere maintained in RPMI 1640

supplemented

with 10% TCGF

(Medical System).

Coculture supernatants were tested twice a week for the presenceof HIV-1 gag

p24 antigen.

In situ PCR. The reaction was carried out

by

a

newly

developed

method offlowcytometry

analysis

of the

product

of

in situ PCR assay. The method was

derived,

with

slight

modifications,

from methods

already

described in reportsfrom

our

(13, 16)

and other

(23, 31)

laboratories and is basedon a

PCR carriedoutin the presence of

digoxigenin-labeled

dUTP toobtaina

digoxigenin-labeled amplicon

ableto reactwithan

antidigoxigenin fluorescein-conjugated

antibody

detectable

by

aflow

cytometric procedure.

Briefly,

the reaction wascarried out as follows. Cells

(5

x

105)

fixed in 1%

paraformaldehyde

in

phosphate-buffered

saline

(PBS;

pH

7.4)

for 2 hat roomtemperatureandwashed three timeswith 500

RI

of PBS were treated with 200

IlI

of pronase

(10 ,ug/ml)

for 5 minat

37°C

inawaterbath. Pronase inactivationwas

performed by placing

the cells at

95°C

for 3 min.The cell

pellet,

obtained

by centrifugation

at

1,500

xgfor 5 minatroom temperature,was

resuspended

in50 ,lofthe

following

PCRmixture: 10 mM Tris HCl

(pH 8.3),

50 mM

KCl,

2.5 mM

MgCl2,

220

,uM

(each)

dATP,

dCTP,

and

dGTP,

215

,uM

dTTP,

7

,uM

digoxigenin-labeled dUTP,

200

pM

each

primer,

and 2.5 U of

Taq polymerase

(Perkin-Elmer

Cetus,

Norwalk, Conn.). Oligonucleotide primers

SK38 and SK39,

whichcan

amplify

a

115-bp region

of the HIV-1 gag gene,were used

(19,

30).

The

amplification

wascarriedout

by

ahot-start

technique

(17)

toprevent

mispriming

and

primer

dimerization.

The

sample

was

subjected

to 35

cycles

of

denaturation,

rean-nealing,

and

extension,

which were

performed

at 94°Cfor 1

min,

55°C

for 1.5min, and

72°C

for 1.5 min, respectively. At the end of the

reaction,

the cellswerewashedtwice with PBS

(200

RI)

and treated with 1%

blocking

reagent

(Boehringer

GmbH, Mannheim, Germany) in PBS (200

[lI)

for 10 minat

370C.

Sampleswere then centrifugedat 1,500 xgfor 5 min and resuspended in 200 ,lI of 1% blocking reagent. Antidigoxigenin

fluorescein-conjugated antibody (Boehringer GmbH) was added (finaldilutionof1:1,000),andsampleswereincubated for 30minat roomtemperature.Finally, after several washings with 200 ,ul of PBS, the samples were analyzed by flow cytometry.

To assess the DNA amplification capability and positive

signal detection in the system employed, samples were

ana-lyzedinparallelwitholigonucleotideprimers GH26 and GH27

(Perkin-Elmer Cetus), which flankaconserved region of the

HLA-DQ(oi) locus, aspreviously described (12). A couple of lambda phage-specific oligonucleotide primers (5'-GATGA

GTTCGTGTCCGTACAACTGG-3', mapping at positions

7131 to 7155, and 5'-GGITATCGAAATCAGCCACAGCG

CC-3', mapping at positions 7606 to 7630) which have no

homologieswith HIV-1orcellular DNAwereusedasnegative controls.

Construction ofa standard curve and positivecontrol for PCR assays. The 8E5/ILAV cell line (11), which contains a

singlecopyof the HIV-1 provirusgenome percell,wasused as apositivecontroltoverifyboth thespecificityandsensitivityof thetest.

Cellswerepassedtwiceaweek in RPMI 1640supplemented with 10% fetal calfserum (GIBCO).

Cells in the mid-log phase ofgrowthwerewashed in PBS and counted inahemocytometer.Toassessthelowest limit of

positive signal detection by the procedure employed and to construct astandardcurvefor PCR assays,8E5/LAV cellswere mixed with uninfected lymphoblastoid A3.01 cells (10) in variousproportionsupto atotal of5 x 105 cells andstoredat

4°C

until tested

by

PCRas previouslydescribed.

