Journal of Virolog~cul Methods,29 (1990) 7 t-80 Elsevier VIRMET 01040

A refined complement-enhanced neutralization test for detecting antibodies to Junin virus Julia G. Barrera Ore’, Kelly T. McKee, Jr?, Joan Spisso’, Bill G. Mahlandt’ and Julio I. Maiztegui3 ‘The Salk Institute,Government Services Division, Swifhuater,Pennsylvania, U.S.A., 2 United States Army Medical Research instituteof Infectious Disease, Fort Detrick, Frederick, Maryland, U.S.A. and 31nstitutoNational de Estudios sobre VirosisHemorrdgicas, Pergamino, Argentina (Accepted

30 March 1990)

Summary A refined, comp~ement~~~ced, plaque-~duction n~u~~ization test was developed for measuring neutralizing antibodies against Junin (Argentine hemorrhagic fever) virus. The assay measured neutralizing antibodies after natural as well as vaccine-induced Junin virus infections. Among vaccinated individuals, titers were 2-4-fold higher than those obtained with conventional assays, without loss of specificity. Enhanced sensitivity was achieved by using a standardized complement source (vs human or animal serum) for virus dilution, incubation of virus-serum mixtures at 36°C for 2 h (vs overnight at 4°C) prior to plaque assay, control of age and density of cell monolayers, and variation in overlay conditions. Junin virus; Argentine hemorrhagic fever; Neutralization;

Complement; Arenavirus

Introduction The Arenaviridae are a family of segmented RNA viruses notable for their capacity to induce chronic infections in certain species of rodents. Four members of the arenavirus taxon ~~ph~ytic cho~~eningit~s, Lassa, Junin, and Machupo) are recognized human pathogens. Serological assays employed for diagnosis of arenavirus; infections such as complement fixation and immunofluorescence tests tend to be broadly cross-reactive among family members (Buchmeier and Oldstone, Cor~sFonde~ce to: K.T. McKee, Jr., Medical Division, USAMRIID, Fort Detrick, Frederick, MD 21701-5011. U.S.A. The views of the authors do not purport to reflect the views of the Department of the Army or the Department of Defense. 0166~0934/90/$03.50 0 1990 Elsevier Science Publishers B.V. (Biomedical Division)

1978; Howard, 1987; Jahrling and Peters, 1985). Plaque-reduction neutralization tests for arenaviruses developed in the late 1960s were shown to enhance both sensitivity and specificity of serological diagnosis (Webb et al., 1969). Recently, such neutralizing antibodies were shown to be directed primarily against surface glycoprotein epitopes (Damonte et al., 1986; Howard, 1987; Jahrling and Peters, 1985). Junin virus is the etiologic agent of a zoonotic disease known as Argentine hemorrhagic fever (AHF). Neutralizing antibodies are generally detectable in patients suffering from natural AHF by the second to third week after onset of symptoms, and may persist for life (Maiztegui, 1975; Damilano et al., Congreso Argentino de Virologia, 1983). The mechanism by which neutralizing antibodies protect in vivo is unclear (Howard, 1987). However, the beneficial effect of immune plasma, particularly that containing certain levels of neutralizing antibody, has been demonstrated clearly in patients with AHF treated within 8 days of disease onset (Enria et al., 1984; Enria et al., 1986; Maiztegui et al., 1979). Recently, a live, attenuated vaccine against AHF, Candid #l, was developed (Barrera Oro and Eddy, 4th International Conference on Comparative Virology, 1982). In the course of assessing the immunogenicity of Candid #l in humans, difficulties in detecting serological responses occasionally were encountered among volunteer recipients (Peters et al., 1987). This finding prompted a reassessment of the standard plaque-reduction neutralization assay being employed for serological measurement, and led to the development of a more sensitive but highly specific modification of the technique.

