Simian immunodeficiency virus resistance of

Journal of General Virology (2000), 81, 2741–2750. Printed in Great Britain ...................................................................................................................................................................................................................................................................................

Simian immunodeficiency virus resistance of macaques infused with interferon β-engineered lymphocytes Franck Matheux,1 Evelyne Lauret,2 Ve! ronique Rousseau,2 Je! ro# me Larghero,1 Bertrand Boson,1 Bruno Vaslin,1 Arnaud Cheret,1 Edward De Maeyer,2† Dominique Dormont1 and Roger LeGrand1 1 CEA, Service de Neurovirologie (DSV/DRM), CRSSA, Institut Paris-Sud sur les Cytokines, BP 6, 92265 Fontenay aux Roses, Cedex, France 2 Equipe de l’Interfe! ron et des Cytokines, UMR 146 CNRS, Institut Curie, Ba# timent 110, Centre Universitaire, 91405 Orsay, France

To test the in vivo anti-simian immunodeficiency virus (SIV) efficacy of interferon (IFN)-β-engineered lymphocytes, peripheral blood lymphocytes harvested from two uninfected macaques were transduced with a retroviral vector carrying a constitutively expressed IFN-β gene and reinfused, resulting in approximately 1 IFN-β-transduced cell out of 1000 circulating cells. The gene-modified cells were well tolerated and could be detected for at least 74 days without causing any apparent side effects. These two animals together with three untreated control macaques were then infected with SIVmac251. The two IFN-β-infused macaques are in good health, 478 days after infection, with a reduced plasma virus load and sustained numbers of CD4M and CD8M cells. Throughout the study, the proportion of IFN-β-transduced cells has been maintained. Of the three control macaques, two were characterized by a high plasma virus load and a decrease in CD4M cells. One was moribund and was sacrificed 350 days after infection and the other now has fewer than 100 circulating CD4M cells/ml. Unexpectedly, the third control macaque, which, like the two IFN-βinfused animals, had a low plasma virus load and a maintenance of CD4M and CD8M cell number, was characterized by a permanent level of serum IFN-β, of unknown origin, already present before SIV infection. Although no definite conclusion can be made in view of the limited number of animals, these data indicate that further exploration is warranted of an IFN-β-based anti-human immunodeficiency virus gene therapy.

Introduction Although antiviral therapy with inhibitors of protease and reverse transcriptase is quite effective in suppressing the replication of human immunodeficiency virus (HIV), its combined use with immune-based strategies able to boost the immune system has been proposed to contribute to the longterm control of HIV infection (Piscitelli et al., 1996 ; Pantaleo, 1997). Anti-HIV gene therapy strategies are currently based on the direct enhancement of the resistance of HIV target cells by intracellular immunization (Baltimore, 1988), but could also Author for correspondence : Evelyne Lauret. Present address : U362 INSERM, Institut Gustave Roussy, PR1, 39 rue Camille Desmoulins, 94800 Villejuif, France. Fax j33 1 42 11 52 40. e-mail elauret!igr.fr † Present address : Les Genie' vres, Augerville la Rivie' re, 45330 Malesherbes, France.

0001-7078 # 2000 SGM

be designed to improve immune responses. In this respect, interferon (IFN)-β is an attractive candidate : it is a natural antiviral protein that inhibits HIV at various stages of the virus cycle, from uptake to release of viral particles (Vieillard et al., 1994 ; Baca-Regen et al., 1994 ; Kornbluth et al., 1990 ; Shirazi & Pitha, 1993 ; Coccia et al., 1994 ; Poli et al., 1989 ; Hansen et al., 1992), and also displays immunomodulatory properties, which could be an important component in controlling HIV replication (Belardelli, 1995). For the reasons outlined above, a gene therapy strategy based on the constitutive expression of IFN-β is being developed (Lauret et al., 1994). Transduction with a retroviral vector, resulting in a low and continuous expression of IFN-β, causes inhibition of HIV replication and increased cell survival of CD4+ cells in peripheral blood from HIV-infected donors (Vieillard et al., 1995, 1997). In addition to its direct antiviral activity, IFN-β also contributes to the restoration of normal immune cell functions. Transduction of peripheral blood CHEB

F. Matheux and others

lymphocytes (PBL) from HIV-infected donors with our IFN-β construct increases IFN-γ and interleukin (IL)-12 production and brings IL-4, IL-6, IL-10 and TNF-α production back to normal levels. It also increases the cytotoxic responses of CD8+ cells against cells expressing HIV proteins and improves the proliferative response to recall antigens (Vieillard et al., 1997 ; Hadida et al., 1999). These results in vitro have been confirmed in a Hu–PBL–SCID mouse model (Vieillard et al., 1999). Furthermore, anti-HIV resistance has been obtained in macrophages transduced with the IFN-β construct (Cremer et al., 2000). As a relevant in vivo model of anti-HIV IFN-β gene therapy, we have resorted to macaques infected with a pathogenic isolate of simian immunodeficiency virus (SIV), SIVmac251 (Matheux et al., 1999). SIV is a non-human primate lentivirus that shares similar genomic organization and biological properties with HIV-1 and HIV-2 (Desrosiers, 1990) and causes a disease in macaques similar to HIV-induced AIDS in humans (Daniel et al., 1985). We have previously described a retroviral vector in which the macaque (Ma) IFN-β coding sequence has been placed under the control of a 0n6 kb fragment of the murine H-2Kb gene promoter (Matheux et al., 1999). PBL obtained from seronegative animals transduced with this construct develop enhanced resistance to SIVmac251 in vitro. We therefore sought to determine whether IFN-βtransduced PBL could survive in vivo without inducing any side effects and exert anti-SIV activity in macaques.

