Neutralization of Residual Infectivity of Venezuelan Equine Encephalomyelitis Virus by Anti-gamma Globulin

Transcription

1 J. gen. Virol. 097o), 6, Printed in Great Britain 36I Neutralization of Residual Infectivity of Venezuelan Equine Encephalomyelitis Virus by Anti-gamma Globulin By N. HAHON Aerobiology and Evaluation Laboratories, Fort Detrick, Frederick, Maryland (Accepted 24 October I969) SUMMARY Residual infectivity detected after the interaction of Venezuelan equine encephalomyelitis virus with specific antiserum was caused mainly by the formation of infective virus +antibody complexes (sensitized virus) that could be neutralized by serum containing antigamma globulin (IgG). The quantities of virus sensitized by antiserum to Venezuelan equine encephalomyelitis and neutralized by anti-igg serum depended on the antibody concentrations of these sera. In contrast to the marked temperature and time-dependence of Venezuelan equine encephalomyelitis virus neutralization by antiserum, neutralization of sensitized virus by anti-igg serum was more rapid, being almost complete within I rain. at 35, and less sensitive to temperature. Virus sensitization preceded neutralization and indicated that infective virus+antibody complexes were formed before virus neutralization began. The neutralization of sensitized virus by anti-igg serum was generally species specific. Differences in the ability of anti-igg, anti-iga, and anti-igm sera to neutralize sensitized virus indicated that the reaction was also influenced by the class specificity of the anti-immunoglobulin. Sensitized virus was partially neutralized by goat antiserum to monovalent Fab fragments of human IgG and, to a lesser degree, by the Fc fragment. Sensitized virus was neutralized by an in vitro mixture of these fragments to almost the same degree as by goat antiserum to intact human IgG. The Fc fragment may, therefore, by involved in virus neutralization. INTRODUCTION Recent studies on the interaction of Venezuelan equine encephalomyelitis virus with antibody showed that neutralization reactions were initially first order and then followed more complex kinetics even in the presence of excess antiserum (Hahon, 1969). This behaviour, resulting in residual infectivity, has been reported with different animal virus systems (Dulbecco, Vogt & Strickland, i956; Thormar, I963; Lafferty, I963a; Granoff, I965). Several mechanisms have been proposed to account for this phenomenon : (i) heterogeneity of virus population (Dulbecco et al. I956); (ii) formation of infectious (amphoteric) virus antibody complexes which resist further neutralization by antibody (Bradish, Farley & Ferrier, 1962; Ashe & Notkins, I966); (iii) non-avidity of antibody (Lafferty, t963 a) ; (iv) dissociation of the virus + antibody complex (Fazekas de St. Groth, Watson & Reid, I958); and (v) pre-aggregation of virus in suspension (Wallis & Melnick, I967). Because there is experimental evidence for each of these mechanisms, the presence of residual infectivity may be due to several factors, acting either alone or together: such factors include the state ~4 vir 6

2 362 N. HAHON of virus, the quality of antibody, and the nature of the host-cell system used for infectivity assay. I report here an investigation of these problems in the neutralization of the residual infectivity of Venezuelan equine encephalomyelitis virus. METHODS Virus. The source and preparation of the TRINIDAD strain of Venezuelan equine encephalomyelitis virus was described by Hahon & Cooke (I967). The attenuated Venezuelan equine encephalomyelitis virus strain (Berge, Banks & Tigertt, I961) was obtained from Dr W. A. Hankins, Fort Detrick, as a suspension derived from infected cultures of chick fibroblasts. Both virus strains were assayed by an immunofluorescent cell-counting procedure (Hahon & Cooke, I967). Virus antisera. Antivirus sera were obtained from rhesus monkeys exposed I month earlier to an aerosol of about Iooo cell-infecting units (c.i.u.) of virus. Human antiserum to the virus was obtained from a convalescent patient. Sera were inactivated at 56o for 3o min. Cell lines. The McCoy cell line was used routinely for the assay of virus; its cultivation and maintenance were described by Hahon, (I969) Other established cell lines and their media for growth were hamster kidney and L-929 cells, Parker's medium I99 with lo ~o foetal calf serum; guinea pig lung cells, basal medium Eagle (BME) with IO % foetal calf serum and o'5 % lactalbumin hydrolysate; Chang's human conjunctiva cells, BME with lo % foetal calf serum. Hamster kidney and L-ga9 cells were maintained with medium I99 and 5 Yo foetal calf serum; guinea pig lung and human conjunctiva cells with BME and 5 % foetal calf serum. Kinetic neutralization. The