Materials and Methods Virus The vaccine strain of Junin virus, Candid #l, neutralization test. This attenuated clone was derived virus, and was passed twice in guinea pigs, 44 times in certified fetal rhesus lung cells in its development

was used for the sensitized from prototype XJ strain Junin in newborn mice, and 19 times as a candidate human vaccine,

Cells VERO cells at passage 140-160 were used for all experiments. Cells were grown in Eagle’s Minimal Essential Medium with Earle’s salts (EMEM), supplemented with non-essential amino acids (NEAA) and 5% heat-inactivated (56°C for 30 min) fetal calf serum (HIFCS). Serum specimens Sera examined here were derived from three sources: the first group (vaccine infections) was obtained from the initial North American human volunteers who

73

received the Candid #1 vaccine under controlled (unblinded) clinical conditions; the second group (vaccine infections) was composed of Argentine citizens who received Candid #1 or placebo during a double-blind, placebo-controlled, study; the third group (natural infections) was obtained from humans hospitalized at the Instituto National de Estudios sobre Virosis Hemorragicas, Pergamiuo, Argentina for suspected AI-IF. Serological tests

Conventional plaque-reduction neutralization tests employed a modification of the methods of Earley et al. (1967) and Webb et al. (1969). In brief, 4-fold dilutions (in Hanks balanced salt solution [ElrBSSJwith 1% 1 M HEPES and 50 1.18per ml of gentamicin) of heat-inactivated (56*C for 30 min) sera were applied to wells of a 96-well microtiter plate (75 r_llper well), beginning at a 1:3 dilution. A working suspension of virus was cmated by diluting Candid #l in HBSS with 1% 1 M HEPES and 50 fig per ml of gentamicin to which fresh guinea pig serum was added (10%). A virus dilution selected to yield 3@60 plaque forming units (PFU) per well in a 24-well cell culture plate was then added in an equal volume to each serum dilution. Controls consisted of equal volumes of app~p~ately diluted virus in serum diluent. After overnight incubation at 4”C, 50 ~1 of each virus-serum mixture was inoculated into duplicate wells of a 24-well tissue culture plate seeded 24 h previously with VERO cells at 105 cells per well. Serial 2-fold dilutions of virus control were inoculated simultaneously into 8-16 wells per dilution. Virus suspensions were adsorbed at room temperature for 1 h with periodic agitation, then overlaid with 0.5% agarose (Seakem ME, FMC Bioproducts, Rockland, ME) in HBSS with 1% 1 M HEPES, 2% HIFCS, and 50 pg per ml of gentamicin (1 ml per well). Incubation of plates at 36°C in a 5% CO, atmosphere for 5 days was followed by application of a second overlay consisting of 0.5% agarose in HBSS with 1% 1 M HEPES, 1% HIFCS, 50 pg per ml of ~gentamicin, and 1:10 000 [final co~cen~ation] of neutral red (1 ml per well). Twenty-four hours later, plaques were counted, and endpoint titers calculated at the 80% plaque reduction level, To perform the refined neutralization test, 4-fold dilutions of heat-inactivated sera for testing were added to wells of a 96”well microtiter plate, using the same serum diluent noted above, beginning at a I.:2 dilution. Working suspensions of virus for the sensitized test were created by diluting Candid #1 in serum diluent (90%) to which freeze-dried guinea pig complement (Low Tox, Cedarlane Laboratories, Homby, Ontario) was added (10%). After thorough mixing, test serum-virus suspensions and virus controls were incubated at 36°C for 2 h in 5% COs. Duplicate samples at each dilution, together with similarly diluted virus controls, were then adsorbed onto 2-day old VERO cell monolayers (24-well plates, seeded at 10’ cells per well) for 1 h at room temperature. An initial 1 ml overlay (0.5% agarose in HBSS with 1% HEPES, 2% HIFCS, and 50 pg per ml gent~i~in) was applied to each well, and plates were incubated at 36°C in 5% CO, for 4 days. A second 1 ml overlay was applied (5% agarose in HBSS with 1% 1 M HEPES, 1% HIFCS, 50 pg per ml gentamicin, and 1:lOOOO[final concentration] neutral

74

red), plates were incubated at 36°C in 5% CO2 for 24 h, and 80.% reduction of plaque counts relative to intratest controls were calculated. Statistical analysis Comparisons of serum antibody titers were made by using paired t-tests and repeated measures analysis of variance (ANOVA) techniques, with statistical analysis system software (SAS Institute, Cary, NC).

