Microbial Pathogenesis 1990 ; 9 : 219-226
Protection of guinea pigs against experimental Argentine hemorrhagic fever by purified human IgG : importance of elimination of infected cells R . H . Kenyon,' R . M . Condie, 2 P . B . Jahrling' and C . J . Peters' 'Disease Assessment Division, Pathogenesis and Immunology Department, United States Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, MD 21702-5011, U.S .A . and 2 The Department of Surgery, University of Minnesota, Minneapolis, MN 55455, U.S .A . (Received 6 March, 1990 ; accepted in revised form 30 June, 1990)
Kenyon, R . H . (Disease Assessment Division, United States Medical Research Institute of Infectious Diseases, Frederick, MD 21701-5011, U .S .A.), R . Condie, P . B . Jahrling and C . J . Peters . Protection of guinea pigs against experimental Argentine hemorrhagic fever by purified human IgG : importance of elimination of infected cells . Microbial Pathogenesis 1990 ; 9 : 219226 . Antibody-containing plasma from patients recovered from Argentine hemorrhagic fever (AHF) is of proven value in treatment of the acute disease, but the possibility of transmitting bloodborne organisms such as HIV and hepatitis viruses detracts from this approach . Purified human immune plasma fractions IgG,, 2, 4, IgG,,2 .3, 4 and F(ab') 2 neutralized Junin virus in vitro . IgG,,2,3.4 and IgG,, 2, 4 lysed (in the presence of complement) cells infected with Junin virus, and protected infected guinea pigs from AHF . However, large quantities of the immune F(ab') 2 fraction from the same plasma pool failed to protect guinea pigs from death, to increase the mean time to death, and to diminish virus load in target organs of infected guinea pigs . This suggests that elimination of infected cells rather than virus neutralization may play a critical role in protection against Junin virus . Key words : Argentine hemorrhagic fever ; F(ab') 2; Junin virus .
Introduction Junin virus (JV) is an arenavirus causing Argentine hemorrhagic fever (AHF), an acute, severe disease with mortality rates of 15 to 30% in untreated cases . Arenaviruses bud from cell membranes and induce virus-specific cell-surface antigens ; hence, infected cells are susceptible to immune recognition, which can trigger responses resulting in cytolysis. At present, the only specific treatment for AHF is administration of whole, immune human plasma .' When immune plasma is given within 8 days of the first signs of human disease, mortality is reduced to less than 1 % . However, treatment after 8 days of disease is usually ineffective .' Individual units of plasma differ with respect to the concentration of antiviral antibodies, making uniformity of treatment difficult . This can be improved by assessing neutralizing antibody ; better standardization is attained by pooling units of immune plasma . Unfortunately, use of pooled immune plasma increases the risk of infection with adventitious, blood-borne agents . 0882-4010/90/100219+08 503 .00/0
© 1990 Academic Press Limited
22 0
R . H . Kenyon et al.
Introduction Junin virus (JV) is an arenavirus causing Argentine hemorrhagic fever (AHF), an acute, severe disease with mortality rates of 15 to 30% in untreated cases . Arenaviruses bud from cell membranes and induce virus-specific cell-surface antigens ; hence, infected cells are susceptible to immune recognition, which can trigger responses resulting in cytolysis . At present, the only specific treatment for AHF is administration of whole, immune human plasma .' .2 When immune plasma is given within 8 days of the first signs of human disease, mortality is reduced to less than 1% . However, treatment after 8 days of disease is usually ineffective .' Individual units of plasma differ with respect to the concentration of antiviral antibodies, making uniformity of treatment difficult . This can be improved by assessing neutralizing antibody ; better standardization is attained by pooling units of immune plasma . Unfortunately, use of pooled immune plasma increases the risk of infection with adventitious, blood-borne agents . Experimental infections of guinea pigs with virulent JV strains mimic the human disease .` Treatment of infected guinea pigs with immune guinea pig plasma is a good model for immunotherapy of the disease .' While early treatment of infected guinea pigs with modest amounts of antibody completely protects the infected animals, late treatment or treatment with low amounts of antibody cures the visceral (hemorrhagic) phase of the disease, but is followed by a late central nervous system (neurological) infection, with high viral titers in the brain, followed by death .' Preliminary experiments in our laboratory have shown that human immune plasma is effective in treating AHF in this guinea pig model . Jahrling' also showed that using guinea pigs is a reasonable alternative to using non-human primates for studies of the efficacy of human immune plasma on infection by Lassa virus, another arenavirus . We determined the feasibility of using purified fractions of immune human plasma for treating the disease in guinea pigs, as purification procedures decrease the risk of transferring adventitious agents .'- " We found that purified IgG,, 2, 4 and IgG,,2,3,4 fractions offered good protection from the disease in guinea pigs ; while the F(ab') 2 portion of the antibody molecules, although neutralizing virus in vitro, failed to protect in vivo .