Flowcytometry. CellswereanalyzedwithaFACScan

(Bec-tonDickinson, Baltimore, Md.) flow cytometer. Laser excita-tionwas15 mW at488nm.

Instrument sensitivity was standardized with calibration beads

(Becton

Dickinson). The percentage of fluorescent

positive

cellswas determined by integration over a range of

0.1%positive counts on identically treated negative controls

(100%

uninfected

PBMC)

oroncellsamplesforwhichthe in situ PCR had been performed with nonspecific primers. At

least10,000cellswereanalyzed for each sample.

Statistical analysis. The results are expressed as means ± standard deviations of the dataobtained intriplicate

experi-ments.Atwo-tailed Student'st testforunpaireddata wasused forstatisticalcomparison.

RESULTS

HIV-1p24 antigen detection and viral isolation. Among the 14patientsof thefirst group,only2 (14%)showed detectable levels ofp24(from12.2to56.9pg/ml)inplasma samples, while viral isolation was positive in five cases (36%). Five of 15

patients

of the second group

(33%)

and 14 of24ofthe third group

(58%)

showedpositivelevels ofp24(from10to53 and from 16 to 560 pg/ml, respectively), while a positive viral isolation result was obtained from 12 of 15 patients in the second group and allpatientsinthe third group(Table 1).

Standardcurve and positive control for PCR assay. Unin-fected lymphoblastoidA3.01 cellsmixed with serially

increas-ing

percentages(0, 6.2, 12.5, 25, 50,and100)of8E5/LAVcells weresubjectedtoamplificationandflow cytometry analysisto

verify the specificity of the method. A standard curve was

plotted as shown in Fig. 1; data are representative of two

on May 15, 2020 by guest

http://jcm.asm.org/

TABLE 1. Hematologicalandvirologicaldatafrom HIV-1-infected patients byCD4+ T-lymphocytecount

No.of Tlymphocytes Detection by: CD4+

CD4+T-cell count No.ofleukocytes Lymphocyte Monocyte (106/liter) p24Ag' cells/PBMC

perliterand

(106liter)

content(%)

co(n%e)n

CD4+ CD8+

(p±ml)

Virus I R eP

patientno.

(pg/mi)DVirusCISPCR6

(%)

isolation (% 36.9 44.8 55.0 40.2 47.6 46.3 41.2 38.6 40.0 28.9 36.0 35.5 35.7 45.9 0.8 6.0 15.7 4.9 5.1 7.2 8.8 6.0 6.0 7.3 9.0 7.2 7.0 5.9 1,080 1,024 945 858 756 638 620 580 576 561 544 527 520 512 696 1,006 2,920 1,014 1,132 836 1,080 980 918 714 833 816 1,200 576 40.9±6.6 6.9 +3.2 696+200 1,051 +566

- 7.3 59.3

- 1.0 40.2

+ 2.8 38.7

- 0.8 40.6

+ 3.0 34.4

- 0.6 38.8

- 0.8 43.4

56.9 + 0.8 39.0

- 2.9 41.5

- 0.8 46.0

12.2 + 2.3 45.2

- 2.1 42.6

+ 1.7 48.4

- 2.2 39.6 2.1+ 1.7 42.7±6 (499-200) x 106

1c 2 3' 4 5c 6 7C 8 9C l0c 11 12c 13 14 15 Mean+ SD <200 x 106

1' 2 3 4 S 6 7C 8 9 10 11 12c' 13' 14C 1Sc 16 17 18 19C 20 21 22c 23C 24'

Mean ±SD

7,700 9,800 8,200 6,900 6,400 4,900 3,800 3,300 4,500 5,100 4,800 4,500 6,100 3,200 8,700 24.8 36.4 33.5 31.7 42.0 45.0 28.0 35.2 39.8 43.9 20.8 46.7 52.7 29.8 23.0 5.8 6.4 8.1 8.6 8.0 6.0 11.0 7.5 10.7 7.4 0.0 7.7 6.5 10.1 3.0 480 432 405 396 378 374 374 348 342 336 326 315 256 240 200 912 2,736 1,944 1,188 1,755 1,232 506 660 1,116 1,792 437 1,140 1,760 530 1,520

