Early appearance of simian immunodeficiency virus (SIV) antigen and antibodies as variables in evaluating antiviral drugs in macaques

43

Journal of Virological Methods, 37 (1992) 43-54 0 1992 Elsevier Science Publishers B.V. / All rights reserved / Ol66-0934/92/$05.00 VIRMET 01293

Early appearance of simian immunodefkiency virus (SIV) antigen and antibodies as variables in evaluating antiviral drugs in macaques E. Ljungdahl-Stihle”,

‘, J. Trojnarb,

B. Gberg” and B. Wahren”,

’

“Department of Virology, National Bacteriological Laboratory and Karolinska Institute, Stockholm (Sweden), bFerring AB, Malmii (Sweden), and ‘Department of Virology, Karolinska Institute, Stockholm (Sweden) (Accepted

IO October

1991)

Summary SIV infection in macaques has become an important animal model for HIV- 1 infection in humans. An antibody assay was therefore developed and compared to a commercially available antigen assay with respect to their usefulness to monitor the course of simian immunodeficiency virus (SIV) infection in cynomolgus monkeys. A peptide, JB6T, consisting of 21 amino acids with the sequence NSWGCAFRQVCHTTVPWVNDS corresponding to a segment in the env protein of human immunodeficiency virus (HIV) type 2 was used as antigen in an enzyme-linked immunosorbent assay (ELISA). JB6T was found to detect IgG and IgM antibodies to viral antigens with high specificity. The earliest anti-SIV IgM antibodies were detected at days 13-l 6, with a maximum at day 20 and subsequently the levels fell. Specific IgG antibody levels increased at day 16-20 after SIV infection and reached a plateau at day 60. The commercially available HIV-l p24/26 antigen test could, due to crossreactivity, be employed to detect SIV antigen delay, peak and duration. Antiviral drugs; gp36/41; IgG/IgM; Monkeys; Peptides; SIV

Correspondence to: E. Ljungdahl-Stihle, Department of Virology, National Bacteriological Laboratory and Karolinska Institute, S-105 21 Stockholm, Sweden.

44

Introduction In Mucaca fascicularis the simian immunodeficiency virus induces a clinical disease which in several aspects resembles human HIV infection (McClure et al., 1989; Putkonen et al., 1989; Lundgren et al., 1990). This virus strain gives immunodeficiency disease and opportunistic infections, changes in Tlymphocyte subsets, humoral immune response and virus-specific antigenemia. Due to the fact that the clinical symptoms in SIV-infected monkeys do not occur until around one year after the primary infection, early laboratory signs such as antigen and antibody appearance may be useful to detect the infection quantitatively. Also methods to measure the antibodies and antigens may be used to evaluate the effect of antiviral compounds in SIV infection. SIV is genetically related to HIV-l but more closely to HIV-2. This has been shown both by serological assays and nucleic acid hybridization studies (Clavel et al., 1986a,b). The conservation of amino acid sequences between SIV/sM and HIV-2 in the gag gene is around 84%, in env it is 76% (Hirsch et al., 1989a). The envelope proteins of HIV-1 are synthesized as a 160-kDa molecule which is subsequently cleaved to glycoproteins 120 and 41. Immunoassays in which whole virus lysates are used as antigens have poor specificity due to partial cross-reactivity against conserved core antigens in HIV-l and HIV-2 (Kanki et al., 1987). To avoid this we developed an immunoassay using a synthetic peptide antigen from a conserved site (Myers et al., 1990) that has the advantage of higher specificity. Peptides around this region were previously proven highly immunoreactive with sera from patients infected with HIV-l (Gnann et al., 1987a; Smith et al., 1987). With HIV-2 peptides from this region HIV-2-infected persons displayed a strong reactivity (Gnann et al., 1987b; Norrby et al., 1987; Broliden et al., 1991). We have separately reported the application of the antibody ELISA, described in the present study, to quantify antiviral efficiency of a series of compounds (Lundgren et al., 1990, 1991).