Methods

Animals. Five male cynomolgus macaques (Macaca fascicularis), weighing 4n5–6 kg and negative for herpesvirus B, filovirus, simian Tlymphotropic virus-1, simian retrovirus-1 and -2, SIV and hepatitis B virus, were used in the study. Two individuals, J813 and J816, subsequently referred to as IFN1 and IFN2, were subjected to a single infusion of IFN-β-transduced lymphocytes. The three others, S62A, S73B and S72C, subsequently referred to as C1, C2 and C3, did not receive transduced lymphocytes and served as controls. Animals were housed in single cages in biosafety level 3 facilities and all experimental procedures were conducted according to the institutional guidelines for animal care (Journal officiel des CommunauteT s EuropeT ennes, L538, 18 December 1986).

IFN-β transduction of PBL. Approximately 100 ml blood was collected in heparin–lithium dry tubes. Buffy coats were obtained by centrifugation (170 g for 15 min). White cells were collected and centrifuged (400 g for 30 min) over a Ficoll density gradient (Eurobio). Plasma and erythrocytes, diluted with an equal volume of PBS, were immediately readministered to the corresponding macaque. PBMC were activated at 10' cells per ml for 3 days in RPMI-1640, 20 % FCS, 2 mM -glutamine (Boehringer Mannheim), 0n2 µM antibiotics (penicillin\ streptomycin\neomycin ; PSN, Life Technologies) and 10 µg\ml concanavalin A (Sigma). After these 3 days, the activated PBL were transduced with the retroviral vector MFG-Kb-MaIFN-β by a subsequent 3 day co-culture on ψ-CRIP packaging cells, as described previously (Matheux et al., 1999). At the end of the co-culture, the different cell populations were transferred two or three times to other plates in order to eliminate by adherence any remaining packaging cells and nonadherent cells were infused into the respective animal. Transduction CHEC

efficacy was estimated, as described hereafter, on PBL maintained in culture for 10 days.

In vivo follow-up of infused macaques. Both infused animals were followed for 21 days before cytapheresis, for 74 days after autologous infusion and for 478 days after SIV infection for weight, haematological analysis, presence of IFN-β-transduced lymphocytes and SIV infection parameters. Axillary lymph nodes were removed under ketamine anaesthesia and disrupted in PBS by using a Potter homogenizer. Peripheral blood was collected in heparin–lithium dry tubes. Mononuclear cells, from blood or lymph nodes, were isolated by standard Ficoll density gradient centrifugation. For each collection, an uninfused animal underwent identical procedures and served as a negative control for DNA extraction and PCR amplification.

SIV challenge. Seventy-four days after infusion, the two macaques that received IFN-β-transduced lymphocytes (IFN1, IFN2) and the three untreated control macaques (C1, C2, C3) were infected intravenously with 4 animal infectious doses (AID ) of a pathogenic SIVmac251 isolate &! (provided by A. M. Aubertin, Universite! Louis Pasteur, Strasbourg, France). In our hands, all but one of 62 inoculated monkeys became infected at this dose. All animals were then followed for at least 478 days after SIV inoculation.

Haematological follow-up. Haematological values and enumeration of blood cells (leukocytes, lymphocytes, polynuclear cells, monocytes, erythrocytes, haematocrit, haemoglobinaemia, platelets) were determined by using an Automat haemocytometer (Coulter Corporation). Estimates of absolute numbers of CD2+, CD4+ and CD8+ cells in the circulation were obtained by using direct immunofluorescence assays (CD4–leu3a–PE and CD8–leu2a–FITC, Becton Dickinson ; CD2– MT910–PE, Dako) and flow cytometry (FACScan, Becton Dickinson) as described previously (Cheret et al., 1996).

Cell sorting. Mononuclear cells were positively separated by using CD4-specific or CD8-specific immunomagnetic microbeads (MiniMACS, Miltenyi) according to the manufacturer’s instructions. Subset purity was evaluated by using second anti-CD4 (OKT4–PE, Ortho-diagnostics) or anti-CD8 (DK25–FITC, Dako) antibodies and quantified by flow cytometry (FACScan).

Analysis of IFN-β-transduced cells. The presence of IFN-βtransduced cells was evaluated, in duplicate or triplicate depending on the amount of sample, by semiquantitative PCR with 200 ng genomic DNA. The amount of DNA used for each sample was normalized after amplifying a fragment of the endogenous IFN-β gene (with 5h ATGACCAACAAGTGTCTCCT 3h as the sense primer and 5h ACAGGCTAGGAGATCTTCA 3h as the antisense primer, delineating a 600 bp sequence) as a quantitative control. Detection of the H-2KbMaIFN-β transgene was performed by using 5h GTTCAGGCAAAGTCTTAGTC 3h (k228 to k209) as the sense primer in the H-2Kb gene promoter and 5h TGAAGATCTCCTAGCCTGT 3h (579 to 560) as the antisense primer in the Ma IFN-β coding sequence. Detection of any remaining murine packaging cells was performed by PCR with a murine α-globin primer set described by Erhart et al. (1987). The PCR amplification products were identified by dot-blot hybridization with an IFN-β probe (Vieillard et al., 1995) and quantified by using a phosphorimager (Molecular Dynamics). The relative intensity of the signals was compared to serial dilutions of lysate derived from plasmid-transfected cells that contained known numbers of IFN-β transgene copies per cell.