35.6 ±9.4 7.1 + 2.8 347 ±74 1,282--643 4,570 5,200 3,900 5,000 3,200 8,400 6,500 3,100 3,700 4,300 2,800 2,700 5,100 3,700 3,400 2,400 3,900 7,800 5,700 3,400 2,100 1,200 3,200 2,100 39.0 51.0 53.1 37.0 40.0 27.7 52.2 42.3 44.0 11.4 40.0 41.1 22.4 28.0 30.0 24.8 19.6 5.4 32.7 27.0 25.7 40.0 14.8 20.1 12.0 7.0 3.0 6.0 6.5 4.7 5.7 5.8 6.0 4.6 7.0 10.9 9.1 8.0 7.9 5.8 7.5 1.2 5.2 3.0 8.0 17.0 4.7 10.4 190 1,390 189 2,079 189 1,197 180 1,296 143 793 138 1,863 136 2,890 130 936 128 1,088 84 420 44 913 44 484 33 726 33 638 30 650 30 360 24 568 20 310 19 1,501 18 648 15 325 10 185 6 444 4 228

32.1 12.8 7.0 +-3.3 76+ 68 914±661

10.0 + 2.4

29.2 + 2.2

+ 10.9

- 1.2

+ 11.3

53.0 + 11.0

+ 9.7 + 9.2 + 3.0 + 11.0 - 1.2 - 9.1 + 7.7

19.2 + 1.3

15.1 + 6.0

6.5 ±4.1 32.5±9.5

+ 5.6

+ 1.2

74.3 + 7.4

423.0 + 8.0

33.7 + 20.0

204.0 + 4.6

+ 2.8

146.0 + 3.0

92.0 + 0.8

151.9 + 2.1

129.0 + 8.4

140.0 + 2.3

+ 15.0

560.0 + 6.3

+ 6.3

179.0 + 7.6

53.0 + 1.3

+ 4.4

+ 2.3

+ 1.0

+ 1.0

16.0 + 3.4

149.0 + 0.8

+ 1.7

4.9±4.7 23.5±7.1

Ag, antigen. bISPCR, in situ PCR.

'Under zidovudine treatment.

>500 x 106 1 2c 3 4 S 6 7 8 9c lbc 11 12 13 14 Mean ± SD

5,010 7,100 8,100 6,400 5,800 4,700 4,800 5,100 4,400 6,000 4,800 4,800 3,800 3,500 39.3 25.3 31.3 35.6 27.8 26.7 53.4 42.3 36.2 27.3 32.7 27.0 18.1 44.1 20.4 31.7 18.3 14.0 22.3 23.6 19.6 13.5 20.8 18.9 41.0 18.2 24.1 30.9 24.7 23.2 23.0 29.9 22.1 14.6 11.8 25.9 31.3 25.1 34.7

on May 15, 2020 by guest

http://jcm.asm.org/

0)-G 60 40

0)

0-20

0 20 40 60 80 100

Positive cells added(%)

FIG. 1. Diagram of results obtained in the experiments with unin-fected lymphoblastoid A3.01cells mixed with seriallyincreasing per-centages(0, 6.2, 12.5, 25, 50, and100)of8E5/LAV cells andsubjected toamplification andflowcytometryanalysis. Dataare representative oftwoindependent experimentsrun intriplicate.

FIG. 3. Example of HIV-positive and -negative cells observed under a UVlamp-excited microscope. A delimited nuclear fluores-cence canbeobserved.

independent experimentsrun intriplicate. The coefficients of

variationateachexperimental pointneverexceeded 5%. Bya

least-fitlinear regression analysisalinearrelation(correlation

coefficient,r=0.998)wasobservedbetweenthepercentageof single HIV-1 provirus-bearingcells addedto eachcell

popu-lationand thepercentageofpositivecells detected.