Materials and Methods Virus SIV/sM, originally isolated from asymptomatic sooty mangabeys (Cercocebus atys) (Fultz et al., 1986) was a gift from Drs. H. McClure and P. Fultz at the Yerkes Regional Primate Research Center, Atlanta, GA (NIH Grant No. RR-00165). A stock of virus was prepared and stored at - 70°C. Antiviral drugs 3’-Azido-3’-deoxythymidine roughs Wellcome, Dartford,

(zidovudine, AZT) was obtained from Bur(FLT) from U.K., 3’-fluoro-3’-deoxythymidine

45

Medivir AB, Huddinge, Sweden and phosphonoformic acid (foscarnet, PFA) from Astra Lakemedel AB, Sodertalje, Sweden. The compounds were dissolved in sterile PBS buffer and stored at +4”C. Animals Macaca fascicularis monkeys were and 8 h later they were infected (Lundgren et al., 1990). Blood samples days 2,4,6,8, 10, 13, 16,20,30,41,49 post-infection.

pretreated with an antiviral compound with 10-100 monkey infectious doses were drawn before any treatment and at and on some animals also around day 60

Virus cultivation

Human peripheral blood lymphocytes (PBL) were separated on IsopaqueFicoll (Pharmacia, Uppsala, Sweden) from fresh, heparinized blood of HIVseronegative healthy individuals. PBL were stimulated with 1 fig/ml of phytohemagglutinin (PHA, Burroughs Wellcome, Dartford, U.K.) for 4 days at a concentration of 1 x lo6 cells/ml of RPMI-1640 (Gibco, Grand Island, NY) medium containing 10% fetal calf serum, 10% interleukin-2 (IL-2) (Amgen Biologicals, Amersham International, U.K.) and penicillin/streptomycin at + 37°C with 5% COZ in humidified air. After centrifugation to remove the PHA, 40 x lo6 lymphocytes were infected with 2 x lo4 reverse transcriptase units of SIV/ SM and cultured in the presence of 2.25 pg/ml of polyprene for 18 days. The culture medium was changed every third day and the cell concentration was adjusted to approximately 106/ml. Formation of syncytia was observed before reverse transcriptase activity became measurable. The cell-free virus pools were frozen at -70°C in aliquots and one new tube was used for each experiment. Virus isolation

Peripheral blood mononuclear cells (PBMC) from the infected monkeys were prepared from heparinized whole blood by density gradient centrifugation on Isopaque-Ficoll (Pharmacia, Uppsala, Sweden) and cocultivated with PHAstimulated human PBMC from two normal blood donors (Albert et al., 1987). Fresh human PBMC were added to the cultures every week. The cultures were maintained in RPMI-1640 supplemented with 10% fetal calf serum (Flow), 5 units/ml recombinant IL-2 (Amersham), 2 pg/ml polyprene (Sigma, St. Louis, MO) and antibiotics. Every 3-4 days the cultures were examined by light microscopy for cytopathic effect and cell culture supernatants were harvested for HIV antigen assay and reverse transcriptase assay. Virus-negative cultures were maintained 4 wk before they were discarded.

