Journal of Medical Virology 38207-213 (1992)

Efficient and Reproducible New Semimicromethod for the Detection and Titration of HIV in Human Plasma M. Andreoni, L. Sarmati, S.G. Parisi, L. Ercoli, and G. Rocchi Infectious Diseases, Department of Public Health and Cellular Biology, University of Rome Tor Vergata, Rome, Italy A semimicromethod was established for isolating human immunodeficiency virus (HIV) in plasma using 48-well plates and a pool of peripheral blood mononuclear cells (PBMC) from several donors as targets for infection, which increases the efficiency of isolation by reducing the effect of variability due t o diverse donor cell susceptibility to HIV infection. The addition of H9 cells t o the PBMC cultures did not affect measurable titers. Nevertheless, it potentiated strongly virus replication in terms of p24 production in the supernatant of the wells with HIV isolates, thus facilitating interpretation of the results. The titration of a virus strain of a known titre and reverse transcriptase activity in parallel provided a constant parameter of efficiency and reproducibility within each experiment, permitting comparison with results from other laboratories. The reproducibility of the method was highly significant (r = 0.97, P < 0.001); 68% of the 22 plasma samples from HIV-infected individuals tested by this method were positive. The presence of plasma HIV titer correlated well ( P < 0.02) with the low count of CD4-t cells of less than 300/mm3, but not with the presence of the p24 antigen in the serum. 0 1992 Wiley-Liss, Inc.

KEY WORDS: HIV viremia titration, plasma viremia, HIV isolation, predictive markers of AIDS, AIDS therapy monitoring

INTRODUCTION In more than a decade of research, much information has been gathered on the clinical features and pathogenic mechanisms of the acquired immunodeficiency syndrome (AIDS). Nevertheless, there is still little information about the factors underlying the progression of the disease and the hematological and virological indices useful for quantifying virus replication in vivo. Yet, given the growing interest in identifying new compounds with antiretroviral activity and the need for clinical trials that could yield positive responses within 8 1992 WILEY-LISS, INC.

a relatively short time, it is essential to have virological markers that would allow a n adequate selection of patients for treatment and the correct monitoring of drug efficiency during treatment. Some indices, such as quantitation of the p24 antigen and CD4-t cell counts, are considered to correlate well with the clinical stage of the disease [Davey and Lane, 19901. They are still regarded as the most reliable parameters for the presence of circulating virus and the host’s immune status, although there is evidence that they have certain limitations. In the very early stages of human immunodeficiency virus (HIV) infection (CDC I1 and I11 groups), the p24 antigen is frequently absent; in the more advanced stages, the amount and the presence of circulating p24 are extremely variable, a t times correlating very poorly with the clinical stage of the disease [Coombs et al., 19891. It has been demonstrated that, at least at some stages of infection, there is no correlation between the presence of p24 in the serum and the risk of developing AIDS, hence asymptomatic seropositive subjects with high CD4+ cell counts (>600/mm3) have a low risk of AIDS regardless of the presence of the p24 antigen [Venet et al., 1990; Mulder et al., 1990; Fahey et al., 19901. It has also been shown that the p24 antibody correlates better with the CDC classification than the p24 antigen [Coombs et al., 19891. The CD4+ cell count representing the best single predictor of the progression to AIDS [Fahey et al., 1990; Jacobson et al., 19911 is a n immunological marker. In any case, CD4+ cells decrease significantly only after some months and can hardly be considered early indices for monitoring patients on antiretroviral therapy. Furthermore, the CD4+ cell count and the presence of the p24 antigen in the serum have limited value as predictors of opportunistic infections in HIV-infected patients with pulmonary symptoms [Orholm et al., 19901. Other indices such as the loss of anti-p24 antibodies, high serum levels of @,-microglobulin and IgA, and the increased production of neopterin have been assessed

Accepted for publication April 15, 1992. Address reprint requests to Dr. M. Andreoni, Department of Public Health and Cellular Biology, University of Rome Tor Vergata, Via 0. Raimondo, 00173 Rome, Italy.