conditions of neutralization varied, but generally a suspension containing virus (lo 8 c.i.u./ml.) was mixed with an equal volume of appropriately diluted monkey antivirus serum. Both reactants were warmed to the incubation temperature of 35 before mixing. At intervals during incubation, 2 ml. of the mixture was withdrawn from the reaction tube and placed immediately into a vial in ice-water. Test samples were then diluted in ice-cold phosphate-buffered saline ph 7.1, free of calcium and magnesium ions, and assayed for residual infectivity. A control mixture, consisting of virus and normal monkey serum of the same concentrations and volumes as the test mixtures, was included in each determination. The fraction of residual infectivity remaining after given times of incubation was determined from the ratio of infectivity in the reaction tube to infectivity in the control tube. Residual infectivity is shown on a logarithmic scale in the figures. Immunological reagents. Normal goat serum was obtained from Pentex Laboratories, Kankalee, Illinois. Goat anti-monkey gamma globulin (IgG) serum and goat antiserum to human, chicken, guinea pig, rabbit, bovine and mouse IgG were obtained from Microbiological Associates, Bethesda, Maryland. Goat antiserum to human albumin, immunoglobulins, and Fab (I and II) and Fc fragments were obtained from Hyland Laboratories, Los Angeles, California. Fragments were prepared by papain digestion of anti-igg serum. All sera were heated at 56o for 30 rain. and diluted in phosphate-buffered saline. Virus sensitization and assay. Virus sensitization refers to the interaction of virus with antivirus serum to form infective virus + antibody complexes (Notkins et al. 1966). Venezuelan equine encephalomyelitis virus was sensitized by mixing lo 8 c.i.u./ml, of virus with an equal volume of a I/5O dilution of monkey antivirus serum or normal monkey serum (control) and incubating at 35 for 1 hr. Sensitized virus, in contrast to initial virus, was susceptible to neutralization by goat anti-monkey IgG serum. To test for sensitization,

3 Neutralization of residual infectivity of VEE virus 363 treated virus was generally mixed with an equal volume of a I/5 dilution of goat anti-monkey IgG serum and incubated at 4 for Io min. A control mixture consisted of treated virus and normal goat serum. Appropriate dilutions of test samples were made in cold phosphatebuffered saline and assayed for residual infectivity. An increase of 0"3 log. units or greater in the depression of infectivity, calculated by subtracting the infectivity (log.) remaining after incubation with goat anti-monkey IgG serum from the infectivity (log.) of the control, indicated significant virus sensitization. 1.0 '~ Relative antibody concentration Fig. I. Neutralization of Venezuelan equine encephalomyelitis virus by monkey antiviral serum. RESULTS Direct neutralization of Venezuelan equine encephalomyelitis by antivirus sera Multiplicity curve Kinetic studies of Venezuelan equine encephalomyelitis virus neutralization by different antibody concentrations showed an initially constant rate of neutralization which then decreased to give a constant level of residual infectivity (Hahon, I969). The initial rate of neutralization was dependent on the antibody/,drus ratio and was most rapid at high antibody concentrations. This dependency was investigated further by determining the residual infectivity when different dilutions of antiserum were mixed with a constant quantity of virus 0o 8 c.i.u./ml.). After incubation at 35 fox" 2 hr, followed by 4 for I hr, reaction mixtures were diluted appropriately and assayed for residual infectivity. When the log. of residual infectivity was plotted against the relative concentration of antibody (Fig. I), the log. depression of infectivity was initially proportional to antibody concentration and then tended to a constant value at higher antibody concentrations. The constant level of residual infectivity that resisted neutralization at high antibody concentration was termed 'the persistent fraction' by Dulbecco et al. 0956). It was previously shown for Venezuelan equine encephalomyelitis virus that residual infectivity was not the consequence of the dissociation a4-o_