46

Peptide synthesis Peptides, 1 l-23 amino acids long, representing the envelope and transmembrane regions of HIV-2 ROD (Myers et al., 1990) were synthesized by the solid-phase method on an automated peptide synthesizer, Applied Biosystem 430 A. They were cleaved from the resins by hydrogen fluoride, chemically cyclized and purified by high-performance liquid chromatography. The peptide JB6T, which corresponds to a conserved region of the transmembrane glycoprotein and has 18 of its 21 amino acids in common with these endogenous in SIV/ so, was regularly used (Table 1). Antibody assay Appearance of antibodies in blood was tested in an enzyme linked immunosorbent assay (ELISA). Microwell plates (Nunc Immunoplate, Nunc, Odense, Denmark) were coated overnight at room temperature with 100 ~1 of a 20+g/ml solution of the synthetic peptides. After coating, the plates were washed 5 times in 0.9% sodium chloride containing 0.05% Tween 20. Hundred ,~l serum diluted 1:20 in phosphate-buffered saline (PBS) with 0.5% bovine serum albumine (BSA), 0.05% Tween 20, 2% goat serum and 0.01% merthiolate were added to each well and allowed to incubate for 20 min at + 37°C. The plates were washed and 100 ,ul affinity-purified goat antihuman IgG conjugated to alkaline phosphatase (Sigma) diluted 1:600 were added. After incubation for 20 min the plates were washed. Hundred ~1 substrate (pnitrophenyl-phosphate, Sigma), 1 mg/ml, in 1 M diethanolamine buffer (pH 9.8) and 0.5 M magnesium chloride were added and the color reaction was developed at + 37°C. The reaction was stopped after 20 min by adding 100 ,~l of 1 M sodium hydroxide to each well. Absorbance was measured at 405 nm (A 405) using a Dynatech Microplate Reader MR 600 (Dynatech Laboratories, Alexandria, VA). The optimal dilution 1:20 was selected. Appearance of IgM antibodies was measured in 10 animals using an affinitypurified goat antihuman IgM conjugated to alkaline phosphatase (Sigma) diluted 1: 1500 and with the same method as above. One series of samples was in addition assayed in a commercial kit (Abbott Recombinant HIV-l/HIV-2 EIA, Abbott, Chicago, IL) using recombinant proteins produced by E. coli representing HIV-l core and envelope and HIV-2 envelope antigens. Antigen assay SIV p24/26 antigen was determined using cross-reactivity of antibodies to SIV with HIV-l in a commercial kit (HIVAG-I, Abbott). The assay was modified from the manufacturer’s description. The plasma samples were not heat inactivated. Triton X-100 was added to samples at a final concentration of 0.5% in order to make more antigens available (Zhang et al., 1988). One series of samples was also assayed in another commercially available kit (HIV Ag

JB14 JB16 JB61 JB15 JB64 JB6 JB6T JB62 JB65

RVTAIEKYLQDQARLNSWGC YLQDQARLNSWGCAFRQVC YLQDQARLNSWGCAFRQVCHW NSWGCAFRQVC NSWGCAFRQVCHTTVPWVN NSWGCAFRQVCHITVPWVNDS NSWGCAFRQVCHTTVPWVNDS SWGCAFRQVCHTTV AFRQVCHTTVPWVN

H2N-RVTAIEKYLQDQARLNSWGCAFRQVCHTTVPWVNDS-COOH

HIV-2 ROD

gp36

H2N-RVTAIEKYLKDQAQLNSWGCAFRQVCHTTVPWPNES-COOH

SIV/SM

Reactivity to HIV-2 peptides

TABLE I

:.: 1:9 2.0

Nonspecific Specific

Specific SpeciJic

Nonspecific Nonspecific Nonspecific Nonspecific Specific

0.0 1.8 1.8 1.8 2.0

0.0 1.4 1.3 1.7 0.5 0.2 0.1 1.0 0.3

Reactivity

Absorbance values in ELISA of serum from infected / uninfected monkeys

48

Assay, Coulter, Hialeah, FL) utilizing a murine monoclonal antibody to HIV core antigen. Using this assay, we could detect SIV/sM antigens in the tissue culture virus stock but not in 8 different sera from infected monkeys. Goldstein et al. (1990) have reported detection of antigens in serum by the Coulter system from one rhesus macaque out of two infected with SIV/sM. Statistics Data are presented as mean values. Difference between antigen or antibody concentration was assessed by the Mann-Whitney U-test. Cut-off values were calculated as the mean of the negative samples plus 3 standard deviations.

Results Selection of a peptide for reactivity with anti-SW

IgG antibodies

Thirteen peptides being subsequences (sometimes modified by cyclization) of transmembrane and envelope sequences of HIV-l and HIV-2 were assayed as antigens in solid-phase ELISA in order to find the peptide with the highest sensitivity to sera from SIV-infected animals and the lowest background to sera from uninfected animals. Nine peptides (covering amino acids 577-612) corresponded to the transmembrane region of HIV-2 (Table 1). One peptide, JBBT, of the 9 HIV2-related peptides tested showed both high values (Adas > 1.8) when tested with seropositive monkey sera and A4a5 levels below 0.1 when tested with sera from noninfected animals. Four peptides represented sequences from gag and gp120 (data not shown).