Results The HPLC elution curves and the molecular masses for the intact IgG and for the F(ab') 2 fraction are shown in Fig . 1 . The HPLC analysis for the intact IgG showed one main peak at 158 kDa and a small dimer peak at 320 kDa . The HPLC analysis for the F(ab') 2 fraction showed one main peak at 96 kDa and no evidence of intact IgG . There was no detectable difference in the plaque-reduction neutralization (PRN) titers against Junin virus of the purified IgG,,2,3,4, IgG,, 2, 4 , and the F(ab') 2 fractions of immune plasma (Table 1) . To ensure full competence of these IgG,, 2 ,4 and IgG,,2,3 .4 molecules in binding complement and thereby effecting lysis of infected cells, we assayed 51 Cr release . The IgG,, 2 ,4 and IgG,,2,3,4 fractions, as well as unfractionated immune whole plasma, lysed JV-infected cells ; however, the F(ab') 2 fraction failed to lyse the infected cells . The half-lives (t„ 2 ) of the immune human F(ab') 2 and IgG,,2,3 .4 fractions in recipient guinea pigs were estimated (Fig . 2) after administration of 6000 therapeutic units (TU) of either IgG,,2,3,4 or F(ab') 2, given subcutaneously . Results indicated a t 12 of 1 .1 days for the F(ab') 2 and 7 .2 days for IgG,,2.3 .4 . The rapid decline in passive titer of the IgG,,2,3,4 after . day 10 was probably due to immune clearance . Immune clearance
Protection of guinea pigs by purified human IgG
221
(a) 0 .15
158
1
0.75 E c O
N a
320
(b) 96 Q
4
0
0 .5
1 .0
K cv
Fig . 1 . HPLC of human anti-Junin virus intact IgG (a) and of the F(ab') 2 fraction (b) . Arrows indicate the molecular masses (kDa) . K ., is the partition coefficient between the liquid and gel phases .
of the F(ab') 2 was probably not an issue as the F(ab') 2 was unlikely to be immunogenic (R . Condie, unpublished data) . We next looked at the ability of the immune fractions to protect guinea pigs infected with JV (Table 2) . Of those infected animals treated with unfractionated whole, human plasma, five of 10 survived, and the mean time to death (29 days post-infection) was significantly increased compared with that of untreated controls (15 days postinfection) . Some of the guinea pigs given one dose of the immune IgG,, 2, 4 or IgG,,2,3,4 also died ; but they too died late (day 23 or later), usually were paralysed, and virus was found almost exclusively in the brain (data not shown) . However, there were no survivors in the infected group of guinea pigs given F(ab') 2 at any of the experimental doses or times of treatment . In all four experimental F(ab') 2 groups, the guinea pigs died on approximately day 15 with a hemorrhagic-type disease . None of the animals in these groups was paralysed . There was no apparent difference in the viral distribution
Table 1 Junin virus plaque-reduction neutralization and cytolysis of Junin virus-infected Vero cells by human immune plasma fractions Fraction Whole plasma (immune) IgG,, 2 , 4 lgG, .2,3.4 F(ab') 2 Whole plasma (non-immune) a
Final dilution = 1 : 25 .