4 \. \ -\ 364 N. HAHON of non-infective virus + antibody complexes since neither dilution nor prolonged incubation resulted in any significant increase in assayable infectivity (Hahon, ~969). Factors influencing the level of residual infectivity To determine whether residual infectivity was due to a persistent virus fraction which constituted a genetically stable subpopulation, a residually infective sample was passed twice intracerebrally in mice. Parallel neutralization tests were then performed with the twice passed and with the original virus suspension using a I/5o dilution of monkey antivirus serum. A similar test was also made with an attenuated strain of Venezuelan equine encephalomyelitis virus. The original and passaged viruses showed almost identical neutralization curves (Fig. 2) and the rate of neutralization of attenuated virus was greater. Residual infectivity for the attenuated virus was less than that for the other viruses. Similar neutraliza- 1.0 o----o Residual infectivity after two mouse brain passages e---e Original virus > o i'%,! "',,',,!.-.x Attenuated virus -E_ '~ 0.1 I i x \ \x x X I I I I o Minutes Fig. 2. Neutralization of Venezuelan equine encephalomyelitis original, passaged, and attenuated viruses by monkey antiserum to original virus. tion tests with original or attenuated viruses mixed with monkey antiserum to attenuated virus gave comparable neutralization curves (Fig. 2). These experiments indicated that the resistance of Venezuelan equine encephalomyelitis virus to neutralization of infectivity, shown by a constant residual infectivity, was not due to the presence of a genetically stable subpopulation. Experiments with other virus, antibody and host-cell systems (Hashimoto & Prince, I963; Granoff, I965; Keller, I968 ) have also failed to demonstrate 'persistent fractions '. When suspensions with residual infectivity were incubated at 35 for 30 min. with additional antiserum of high concentration, there was no further depression of infectivity. Similarly, the level of residual infectivity was unaffected after incubation with RNase (2 #g./ml.). The latter, in agreement with findings with Newcastle disease virus (Granoff, ~965), indicated that residual infectivity was not caused by the presence of infective RNA.

5 Neutralization of residual infectivity of VEE virus 365 A report that residual infectivity was caused by aggregated virus (Wallis & Melnick, ]967) prompted an investigation of this factor. Kinetic neutralization tests were made with ~/to dilutions of monkey antivirus serum mixed with equal volumes of either unfiltered virus, sonically treated virus (20 kcyc./sec., 2 rain., Ultrasonic Probe BP-I, Blackstone Ultrasonics, Inc., Sheffield, Penns.), or virus filtered through 0.2 or o.[ #m. porosity membranes (Millipore Corp., Bedford, Mass.) by the procedure ofwallis & Melnick (]967). Appropriate mixtures of virus with normal serum served as controls. Reaction conditions were as before. Deviations from initial linearity (Fig. 3) first occurred with unfiltered and sonically treated viruses and resulted in similar depressions of infectivity. For filtered virus preparations the L ~ Unfiltered Wrus \\ ',~... e-... -o Sonically-treated virus o~ 0.01 ~,. x~.~..x... x Filtered (0-2/t) ",.. virus " o Filtered (0.1/~) virus i 5 i 10 I 15 t 20 t Minutes Fig. 3. Neutralization of Venezuelan equine encephalomyelitis virus with prior filtration or sonic treatment (2o kcyc./sec, for 2 rain.) of virus. linearity of the initial reaction continued longer and the residual infectivity approached a constant level which was lower than that for the unfiltered or sonically treated preparations. These observations indicate that pre-aggregation of Venezuelan equine encephalomyelitis virus was a factor controlling the level of residual infectivity but not, as will be shown, the major factor. The level of residual infectivity depends on the cell type employed for virus assay (Friedman, ~954; Kjell6n & Schlesinger, ~959)- Different host-cell systems were used to assay virus infectivity and to determine their effect on the level of residual infectivity. A kinetic neutralization test was made using a I/5o dilution of monkey antivirus serum; reaction samples taken at different incubation times were then assayed in parallel on five cell lines: McCoy, L-929, hamster kidney, Chang's human conjunctiva, and guinea-pig lung. The human conjunctiva cell line was included because it was reported to be less capable of supporting Venezuelan equine encephalomyelitis virus growth (Hardy, I959). The kinetic

6 366 N. HAHON neutralization curves (Fig. 4) were similar with all host-cell systems and the levels of residual infectivity differed by less than 5 %- Dulbecco et al. (1956) also failed to demonstrate any difference in the shape and slope of multiplicity curves with the limited cell types used for assay of infectivity. The use of other host cells for the assay of infectivity might, however, influence the level of residual infectivity. Neutralization of residual infectivity by anti-igg serum The residual infectivity found when lactic dehydrogenase or herpes simplex viruses were incubated with their antisera (Ashe & Notkins, I966; Notkins et al. 1966) was due to infectious virus+antibody complexes (sensitized virus) that