208

as predictors of progression to AIDS, but they appear to be surrogate markers with little predictive value unless their variation is associated with a significant decrease in CD4+ cells [Fahey et al., 19901. Recently, methods have been developed for quantitating virus replication in vivo. Coombs et al., [19891 and Ho et al. [1989] reported a method of plasma HIV titration, using PBMC from seronegative donors as targets for infection. However, the rate of isolation in homogeneous groups varied widely [Bremer et al., 1991; Merigan e t al., 19911. This suggests poor reproducibility of the technique, which might be due to the different capacity of PBMC from different subjects to replicate HIV [Yamada et al., 19901. We describe a method for quantitating plasma HIV, using 48-well plates, using a pool of PBMC from several subjects as targets for infection.

MATERIALS AND METHODS Viruses An HTLV-IIIB strain derived from H9iHTLV-IIIB cells, a chronically infected human tumor cell line [Popovic et al., 19841, and a PCi90 strain isolated from a patient with AIDS were used. For the preparation of HTLV-IIIB virus stock, the number of cells was adjusted to a cell concentration of 2 x lo5 cellsiml and cultured for 3 days a t 37°C. The PCi90 strain was isolated from a patient with AIDS by coculture of PBMC with 3-day-old 0.5% phytohemagglutinin (PHA)-stimulated PBMC from a healthy donor. This strain was grown only in PHA-stimulated PBMC cultures. The virus stocks were prepared by centrifuging the cells a t 1,300g for 10 min a t 4°C and the subsequent filtering of cell culture supernatants using a 0.22-pm Millipore membrane filter (Nalgene, Rochester, NY). These preparations were stored in liquid nitrogen until use. The two virus stocks were assayed for particle-associated reverse transcriptase activity [Koyanagi e t al., 19861, which was 13,500 cpmiml for HTLV-IIIB and 8,600 cpmiml for the PCI90 strains. The titers of these virus preparations were lo3 26 and lo1 7 5 , respectively, expressed in tissue culture infectious doses 50% (TCID,,) IKarber, 19311 and assayed in H9 cells [Hartshorn et al., 1987 I. Patients Blood samples were obtained from 22 HIV-l-seropositive adults (16 men and 6 women) who had never received antiretroviral treatment; 12 were asymptomatic and 10 were AIDS patients. The mean CD4-t lymphocyte counts in these two groups were 636 and 260imm3, respectively. Plasma A 20-ml sample of heparinized blood (final heparin concentration 20 Uiml) was obtained from each patient. Blood samples were kept at room temperature and centrifuged at 200g for 15 min to separate the plasma and cellular fractions within 4 h r of sampling. The plasma

Andreoni et al. fractions were then centrifuged a t 1,OOOg at 4°C for 10 min; 8-10 ml of plasma were removed and passed through a sterile 0.45-km filter (Nalgene, Rochester, NY) to ensure the removal of any residual cellular o r platelet debris. Aliquots of 1.5 ml of filtered plasma were stored in liquid nitrogen.