Measurements of SIV replication. Anti-SIV antibodies were determined by using an HIV-2 antigen-detection ELISA (ELAVIAII ; Sanofi Diagnostics Pasteur) as reported previously (Le Grand et al., 1994).

Anti-SIV gene therapy with IFN-β

RT–PCR analysis. Total RNA was extracted by using the TriReagent kit (Sigma) and DNase-treated before synthesizing cDNA using the First Strand cDNA synthesis kit (Pharmacia Biotech). Transcription from the construct was evidenced by using primers absent from the endogenous gene, a sense primer located downstream from the H2Kb promoter and upstream from the IFN-β start codon (5h CTCAGAAGTCGTGGTCGACT 3h) and an antisense primer located downstream of the IFN-β stop codon (5h ATCAATTCGAGCTCGGTACC 3h), generating a 660 bp fragment (45 cycles). Control glyceraldehyde-3phosphate dehydrogenase (GAPDH) transcripts were amplified with a sense primer located in the second exon (5h ACTGGCGTCTTCACCACCATGGA 3h) and an antisense primer in the eighth exon (5h GGTCCACCACCCTGTTGCTGTAGC 3h), generating a 932 bp fragment (30 cycles).

Results Transduction and infusion of IFN-β-transduced PBL

To transfer the IFN-β gene into PBL from two SIV-negative macaques, we used a retroviral vector containing the macaque IFN-β coding sequence placed under the control of a 0n6 kb fragment of the murine H-2Kb gene promoter (Matheux et al., 1999). After transduction, the lymphocytes were readministered ; 4i10) PBL for animal IFN1 and 6i10) PBL for animal IFN2. Semiquantitative PCR analysis revealed a mean of 0n2 IFN-β transgene copies per infused cell for both macaques. Hence, the estimated numbers of IFN-β-transduced PBL introduced into the animals were 8i10( and 12i10( for macaques IFN1 and IFN2. In vivo persistence of IFN-β-transduced PBL

After inoculation of IFN-β-transduced cells, PBL samples were drawn at regular intervals for 74 days for PCR analysis. To ensure that the IFN transgene signal did not result from the presence of murine packaging cells, we performed PCR analysis to detect murine DNA sequences. A murine signal was found

PBL (×10–3 per µl blood) CD4+ cells (per µl blood)

–20

(a)

8 6 4 2 –10

0

3000

10

20

30

40

50

60

70

80

10

20

30

40

50

60

70

80

10

20

30

40

50

60

70

80

10

20 30 40 50 60 Time after infusion (days)

70

(b)

2000

1000 0

CD8+ cells (per µl blood)

IFN titration. In view of the absence of a commercially available ELISA for measuring macaque IFN-β, IFN was measured by its biological activity on human WISH cells with vesicular stomatitis virus (VSV) as the challenge (Matheux et al., 1999). Before titration, sera were inactivated by exposure to 56 mC for 30 min but, in spite of this, the presence of small amounts of IFN could not be measured in some serum samples due to the toxicity of low serum dilutions (1 : 3, 1 : 6) on WISH cells. Because of the non-specific acid-labile anti-VSV activity present in the serum (Fall et al., 1995), the IFNs were characterized after acid treatment (pH 2). IFN-α was characterized as acid-stable IFN-α by neutralization with a rabbit antihuman (Hu) IFN-α polyclonal serum that recognized Ma IFN-α and IFNβ by neutralization with a rabbit anti-Hu IFN-β polyclonal serum that cross-reacted with Ma IFN-β.

10

IFN-β (copies per 106 cells)

To reveal anti-SIV antibodies, we used a peroxidase-labelled goat antimonkey IgG (Organon Teknika). The ELISA was performed with a plasma dilution of 1 : 100. Titres were determined by reference to a standard titrated positive pool from sera of SIV-infected macaques. The p27 nucleocapsid protein was detected in undiluted plasma by using the Coulter SIV core Ag assay (Retrovirology Coulter Corporation). SIV RNA levels were determined in plasma by using the SIVmac-bDNA assay (Chiron Diagnostics).

6000

(c)

4000 2000 0

0

10 000

(d)

1000

100

10

0

80

Fig. 1. Analysis of macaques IFN1 (#) and IFN2 ( ) infused with IFN-βtransduced lymphocytes before SIV infection. The proportion of IFN-βtransduced lymphocytes was estimated by semiquantitative PCR, as described in Methods. (a) Number of total lymphocytes in the circulation. (b) Number of CD4+ T cells in the circulation. (c) Number of CD8+ T cells in the circulation. (d) Proportion of IFN-β-transduced cells in the circulation.

only the day after infusion in the PBL of one macaque (IFN1). No more murine DNA could be detected thereafter in either macaque and the presence of IFN-β-transduced cells could be demonstrated in both animals (Fig. 1 d). The proportion of IFNβ-transduced cells increased in the circulation during the first 2 weeks following infusion, 15-fold in macaque IFN1 and 4-fold in macaque IFN2, reaching a peak of 600 and 6000 transduced cells per 10' PBL for macaques IFN1 and IFN2, respectively. CHED

F. Matheux and others

SIV p27 in plasma (ng/ml)

3

3

(a)

2

2

1

1

0

0 0

SIV RNA in plasma (keq/ml)

106

10

30

40

105

105

104

104

103

103

102

102

101

101

105

10

20

30

40

10

20

30

40

0

10

20

30

40

105

(c)

104

104

103

103

102

102

101

101

0

0 106

(b)

0

Serum anti-SIV antibodies

20

20

40

60

80 0 20 Time after SIV infection (days)

40

60

80

Fig. 2. Virus parameters in control and infused macaques during primary infection. Macaques C1 (>), C2 ($), C3 ( ), IFN1 (#) and IFN2 ( ) were studied. The p27 nucleocapsid protein was detected in undiluted plasma by using the Coulter SIV core Ag assay. SIV RNA levels (keq, thousands of RNA equivalents) were determined in plasma by using the SIVmac-bDNA assay.