could be neutralized by goat antimouse IgG serum. To extend this type of study to the Venezuelan equine encephalomyelitis g 7.0 ormal monkey _~Zu m ~ 6.0 > u.e "C3, Minutes "~ 4.0 u Neutralization of sensitized virus VEE monkey antiserum Total depression of infectivity I ~ 3.0 ~, ~. ~ Goat anti-monkey I ~ d e d ~ 2.01, -,,,,,,, ~ -]---?, Minutes Fig. 5 Fig. 4 Fig. 4. Levels of residual infectivity in neutralization of Venezuelan equine encephalomyelitis virus assayed on different cell lines. Fig. 5. Neutralization of Venezuelan equine encephalomyelitis virus by monkey antiserum and the neutralization of residual infectivity by goat anti-monkey IgG serum. Dilutions of 1/5o monkey antiviral or normal sera were mixed with equal volumes of lo 8 c.i.u./ml, of virus suspension and incubated at 35. Reaction mixtures were sampled at different times and assayed for residual infectivity. Samples were also mixed with equal volumes of i/5 dilutions of either goat anti-monkey IgG or normal goat sera, held in ice-water for io min., and assayed for depression of residual infectivity. virus + antibody system, dilutions of 1/5 monkey antivirus or normal sera were mixed with equal volumes of io s c.i.u./ml, of virus suspension and then incubated at 35. Reaction mixtures were sampled at different times and assayed for residual infectivity. Samples were also mixed with equal volumes of I/5 dilutions of either goat anti-monkey IgG serum or normal goat serum to test for sensitized virus as described earlier. The initial rate of neutralization (Fig. 5) was constant and then decreased to give a constant residual infectivity. The

7 Neutralization of residual infectivity of VEE virus 367 extent of virus neutralization by antiserum to Venezuelan equine encephalomyelitis was about ~ log. unit. While the addition of normal goat serum to virus + antiserum mixtures did not alter the kinetic curve, the addition of goat anti-monkey IgG serum reduced residual infectivity by more than 3 log. units. The total neutralization of virus infectivity was in excess of 4 log. units. Thus the residual infectivity was due to infective virus+ antibody complexes that could be neutralized by antibody directed against the antivirus globulin but not by that against the virus. Neutralization of sensitized virus" by anti-igg serum Effect of antibody concentration on virus sensitization This was first investigated by mixing equal volumes of different dilutions of monkey antivirus serum with 5 x io s c.i.u./ml, of virus and incubating the mixtures at 35 for I hr. Portions of the mixtures were assayed for residual infectivity; other portions were mixed with an equal volume of a I/tO dilution of goat anti-monkey IgG serum, incubated at 35 for ~ hr, and then assayed for residual infectivity. Normal monkey or goat sera served as controls. The I/IO dilution of antiserum to Venezuelan equine encephalomyelitis (Table I) Table i. Effect of antibody concentration on Venezuelan equine encephalomyelitis virus sensitization First incubation with VEE Second incubation with goat anti-monkey monkey antiserum* IgG serum Dilutions of c ~, c- ~ VEE monkey Infective Depression of infective Depression of Total depression antiserum u./ml, infectivityt u./ml, infectivity of infectivity log. log. log. log. log. i/lo o 4" /5o 6.8 I'4 3"9 3'7 5"I I/I00 7"2 PO 3'9 3"7 4'7 1/500 7"3 O'9 5"9 I'7 2"6 I/I000 7'4 0' " /5000 8"0 0'2 6'0 I'6 1"8 I / I o (control) 8-2 o.o 7"6 o'o o'o * Incubation with each phase was 35 for I hr. "]" Depression of infectivity was calculated by subtracting the log10 infectivity remaining after incubation with Venezuelan equine encephalomyelitis monkey antiserum or goat anti-monkey IgG serum from the log10 infectivity of controls (normal monkey or goat sera). showed a greater depression of infectivity (2.o log.) than did higher dilutions. The addition of goat anti-monkey lgg serum increased the depression of infectivity 3"7 log. units with both I/5o and I/Ioo dilutions of antiserum to Venezuelan equine encephalomyelitis. In the presence of higher Venezuelan equine encephalomyelitis antiserum dilutions 0/5oo to 7/5ooo), where little virus neutralization occurred, the depression of infectivity was increased by more than I'5 log. units by anti-igg serum. Virus sensitization, therefore, had also occurred with the high Venezuelan equine encephalomyelitis antiserum dilutions. With the ~/xo dilution of antivirus serum, anti-igg serum was inhibited in its neutralization of sensitized virus. That excess unbound IgG in the reaction mixture can prevent anti-igg from reacting with sensitized virus has been suggested as the cause of the observed inhibition (Ashe & Notkins, J966). In general, however, greater concentrations of antivirus serum sensitized more virus.

