Tethering KSRP, a decay-promoting AU-rich element-binding protein, to mRNAs elicits mRNA decay

Abstract

Inherently unstable mRNAs contain AU-rich elements (AREs) in their 3' untranslated regions that act as mRNA stability determinants by interacting with ARE-binding proteins (ARE-BPs). We have destabilized two mRNAs by fusing sequence-specific RNA-binding proteins to KSRP, a decay-promoting ARE-BP, in a tethering assay. These results support a model that KSRP recruits mRNA decay machinery/factors to elicit decay. The ability of tethered KSRP to elicit mRNA decay depends on functions of known mRNA decay enzymes. By targeting the Rev response element of human immunodeficiency virus type 1 by using Rev-KSRP fusion protein, we degraded viral mRNA, resulting in a dramatic reduction of viral replication. These results provide a foundation for the development of novel therapeutic strategies to inhibit specific gene expression in patients with acquired or hereditary diseases.

FIG. 1.

Tethering KSRP to a non-ARE-containing mRNA triggers mRNA decay. (A) Schematic diagram of constructs expressing human β-globin reporter mRNAs. Exons and introns are shown as boxes and lines, respectively. The fragment containing part of the coding region and 3′ UTR of GAPDH inserted is shown as a gray box. The six copies of the MS2 binding site are depicted as black boxes. (B and C) Northern blot analysis of the steady-state levels of reporter mRNAs in HeLa-TO cells expressing various effectors. Numbers below lanes represent the levels of GB-6bs normalized to that of GB-GAPDH. GB-6bs levels in cells transfected with pcDNA were set at 100%. Mean values with standard deviations (SDs) are shown. (D) Immunoblot analysis shows expression of transfected proteins by use of anti-FLAG (lanes 1 to 5) or anti-Xpress (lanes 6 to 9) antibody. (E) HeLa-TO cells were transfected with a construct expressing GB-6bs mRNA, under the control of a Tet-regulatory promoter, a construct constitutively expressing GB-GAPDH mRNA, and constructs expressing either MS2 or MS2-KSRP. The decay of GB-6bs mRNA was analyzed (left panel). Varied levels of GB-GAPDH mRNA are due to loading variations but not degradation of the mRNA. Signals of GB-6bs were quantitated by a phosphorimager, normalized to that of GB-GAPDH, and plotted as mean values ± SDs against time (right panel). (F) HeLa-TO cells were transfected with a reporter expressing GB-GAPDH mRNA, under the control of a Tet-regulatory promoter, and constructs expressing either MS2 or MS2-KSRP. The decay of GB-GAPDH mRNA and levels of endogenous β-actin mRNA were analyzed (left panel). Signals of GB-GAPDH were quantitated, normalized to β-actin mRNA levels, and plotted as mean values ± SDs against time (right panel). Dox, doxycycline.

FIG. 2.

The central KH motifs or the C-terminal domain of KSRP is capable of triggering mRNA decay. (A) Schematic diagram of KSRP and different fragments used to prepare expression vectors. The N and C termini are shown as open boxes, and the four KH motifs are depicted as gray boxes. Our KSRP clone lacks the first 46 amino acids (indicated as dashed lines). (B and C) Northern blot analysis of the steady-state levels of reporter mRNAs in HeLa-TO cells expressing various effectors (top panels). The levels of GB-6bs mRNA were quantified as described in the legend for Fig. 1. Mean values with standard deviations (SDs) are shown. Expression of transfected proteins analyzed by anti-FLAG or anti-Xpress (B, lane 6) immunoblotting is shown (bottom panels). (D and F) The decay of GB-6bs mRNA was analyzed in cells coexpressing (D) different KH motifs or (F) either the N or the C terminus of KSRP fused with the MS2 coat protein. Varied levels of GB-GAPDH mRNA are due to loading variations but not degradation of the mRNA. (E and G) Signals of GB-6bs shown in panels D and F (for panels E and G, respectively) were quantitated, normalized to that of GB-GAPDH, and plotted as mean values ± SDs against time. Dox, doxycycline.

FIG. 3.