Results Sequential serum specimens from 53 North American Candid #l vaccinees were tested using the refined complement-enhanced neutralization assay (Fig. 1, panel A). Antibodies were detected at or near peak values by 1 month postvaccination, remained at relatively constant levels through the first 9 months of followup, then fell gradually over the ensuing year. Duplicate aliquots of the same sera previously had been assayed using the classical neu~~ization test (Fig. 1, panel B). While the kinetics of antibody production by the vaccinees appeared similar by the two techniques through the first 9 months of followup (a time period during which the number of duplicate samples available was sufficient to make legitimate comparisons), titers consistently were measured at serum dilutions 2-4fold higher using the refined technique (P 1:6) was undetectable by the conventional test in 12 individuals, while 10 were judged non-responders (same criteria) using the sensitized test. Concordance for non-response was noted in 9 cases; 1 seroconversion by conventional testing was not confirmed by the enhanced test, while 3 conversions occurred using the sensitized test which were not detected by the conventional technique. When responses among seroconverters only were considered, differences in geometric mean titers (GMT) were even more pronounced than those seen for the total sample (P co.03 for all months except month 7) (dashed lines in Fig. 1). Data represented in the figure were collected from volunteers vaccinated between October 1985 and June 1987. Conventional neutm~zation tests generally were performed contemporaneously with post-vaccination serology, months to years before repeat testing by the refined method. Sample availability precluded simultaneous retesting of the vast majority of these sera; however, a small number of specimens were of sufficient volume to allow for such a direct comparison (Table 1). Results of this study confirmed the previous findings, demonstrating .that the refined neu~aIization test offered improved sensitivity for antibody detect&m at low (sera #J23 and #J38) as well as high (serum #J40) levels. The absence of detectable antibody in serum from one vaccinee (#J22) by either method was reassuring with respect to specificity. Conduct of a blinded, placebo-controlled safety study with Candid #l afforded

sb 0

0123466786

10

MONTH

1‘

12

13

14

16

16

17

16

16

20

POST-VACCINATION

Fig. 1. Sequential neu~i~ng antibody titers for 53 North American volunteers receiving Candid #l Junin virus vaccine. Only those sera for which duplicate determinations by conventional and refined neutralization tests were run (see text) are displayed. Panel A: antibody titers &ten&red by refined complement-enhanced serum dilution plaque-reduction assay. Solid line depicts GMT for total sample; dashed line depicts GMT for seroconverters only. Panel B: antibody titers determined by conventional complement-enh~~d serum dilution plaque-reduction assay. Solid line depicts GMT for total sample; dashed line depicts GMT for seroconvetters only.

an opportunity to assess more directly the operational characteristics of the refined neutralization test. Among 82 participants in this study, 55 received vaccine and 27 received placebo. Neutralization testing of sera was performed and results recorded prior to study ~blin~ng. We were able to detect neutralizing antibodies at titers 1:4 or greater in 53155 vaccinees (96.4% sensitivity) during the first 3 months

TABLE 1 Comparison of conventional and refined neutralization tests for measuring Junin virus antibodies in North American volunteers immunized with Candid #l vaccine Serum

Antibody titer Conventional

522 523 535 536 537 538 539 540

A refined complement-enhanced neutralization test for detecting antibodies to Junin virus.

A refined, complement-enhanced, plaque-reduction neutralization test was developed for measuring neutralizing antibodies against Junin (Argentine hemo...
789KB Sizes 0 Downloads 0 Views