8 368 N. HAHON Effect of concentration of goat anti-monkey IgG Virus was first sensitized by mixing io 8 c.i.u./ml, of virus with an equal volume of I/5O dilution of ant/virus monkey serum and incubating at 35 for I hr. Equal volumes of sensitized virus suspension were then mixed with different dilutions of goat antiserum to monkey IgG or normal goat serum (control), incubated at 35 for I hr, and assayed for residual infectivity. The depression of residual infectivity due to neutralization of sensitized virus increased with the concentration of anti-igg serum (Table z). Table 2. Neutralization of sensitized Venezuelan equine encephalomyelitis virus by different concentrations of goat anti-monkey IgG antibody Infective Depression of Dilution of goat u./ml, infectivity anti-monkey IgG serum log. log. 1/2"5 2"4 4"3 1/5 3"I 3"6 I/I2'5 4'I 2.6 1/25 4" /5o 4"7 2.0 I/250 5"3 I-4 1/2-5 (normal goat serum) 6'7 o.o Table 3. Effect of time and temperature on the neutralization of sensitized Venezuelan equine encephalomyelitis virus by goat anti-monkey IgG serum 35 o A z. Infective Depression of Infective Depression of Time in u./ml, infectivity u./ml, infectivity min. log. log. log. log. i 4"4 2"3 4'0 2"5 3 4"5 2"2 4"2 2'3 5 4"4 2'3 4"3 2"2 I5 4'4 2"3 4"3 2'2 Control 6"7 o'o 6" 5 o.o Effect of time and temperature on the neutralization of sensitized Venezuelan equine encephalomyelitis virus by goat anti-monkey IgG serum Virus was sensitized as described earlier and then mixed with anti-igg serum. Mixtures were incubated at 35 or o, sampled at prescribed times, and then assayed for residual infectivity. Sensitized virus was neutralized by anti-igg serum almost completely within I min. at 35 but less rapidly at o (Table 3). Time relationship between neutralization and sensitization of Venezuelan equine encephalomyelitis virus by ant/viral serum A kinetic neutralization test was made in the usual manner with the same concentrations of reactants as in the preceding experiment. Reaction mixtures were sampled at different times and assayed for sensitized virus. Virus sensitization preceded virus neutralization (Table 4). During the 1"to 4 rain. interval of incubation, virus neutralization continued while the quantity of sensitized virus remained constant.

