ECENTLY several laboratories have discovered genes or cDNAs encoding proteins that are 35-55 % identical to conventional actin isoforms. The initial discoveries, largely serendipitous, were unexpected, as for some time it had been thought that actin was a unique protein.These actin-related proteins apparently play profound roles in cells. Those discovered to date have been shown to be essential or important in yeasts (Lees-Miller et al., 1992a;Schwob and Martin, 1992;Muhua et al., 1994;Harata et al., 1994), and one is known to copolymerize with actin in vivo (Schafer et al., 1994) to form a short, discrete microfilament. The emerging consensus is that there are only a few such proteins, and that they are highly conserved in all eukaryotes, from protozoa to yeasts to mammals.Functional studies of the actin-related proteins are proceeding rapidly. It was suspected from the time of their initial discovery that actin-relatod proteins would have distinct cellular roles, for while their similarities to actins are striking, the novel proteins are much less abundant, and in many instances have peptide insertions never found within actin isoforms. Indeed, the best characterized actin-related protein, referred to by various authors as actin-RPV, centractin, Saccharomyces cerevisiae Act5 protein, and S. cerevisiae Act3 protein (Lees-Miller et al., 1992b; Meyer, 1992, 1994;Muhua et al., 1994), has been found in vertebrates to form a short filament that facilitates movemonts of vesicles on microtubules by cytoplasmic dynein (Lees-Miller et al., 1992b;Paschal et al., 1993;Schafer et al., 1994), and in yeast and filamentous fungi to implement mitotic spindle orientation and nuclear migration (Muhua et al., 1994; Plamann et al., 1994). In Acanthamoeba, two different actin-related proteins are found in a complex, four members of which bind to profilinSepharose (Machesky ot al., 1994). The importance of these actin relatives now seems well established, but much remains to be learned.In the course of analyzing primary sequences of actinrelated proteins from several sources we have noted that Address all correspondence to E. Fyrberg, Department of Biology, Johns Hopkins University, 144 Mudd Hall, 3400 N. Charles Street, Baltimore, MD 21218-2685. Tel.: (410) 516-5213. three families are widely distributed. As might be expected, very similar, and in some cases identical, proteins have been given a variety of names. Both the random naming and the fact that actin-related proteins are not designated so as to differentiate them from conventional actins have created confusion within the literature. Therefore, we feel that it is essential to devise a more unifying nomenclature for actinrelated proteins, and also to point out the apparent subsets of equivalent proteins defined by us independently using phylogenetic tree analysis programs, such as CLUSTAL V.We propose designating actin-related proteins with the Arp (actin-related protein) prefix. Of the currently known families, the one that is most similar to conventional actin (•4...