Thereafter, a mean level of 300 Ma IFN-β-transduced cells per 10' circulating cells was maintained in macaque IFN1 until day 66 whereas, in macaque IFN2, the number of transduced cells, close to 6000 per 10' cells on day 7, declined gradually to reach 200 Ma IFN-β-transduced cells per 10' circulating cells 74 days after infusion. To investigate the potential side effects of the IFN-βCHEE

transduced cells, we followed the behaviour, haematological values and weight of the animals. None of these altered during the 74 days of observation, except for a slight and transient decrease of the leukocyte counts in the peripheral blood which affected the CD4+ and CD8+ lymphocyte populations equally (Fig. 1 a–c). Other haematological parameters were not affected.

Anti-SIV gene therapy with IFN-β (a)

SIV RNA copies (keq/ml)

1000000 100000 10000

100 cut-off 1

10

cut-off 2

0 CD4+ lymphocytes (per µl blood)

(a)

1000

0

1800

CD8+ lymphocytes (per µl blood)

60

90

120 150 180 210 240 270 300 330 360 390

60

90

120 150 180 210 240 270 300 330 360 390 420 450 480

60

90

120 150 180 210 240 270 300 330 360 390 420 450 480

60

90

120 150 180 210 240 270 300 330 360 390 420 450 480 Time after SIV infection (days)

(b)

1600 1400 1200 1000 800 600 400 200 0 0 5000

30 (c)

4000 3000 2000 1000 0

0

0

10000 IFN-β (copies per 106 cells)

30

30 (d)

1000

100 0 10

0

30

Fig. 3. Analysis of control and infused macaques after SIV infection. Macaques C1 (>), C2 ($), C3 ( ), IFN1 (#) and IFN2 ( ) were studied. The p27 nucleocapsid protein was detected in undiluted plasma by using the Coulter SIV core Ag assay. SIV RNA levels were determined in plasma by using the SIVmac-bDNA assay. The proportion of IFN-β-transduced lymphocytes was estimated by semiquantitative PCR, as described in Methods. (a) SIV RNA copies in the plasma (keq, thousands of RNA equivalents) ; (b) number of CD4+ T cells in the circulation ; (c) number of CD8+ T cells in the circulation ; (d) proportion of IFNβ-transduced cells in the circulation.

CHEF

F. Matheux and others

Table 1. Distribution of Ma IFN-β-transduced cells among circulating and lymph node lymphocytes The proportion of IFN-β-transduced lymphocytes was estimated by semiquantitative PCR, as described in Methods. PBMC, Peripheral blood mononuclear cells ; LNMC, lymph node mononuclear cells. , Not done. Day 50 post-infusion Cell population Macaque IFN1 Total PBMC CD4+ PBMC CD8+ PBMC Total LNMC CD8+ LNMC Macaque IFN2 Total PBMC CD4+ PBMC CD4− PBMC CD8+ PBMC CD8− PBMC Total LNMC CD4+ LNMC CD8+ LNMC

Purity (%)

IFN-β copies per 106 cells

Purity ( %)

IFN-β copies per 106 cells

    

    

– 80 99 – 90

250 100 250 1000 400

– 96 99 88 95   

2300 400 2800 2500    

–   95  –  93

1000   50  1000 200 200

Evolution of the virus parameters

To determine the anti-SIV resistance of the two IFN-βinfused macaques and of the three untreated animals used as controls (C1, C2, C3), they were infected intravenously with SIVmac251 (4 AID ). The course of SIV infection was studied &! by measuring the following virus parameters : plasma p27 antigen, plasma SIV RNA copies and serum antibodies directed against SIV antigens (Fig. 2). Between days 12 and 17 after infection, the three control macaques developed a peak of p27 antigenaemia, higher for macaque C2 (2n5 ng\ml) and lower for macaques C1 and C3 (0n5 and 0n4 ng\ml), whereas macaques infused with IFN-β-transduced cells maintained a level of p27 antigenaemia below 0n25 ng\ml (Fig. 2 a). With the more sensitive Chiron b-DNA assay, our results showed that, between days 10 and 25, the peak numbers of plasma SIV RNA copies in control macaques reached values of 5i10) copies\ml for macaque C2, 4n1i10( copies\ml for macaque C1 and 2n7i10( copies\ml for macaque C3, while the peak values measured in the infused macaques were lower, 4n7i10' and 8i10' in macaques IFN2 and IFN1, respectively (Fig. 2 b). All of the SIV-infected macaques seroconverted within 1 month of SIV inoculation, with anti-SIV antibody titres reaching 5i10% within 2 months of inoculation (Fig. 2 c). When the numbers of SIV RNA copies in the plasma were measured during chronic infection (Fig. 3 a), SIV RNA was detected in two control macaques, amounting to between 10( and 8i10( SIV RNA copies\ml for macaque C2 and between CHEG

