The concerted activity of four influenza virus proteins, PB1, PB2, PA, and NP is necessary and sufficient for transcription and replication of the viral genome in the nucleus of the cell. The three P proteins form a heterotrimeric complex in virions and the nuclei of infected cells. Biochemical analyses have shown specific interactions between PB1 and PA as well as PB1 and PB2, indicating that PB1 is the backbone of the complex. To identify domains of PB1 involved in binding PA, a two-hybrid system adapted for mammalian cells (CV-1) was implemented. First, we demonstrate the ability of PB1 and PA to interact efficiently and specifically in reciprocal combinations of two-hybrid reporter moieties, suggesting that transcription factor module fusion did not interfere sterically or allosterically with interaction between PB1 and PA. Subsequent analyses with a set of chimeric proteins with truncations of the PB1 C termini, N termini, or internal sequences led to the identification of a region at the N terminus of PB1 responsible for binding PA. Forty-eight amino acids at the N terminus of PB1 were sufficient for binding PA in vivo with the same efficiency as the complete PB1 protein. This region of PB1 responsible for binding PA does not overlap with other previously described PB1 functional domains involved in nuclear transport and RNA polymerization. We propose to name this region of interaction with PA domain alpha, to differentiate it from other functional domains described for PB1.