Preparation of PBMC for HIV Isolation Lymphocytes were obtained from 46 HIV-antibodynegative blood donors. Whole blood containing preservative-free heparin (Sigma Chemical, St. Louis, MO) a t a final concentration of 20 Uiml was diluted 1:1with sterile phosphate-buffered saline (PBS). PBMC were separated by Ficoll-Hypaque gradient centrifugation and washed twice with PBS; the cell pellets were resuspended in medium A (RPMI-1640, penicillin [ l o 0 Uiml], streptomycin [ 100 kgimll, L-glutamine 10.3 mgi ml] and 20% heat inactivated fetal calf serum). Mononuclear cells were counted in a hemocytometer, distributed into T-75 flasks (Falcon, Basel, Switzerland) to a final concentration of 2 x lo6 cellsiml of medium A, and stimulated with PHA (5 pgiml) for 3 days in 5% CO, at 37°C. Stimulated PBMC were washed twice with PBS, counted, distributed into T-25 flasks (Falcon, Basel, Switzerland) to a final concentration of lo6 cellsiml of medium A containing 2 pgiml of Polybrene (Aldrich Chemical Co., Milwaukee, WIj, and maintained in 5% CO, at 37°C for 30 min. J u s t prior to infection, the PBMC were washed twice with PBS. PBMC from various donors were used in experiments performed on different days. Target Cell Line Clone H9, a n HTLV-I11permissive T-cell line derived from a parental human leukemic T-cell line, was maintained in a n RPMI-1640 medium supplemented with 10% fetal calf serum, 100 Unitsiml of penicillin, 100 pgiml of streptomycin and 0.3 mg/ml of L-glutamine; viable cells (counted by the Trypan blue dye exclusion method) were adjusted by subculture to 3 x 10" cellsiml twice a week. Semimicromethod for Plasma HIV Titration Experiments on cell concentrations. One ml of each dilution (from lop1 to lop6)from two virus stocks was added to 15-ml polyethylene tubes (P.B.I. International, Milan, Italy) on pellets of 4 x lo6, 6 x lo6, 8 x lo6 and 16 x lo6 PBMC treated a s described above. After incubation in 5%CO, a t 37°C for 2 hr, the PBMC were washed twice with medium A; 4 ml of medium B (medium A with 10 Uiml of recombinant interleukin 2 (IL-2) [Biomedical Products Division, Collaborative Research Incorporated, Bedford, MA] and 2 pgiml of Polybrene were added to each tube. Then 1ml of cell suspension was inoculated in each well of a 48well plate (Constar, Cambridge, MA). Each viral dilution was inoculated in 4 wells. The plates were incubated in 5% CO, a t 37"C, half the volume of supernatant in each well was replaced every 3-4 days by fresh medium B, and 5 x 10" healthy donor PBMC

HIV Plasma Viremia Titration treated as described above were added to the cultures on days 7, 14, and 21. On day 28, the culture supernatants were assayed for the p24 antigen and the cells were removed from each well for evaluation of their viability by Trypan blue staining. Testing of PBMC from different donors for capacity to develop H N . The HIV titer, expressed in TCID,,, was determined in plasma samples from 8 AIDS patients and in two virus stocks a s described above, using a final density of 1.5 x lo6 cells/well; 5 x lo5 healthy donor PBMC were added to the culture 7, 14, 21, and 28 days later. Each titration was carried out in fourfold using PBMC from each of the three donors individually and combined. Plasma samples were tested at dilutions of 10°-10-3. Culture supernatants from each well were collected and assayed for the p24 antigen immediately before the addition of PBMC on days 7,14,21, and 28, and subsequently on day 35. H N yield amplification. The titers of the two virus stocks were determined by the method described previously using a mixture of 1.5 x lo6 PHA-stimulated PBMC/well from three donors. The titrations were carried out following four different schedules requiring the weekly addition of PBMC and H9 cells in various concentrations (Fig. 4). Culture supernatants were collected from the wells and assayed for the p24 antigen on day 28.