days post infection Fig. 1. Mean antibody

response

in JB6T ELISA of 18 SIV/ sM-infected compounds.

monkeys

not treated

with antiviral

49

These peptides have previously been used to detect antibodies in both HIV-l and HIV-2 infected persons (Broliden et al., 1991). They gave moderate AdO5 values with SIV-infected monkey sera but 3 of them gave even higher values with negative sera from noninfected animals. JB6T was therefore chosen for further serological assays. IgG and IgM antibody responses in SW-infected monkeys Sera from 18 monkeys infected with SIV/ sM, but without antiviral treatment, were analysed in the JB6T-ELISA. The antibody levels increased during the second week after infection, were high at day 20 and reached a maximum at day 60 (Fig. 1). The earliest IgG responders had detectable specific IgG 16 days after infection. Antibody determination did not significantly separate the control group from the groups treated with zidovudine or FLT (Fig. 2). The samples were also assayed in a recombinant HIV-l/HIV-2 EIA (Abbott) in which they were found not to react with HIV-l core or envelope, or with HIV-2 envelope antigens in that kit. Samples from 9 monkeys were analysed in the JB6T peptide ELISA in order to detect IgM antibodies. In Fig. 3 the kinetics of specific IgM (Fig. 3A) and IgG (Fig. 3B) antibodies of 3 animals treated with foscarnet in concentrations of 3 x 65 mg/kg/day and 6 untreated animals are compared. IgM was demonstrable in samples from the untreated animals at days 13-16. The peak levels of IgM appeared at day 20 and then the levels fell until day 60 after 2-

E

1.5-

C 3 8 C

l-

2 i 9

0.5-

0-w 0

10

! 20

I 30

I 40

50

days post infection treatment Fig. 2. Comparison between mean SIV IgG antibody values of infected monkeys treated with FLT 3 x 5 mg/kg/day (0, n=4), zidovudine 3 x 5 mg/kg/day (A, n=4) and saline-ireated controls (- - -, n =4). Antiviral compounds were administered for 10 days.

50

infection. At this time-point the maximum level of IgG was attained. Sera from monkeys treated with foscarnet were shown to have a delayed IgM and IgG response of around 3 days compared with the animals without antiviral treatment (P= 0.05) (Fig. 3A,B). Detection of viral replication by antigen assay The mean serum antigen content of 19 SW-infected monkeys not treated with an antiviral drug is shown in Fig. 4. SIV antigen was detectable in all sera. The levels rose at day 8 to a peak at day 10 after infection and all values were

0

IO

20

30

40

50

60

70

50

60

70

days post infection

0

lo

20

30

40

days post infection Fig. 3. Nine SW/s,-infected monkeys were analysed for (A) IgM and (B) IgG responses. Three animals treated with foscarnet 3 x 65 mg/kg/day (A) and 6 animals treated with saline (0).

51

‘1

\

0

\ ‘9,_,_-a--**__-e

w** 0

\

I IO

1 20

, 30

40

50

I 60

days post infection Fig. 4. Mean SIV antigen

values in sera from 19 SW/ sM-infected

monkeys

not treated with antiviral

drugs.

again below the cut-off value at day 30. Isolation attempts were performed in 10 animals. At day 6 one of the 10 animals had detectable virus, at day 10 none and at day 20 all 10 animals.

Discussion It is an important issue to identify primates infected with SIV. Not only for research for AIDS on animal models, but also at primate centers a method for screening large numbers of primates is needed. In this paper we report an ELISA with a peptide which can be synthesized in large quantities in order to make standardized assays. Nine peptides corresponding to glycoprotein 36 of HIV-2 were synthesized ranging over amino acids 577-612. Three of them - JB6, JB6T and JB65 showed high absorbance values when they were tested with seropositive monkey sera and low values when tested with sera from uninfected animals. The peptides JB6 and JB6T both contain an amino-acid sequence, WGCAFR, which has previously been proposed to be essential in the reaction of HIV-2positive sera to the peptide (Norrby et al., 1989). The present findings support this proposal. Although a small synthesized peptide does not possess conformational properties, the two peptides were chemically cyclized in order to equal the native protein (Smith et al., 1987; Johnson et al., 1988). The two cysteine residues (env amino acid 603 + 609 in HIV-l) have been shown to be essential for antigenicity in an HIV-l peptide ELISA and are also conserved in HIV-2 (Gnann et al., 1987b). Another peptide, JB64, lacks the last two amino