mg protein/ml 54.2 25 .5 25 .4 25 .5
1 /PRN titer
% 51 Cr release in presence of Aba+C
64 128 128 128 < 4
37 .7 37 .4 35.3 6 .2 4.5
222
R . H . Kenyon et al. 150
100
10
Protection of guinea pigs against experimental Argentine hemorrhagic fever by purified human IgG : importance of elimination of infected cells R . H . Kenyon,' R . M . Condie, 2 P . B . Jahrling' and C . J . Peters' 'Disease Assessment Division, Pathogenesis and Immunology Department, United States Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, MD 21702-5011, U.S .A . and 2 The Department of Surgery, University of Minnesota, Minneapolis, MN 55455, U.S .A . (Received 6 March, 1990 ; accepted in revised form 30 June, 1990)
Kenyon, R . H . (Disease Assessment Division, United States Medical Research Institute of Infectious Diseases, Frederick, MD 21701-5011, U .S .A.), R . Condie, P . B . Jahrling and C . J . Peters . Protection of guinea pigs against experimental Argentine hemorrhagic fever by purified human IgG : importance of elimination of infected cells . Microbial Pathogenesis 1990 ; 9 : 219226 . Antibody-containing plasma from patients recovered from Argentine hemorrhagic fever (AHF) is of proven value in treatment of the acute disease, but the possibility of transmitting bloodborne organisms such as HIV and hepatitis viruses detracts from this approach . Purified human immune plasma fractions IgG,, 2, 4, IgG,,2 .3, 4 and F(ab') 2 neutralized Junin virus in vitro . IgG,,2,3.4 and IgG,, 2, 4 lysed (in the presence of complement) cells infected with Junin virus, and protected infected guinea pigs from AHF . However, large quantities of the immune F(ab') 2 fraction from the same plasma pool failed to protect guinea pigs from death, to increase the mean time to death, and to diminish virus load in target organs of infected guinea pigs . This suggests that elimination of infected cells rather than virus neutralization may play a critical role in protection against Junin virus . Key words : Argentine hemorrhagic fever ; F(ab') 2; Junin virus .
Introduction Junin virus (JV) is an arenavirus causing Argentine hemorrhagic fever (AHF), an acute, severe disease with mortality rates of 15 to 30% in untreated cases . Arenaviruses bud from cell membranes and induce virus-specific cell-surface antigens ; hence, infected cells are susceptible to immune recognition, which can trigger responses resulting in cytolysis. At present, the only specific treatment for AHF is administration of whole, immune human plasma .' When immune plasma is given within 8 days of the first signs of human disease, mortality is reduced to less than 1 % . However, treatment after 8 days of disease is usually ineffective .' Individual units of plasma differ with respect to the concentration of antiviral antibodies, making uniformity of treatment difficult . This can be improved by assessing neutralizing antibody ; better standardization is attained by pooling units of immune plasma . Unfortunately, use of pooled immune plasma increases the risk of infection with adventitious, blood-borne agents . 0882-4010/90/100219+08 503 .00/0
© 1990 Academic Press Limited
22 0
R . H . Kenyon et al.
Introduction Junin virus (JV) is an arenavirus causing Argentine hemorrhagic fever (AHF), an acute, severe disease with mortality rates of 15 to 30% in untreated cases . Arenaviruses bud from cell membranes and induce virus-specific cell-surface antigens ; hence, infected cells are susceptible to immune recognition, which can trigger responses resulting in cytolysis . At present, the only specific treatment for AHF is administration of whole, immune human plasma .' .2 When immune plasma is given within 8 days of the first signs of human disease, mortality is reduced to less than 1% . However, treatment after 8 days of disease is usually ineffective .' Individual units of plasma differ with respect to the concentration of antiviral antibodies, making uniformity of treatment difficult . This can be improved by assessing neutralizing antibody ; better standardization is attained by pooling units of immune plasma . Unfortunately, use of pooled immune plasma increases the risk of infection with adventitious, blood-borne agents . Experimental infections of guinea pigs with virulent JV strains mimic the human disease .` Treatment of infected guinea pigs with immune guinea pig plasma is a good model for immunotherapy of the disease .' While early treatment of infected guinea pigs with modest amounts of antibody completely protects the infected animals, late treatment or treatment with low amounts of antibody cures the visceral (hemorrhagic) phase of the disease, but is followed by a late central nervous system (neurological) infection, with high viral titers in the brain, followed by death .' Preliminary experiments in our laboratory have shown that human immune plasma is effective in treating AHF in this guinea pig model . Jahrling' also showed that using guinea pigs is a reasonable alternative to using non-human primates for studies of the efficacy of human immune plasma on infection by Lassa virus, another arenavirus . We determined the feasibility of using purified fractions of immune human plasma for treating the disease in guinea pigs, as purification procedures decrease the risk of transferring adventitious agents .'- " We found that purified IgG,, 2, 4 and IgG,,2,3,4 fractions offered good protection from the disease in guinea pigs ; while the F(ab') 2 portion of the antibody molecules, although neutralizing virus in vitro, failed to protect in vivo .