Plasma HIV Titration A 48-well plate method for plasma HIV titration was carried out using a mixture of PBMC from three donors as a target for HIV isolation. This method was used to determine the virus titer in 22 plasma samples from seropositive subjects and in two virus stocks. One ml of each plasma dilution (10°-10-3) and of each virus stock dilution (10-1-10p6) was added to 15 ml polyethylene tubes containing pellets of 6 x lo6 PBMC, treated as described above, derived from a mixture of equal parts of PBMC from three donors. The PBMC were incubated in 5%CO, at 37°C for 2 h r and washed twice in medium A. After the addition of 4 ml of medium B to each tube, 1 ml/well of the cell suspension was inoculated in 4 wells of a 48-well plate. The plates were incubated in 5% CO, at 37"C, replacing one-half the supernatant volume by fresh medium B every 3 - 4 days; 5 x lo5 stimulatedPBMC were added to cultures on the 7th day and 4 x lo5 PHA-stimulated PBMC + lo5 H9 cells were added on the 14th and 21st days. Culture supernatants were assayed for p24 antigen on day 28. Assay for the p24 Antigen Culture supernatants were assayed for HIV-1 p24 core antigen using a commercial enzyme immunoassay (Abbott Laboratories, North Chicago, IL). Samples were considered positive when the HIV antigen levels were detectable in culture supernatants at a n optical density of >0.1 compared with positive and negative controls according to the criteria provided by the manufacturer. All positive results were confirmed by a standard neutralization assay according to the manufactur-

209 4.5

I

4.25

--4.5

I

6

I PBMC/well (millions)

Fig. 1. Titers of the HTLV-IIIB and PC190 virus strains expressed as log,,TCID,,. The test was performed on 48 well plates using 4 different concentrations of PBMC per well.

er's instructions (HIVAG-1 Blocking antibody test, Abbott Laboratories, North Chicago, IL).

Statistical Analysis Statistical analyses of correlation coefficients were carried out by Student's t-test and by Fisher's exact test (P two-tailed). RESULTS Effect of Cell Density on Isolation of HIV To identify the optimal cell concentration for the isolation of HIV in 48-well plates, preliminary studies were carried out using two virus strains, HTLV-IIIB and PCI90, and 4 different cell concentrations. The results (Fig. 1) show that, after 28 days in culture, the TCID,, of HTLV-IIIB varied with the PBMC concentration per well from 103.5/mlto 104.5/mland the TCID,, of PC/90 from 101-75/mlto 102,5/ml.The analysis of cell viability on day 28 by vital staining with Trypan blue (data not reported) showed a live/dead cell ratio of 1:1.5 for the lowest PBMC concentrations of 1 x lo6 and 1.5 x lo6 cells/well, compared to a cell ratio of 1:3 and 1:5 for concentrations of 2 x lo6 and 4 x lo6 cellsiwell, respectively. Based on these results and on published data on the optimal conditions for HIV isolation [Castro et al., 19881, the concentration of 1.5 x lo6 cellsiwell was chosen for our semimicromethod. Variation in PBMC Preparations From Different Donors to Isolate HIV The capacity of PBMC preparations from different donors to isolate HIV was tested by viral titration in 8 plasma samples from AIDS patients and in the two virus stocks (Fig. 2). The PBMC from donor C were not efficient for the purpose in plasma 1 and those from donors B and C were not efficient in plasma 3. For all the plasma samples and for both virus stocks, the highest titers were obtained with PBMC from donor A, yielding titers 2 logarithms higher than those from the other two donors.

210

Andreoni et al. 1

5 \I

PL I

I

I

I

I

I

I

I

PL 2

PL 3

PL 4

PL 5

PL 6

PL 7

PL 8

I

J

1

HTLV-1118 P C / 9 0

Fig. 2. Comparison of the HIV titers in 8 plasma samples and in two virus stocks obtained in 4 experiments, using PBMC from each of the 3 donors individually (A, B, and C) and combined (ABC) as targets for infection. PL, plasma; n.d., not detectable.

The pool of PBMC from the 3 donors compared with donor A's PBMC alone yielded a higher virus titer in the HTLV-IIIB strain and in plasma samples 4 and 7, equal titers in plasma sample 2, and titers of one logarithm lower in plasma samples 1 , 3 , 5 , and 8, and in the PCI90 strain. HIV was not isolated from plasma sample 6 by any of the PBMC preparations used.