Figure 2A to C show some representative histograms for

A B C

A7

&A

NEG POS NEG POS NEG POS LOG (Green fluorescence)

0

0

0)

>)

D E

1 2 3 4 0 1 2 3 4

LOG(Green fluorescence)

FIG. 2. Thetophistograms(A, B, and C) showsome

representa-tiveresults obtained in theexperiments with 8E5/LAV cells mixed with variouspercentagesofuninfected A3.01cellstoassessthe lower limit

ofsensitivity of the method employed: 0.05% 8E5/LAV cells plus 99.95% A3.01 cells (A); 40.00% 8E5/LAV cells plus 60.00% A3.01 cells(B);and90.00%8E5/LAVcellsplus 10.00%A3.01cells(C). The bottomhistograms (DandE) show thepercentagesof HIV-1-infected PBMCobserved in twowidely different HIV-1-seropositive patients: 0.80% positive PBMC in a patient with an absolute CD4+ T-lymphocytecountofmorethan 500/,ul (D)and20.00% positivePBMC inapatientwithanabsoluteCD4+T-lymphocytecountof fewer than 200/,ul (E).

mixed cell populations containing from 0.05 to 100% HIV-1 DNA-positive cellstoascertain thesensitivity of thetestinour experimental conditions. A negative PCR signal was consis-tently obtained when 8E5/LAV cells were present at a fre-quency of<0.1%, whichmeansthat the lowerlimitforpositive signal detection by the procedure was about 10 proviral genomes in abackground of 10,000 cells.

No positive signal over thebackground was ever observed when DNApolymerasewasomittedorlambdaphage-specific

primers were used. Nonspecific fluorescein-conjugated anti-digoxigenin antibody stainingwas notdetected.Figure3shows anexample of positive and negative cells with stained nuclei.

Percentage of HIV-1 provirus genome-containing PBMC detected byin situ PCR. As expected, all HIV-1-seropositive

patients showed detectable levels ofHIV-1 provirus DNA in PBMC, with percentagesofpositive cellsranging from 0.6to

20%.Tworepresentative histograms from patients with diver-gent values areillustrated in Fig. 2Dand E.

The percentages of infected PBMC differed in the three categories of HIV-1-seropositive patients studied. The first group ofsubjects showed a meanvalue of infected PBMC of 2.1% ± 1.7% (with a range ofpositivity from 0.6 to 7.3%), while the other twocategoriespresented mean values of 6.5% + 4.1% and4.9% ± 4.7%(with ranges ofpositivity from1.2 to

11.3% and from 0.8to 20.0%), respectively(Table 1). These differences were significant between the first and second groups of patients (P < 0.001) and between the first and the third (P<0.05). The differences remained significant when the percentagesof infected PBMCwere normalized for theCD4+ cells(T4lymphocytes plus monocytes) estimatedto be presentineachsampleanalyzed by the flow cytometer (Fig. 4).

Figure 4showsthe normalized percentage of HIV-1 provi-rus-containing CD4+ cells and p24 antigenemia (picograms per milliliter) of each subject plotted against the absolute numberofCD4+ T

lymphocytes

per microliter.

Ascanbeseen, thepercentage of virus-infected CD4+cells and the absolute values ofp24 antigenemiawere consistently higher for patients with a CD4+ T-cell count below 400/,ul than forpatients having higherCD4+ T-cellcounts.

No significant differences in the percentages of HIV-1-infected CD4+ cells were observed between the patients

on May 15, 2020 by guest

http://jcm.asm.org/

600

-U AZT-untreated

500 - *AZT-treated

1

400 0l.

'

300-.

200-0.

100 ±

00

-1200

0p24 Ag

U

* *

U*

0 * a

1000 800 600 400 :

T4lymphocytes absolutecounts/pl

FIG. 4. Behavior of in situ PCR and flow cyt normalized for CD4 counts on PBMC of zidi untreated patients and p24 antigenemia value absolutecountspermicroliter.

undergoingzidovudinetherapyand thoseun ofbloodsampling.

DISCUSSION For different technical approaches, cont been reported on the number of infected PE

lymphocytes during the courseof HIV-1 inf

28).Theuncertaintyof results and technical far prevented the routine use of virus l disease staging and tailoring of adequate gies.

Byslight modifications of methodsdescrib our (13, 16) and other (23, 31) laboratoriev

out a procedure which allows the flow cytc individual cells containing HIV-1 proviral amplified by in situ PCR. The procedure

synthesisofadigoxigenin-tagged amplicon,t

digoxigenin-labeled dUTP, and its subsequ fluorescein-conjugated antidigoxigenin anti dure isabletodetectasfewas 10 HIV-1 pi abackground of10,000 cells.

