Human enteroviruses and rhinoviruses rapidly and selectively abolish translation from cellular mRNA upon infection of susceptible cells. Expression of the poliovirus 2A protease (PV 2Apro) is sufficient to cause host translation shutoff through cleavage of elF4G (formerly p220, elF4 gamma) either directly or indirectly through activation of a cellular factor. Evidence exists for both direct and indirect cleavage mechanisms; however, factors presumed to participate in an indirect mechanism have not yet been purified or defined. Here we show that the dominant elF4G cleavage activity in lysates from infected HeLa cells was separable from PV 2Apro by size exclusion chromatography. 2Apro separated into two peak fractions which contained activity which cleaved a peptide substrate derived from the poliovirus polyprotein. These peak 2Apro fractions did not cleave elF4G or an elF4G-derived peptide, as expected, due to the poor efficiency of direct cleavage reactions. Conversely, fractions which contained peak elF4G cleavage activity and only trace amounts of 2Apro efficiently cleaved a peptide substrate derived from the previously mapped elF4G cleavage site and also cleaved a peptide derived from the poliovirus 1D2A region. The dominant elF4G cleavage activity was highly purified through four chromatography steps and found to be devoid of all traces of 2Apro or its precursors. Quantitation of 2Apro from lysates of infected cells showed that during infections in HeLa cells, 2Apro does not reach molar excess over elF4G, as previously shown to be required for direct elF4G cleavage in vitro. Further, infection of HeLa cells in the presence of 2 mM guanidine-HCl, a potent inhibitor of viral RNA replication, suppressed accumulation of 2Apro and its precursor 2ABC below detectable levels but was unable to delay the onset of elF4G proteolysis in vivo. The elF4G cleavage activity was still easily detectable in in vitro assays using fractions from guanidine-treated cells. Thus, the data suggest that poliovirus utilizes two catalytic activities to ensure rapid cleavage of elF4G in vivo. Although it was not directly measurable here, 2Apro likely does cleave a portion of elF4G in cells. However, the data suggest that a cellular factor which can be activated by small quantities of 2Apro constitutes the bulk of the elF4G-specific cleavage activity in infected cells and is responsible for the rapid and efficient elF4G cleavage activity observed in vivo.