Optimum Culture Time for HIV Titration Of the 8 plasma samples analyzed (Fig. 31, two (pl 3 and pl 7) had a stable virus titer from day 14 onward, one (pl8) from day 21 onward, and three (pl 1, p l 4 , and pl 5) from day 28 onward. The virus titer increased in the last week of culture (day 35) in only one plasma sample (pl 2) and by not more than one-half a logarithm. No HIV was isolated from plasma sample 6 up to day 35. One plasma sample (pl I), which was negative on day 14, had a virus titer of 10°.5/ml on days 28 and 35. Analysis of results from virus strains revealed a stable titer of HTLV-IIIB on day 14, while PCI90 showed a n increase up to day 21.

7

14

-P L I

20

+PL2

.E-+:

ad

+ P L 3

-PL4

'cPL5

r!*_1. 1 _ -_Ap

14

7

Use of Cell Line H9 for Amplifying HIV Replication The effect of the addition of a cell line capable of potentiating virus replication (Fig. 41, showed that none of the experimental schedules providing for the combined use of cell line H9 together with PBMC was better in terms of the ability to isolate HIV compared with the schedule for PBMC alone. In two experiments where PBMC were not added on days 14 (Exp. 3) and 21 (Exp. 41, the titers of the HTLV-IIIB and PCI90 strains were lower than when PBMC were added (Exp. 11, although the differences were one logarithm only. The

35

28

21

DAYS OF CULTURE

2.9

21

35

DAYS OF CULTURE -PLd

+PL7

PLB

-

HTLV-1118

* PC/90

Fig 3. Virus titers in 8 plasma samples and in two virus strains (HTLV-IIIBand PCISO) calculated evaluating the p24 antigen level on the harvesting of cultures on days 7 , 1 4 , 2 1 , 2 8 ,and 3 5 , 5 x lo5 PHAstimulated PBMC were added to the cultures on days 7 , 1 4 , 2 1 , and 28 PL, plasma, n d , not detectable

addition of a mixture of PBMC and H9 on days 14 and 21 (Exp. 2) did not increase the titer of either strain but did ensure higher p24 values in the supernatants of the wells positive for isolation (data not shown).

211

HIV Plasma Viremia Titration E1p.A

5.5

5.25

~0.97

z

p(O.001

-

EXP. I

Exp. 3

Exp. 2

HTLV-III B

l

Exp. 4

*

D

s

o

p(0.02

0

*

*

+

n.d.

0

*

I

zoo

400

*

+ 600

+ ASYMPTOMATIC

+ ooo

1000

*

o

n.ZI m ,

I PUPO

Fig. 4. Titers of virus strains HTLV-IIIB and PC/90 obtained in experiments designed to determine the effect of adding cell line H9 to PBMC. Four different experiment schedules were used for the addition of target cells. Exp. 1: 5 x lo5 PBMCiwell on days 7, 14, and 21. Exp. 2: 5 x lo5 PBMCiwell on day 7 and 4 X lo5 PBMC + lo5 H9 cellsiwell on days 14 and 21. Exp. 3: 5 x lo5 PBMC/well on day 7 and 2 X lo5 H9 cellsiwell on days 14 and 21. Exp. 4: 5 X lo5 PBMCiwell on days 7 and 14, and 2 x lo5 H9 cellsiwell on day 21.

3

o

+ iaoo

MOO

AIDS

Fig. 5. Comparison of plasma virus titers and CD4 counts in 10 AIDS patients and in 12 asymptomatic subjects. n.d., not detectable.