Day 552 post-infusion

5i10% and 10& copies\ml for macaque C1. Macaque C2 was moribund on day 380 and was sacrificed. SIV RNA remained undetectable for control animal C3 and the two infused macaques throughout the study (lower limit of detection, 1500 copies\ml). Evolution of CD4M and CD8M cells

For all animals, control and infused, SIV infection led to a decline in the absolute numbers of CD8+ and CD4+ lymphocytes 14 days after SIV challenge (Fig. 3 b, c) and these values returned to normal 30 days after the onset of infection. At 293 days after SIV inoculation, two of the three control animals, C1 and C2, had reduced numbers of circulating CD4+ lymphocytes compared with their baseline. The other control animal, C3, and the two infused animals still have sustained numbers of circulating CD4+ lymphocytes at the time of writing (day 478) (Fig. 3 b). The number of circulating CD8+ cells of control macaques C1 and C2 declined in the circulation during chronic infection in parallel with the decrease of the number of CD4+ T cells (Fig. 3 c). Evolution of IFN-β-transduced cells in peripheral blood

The day after SIV challenge, the numbers of circulating IFN-β-transduced PBL were 150 (IFN1) and 160 (IFN2) per 10' cells. During the first month following SIV inoculation, the level of IFN-β-transduced cells decreased to reach a nadir of 30

Anti-SIV gene therapy with IFN-β

IFN-β-transduced PBL per 10' circulating cells 25 days after SIV challenge in macaque IFN1 (Fig. 3 d). In macaque IFN2, this proportion increased transiently to reach 700 Ma IFN-βtransduced PBL per 10' circulating cells 20 days after SIV inoculation and then decreased to reach a nadir of 160 Ma IFNβ-transduced PBL per 10' circulating cells 5 days later. These proportions increased thereafter to reach 300 (IFN1) and 3000 (IFN2) Ma IFN-β-transduced cells per 10' cells 40 days after SIV infection (Fig. 3 d). During the 16-month follow-up following the primary infection, both macaques maintained a level of IFN-βtransduced cells, ranging from 150 to 500 per 10' circulating cells in macaque IFN1 and from 250 to 1000 per 10' circulating cells in macaque IFN2. Distribution of IFN-β-transduced cells

To follow the distribution of IFN-β-transduced cells among lymph nodes, a biopsy of an inguinal lymph node was obtained from macaque IFN1 the day after infusion, which revealed a concentration of IFN-β-transduced cells similar to that in the circulation (45 per 10' cells). A similar analysis was performed 66 and 478 days after SIV inoculation. In macaque IFN1, on day 66, when the proportion of IFN-β-transduced cells was 360 per 10' cells in the circulation, 10-fold more IFN-β-transduced cells were found among lymph node cells (3500 per 10' cells). In the same animal, 478 days after SIV infection, the proportion of IFN-β-transduced cells was four times higher in the lymph node as compared with the proportion in the peripheral blood, whereas there was no significant difference in the proportion of IFN-β-transduced cells between the lymph node and circulating cells in macaque IFN2 at this time (Table 1). The distribution of IFN-β-transduced cells among CD4+ and CD8+ populations was analysed by cell-sorting experiments performed on PBMC obtained from macaque IFN2 on day 50 after infusion, with population purities of 88–99 % (Table 1). PCR analysis of these sorted cells indicated that most of the transduced circulating lymphocytes in this monkey were of the CD8+ phenotype. Such an analysis was also performed on day 478 after SIV infection on circulating and lymph node cells of both animals. In contrast to the results obtained 50 days after infusion, IFN-β-transduced cells did not consist mainly of the CD8+ phenotype and were equally distributed among CD4+ and CD8+ T cells (Table 1). Serum IFN subtypes

During primary infection, a peak of IFN-α production was observed in three macaques, high in macaques C1 and IFN1 (2000 U\ml) and lower in macaque C3 (120 U\ml) (Table 2). Thereafter, IFN-α could only be detected at a low level (between 60 and 120 U\ml) in some serum samples of macaques C2 and IFN2. When analysing IFN-β production, we observed the presence of IFN-β in some serum samples of the two infused monkeys, ranging from 60 to 120 U\ml, and

Table 2. IFN subtypes present in the sera of control and infused macaques before and after SIV infection IFN was measured as described in Methods. p.i., Post-infection ; , not done ; , not determined due to the toxicity of the serum. Amount of IFN (U/ml) Time p.i. (days) IFN-α k388 k88 k4 7 14 21 28 194 279 293 347 386 IFN-β k388 k88 k4 7 14 21 28 194 279 293 347 386

C1

C2

C3

IFN1

IFN2

60  60  2000 60  60    

60  60 60  60 120  60 120 120 

60  60  120  60   60  60

    2000 60  60 60   

   120   60  60 60 120 120

 60  60  60   60 60  

 60  60 60  60 60   60 

 120 120 120  120  120   120 120

      60  120   

    60   60   60 

IFN-β transgene

1

2

3

4

GAPDH Fig. 4. RT–PCR analysis of the expression of the IFN-β transgene in PBL 478 days after SIV infection. PBL from animals C1 (lane 1), C3 (2), IFN1 (3) and IFN2 (4) were analysed. Total RNA was extracted by using the TriReagent kit (Sigma) and DNase-treated before synthesizing cDNA. Transcription from the construct was evidenced by using a sense primer located downstream of the H-2Kb promoter and upstream of the IFN-β start codon and an antisense primer located downstream of the IFN-β stop codon, generating a 660-bp fragment (45 cycles) as described in Methods. GADPH transcripts serve as an internal control.