9 Cross-specificity Neutralization of residual infectivity of VEE virus 369 Tests were made on the capacity of goat antiserum to IgG of different animal species to neutralize virus that had been sensitized by monkey antiviral serum. Sensitized virus was mixed with different anti-igg sera, incubated at 35 for IO rain., and then assayed for residual infectivity. Significant neutralization of sensitized virus occurred only with goat anti-monkey serum and with goat anti-human IgG serum (Table 5). Antiserum to IgG of other animal species neutralized 0.2 log. units, or less of sensitized virus. Table 4. Sensitization preceding neutralization of Venezuelan equine encephalomyelitis virus by antibody Depression of infectivity Neutralization of Sensitization of Time in rain. original virus log. original virus log. o'5 o'3 4"3 r 0'4 4' "3 4 0"7 4'3 Normal monkey or goat serum o.o o.o Table 5. Neutralization of sensitized Venezuelan equine encephalomyelitis virus by goat anti-igg serum directed against different species Infective Depression of Goat serum u./ml, log. infectivity log. Anti-monkey IgG Anti-human IgG 3.1 5"7 3"6 I.o Anti-chicken IgG Anti-guinea pig IgG 6"7 6.6 o-o o.l Anti-rabbit IgG 6"5 0.2 Anti-bovine lgg Anti-mouse IgG 6"5 6"5 0-2 o.2 Normal 6"7 o-o To explore cross-neutralization further, virus sensitized with either monkey or human antivirus serum was reacted with goat anti-monkey or anti-human IgG serum. Sensitized virus was almost equally neutralized by homologous and heterologous anti-igg sera. The exception was virus sensitized with monkey antivirus serum and reacted with anti-human IgG serum: the depression of infectivity was z.o log. units less in this combination than in others. Neutralization of sensitized virus by different classes of immunoglobulins Virus was first sensitized with human antivirus serum and then reacted with different goat anti-human immunoglobulins. Most neutralization of sensitized virus was observed with antmgg and anti-iga sera (Table 6). Virus was not neutralized by either anti-lgm or antialbumin sera. These results are similar to those reported by Notkins et al. (I968) who noted neutralization of lactic dehydrogenase virus by anti-mouse IgF serum. An experiment was made to determine whether antiserum to monovalent fragments of IgG could neutralize sensitized virus. Virus was sensitized with human antivirus serum and

10 370 N. HAHON then incubated with goat antiserum to either Fab or Fc fragments of human IgG. A mixture of antisera to Fab and Fc fragments was also reacted with sensitized virus. Antibody to monovalent Fab[fragments could neutralize sensitized virus (Table 7). Antibody to the Fc fragment also neutralized sensitized virus, but about ten times less effectively than antibody to Fab fragments. The mixture of antisera to both Fab and Fc fragments neutralized almost as much virus as intact anti-igg serum. The Fc fragment, devoid of combining sites but containing structural features necessary for other biological functions (Fahey, I962), may play a role in virus neutralization. Table 6. Neutralization of sensitized (human antiserum) Venezuelan equine encephalomyelitis virus by different classes of goat anti-human immunoglobulins Infective Depression of Goat antiserum to u./ml, infectivity human immunoglobulin log. log. IgG 3"3 3"6 IgA 45 2"4 IgM 69 o.o Albumin 7o o.o Normal goat serum 69 o.o Table 7. Neutralization of sensitized (human antiserum) Venezuelan equine encephalomyelitis virus by goat antiserum to papain digested fragments of IgG Infective Depression of Goat antiserum to u./ml, infectivity human IgG fragments log. log. Fc 6q, , o-6 Fab 5q, 5 "I 1.6, I. 5 Fc+Fab 3"9, 4.0 2'8, 2'6 IgG 3'7, 4"o 3"o, 2'6 Normal goat serum 6"7, 6-6 o.o, o.o DISCUSSION These studies indicate that the residual infectivity detected when Venezuelan equine encephalomyelitis virus interacts with neutralizing antibody is partly the result of preaggregation of virus particles but is not related to heterogeneity of virus population, infective RNA, or to the host-cell systems here employed for virus assay. Dissociation of virus antibody complexes was not a factor because neutralized virus was not reactivated by dilution under physiological conditions (Hahon, 1969). The residual infectivity of Venezuelan equine encephalomyelitis virus was mainly due to infective virus+antibody complexes (sensitized virus) that could be neutralized by anti-igg antibodies which were inactive against the virus itself. Virus sensitized with Venezuelan equine encephalomyelitis monkey antiserum was further neutralized by goat anti-human IgG serum but not by anti-igg sera of different animal species. Sensitized virus formed with monkey or human antivirus serum was almost equally neutralized by homologous and heterologous anti-igg sera. The cross-reactivity is probably the consequence of a common primate antigenic determinant. In contrast to the marked temperature and time-dependence of Venezuelan equine encephalomyelitis virus neutralization by antiserum (Hahon, I969), neutralization of sensitized virus by anti-igg serum was rapid and less sensitive to temperature. Previous studies