By this method, we examined the perc infected PBMC in three groups (group

CD4+ Tcells per

pul;

group2,between 200 cells per

pul,

group3,fewer than 200 CD4+ HIV-1-infected patientsindifferent stages

compared results with those of p24 antig

isolation assays.

p24 antigenemia clearlycorrelated with th of CD4+ Tcells per microliter and the ste although itwasonly detected inalow perc4 Virus isolation was constantly positive in E

withCD4+T-cellcountsbelow

500/pul

(Tab]

with those of other authors(8, 9, 14),ourd the numberof CD4+ Tcells declines,the HIV-1 p24 is higher and a greater numbe HIV-1 p24 antigendetectable inplasma sai

Flow cytometry analysis of theproduct c assayperformedonPBMC showed detectal ralDNAin allHIV-1-positivesubjects studit ofpositivecells

significantly

correlated with

100 antigenemia, and virus isolation when oligosymptomatic 90 (group 1) andsymptomatic (group 2 and 3) patients were taken * m intoconsideration

(Table

1).

-80 In agreement with the data reported by Hsia and Spector o 70 3

(13),

the PBMC from

patients

with CD4+ T-cell counts at between 200 and499/,ul frequently containedahigher percent-60 , age (and ahigher absolute number) of HIV-1-infected cells X-50 c,> O than thosewith

higher

orlower CD4+ T-cell numbers.

o cu No significant differences were observed between patients

40 a^

undergoing

zidovudine treatment and those untreated in the

*

030

^ D three groups studied. This lack of correlation may be dueto

0go

8X + the limited number ofsubjects analyzed as wellastothelength

*-20 9 of the treatment periodcombinedwith different immunologi-° '@ - 10 cal parameters atthe time ofblood

sampling.

Thusour data tPi : suggesttheneed for further more specific longitudinal studies 0 on HIV-1-positive patients monitored from the beginning of

200 0 therapyonwards.

Although the lower percentage ofHIV-1-infected PBMC in

tometry

(ISPCR-FC) patients withfewer than 200 CD4+ T cellsper

pl

compared

ovudine-treated and with patients with CD4+ T-cell counts between 200 and499/ul

s plotted versus T4 is not statistically significant and may be explained by the consistent lossofCD4+ Tlymphocytesinthe former group of subjects, our data demonstrate that the value around 500 CD4+Tcells per

[lI

isacritical threshold below which asharp treated at the time increase in PBMC virus load is observed, while a number of CD4+ T cells per microliter below 200 is associated with evident virusreplication,asdemonstratedby the higher levels ofp24antigenemia (Fig. 4).

In mostpatients,however, the percentage of HIV-1-infected trasting data have PBMC wasrelatively low, thus indicating that, besides HIV-1 3MCand CD4+ T infection of individual cells, additional mechanisms may con-fection(2, 4, 5, 14, tribute to thedepletion of CD4+ T cells observed in vivo.

difficulties haveso Inconclusion, we have developed a relatively simple proce-oad estimates for durewhich allows flow cytometry analysis of the product ofan therapeutic strate- in situ PCR assay. With this technique, we analyzed the percentage of HIV-1-infected PBMC in 53 seropositive

sub-led

in reportsfrom jects, showing a fairly good correlation with CD4+ T-cell s,we haveworked counts,thedifferent stages of the disease, and other markers of )metric analysisof HIV-1 infection. Moreover, the possibility of easily detecting DNA sequences the levels of transcriptionally active and quiescent viruses is based on the makes this a valuable method for monitoring the evolution of byincorporationof HIV-1 infection and providingausefuladjunct intailoring of tent exposure to a therapeuticstrategies.

ibody. This

proce-roviralgenomesin

entage of HIV-1-1, more than 500 and 499CD4+ T Tcells per

RI)

of ofthediseaseand

Yenemia and virus

eabsolute number age of the disease,

entageofpatients. almost all patients

le1).In agreement atashowedthat as relative amountof rof subjects show mples.

Df the in situ PCR ble levels of

provi-ed. The percentage disease status,p24

ACKNOWLEDGMENTS

This research was supported by Progetto AIDS of the Italian

MinistryofHealth, Rome,

Italy,

and

by

CNRtarget

project

Fat.Ma. REFERENCES

1. Bagasra, O., S. P.Hauptman,H. W.Lischner,M.Sachs,and R.J. Pomerantz. 1992. Detection of human immunodeficiency virus type 1 provirus in mononuclear cellsbyin situpolymerasechain reaction. N.Engl.J.Med.326:1385-1391.

