In Friend murine erythroleukemia, the membrane glycoprotein (gp55) encoded by the spleen focus-forming virus (SFFV) env gene forms a disulfide-bonded dimer that binds to erythropoietin receptors (EpoR) to stimulate erythroblastosis. This mitogenic interaction is blocked in mice homozygous for the Fv-2r resistance allele and also in nonmurine species. Recently, we found that SFFV mutants with deletions in one domain (called ecotropic) of gp55 overcome Fv-2r resistance, suggesting that Fv-2r resistance is targeted against this domain. Thus, the SFFV glycoprotein structure optimal for activating EpoR depends on the mouse genetic background in a manner amenable to classic genetic analysis. Moreover, in contrast to gp55, these same partially deleted mutant glycoproteins activate human EpoR in a cell culture system, raising the possibility that Fv-2-related proteins may restrict SFFV not only in mice but also in other species. Evidence suggests that Fv-2 proteins may interact with EpoR as """"""""guardians"""""""", and that the efficiency of this defense depends on the structures of gp55, of polymorphic Fv-2 proteins, and of the species EpoR. We propose: (l) Determine whether SFFV mutants with deletions in the ecotropic domain cause erythroblastosis in nonmurine species, and identify the critical target sequence in gp55 for Fv-2-mediated host resistance. (2) One SFFV mutant (Pvu-delta) with such a deletion causes extensive erythroblastosis in both DBA/2 (Fv-2s) and in congenic D2.R16 (Fv-2r mice. However, Pvu- delta is surprisingly not pathogenic in several mouse strains that are susceptible to wild-type SFFV. By genetic crosses, identify mouse genes that restrict Pvu-delta but not wild-type SFFV. (3) It has been hypothesized that SFFV may have originated from a type of mouse retrovirus (called MCF) by a series of env mutations. Recently, we developed a means to isolate novel SFFVs. From a single mouse we isolated an unusual MCF that activates EpoR plus a series of intermediates in its evolution to form a new SFFV. Clone and sequence all of these viruses in order to definitively trace the SFFV lineage. Do SFFVs evolve differently in distinct genetic backgrounds? (4) Clone cDNAs for erythroblast proteins that bind to EpoR and determine their role(s) in controlling susceptibility to Friend erythroleukemia. (5) Determine whether EpoR-associated proteins contribute to human erythropoietic diseases. Friend erythroleukemia has been the principal model system for analyzing genetic control of cancer. Our background studies and understanding of the primary pathogenic mechanism now provides an opportunity to elucidate molecular mechanisms for host resistance to an oncoprotein and for regulation of hemopoietin receptors in normal and diseased animals.