RT–PCR analysis performed on day 478 after SIV infection revealed the presence of small amounts of transcripts specific for the transgene in PBL of macaques IFN1 and IFN2 (Fig. 4, CHEH

F. Matheux and others

lanes 3 and 4, respectively). No IFN-β could be detected at any time in the serum of control macaques C1 and C2. However IFN-β was regularly detected at 60–120 U\ml in the serum of macaque C3. This surprising observation prompted us to look for the possible presence of IFN-β in the sera of all the monkeys before SIV infection. This revealed the presence of IFN-β in the serum of macaque C3 at a level identical to that observed after infection, but not in the sera of the other monkeys. RT–PCR analysis revealed that, in the three control macaques, circulating cells did not produce detectable levels of IFN-β (data not shown).

Discussion The IFN-β-transduced lymphocytes infused into the two cynomolgus macaques (IFN1 and IFN2) were well tolerated and could still be detected in the circulation 552 days after infusion. After SIVmac251 infection of these infused macaques, virus replication was limited and disease progression was delayed, as opposed to two control, untreated animals. However, in the third control macaque, SIV replication was limited to an extent comparable to the IFN-β-infused macaques ; this animal was characterized by constitutive IFN-β production, already present before SIV infection. The continuous presence of IFN-β in the serum of macaque C3 points to the existence of a site of production releasing this cytokine constantly into the circulation. That no IFN-β mRNA could be detected in the lymphocytes of this macaque is in line with the fact that lymphocytes produce only IFN-α and not IFN-β. In view of the stable level of IFN-β throughout the experiment in the serum of macaque C3, it is rather unlikely that this IFN-β production results from a chronic virus infection, which would give rise to a mixture of IFN-α and -β. Most studies on the cellular sources of IFN-β production have been done in mice and the only cell type in which evidence for constitutive IFN-β production has been reported is the macrophage (De Maeyer & De MaeyerGuignard, 1988). There are no data available in the literature concerning the presence of IFN-β in the serum of macaques. In humans, one study reported that no IFN-β was present in the serum of uninfected individuals (1n72p3n48 U\ml, n l 77 ; Minagawa et al., 1989). SIV replication was reduced in IFN-β-infused macaques as well as in the control macaque characterized by permanent IFN-β production and these three macaques maintained normal levels of CD4+ and CD8+ T cells during the 478 days of observation, as opposed to the two other control macaques. We are aware that the number of animals used in our study is not sufficient to prove that IFN-β is responsible for this antiSIV resistance and that several hypotheses such as genetic influences (Michael, 1999) or chemokines controlling infection (Lehner et al., 1996 ; Wang et al., 1998) could be invoked to explain these data, but our results are in line with our hypothesis. The production of IFN-β, directly in the lymph nodes, could enhance the antiviral resistance of the CD4+ T CHEI

transduced cells directly and, furthermore, could induce a protective bystander effect on neighbouring untransduced cells, as demonstrated in vitro by the protection of neighbouring untransduced cells in culture (Matheux et al., 1999). Secondly, CD8+ cells are important immune effector cells for the control of early viraemia in primate lentivirus infection, as highlighted by at least two studies (Jin et al., 1999 ; Schmitz et al., 1999) that reported that depletion of CD8+ cells during infection led to uncontrolled viraemia. There is convincing evidence that IFN-β can boost the production and activity of CD8+ memory cells, including cells displaying specific cytotoxic activity against HIV-expressing cells, during primary and chronic HIV infection. In this respect, Tough et al. (1996) have shown that, during the primary response to viruses, type I IFN acts as adjuvant and augments both the intensity of the response and the survival of early memory cells. Furthermore, during chronic infection, the survival of pre-existing memory CD8+ T cells does not require continuous T cell receptor contact with specific antigen but simply intermittent contact with type I IFN released during intercurrent virus infections (Tough et al., 1996). Recently, Zhang et al. (1998) have demonstrated that this property is mediated by IL-15. IFN-β is not only able to maintain the survival of the memory CD8+ cells, but has been demonstrated to enhance HIV-specific CTL activity mediated by RANTES via the chemokine receptor CCR3 (Hadida et al., 1999). Finally, IFN-β triggers the release of IFN-γ and IL-12, essential for controlling HIV replication (Clerici & Shearer, 1994 ; Vieillard et al., 1997). We have therefore transduced both CD4+ and CD8+ cells, because of the benefits provided by the presence of the transgene in both cell types. Because of the well-known inhibitory effects of IFN on cell proliferation and haematopoiesis, as reviewed in De Maeyer & De Maeyer (1988), we looked into the possible effects of the presence of IFN-β-transduced cells. We have assessed previously that macaque lymphocytes that constitutively express small amounts of IFN-β in vitro (3–10 U IFN-β per 10% cells per 2 days) were not impaired in their proliferative potential and replicated to the same extent as untransduced lymphocytes (Matheux et al., 1999). Furthermore, the control macaque C3, characterized by the presence of about 100 U\ml IFN-β in the circulation, did not display any impairment of haematological values before SIV infection, indicating that this level of IFN-β production is compatible with normal haematopoiesis. In summary, of the five infected macaques, the three characterized by continuous IFN-β production (IFN1, IFN2 and C3) were more resistant to SIV infection. We are well aware that the number of animals used and the number of transduced lymphocytes were low and believe that further exploration of an IFN-β-based anti-HIV gene therapy is warranted before reaching definite conclusions. To this purpose, we are currently directing our efforts towards the construction of high-titre vectors, with the aim of increasing the proportion of vectortransduced HIV target cells.