11 Neutralization of residual infectivity of VEE virus 371 showed that sensitized lactic dehydrogenase (Notkins et al. 1968) and herpes simplex viruses (Ashe & Notkins, 1966 ) were neutralized by anti-lgg serum within I rain. and within 15 sec., respectively, at 4. Differences in temperature did not affect the reaction significantly. The formation of infective Venezuelan equine encephalomyelitis virus + antibody complexes preceded neutralization of infectivity as has been reported with other systems (Notkins et al. i968; Yoshino & Taniguchi, 1965). Aggregation of sensitized Venezuelan equine encephalomyelitis virus particles as a factor in the inhibition of virus neutralization is not supported by this or other studies (Ashe & Notkins, 1966; Notkins et al. 1968; Westaway, 1968). Antibodies to intact IgG, and to monovalent fragments of IgG, neutralized sensitized Venezuelan equine encephalomyelitis virus rapidly and effectively. It is relevant that monovalent Fab fragments of papain-digested antisera to influenza or herpes simplex viruses, which are not prone to aggregation, sensitized these viruses and reduced their neutralization by undigested antivirus sera (Lafferty, i963b; Notkins et ai. I968). The neutralization of low concentrations of sensitized Venezuelan equine encephalomyelitis virus by antmgg serum excluded lattice formation as a cause of inhibition of virus neutralization. Monovalent Fab fragments of goat anti-mouse antibody or goat anti-rabbit IgG antibody neutralize lactic dehydrogenase virus (Notkins et al. I968) and poliovirus (Keller, 1968 ). Lactic dehydrogenase virus was not neutralized by the Fc fragment. Although the Fc fragment does not contain antigen-binding sites and has not been specifically related to antigen-binding activities (Keller, 1966), preparations of the fragment do show virus neutralizing activity which may be due to contamination by other active fragments (Cremer et al. 1964). In this study on Venezuelan equine encephalomyelitis virus neutralization, sensitized virus was partially neutralized by goat antiserum to monovalent fragments of human IgG and, to a lesser degree, by the Fc fragment. Venezuelan equine encephalomyelitis virus sensitized with human antivirus serum was also neutralized by a mixture of goat antisera to Fab and Fc fragments of human IgG (Table 7) to almost the same degree as that neutralized by goat antiserum to intact human IgG. Contamination of the antiserum to the Fc fragment with anti-fab fragment antibodies does not satisfactorily explain the present findings. When depressions of infectivity of sensitized virus by antiserum to the Fab or the Fc fragments were totalled, the result was less than that due to a mixture of the antisera. Indirectly, these results suggest a role for the Fc fragment in virus neutralization. Observations by others of the inability of Fab fragments to establish and maintain neutralization of virus (Cremer et al. 1964; Kjell6n, 1965 ; Vogt et al. 1964), suggest that the Fc fragment may complete the structural configuration necessary for antibody to exert its full neutralizing activity. Effective virus neutralization, therefore, may not only be a function of the antigenbinding sites of the antibody but may also depend on the entire structure of the IgG molecule (Keller, 1966 ). These experiments support the hypothesis (Notkins et al. 1966, I968; Keller, 1968) that a secondary reaction of antivirus IgG with anti-igg is involved in the interaction of anti-igg with sensitized virus. Neutralization of infective virus + antibody complexes by anti-lgg is effected either by an increase in size of the complexes or by the formation of bridges that block infective sites on the virus particle. The inefficiency of monovalent Fab antibody fragments in blocking the reactive sites on the infective virus + antibody complex may be related to their smaller size or to the need for the Fc antibody fragment to complete an effective configuration on the complex. The excellent technical assistance of W. D. Zimmerman is gratefully acknowledged.