52

acids, aspartic acid and serine, and had a decreased ability to bind antibodies. When tested with monkey sera, none of the HIV-l peptides were sufficiently specific to merit further examination. The IgG antibody levels of SIV-infected monkeys were elevated at day 20 after infection. In humans Gaines et al. (1988) have reported the appearance of IgG antibodies at 7-15 days after onset of the disease. It is, however, difficult to compare the experimental primary infection measured as serum antibody response after inoculation of a laboratory animal with the response in human primary infection relative to onset of the disease. Our antibody assay has repeatedly been used in studies on the antiviral effects of a number of compounds in more than 200 SIV-infected monkeys and all animals have developed IgG antibody response. The viral replication was measurable as antigen appearance in blood at day 8 in untreated animals, had a maximum at day 10 and disappeared around day 20 after infection. However, it can not be excluded that the antigen assay used here had a too low sensitivity to detect a small persisting antigen concentration. It has been reported that the reverse transcriptase activity could be detected earliest at day 30 in one isolate of SIV/~M (Hirsch et al., 1989b). Thus, viral antigen could be measured at least a few days before viral isolation was feasible. The present model of monitoring SIV antibody and antigen responses appears relevant and reproducible in order to characterize a primary infection by IgM and IgG responses and to estimate delayed or decreased viral production.

References Albert, J., Gaines, H., Siinnerborg, A., Nystrom, G., Pehrson, P.O., Chiodi, F., von Sydow, M., Moberg, L., Lidman, K. and Christensson, B. (1987) Isolation of human immunodeticiency virus (HIV) from plasma during primary HIV infection. J. Med. Virol. 23, 67-73. Broliden, P.-A., RudCn, U., Ouattara, AS., de The, G., Solver, E., Trojnar, J. and Wahren, B. (1991) Specific synthetic peptides for detection of and discrimination between HIV-l and HIV-2 infection. J AIDS 4, 952-958. Clavel, F., Guttard, D., Brun-V&net, F., Chamaret, S., Rey, M.-A., Santos- Ferreira, M.O., Laurent, A.G., Dauguet, C., Katlama, C. and Rouzioux, C. (1986a) Isolation of a new human retrovirus from West African patients with AIDS. Science 233, 343-346. Clavel, F., Guyader, M., Guetard, D., Salle, M., Montagnier, L. and Alizon, M. (1986b) Molecular cloning and polymorphism of the human immune deficiency virus type 2. Nature 324, 691695. Fultz, P.N., McClure, H.M., Anderson, D.C., Swenson, R.B., Anand, R. and Srinivasand, A. (1986) Isolation of a T-lymphotropic retrovirus from naturally infected sooty mangabey monkeys (Cercocebus atys). Proc. Natl. Acad. Sci. USA 83, 52865290. Gaines, H., von Sydow, M,, Parry, J.V., Forsgren, M., Pehrson, P.O., Siinnerborg, A., Mortimer, P.P. and Strannegard, 0. (1988) Detection of immunoglobulin M antibody in primary human immunodeliciency virus infection. AIDS 2, 1 I-15. Gnann, J.W.J., Schwimmbeck, P.L., Nelson, J.A., Truax, A.B. and Oldstone, M.B.A. (1987a) Diagnosis of AIDS by using a 12-amino acid peptide representing an immunodominant epitope of the human immunodeliciency virus. J. Infect. Dis. 156, 261-267.

