SUMMARYAt physiological ionic strength, monoclonal antibody 35-1f4 has previously been shown to neutralize poliovirus type 1 by antibody-mediated polymerization. At low ionic strength, this antibody neutralized the virus by a hit-and-run mechanism: the virions were converted to non-infectious, empty capsids devoid of antibodies. These empty capsids resembled those formed by thermal denaturation of native polio virions in their sedimentation coefficient (80S), antigenicity (H) and isoelectric pH (6.3).Neutralizing antibodies appear to interact with poliovirus in different ways. Some monoclonal antibodies form stable complexes with single virions, whose infectivity is consequently reduced (Icenogle et al., 1983); our 35-1f4 antibody (Brioen et al., 1982) causes the virions to aggregate (Brioen et al., 1983), thereby lowering the infectivity of the virus preparation without irreversible damage to the virions (Thomas et al., 1985). In both mechanisms, the antibody remains bound to the (clusters of) neutralized virions. In the present paper, we show the existence of yet another mechanism of the hit-and-run type : the virion is disrupted to a noninfective, empty capsid and the antibody is released.The rate of neutralization of T-even phages has been shown to decrease with increasing ionic strength (Jerne & Skovsted, 1953;Cann & Clark, 1956). We are not aware of similar findings with animal viruses. As we will show in the present paper, the ionic strength influences not only the degree of poliovirus neutralization, but also the nature of the antibody-virion interaction.Nine solutions were prepared with three levels of ionic strength, the highest of which corresponded to physiological saline (Table 1). At each level, either Mg -'÷, Ca 2÷ or EDTA was present at a concentration of 0.1 mM. Radiolabelled poliovirus type 1 was reacted with monoclonal antibody 35-1f4 in each of these media. After incubation, the mixtures were centrifuged in order to pellet virus aggregates, but not individual virions or small complexes. At physiological ionic strength (Table 1, lines 1 to 3), more than 90~o of the virus radioactivity was pelleted as expected from previous results (Brioen et al., 1983;Thomas et al., 1985). When the ionic strength was reduced from p. = 0.13 to ~t = 0.002 (lines 4 to 6), the aggregates represented only about 50~o of the input when Ca 2+ or Mg 2÷ was present, and even less (25 ~) in the presence of EDTA. When the ionic strength was further reduced by omission of the basal salts mixture (lines 7 to 9), the sedimentable material was down to 19 to 27~ of the input radioactivity. The results clearly show that the ionic strength is a major factor determining the extent of virus aggregation. The additional influence of EDTA was only evident at I1 = 0-002.Neutralization experiments were carried out in PBS (150 mM-NaCI, 13 mM-KC1, 0.3 mMCaC12, 0.5 mM-MgC12, 10raM-sodium phosphate pH 7.3) or in 0.1 mM-EDTA pH 7.3. The residual infectivity was measured in both media, and the results are given in Table 2. Even though the vi...