2. Bagasra, O., and R. J. Pomerantz. 1993. Human immunodefi-ciency virus type Iprovirus isdemonstrated inperipheral blood monocytesin vivo: a studyutilizingan insitupolymerase chain reaction.AIDS Res. Hum.Retroviruses9:69-76.

3. Bagasra, O., T. Seshamma, J. W.Oakes, andR.J. Pomerantz. 1993.High percentage ofCD4-positive lymphocytesharbor the HIV-1provirusin the blood of certain infected individuals. AIDS 7:1419-1425.

4. Bieniasz, P.D.,K.Ariyoshi,M. A.Bourelly,S.Bloor,R. B.Foxall, E. C. Harwood, and J. N. Weber. 1993. Variable relationship between proviral DNA load and infectious virus titre in the peripheralblood mononuclear cells of HIV-1-infected individuals. AIDS7:803-806.

5. Brinchmann,J. E., J. Albert, and F. Vartdal. 1991. Fewinfected

on May 15, 2020 by guest

http://jcm.asm.org/

CD4+ T cells but a high proportion of replication-competent provirus copies in asymptomatic human immunodeficiency virus type 1 infection. J. Virol. 65:2019-2023.

6. Centers for Disease Control. 1992. 1993 revised classification system for HIV infection and expanded surveillance case defini-tion for AIDS among adolescents and adults. Morbid. Mortal. Weekly Rep. 41:1-19.

7. Chou, Q., M. Russell, D. E. Birch, J. Raymond, and W. Bloch. 1992. Prevention of pre-PCR mis-priming and primer dimeriza-tionimproves low-copy-number amplifications. Nucleic Acids Res. 20:1717-1723.

8. Embretson, J., M. Zupancic, J. L. Ribas, A. Burke, P. Racz, K. Tenner-Racz, and A. T. Haase. 1993. Massive covert infection of helper T lymphocytes and macrophages by HIV during the incubation period of AIDS. Nature (London) 362:359-362. 9. Folks, T., S. Benn, A. Rabson, T. Theodore, M. D. Hoggan, M.

Martin, M.Lightfoote, and K. Sell. 1985. Characterization of a continuousT-cell linesusceptible to thecytopathiceffects of the acquired immunodeficiency syndrome(AIDS)-associated retrovi-rus. Proc.Natl. Acad. Sci. USA 82:4539-4543.

10. Folks, T. M., D. Powell, M. Lightfoote, S. Koenig, A. S. Fauci, S. Benn, A. Rabson, D.Daugherty, H. E. Gendelman, M. D. Hoggan, etal.1986.Biologicaland biochemicalcharacterization of a cloned Leu-3-cell surviving infection with the acquired immune deficiency syndrome retrovirus. J. Exp. Med. 164:280-290.

11. Furlini, G., M. C. Re, E. Ramazzotti, M. Vignoli, D. Zaccarelli, G. Zauli, and M. La Placa. 1992. Absence of human immunodefi-ciency virus type 1 infection in blood donors with doubtful serum reactivity by Western blot analysis. Immunol. Infect. Dis. 2:157-160.

12. Gibellini, D., M. Zerbini, M. Musiani, S. Venturoli, G. Gentilomi, and M. La Placa. 1993. Microplate capture hybridization of amplified parvovirusB19 DNAfragment labeled withdigoxigenin. Mol. Cell. Probes 7:101-106.

13. Hsia, K., and S. A. Spector. 1991. Humanimmunodeficiency virus DNAis present in a high percentage of CD4+ lymphocytes of seropositive individuals. J. Infect. Dis. 164:470-475.

14. Lafeuillade, A., C. Tamalet, P. Pellegrino, C. Tourres, N. Yahi, C. Vignoli, R. Quilichini, and P. de Micco. 1993. High viral burden in lymph nodes during the early stage of HIV-1 infection. AIDS 7:1527-1528.

15. Lalli, E., D. Gibellini, S. Santi, and A. Facchini. 1992. In situ hybridization in suspension and flow cytometry as a tool for the study of geneexpression.Anal. Biochem. 207:298-303.