Results The HPLC elution curves and the molecular masses for the intact IgG and for the F(ab') 2 fraction are shown in Fig . 1 . The HPLC analysis for the intact IgG showed one main peak at 158 kDa and a small dimer peak at 320 kDa . The HPLC analysis for the F(ab') 2 fraction showed one main peak at 96 kDa and no evidence of intact IgG . There was no detectable difference in the plaque-reduction neutralization (PRN) titers against Junin virus of the purified IgG,,2,3,4, IgG,, 2, 4 , and the F(ab') 2 fractions of immune plasma (Table 1) . To ensure full competence of these IgG,, 2 ,4 and IgG,,2,3 .4 molecules in binding complement and thereby effecting lysis of infected cells, we assayed 51 Cr release . The IgG,, 2 ,4 and IgG,,2,3,4 fractions, as well as unfractionated immune whole plasma, lysed JV-infected cells ; however, the F(ab') 2 fraction failed to lyse the infected cells . The half-lives (t„ 2 ) of the immune human F(ab') 2 and IgG,,2,3 .4 fractions in recipient guinea pigs were estimated (Fig . 2) after administration of 6000 therapeutic units (TU) of either IgG,,2,3,4 or F(ab') 2, given subcutaneously . Results indicated a t 12 of 1 .1 days for the F(ab') 2 and 7 .2 days for IgG,,2.3 .4 . The rapid decline in passive titer of the IgG,,2,3,4 after . day 10 was probably due to immune clearance . Immune clearance
Protection of guinea pigs by purified human IgG
221
(a) 0 .15
158
1
0.75 E c O
N a
320
(b) 96 Q
4
0
0 .5
1 .0
K cv
Fig . 1 . HPLC of human anti-Junin virus intact IgG (a) and of the F(ab') 2 fraction (b) . Arrows indicate the molecular masses (kDa) . K ., is the partition coefficient between the liquid and gel phases .
of the F(ab') 2 was probably not an issue as the F(ab') 2 was unlikely to be immunogenic (R . Condie, unpublished data) . We next looked at the ability of the immune fractions to protect guinea pigs infected with JV (Table 2) . Of those infected animals treated with unfractionated whole, human plasma, five of 10 survived, and the mean time to death (29 days post-infection) was significantly increased compared with that of untreated controls (15 days postinfection) . Some of the guinea pigs given one dose of the immune IgG,, 2, 4 or IgG,,2,3,4 also died ; but they too died late (day 23 or later), usually were paralysed, and virus was found almost exclusively in the brain (data not shown) . However, there were no survivors in the infected group of guinea pigs given F(ab') 2 at any of the experimental doses or times of treatment . In all four experimental F(ab') 2 groups, the guinea pigs died on approximately day 15 with a hemorrhagic-type disease . None of the animals in these groups was paralysed . There was no apparent difference in the viral distribution
Table 1 Junin virus plaque-reduction neutralization and cytolysis of Junin virus-infected Vero cells by human immune plasma fractions Fraction Whole plasma (immune) IgG,, 2 , 4 lgG, .2,3.4 F(ab') 2 Whole plasma (non-immune) a
Final dilution = 1 : 25 .
mg protein/ml 54.2 25 .5 25 .4 25 .5
1 /PRN titer
% 51 Cr release in presence of Aba+C
64 128 128 128 < 4
37 .7 37 .4 35.3 6 .2 4.5
222
R . H . Kenyon et al. 150
100
10