12 372 N. HAHON REFERENCES ASHE, W. K. & NOTK1NS, A. L. (1966). Neutralization of an infectious herpes simplex virus-antibody complex by anti-y-globulin. Proceedings of the National Academy of Science of the United States' of America 56, 447. BERGE, T. O., ~ANKS, t. S. & TIGERT:r, W. D. (I 96I). Attenuation of Venezuelan equine encephalomyelitis virus by in vitro cultivation in guinea-pig heart cells. American Journal of Hygiene 73, zo9. RRADlS., C. J., FARLEY, J. O. & terrier, n. E. Y. (I962). Studies on the nature of the neutralization reaction and the competition for neutralizing antibody between components of the virus system of foot-andmouth disease. Virology x8, 378. CREMER, N. E., RIGGS, J. L., FUJIMOTO, F. Y., HAGENS, S. 3., OTA, M. 1. & LENNETTE, E. H. (1964). Neutralizing activity of fragments obtained by papain digestion of viral antibody. Journal of Immunology 93, 283. DULBEC O, R., VOGT, M. & STRICKLAND, A. G. R. (1956). A study of the basic aspects of neutralization of two animal viruses, Western equine encephalitis and poliomyelitis virus. Virology 2, 162. rahey, J. L. (I962). Heterogeneity of y-globulins. Advances in Immunology 2, 4t. razekas DE ST GROT., S., WATSON, ~. S. & REID, A. r. (I958). The neutralization of animal viruses. L A model of virus-antibody interaction. Journal oflmmunology 8o, 215. FRIEDMAN, M. (I954). The effect of the host on the properties and sensitivity of a bacterial virus. Journal of Bacteriology 68, 274. GRANOFF, A. (1965). The interaction of Newcastle disease virus and neutralizing antibody. Virology 25, 38. HAHON, Y. (1969). The kinetics of neutralization of VEE virus by antiserum and the reversibility of the reaction. Journal of General Virology 4, 77. HAHON, Y. & COOKE, K. O. (1967). Primary virus-cell interactions in the immunofluorescent assay of Venezuelan equine encephalomyelitis virus. Journal of Virology i, 317. HARDY, r. M. (I959). The growth of Venezuelan equine encephalomyelitis virus in various tissue cultures. American Journal of Hygiene 7 o, 2 t. HASHIMOTO, N. & PRINCE, A. M. (I963). Kinetic studies on the neutralization reaction between Japanese encephalitis virus and antiserum. Virology 19, 261. KELLER, R. (1966). The stability of neutralization of poliovirus by native antibody and enzymatically derived fragments. Journal of Immunology 96, 96. KELLER, R. (I968). Studies on the mechanism of the enzymatic reactivation of antibody-neutralized poliovirus. Journal of Immunology loo, Io7 I. KJELL~N, L. (I964). Reactions between adenovirus antigens and papain digested rabbit immune globulin. Archiv fiir die gesamte Virusforschung i4, 189. KJELL~N, L. (1965). On the capacity of pepsin-digested antibody to neutralize adenovirus infectivity. Immunology 8, 557. KJELL~N, L. E. & SCHLESlNGER, R. W. (1959). Influence of host cell on residual infectivity of neutralized vesicular stomatitis virus. Virology 7, 236. LAFFERTY, K. J. (1963 a). The interaction between virus and antibody. I. Kinetic studies. Virology 2I, 6L LAFFERTY, K. J. (1963 b). The interaction between virus and antibody. II. Mechanism of the reaction. Virology 2I, 76. NOTKINS, A. L., MAGE, M., ASHE, W. K. & MAHAR, S. (I968). Neutralization of sensitized lactic dehydrogenase virus by anti-y-globulin. Journal of Immunology ioo, 3 I4. NOTKINS, A. L., MAHAR, S., SCIaEELV, C. & GOFVMAN, J. (1966). Infectious virus-antibody complex in the blood of chronically infected mice. Journal of Experimental Medicine 124, 8 I. THORMAR, H. (I963). Neutralization of visna virus by antisera from sheep. Journal of Immunology 9 o, 185. VOGT, A., KO~P, R., MAAS, G. & REICH, L. (I964). Poliovirus type t : Neutralization by papain-digested antibodies. Science, New York I45, I447. WALLIS, C. & MELNICK, J. L. (1967). Virus aggregation as the cause of the non-neutralizable persistent fraction. Journal of Virology i, 478. WESTAWAY, E. 6. (I968). Antibody responses in rabbits to the group B arbovirus Kunjin: Serologic activity of the fractionated immunoglobulins in homologous and heterologous reactions. Journal oflmmunology IOO, 569. YOSmNO, K. & TANIGUCHI, S. (1965). Studies on the neutralization of herpes simplex virus. III. Mechanism of the antibody-potentiating action of complement. Virology 26, 6I. (Received 5 May ~969)