16. Levy, J. A. 1993.HIVpathogenesis and long term survival. AIDS 7:1401-1410.

17. Nuovo, G. J., M. Margiotta, P.MacConnell, andJ. Becker. 1992. Rapid in situ detection of PCR amplified HIV-1 DNA. Diagn. Mol. Pathol.2:98-102.

18. Ou, C. Y., S. Kwok, S.W.Mitchell,D. H.Mack, J. J. Sninsky, J.W. Krebs,P.Feorino, D. Warfield, and G. Schochetman. 1988.DNA amplification for direct detection ofHIV-1in DNA ofperipheral

bloodmononuclear cells.Science 239:295-297.

19. Pantaleo, G., C.Graziosi, L. Butini, P. A. Pizzo, S. M. Schnittman, D. P.Kotler, and A. S. Fauci. 1991.Lymphoid organs function as major reservoirs for human immunodeficiency virus. Proc. Natl. Acad. Sci. USA88:9838-9842.

20. Pantaleo, G., C. Graziosi, J. F. Demarest, L. Butini, M. Montroni, C. H. Fox, J. M. Orenstein,D. P.Kotler, andA. S. Fauci. 1993. HIVinfection is active andprogressiveinlymphoidtissue during the clinically latent stage of disease. Nature (London) 362:355-358.

21. Pantaleo,G., C. Graziosi, andA. S.Fauci. 1993.Newconcepts in theimmunopathogenesisof humanimmunodeficiencyvirus infec-tion. N. Engl. J. Med. 328:327-335.

22. Patterson, B. K., M. Till, P.Otto, C. Goolsby,M.R.Furtado, L. J. McBride, and S. M.Wolinsky. 1993. Detection of HIV-l DNA and messenger RNA in individual cells by PCR-driven in situ hybridization and flow cytometry. Science260:976-979.

23. Poznansky, M. C., B. Walker, W. A. Haseltine, J. Sodroski, and E. Langhoff. 1991. Arapidmethod forquantitatingthefrequencyof peripheral blood cells containing HIV-1 DNA. J. Acquired Im-muneDefic. Syndr. 4:368-373.

24. Re, M. C., G. Furlini, andM.La Placa.1989. Rapid detection of HIV-1inclinical samples by co-culture with heat-shocked cells. J. Virol. Methods 26:313-317.

25. Rosenberg, Z. F.,andA. S. Fauci. 1989.Theimmunopathogenesis of HIV infection. Adv. Immunol.47:377-431.

26. Saksela, K., C. Stevens, P. Rubinstein, and D. Baltimore. 1994. Human immunodeficiency virus type 1 messenger RNA expres-sion in peripheral blood cells predicts disease progression inde-pendently of the numbers of CD4+ lymphocytes. Proc. NatI. Acad. Sci. USA 91:1104-1108.

27. Schnittman, S. M., J. J. Greenhouse, M. C. Psallidopoulos, M. Baseler, N. P. Salzman, A. S. Fauci, and H. C. Lane. 1990. Increasingviral burden in CD4+ Tcellsfrompatientswith human immunodeficiencyvirus (HIV) infection reflects rapidly progres-sive immunosuppression and clinical disease. Ann. Intern. Med. 113:438-443.

28. Schnittman, S. M., M. C.Psallidopoulos, H. C. Lane,L. Thomp-son,M.Baseler, F.Massari, C.H.Fox, N. P.Salzman, andA. S. Fauci.1989. The reservoir for HIV-1 in humanperipheralblood is a Tcell that maintainsexpressionof CD4.Science 245:305-308. 29. Simmonds, P., P. Balfe, J. F. Peutherer, C. A. Ludlam, J. 0.

Bishop, andA.J.Brown. 1990. Human immunodeficiency virus-infected individuals contain provirus in small numbers of periph-eralmononuclear cells andatlowcopy numbers. J. Virol. 64:864-872.

30. Wain-Hobson, S., P. Sonigo, 0. Danos, S. Cole, and M.Alizon. 1985.Nucleotide sequence of the AIDSvirus,LAV.Cell 40:9-17. 31. Yang, G., S. Garhwal, J. C. Olson, and G. N. Vyas. 1994. Flow cytometric immunodetection of human immunodeficiency virus type 1 proviralDNAbyheminested PCR anddigoxigenin-labeled probes. Clin. Diagn. Lab. Immunol. 1:26-31.

on May 15, 2020 by guest

http://jcm.asm.org/