Application of the Method for Plasma HIV Titration When plasma HIV titration was used in 22 seropositive subjects (12 asymptomatic and 10 AIDS patients), plasma virus titers were obtained in 68% of cases, that is, 50% of asymptomatic subjects and 90% of AIDS patients. Positivity in this respect did not correlate with the presence of the p24 antigen in the serum (data not shown), but it did correlate directly and significantly (P < 0.02) with a CD4+ count lower than 300/mm3 (Fig. 5). The titration test was repeated with two virus stocks on 8 different days, using pools of PBMC from different donors for each experiment. The results obtained ranged from 105.25/mlto 104.5/mlfor HTLV-IIIB and from 102.5/ml to 101.25/ml for PC/90 (data not shown). Moreover, in two of the 8 experiments, titration of HIV in 8 plasma samples obtained from AIDS patients had few or insignificant variations (Fig. 6 ) .Thus, the reproducibility of this method is satisfactory (correlation coefficient r = 0.97, 95% confidence limits: 0.88 < R < 0.99, P < 0.001).

1

1

2 3 4 Log ,o K I D 50/ml

5

Exp.B

Fig. 6. Comparison of HIV titers in 8 plasma samples and in two virus stocks (HTLV-IIIB and PCi90) obtained in two experiments ( A and B) by using two pools of PBMC on different days. 0 , plasma; +, PC/90; *, HTLV-IIIB; n.d., not detectable.

DISCUSSION The aim of the study was to establish a plate micromethod with a capacity to isolate and titrate HIV that would be reproducible and applicable to semiautomatic systems. All cell concentrations proved efficient for obtaining a virus titer. The surface cell density used ranged according to the experimental conditions from 1 to 4 x lo6 cells/ml/cm2. Surface cell density is the main factor affecting the time and amount of virus replication in culture, and the optimal yield in terms of HIV isolation is obtained with a surface cell density of 1.5 x lo6 cells/ml/cm2 [Castro et al., 19881. The study of cell viability after 28 days of culture and the results of the virus titrations showed that the concentration of 1.5 X lo6 celldm1 was also optimal in our plate method. The experiments designed to test the efficiency of PBMC from different donors for isolating HIV showed considerable variability in different preparations. On three occasions the use of inefficient PBMC resulted in the failure to isolate virus. Our findings confirm, therefore, t h a t the capacity of PBMC to become infected with HIV and to reproduce the virus varies with cell origin. It is reported that the differences do not depend on the percentage of CD4+ lymphocytes or on the capacity of PBMC to respond to PHA stimulation [Yamada et al., 19901. These findings led to the conclusion that the data obtained from plasma virus titrations carried out a t different times on PBMC from different donors are not comparable. To resolve this problem and to improve the reproducibility of the system, we used a mixture of PBMC from three donors. The virus titers obtained with the resulting pool were similar to those obtained with lymphocytes from the donor who proved to be the most efficient (donor A). PBMC mixtures are widely used for the isolation of HIV from body fluids and from infected cells [Popovic et al., 19841. Indeed, the standard method of HIV isolation provides for the addition, a t intervals, of fresh allogeneic PBMC from seronega-

212

Andreoni et al.

tive subjects to the culture system. Lymphoproliferation and cytotoxicity activation t h a t occur in mixed lymphocyte cultures do not seem to reduce the capacity of PBMC to become infected and to produce the virus, but rather they appear to increase it. Moreover, i t has been demonstrated that in a mixed lymphocyte culture allogeneic PBMC from seronegative subjects supply a n antigenic stimulus capable of activating HIV production by infected PBMC [Margolick et al., 19871. The plasma virus titration test was stopped on day 28 in order not to prolong the culture time unduly. It was found that when culture was prolonged to 35 days, the virus titer increased in only one case and by not more than one-half a logarithm. The use of a cell line known to be susceptible to HIV infection, added on days 14 and 21 together with activated PBMC, did not increase the titer of the two virus stocks, but strongly potentiated virus replication in terms of the production of p24 in the culture medium in all wells in which HIV was detected. The use of the H9 cell line, providing a significant amplification of the level of the production of p24, allowed a more precise interpretation of the results obtained by enzyme-linked immunosorbent assay (ELISA). In order to evaluate the sensitivity and reproducibility of the system, the titration of a standard virus preparation of known titer and transcriptase activity in parallel with every plasma HIV titer determination is essential, also providing a standard of comparison for titrations carried out a t different times and a reference for comparing results from different laboratories. The low percentage (50%)of virus isolated from the plasma of 12 asymptomatic seropositive subjects obtained by using the semimicromethod confirms published reports IEscaich et al., 1991; Rouzioux et al., 1991; Groupe Viremie Quantitative, 19911. This may be explained by the small quantity of circulating free virus present during the early stages of the disease [Clark et al., 1991J or the presence of virus strains with a lower capacity for replication [Valentin et al., 19901 andlor different tropism, i.e. monocytotropic variants [Balachandran et al., 1991; Schuitemaker et al., 19911. Given its excellent reproducibility and standardization, the method of plasma HIV titration as described here may be used for testing the efficacy of antiviral drugs in HIV-infected patients.