Anti-SIV gene therapy with IFN-β E. Lauret and F. Matheux contributed equally to this work. This work was supported by the Centre National de la Recherche Scientifique, the Institut Curie, the Agence Nationale de Recherches sur le SIDA (ANRS, Paris, France), the Commissariat a' lhEnergie Atomique (CEA, Paris, France), the Centre de Recherches du Service de Sante! des Arme! es (CRSSA, La Tronche, France) and ‘ Tous ensemble contre le SIDAh (SIDACTION, Paris, France). Franck Matheux was the recipient of a fellowship from the Association Française contre les Myopathies (AFM, Evry, France). We thank Sophie Peyramaure for expert technical assistance, Philippe Caufour and Olivier Neildez for their help during the study, Juana Wietzerbin for providing us with antibodies to Hu-IFN-γ and Ion Gresser, William Vainchenker and Jean Louis Virelizier for critically reading the manuscript.

infectivity by progeny virus from alpha interferon-treated human immunodeficiency virus type 1-infected T cells is associated with defective assembly of envelope gp120. Journal of Virology 66, 7543–7548. Jin, X., Bauer, D. E., Tuttleton, S. E., Lewin, S., Gettie, A., Blanchard, J., Irwin, C. E., Safrit, J. T., Mittler, J., Weinberger, L., Kostrikis, L. G., Zhang, L., Perelson, A. S. & Ho, D. D. (1999). Dramatic rise in plasma

viremia after CD8+ T cell depletion in simian immunodeficiency virusinfected macaques. Journal of Experimental Medicine 189, 991–998. Kornbluth, R. S., Oh, P. S., Munis, J. R., Cleveland, P. H. & Richman, D. D. (1990). The role of interferons in the control of HIV replication in

macrophages. Clinical Immunology & Immunopathology 54, 200–219. Lauret, E., Vieillard, V., Rousseau, V., De Maeyer-Guignard, J. & De Maeyer, E. (1994). Exploring interferon beta for gene therapy of HIV

infection. Research in Immunology 145, 674–677.

References Baca-Regen, L., Heinzinger, N., Stevenson, M. & Gendelman, H. E. (1994). Alpha interferon-induced antiretroviral activities : restriction of

viral nucleic acid synthesis and progeny virion production in human immunodeficiency virus type 1-infected monocytes. Journal of Virology 68, 7559–7565. Baltimore, D. (1988). Gene therapy. Intracellular immunization. Nature 335, 395–396. Belardelli, F. (1995). Role of interferons and other cytokines in the regulation of the immune response. APMIS 103, 161–179. Cheret, A., Le Grand, R., Caufour, P., Dereuddre-Bosquet, N., Matheux, F., Neildez, O., Theodoro, F., Maestrali, N., Benveniste, O., Vaslin, B. & Dormont, D. (1996). Cytokine mRNA expression in mononuclear

cells from different tissues during acute SIVmac251 infection of macaques. AIDS Research and Human Retroviruses 12, 1263–1272. Clerici, M. & Shearer, G. M. (1994). The Th1–Th2 hypothesis of HIV infection : new insights. Immunology Today 15, 575–581. Coccia, E. M., Krust, B. & Hovanessian, A. G. (1994). Specific inhibition of viral protein synthesis in HIV-infected cells in response to interferon treatment. Journal of Biological Chemistry 269, 23087–23094. Cremer, I., Vieillard, V. & De Maeyer, E. (2000). Retrovirally mediated IFN-β transduction of macrophages induces resistance to HIV, correlated with up-regulation of RANTES production and down-regulation of C-C chemokine receptor-5 expression. Journal of Immunology 164, 1582–1587. Daniel, M. D., Letvin, N. L., King, N. W., Kannagi, M., Sehgal, P. K., Hunt, R. D., Kanki, P. J., Essex, M. & Desrosiers, R. C. (1985). Isolation

of T-cell tropic HTLV-III-like retrovirus from macaques. Science 228, 1201–1204. De Maeyer, E. & De Maeyer-Guignard, J. (1988). Interferons and Other Regulatory Cytokines. New York : John Wiley. Desrosiers, R. C. (1990). The simian immunodeficiency viruses. Annual Review of Immunology 8, 557–578. Erhart, M. A., Piller, K. & Weaver, S. (1989). Polymorphism and gene conversion in mouse α-globin haplotypes. Genetics 115, 511–519. Fall, L. S., MhBika, J. P., Le Coq, H., Fouchard, M., Astgen, A., Bizzini,

