Phylogenetic analysis of nucleoproteins suggests that human influenza A viruses emerged from a 19th-century avian ancestor.

The nucleoprotein (NP) is one of the main determinants of species specificity of influenza A viruses. Using 25 NP sequences we have constructed evolutionary trees by the strict-parsimony procedure of Fitch (197 1). In contrast to the evolutionarygene tree, the tree based on amino acid sequences unravels remarkable differences between avian and human NPs, differences which are best explained by a strong differential selection pressure on the human NPs. It is speculated that this selection pressure is caused by a change of the host and the (T-cell) immune response. A cautious extrapolation of the tree suggests that the human influenza A virus NPs evolved 150 years ago from an avian ancestor. Furthermore, the ancestral relation between the NPs of influenza A, B, and C viruses was analyzed. The influenza A virus has a genome consisting of eight RNA segments. Its NP gene seems to play the major role in host range: (1) Rescue of NP ts mutants of fowl plague virus (H7N1, FPV) in chicken cells is possible by avian but not by human H3N2 strains. However, the formation of FPV reassortants with the NP gene of the human Hong Kong virus (H3N2) can easily be achieved in dog kidney cells. These FPV reassortants with the Hong Kong NP do not multiply in chicken cells and are nonpathogenic for chickens (Scholtissek et al. 1978, 1985). Replacement of the other FPV genes by those of the human Hong Kong virus leads to FPV reassortants that replicate well in chicken cells (Scholtissek et al. 1976). Thus, by specific replacement of the NP gene the host range has been changed. (2) This view is strengthened by the observation that the sole replacement of the NP gene of a human influenza A strain by the NP gene of an avian virus is sufficient for an efficient attenuation of the human virus for monkeys (Snyder et al. 1987). In nature, avian influenza A viruses do not spread in the human population, and human viruses do not spread in birds. In the laboratory, however, ducks can become infected artificially by human H3N2 viruses. These viruses induce relatively high antibody titers in the birds, but the viruses are not excreted. In contrast, ducks infected with duck H3N2 viruses induce only very low antibody titers and excrete these viruses (Scholtissek et al. 1985), which in this way can spread in the duck population. The NP genes-and not other genesare responsible for keeping the two large separately evolving reservoirs of influenza A viruses-that in humans and that in water birds (Hinshaw and Webster 1982; Kawaoka et al. 1988)-apart. Therefore it was of interest to search for a common ancestor of these viruses. Furthermore, we wanted to know to what extent the NPs of influenza B and C viruses were related to the NP of influenza A viruses. A thorough comparison of sequences of the NPs of many strains should give an answer. The sequences of many NP genes of influenza A viruses have been determined (Winter and Fields 198 1; Huddleston and Brownlee 1982; van Rumpuy et al. 1982;