ACKNOWLEDGMENTS This work was supported by the National Research Council (C.N.R.), “Prevention and Control of Disease Factors (FATMA)” Project, contract 91.00268.PF41, and by the Ministry of Health “IV AIDS Research Project,” contract 6203-033. The authors would like to thank Simona Bach for her help in the preparation of the manuscript.

REFERENCES Balachandran R, Thampatty P, Enrico A, Rinaldo C, Gupta P (1991): Human immunodeficiency virus isolates from asymptomatic ho-

mosexual men and from AIDS patients have distinct biologic and genetic properties. Virology 180:229-238. Bremer WJ, Hollinger FB, ACTG Virology Laboratories, Lynch C, McCreedy B (1991): HIV detection and quantitation procedures suitable for virologic studies and endpoint determination in clinical trials. Results of a multicenter study. In Andrieu J M (ed): “Viral Quantitation in HIV Infection.” Paris: John Libbey Eurotext, pp 141-145. Castro BA, Weiss CA, Wiviott LD, Levy J A (1988):Optimal conditions for recoverv of human immunodeficiencv virus from uerioheral blood mononuclear cells. Journal of“ Clinical Mi’crodiology 26:2371-2376. Clark SJ, Saag MS, Don Decker W, Campbell-Hill S, Roberson JL, Veldkamp PJ, Kappes JC, Hahn BH, Shaw GM (1991):High titres of cytopathie virus in plasma of patients with symptomatic primary HIV-1 infection. New England Journal of Medicine 324:954960. Coombs RW, Collier AC, Allain TP, Nikora B, Leuther M, Gjerset GF, Corey L (1989):Plasma viremia in human immunodeficiency virus infection. New England Journal of Medicine 321:162&1632. Davey RT, Lane HC (1990): Laboratory methods in the diagnosis and prognostic staging of infection with human immunodeficiency virus type 1.Review of Infectious Diseases 12:912-930. Escaich S, Ritter J , Rougier P, Lepot D, Lamelin J P , Sepetjan M, Trepo C (1991): Detection of HIV plasma viremia by culture and polymerase chain reaction (PCR) at different stages of HIV infection. Abstract book VII International Conference on AIDS, Florence, June 16-21,1991, Abstract No. M.A. 1109,1:119. Fahey JL, Taylor JM, Detels R, Hofmann B, Melmed R, Nishanian P, Giorgi J V (1990): The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1. New England Journal of Medicine 322166172. Groupe Viremie Quantitative: A.C. 11-ANRS; Rouzioux C, Puel J , Agut H, Brun-Vezinet F, Ferchal F, Fleury H, Tamalet C (1991): Quantitative plasma viremia: Comparison of two techniques. Abstract book VII International Conference on AIDS, Florence, June 16-21, 1991, Abstract No. M.A. 1107, 1:118. Hartshorn KL, Vogt MW, Chou T-C, Blumberg RS, Byington R, Schooley RT, Hirsch MS (1987): Synergistic inhibition of human immunodeficiency virus in vitro by azidothymidine and recombinant alpha A interferon. Antimicrobial Agents and Chemotherapy 31:168-172. Ho DD, Moudgil T, Alam M (1989):Quantitation ofhuman immunodeficiency virus type 1in the blood of infected persons. New England Journal of Medicine 321:1621-1625. Jacobson MA, Bacchetti P, Kolokathis A, Chaisson RE, Szabo S, Polsky B, Valainis GT, Mildvan D, Abrams D, Wilber J, Winger E, Sacks HS, Hendricksen C, Moss A (1991): Surrogate markers for survival in patients with AIDS and AIDS related complex treated with zidovudine. British Medical Journal 302:73-78. Karber G (1931): Beitrag zur kollektiven behandlung pharmakologisker reihenversuche. Archiv fur Experimentelle Pathologie und Pharmakologie 162:480483. Koyanagi Y, Harada S, Yamamoto N (1986): Establishment of a high production system for AIDS retroviruses with a human T-leukemic cell line MOLT-4. Cancer Letters 30:299-310. Margolick JB, Volkman DJ, Folks TM, Fauci AS (1987):Amplification of HTLV-IIIILAV infection by antigen-induced activation of T cells and direct suppression by virus of lymphocyte blastogenic responses. Journal of Immunology 138:1719-1723. Merigan TC, Clark G, Katzenstein D, Holodniy M (1991):Correlation between Human Immunodeficiency Virus (HIV) load, stage of disease and dideoxy drug responses using quantitative RNA and DNA Polymerase chain reaction (PCR) and viremia measurements. In Andrieu JM (ed):“Viral Quantitation in HIV Infection.” Paris: John Libbey Eurotext, pp 147-152. Mulder JW, de Wolf F, Goudsmit J, Cload PA, Coutinho RA, Fiddian AP (1990):Long-term zidovudine treatment of asymptomatic HIV1-infected subjects. Antiviral Research 13:127-138. Orholm M, Nielsen TL, Nielsen JO, Lundgren J D (1990):CD4 lymphocyte counts and serum p24 antigen of no diagnostic value in monitoring HIV-infected patients with pulmonary symptoms. AIDS 4:163-166. Popovic M, Sarngadharan MG, Read E, Gallo RG (1984): Detection, isolation and continuous production of cytopathic retroviruses (HTLV-111) from patients with AIDS and pre-AIDS. Science 224:497-500.

