Sequence and structural homologies among type I and type II interferons

Abstract

Interferons are proteins with antiviral, antitumour and immunomodulator activities, which are secreted in response to various inducers¹. The interferons have been classified into two categories on the basis of their biological and physical properties. Type I interferons include fibroblast interferon (IFN-β)^2,3 and the leukocyte family of interferons which is composed of at least 10 subspecies^4–8. Each member of the type I interferons contains ∼165 amino acids, is acid-stable, and competes for the same receptors⁹; furthermore, identical amino acids occupy invariant positions in 23% of their amino acid sequences⁷. In contrast, type II interferon (IFN-γ) is produced in response to mitogens and antigenic stimuli¹⁰, contains ∼146 amino acid residues¹¹, is not acid-stable, and displays no measurable binding to type I interferon receptors⁹. The nucleotide sequence of the cDNA coding for IFN-γ has recently been reported¹¹, and no statistically significant sequence homology was detected between the deduced amino acid sequences of IFN-γ and any of the type I interferons^11,12. To determine whether there might be structural similarities between IFN-γ and the type I interferons, we have conducted a predictive analysis of the secondary structure of these proteins^13,14. IFN-γ as well as each of the type I interferons contain a segment with a high potential to form an amphiphilic α-helix of approximately the same length and hydrophobic/hydrophilic balance (Fig. 1). As we report here, the identification of these segments generates an alignment of sequences which reveals previously undetected sequence homologies among the type I and II interferons (Fig. 2). Thus, we propose that there is a common evolutionary ancestor for both type I and type II interferons.