Le Grand, R., Clayette, P., Noack, O., Vaslin, B., Theodoro, F., Michel, G., Roques, P. & Dormont, D. (1994). An animal model for antilentiviral

therapy : effect of zidovudine on viral load during acute infection after exposure of macaques to simian immunodeficiency virus. AIDS Research and Human Retroviruses 10, 1279–1287. Lehner, T., Wang, Y., Cranage, M., Bergmeier, L. A., Mitchell, E., Tao, L., Hall, G., Dennis, M., Cook, N., Brookes, R., Klavinskis, L., Jones, I., Doyle, C. & Ward, R. (1996). Protective mucosal immunity elicited by

targeted iliac lymph node immunization with a subunit SIV envelope and core vaccine in macaques. Nature Medicine 2, 767–775. Matheux, F., Le Grand, R., Rousseau, V., De Maeyer, E., Dormont, D. & Lauret, E. (1999). Macaque lymphocytes transduced by a con-

stitutively expressed interferon beta gene display an enhanced resistance to SIVmac251 infection. Human Gene Therapy 10, 429–440. Michael, N. L. (1999). Host genetic influences on HIV-1 pathogenesis. Current Opinion in Immunology 11, 466–474. Minagawa, T., Mizuno, K., Hirano, S., Asano, M., Numata, A., Kohanawa, M., Nakane, A., Hachimori, K., Tamagawa, S., Negishi, M. and others (1989). Detection of high levels of immunoreactive human

beta-1 interferon in sera from HIV-infected patients. Life Sciences 45, iii–vii. Pantaleo, G. (1997). How immune-based interventions can change HIV therapy. Nature Medicine 3, 483–486. Piscitelli, S. C., Minor, J. R., Saville, M. W. & Davey, R. T., Jr (1996).

Immune-based therapies for treatment of HIV infection. Annals of Pharmacotherapy 30, 62–76. Poli, G., Orenstein, J. M., Kinter, A., Folks, T. M. & Fauci, A. S. (1989).

Interferon-alpha but not AZT suppresses HIV expression in chronically infected cell lines. Science 244, 575–577. Schmitz, J. E., Kuroda, M. J., Santra, S., Sasseville, V. G., Simon, M. A., Lifton, M. A., Racz, P., Tenner-Racz, K., Dalesandro, M., Scallon, B. J., Ghrayeb, J., Forman, M. A., Montefiori, D. C., Rieber, E. P., Letvin, N. L. & Reimann, K. A. (1999). Control of viremia in simian immuno-

interferon beta after gene transduction enhances human immunodeficiency virus type 1-specific cytotoxic T lymphocyte activity by a RANTES-dependent mechanism. Human Gene Therapy 10, 1803–1810.

deficiency virus infection by CD8j lymphocytes. Science 283, 857–860. Shirazi, Y. & Pitha, P. M. (1993). Interferon alpha-mediated inhibition of human immunodeficiency virus type 1 provirus synthesis in T-cells. Virology 193, 303–312. Tough, D. F., Borrow, P. & Sprent, J. (1996). Induction of bystander T cell proliferation by viruses and type I interferon in vivo. Science 272, 1947–1950. Vieillard, V., Lauret, E., Rousseau, V. & De Maeyer, E. (1994). Blocking of retroviral infection at a step prior to reverse transcription in cells transformed to constitutively express interferon beta. Proceedings of the National Academy of Sciences, USA 91, 2689–2693.

Hansen, B. D., Nara, P. L., Maheshwari, R. K., Sidhu, G. S., Bernbaum, J. G., Hoekzema, D., Meltzer, M. S. & Gendelman, H. E. (1992). Loss of

Vieillard, V., Lauret, E., Maguer, V., Jacomet, C., Rozenbaum, W., Gazzolo, L. & De Maeyer, E. (1995). Autocrine interferon-beta synthesis

B., Gringeri, A., Santagostino, E., Burny, A., Zagury, D. & Chams, C. (1995). Biological effect of active anti-IFN-α immunization in HIV-

infected patients. Biomedicine & Pharmacotherapy 49, 422–428. Hadida, F., De Maeyer, E., Cremer, I., Autran, B., Baggiolini, M., Debre! , P. & Vieillard, V. (1999). Acquired constitutive expression of

CHEJ

F. Matheux and others for gene therapy of HIV infection : increased resistance to HIV-1 in lymphocytes from healthy and HIV-infected individuals. AIDS 9, 1221–1228. Vieillard, V., Cremer, I., Lauret, E., Rozenbaum, W., Debre, P., Autran, B. & De Maeyer, E. (1997). Interferon beta transduction of peripheral

blood lymphocytes from HIV-infected donors increases Th1-type cytokine production and improves the proliferative response to recall antigens. Proceedings of the National Academy of Sciences, USA 94, 11595–11600. Vieillard, V., Jouveshomme, S., Leflour, N., Jean-Pierre, E. Debre, P., De Maeyer, E. & Autran, B. (1999). Transfer of human CD4+ T

lymphocytes producing beta interferon in Hu-PBL-SCID mice controls human immunodeficiency virus infection. Journal of Virology 73, 10281– 10288.

CHFA

Wang, Y., Tao, L., Mitchell, E., Bogers, W. M. J. M., Doyle, C., Bravery, C. A., Bergmeier, L. A., Kelly, C. G., Heeney, J. L. & Lehner, T. (1998).

Generation of CD8 suppressor factor and β chemokines, induced by xenogeneic immunization, in the prevention of simian immunodeficiency virus infection in macaques. Proceedings of the National Academy of Sciences, USA 95, 5223–5228. Zhang, L., He, T., Huang, Y., Chen, Z., Guo, Y., Wu, S., Kunstman, K. J., Brown, R. C., Phair, J. P., Neumann, A. U., Ho, D. D. & Wolinsky, S. M. (1998). Chemokine coreceptor usage by diverse primary isolates of

human immunodeficiency virus type 1. Journal of Virology 72, 9307–9312.

Received 4 April 2000 ; Accepted 24 July 2000