HIV Plasma Viremia Titration Rouzioux C, Puel J, Brun-Vezinet F, Fleury H, Agut H, Aubert J , Buffet-Janvresse C, Deforges L, Ferchal F, Gochi K, Ingrand D, Jung M, Masquelier B, Schmitt MP, Tamalet C, Huraux JM (1991): A multicenter assessment of HIV quantitative techniques for plasma viremia and IC50 determination. In Andrieu J M (ed): “Viral Quantitation in HIV Infection.” Paris: John Libbey Eurotext, pp. 21-26. Schuitemaker H, Kootstra NA, de Goede REI, de Wolf F, Miedema F, Tersmette M (1991):Monocytotropic human immunodeficiency virus type 1(HIV-1)variants detectable in all stages ofHIV-1 infection lack t-cell line tropism and syncytium-inducing ability in primary T-cell culture. Journal of Virology 65:356-363.

213 Valentin A, Albert J, Fenyo EM, Asjo B (1990): HIV-1 infection of normal human macrophage cultures: Implication for silent infection. Virology 177:790-794. Venet A, Tourani JM, Beldjord K, Erne D, Even P, Andrieu JM (1990): Actuarial rate of clinical and biological progression in a cohort of 250 HIV-1-seropositive subjects. Laennec HIV Study Group. Clinic of Experimental Immunology 80:151-155. Yamada 0, Matsumoto T, Sasoaka R, Kurimura T (1990): Variations in growth capacity of HIV in peripheral blood mononuclear cell preparations from different individuals. AIDS 4:35-40.