Number and location of AUUUA motifs: role in regulating transiently expressed RNAs

Abstract

Abstract Many RNAs coding for either cytokines or oncogenes are unstable and have a short half-life (t1/2). The AUUUA motif is a highly conserved sequence and is repeated three or more times in the 3′ untranslated region (3′UTR) of RNAs encoding many of these short-lived cytokines and oncogenes. These sequences can confer instability. In this study, we investigated the role of number and location of AUUUA motifs in stabilization of RNA. We introduced 1xATTTA, 2xATTTA, ATTTTTTTA (second adenosine of 2xATTTA was substituted with a thymidine), 3xATTTA, 5xATTTA, 7xATTTA [AT-rich sequence from granulocyte-macrophage colony- stimulating factor [GM-CSF] gene (AT-62)], and GC-62 (GC sequences were substituted for ATTTA sequences in the 7xATTTA) into the 3′UTR of rabbit beta-globin (R beta G) gene. This construct also contained the neomycin-resistance gene. These expression vectors were transfected into human lung fibroblasts (W138), which constitutively expressed low levels of GM-CSF mRNA. Stable transfectants were selected by growth in G418. Northern blot analysis of actinomycin D-treated, stably transfected cells demonstrated that the number of AUUUA sequences correlated with rapidity of turnover of the chimeric R beta G mRNA. The rank order of stability was GC-62 = 1xATTTA = 2xATTTA (no RNA decay at 4 hours) > 3xATTTA = 5xATTTA (t1/2, 4 hours) > 7xATTTA (t1/2, 2 hours). Stability of mRNA of R beta G also was reduced (t1/2, 2 to 4 hours) when AT-62 was introduced into the second exon of R beta G gene. In these same cells, the t1/2 of GM-CSF RNA was approximately 10 to 15 minutes, suggesting that the AUUUA motifs cannot alone account for the rapid degradation of this cytokine mRNA. Phorbol diesters, including 12- 0-tetradecanoyl phorbol 13-acetate (TPA), stabilize a variety of transiently expressed RNAs, including GM-CSF RNA. We found that TPA markedly increased (> 30-fold) the accumulation of GM-CSF RNA. In contrast, TPA was unable to stimulate the levels of the chimeric R beta G when either 1x, 2x, 3x, or 5xATTTA motifs were fused to 3′UTR, or when either AT-62 or GC-62 control sequences were fused to the second exon. The chimeric beta-globin construct with either AT-62 or ATTTTTTTA in the 3′UTR had only an approximately twofold to threefold increase in accumulation.