Most mutations in the rBAT subunit of the heterodimeric cystine transporter rBAT-b(0,+)AT cause type I cystinuria. Trafficking of the transporter requires the intracellular assembly of the two subunits. Without its partner, rBAT, but not b(0,+)AT, is rapidly degraded. We analyzed the initial biogenesis of wild-type rBAT and type I cystinuria rBAT mutants. rBAT was degraded, at least in part, via the ERAD pathway. Assembly with b(0,+)AT within the endoplasmic reticulum (ER) blocked rBAT degradation and could be independent of the calnexin chaperone system. This system was, however, necessary for post-assembly maturation of the heterodimer. Without b(0,+)AT, wild-type and rBAT mutants were degraded with similar kinetics. In its presence, rBAT mutants showed strongly reduced (L89P) or no transport activity, failed to acquire complex N-glycosylation and to oligomerize, suggesting assembly and/or folding defects. Most of the transmembrane domain mutant L89P did not heterodimerize with b(0,+)AT and was degraded. However, the few [L89P]rBAT-b(0,+)AT heterodimers were stable, consistent with assembly, but not folding, defects. Mutants of the rBAT extracellular domain (T216M, R365W, M467K and M467T) efficiently assembled with b(0,+)AT but were subsequently degraded. Together with earlier results, the data suggest a two-step biogenesis model, with the early assembly of the subunits followed by folding of the rBAT extracellular domain. Defects on either of these steps lead to the type I cystinuria phenotype.