53 Gnann, J.W.J., McCormick, J.B., Mitchell, S., Nelson, J.A. and Oldstone, M.B.A. (1987b) Synthetic peptide immunoassay distinguishes HIV type 1 and HIV type 2 infections. Science 237, 1346 1349. Goldstein, S., Engle, R., Olmsted, R.A., Hirsch, V.M. and Johnson, P.R. (1990) Detection of SIV antigens by HIV-l antigen capture immunoassays. J. AIDS 3, 988102. Hirsch, V.M., Olmsted, R.A., Murphey-Corb, M., Purcell, R.H. and Johnson, P.R. (1989a) An African primate lentivirus (SIV/sm) closely related to HIV-2. Nature 339, 389-392. Hirsch, V.M., Dapolito, G., McGann, C., Olmsted, R.A., Purcell, R.H. and Johnson, P.R. (1989b) Molecular cloning of SIV from Sooty Mangabey monkeys. J. Med. Primatol. 18, 2799285. Johnson, P.R., Parks, D.E., Norrby, E., Lerner, D.A., Purcell, R.H. and Chanock, R.M. (1988) Site-directed ELISA identities a highly antigenic region of the simian immunodeficiency virus transmembrane glycoprotein. AIDS Res. Hum. Retrovir. 4, 159-164. Kanki, P.J., M’Boup, S., Ricard, D., Barin, F., Denis, F., Boye, C., Sangare, L., Travers, K., Albaum, M. and Marlink, R. (1987) Human T-lymphotropic virus type 4 and the human immunodeficiency virus in West Africa. Science 236, 827-831. Lundgren, B., Ljungdahl-Stable, E., B&tiger, D., Benthin, R., Hedstrom, K.-G., Norrby, E., Putkonen, P., Wahren, B. and bberg, B. (1990) Acute infection of Cynomolgus monkeys with simian immunodellciency virus (SIVsm) as a model to evaluate antiviral compounds. Effects of 3’-azido, 3’-deoxythymidine, Foscarnet and 2’,3’-dideoxycytidine. Antivir. Chem. Chemother. l(5), 299-306. Lundgren, B., B&tiger, D., Ljungdahl-Stable, E., Norrby, E., Stihle, L., Wahren, B. and Gberg, B. (1991) Antiviral effects of 3’-fluorothymidine and 3’-azidothymidine in Cynomolgus monkeys infected with simian immunodeticiency virus. J. AIDS 4, 489498. McClure, H.M., Anderson, DC., Fultz, P.N., Ansari, A.A., Lockwood, E. and Brodie, A. (1989) Spectrum of disease in Macaque monkeys chronically infected with SIV/SMM. Vet. Imm. Immunopath. 21, 13-24. Myers, G., Josephs, S., Rabson, A., Smith, T. and Wong-Stahl, F. (Eds.) (1990) Database for Human Retroviruses and AIDS. Los Alamos Natl. Lab. Los Alamos, NM. Norrby, E., Biberfeld, G., Chiodi, F., von Gegerfeldt, A., Naucler, A., Parks, E. and Lerner, R. (1987) Discrimination between antibodies to HIV and to related retroviruses using site-directed serology. Nature 329, 248-250. Norrby, E., Chiodi, F., Whalley, A., Parks, E., Naucler, A., Costa, C.M., Thorstensson, R. and Biberfeld, G. (1989) Site-directed enzyme-linked immunosorbent assay with a synthetic simian immunodeticiency virus SIVmac peptide identifying antibodies against the HIV-2 transmembrane glycoprotein. AIDS 3, 17-20. Putkonen, P., Warstedt, K., Thorstensson, R., Benthin, R., Albert, J., Lundgren, B., Gberg, B., Norrby, E. and Biberfeld, G. (1989) Experimental infection of Cynomolgus monkeys (Mucnca fusciculuris) with simian immunodeficiency virus (SIVsm). J. AIDS 2, 359-365. Smith, R., Naso, R.B., Rosen, J., Whalley, A., Horn, Y.-L., Hoey, K., Kennedy, C.J., McCutchan, J.A., Spector, S.A. and Richman, D.D. (1987) Antibody to a synthetic oligopeptide in subjects at risk for human immunodeflciency virus infection. J. Clin. Microbial. 25, 1498-1504. Zhang, J., Martin, L.N., Watson, E.A., Montelaro, R.C., West, M., Epstein, L. and Murphey-Corb, M. (1988) Simian immunodeliciency virus/delta-induced immunodeficiency disease in Rhesus monkeys: relation of antibody response and antigenemia. J. Inf. Dis. 158, 1277-1286.