MS2-KSRP associates with human mRNA decay enzymes. (A) HeLa-TO cells were transfected with constructs expressing FLAG-tagged MS2 or MS2-KSRP. RNase-treated cell extracts were subjected to anti-FLAG immunoprecipitation. (B) HeLa-TO cells were cotransfected with constructs expressing FLAG-tagged MS2-KSRP_N, MS2-KSRP_C, or MS2-KH_1-4. Cell extracts were subjected to anti-FLAG immunoprecipitation. For both panels, the precipitates were analyzed by anti-PARN, anti-DCP2, anti-RRP4, or anti-FLAG immunoblotting. A 2.5% input for immunoprecipitation reactions was also analyzed. The asterisk indicates the heavy chain of anti-FLAG immunoglobulin G, which comigrates with MS2-KH_1-4 and cross-reacts with the secondary antibody. IP, immunoprecipitate.

FIG. 4.

mRNA decay triggered by tethered KSRP requires human mRNA decay enzymes. (A and B) HeLa-TO cells were transfected with siRNAs targeting chloramphenicol acetyltransferase (CAT), PARN, RRP46, or DCP2 (A) and siRNAs targeting CAT, RRP4, or PM-Scl100 (B). Total extracts were subjected to immunoblot analysis with anti-PARN, anti-RRP46, anti-DCP2, or anti-α-tubulin antibody (A) and with anti-RRP4, anti-PM-Scl100, or anti-α-tubulin antibody (B). (C) HeLa-TO cells were transfected with siRNAs targeting CAT or CCR4. poly(A)⁺ mRNAs were subjected to Northern blot analysis with probes against CCR4 or GAPDH. (D to G) The levels of β-globin mRNAs were analyzed in cells coexpressing MS2 (lanes 1), MS2-KSRP (D, lanes 2 to 11, and E, lanes 2 to 5), MS2-KH_1-4 (F, lanes 2 to 9), or MS2-KSRP_C (G, lanes 2 to 9) along with siRNAs targeting different mRNA decay enzymes as indicated (top panels). The levels of GB-6bs mRNA were quantified as described in the legend for Fig. 1. Mean values with standard deviations (SDs) are shown. The expression of MS2-KSRP (D and E), MS2-KH_1-4 (F), and MS2-KSRP_C (G) in siRNA-treated cells was analyzed by anti-FLAG immunoblotting (bottom panels). (H) The decay of GB-6bs mRNA was analyzed in cells coexpressing MS2-KSRP along with siRNAs targeting CAT, PARN, DCP2, or RRP46. Varied levels of GB-GAPDH mRNA are due to loading variations. Signals of GB-6bs were quantitated, normalized to that of GB-GAPDH, and plotted as mean values ± SDs against time. Dox, doxycycline.

FIG. 5.

Overexpression of non-RNA-binding KSRP polypeptide impairs ARE-directed mRNA decay. (A) HeLa-TO cells were transfected with vectors expressing different FLAG-tagged KSRP polypeptides. Cytoplasmic extracts were prepared and incubated with ³²P-labeled ARE^tnf RNA, and UV cross-linking assays were performed. The UV cross-linking reactions were immunoprecipitated with anti-FLAG agarose (lanes 2, 4, 6, 8, 10, and 12), and precipitates were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography (top panel). A portion (25%) of each UV cross-linking reaction was loaded onto lanes 1, 3, 5, 7, 9, and 11. The immunoprecipitates were also analyzed by anti-FLAG immunoblotting (bottom panel). (B) HeLa-TO cells were transfected with constructs expressing FLAG-tagged KSRP polypeptides. Cell extracts were subjected to anti-FLAG immunoprecipitation. The precipitates were analyzed by anti-RRP4 or anti-FLAG immunoblotting. A 5% extract used for immunoprecipitation reactions was also analyzed. The asterisk indicates the heavy chain of anti-FLAG immunoglobulin G. (C and D) The decay of GB-ARE^tnf (C) and GB-ARE^fos (D) mRNAs in HeLa-TO cells expressing various KSRP polypeptides was analyzed. Signals of GB-ARE^tnf and GB-ARE^fos mRNAs were quantitated and normalized with that of GB-GAPDH. The calculated half-lives (n = 3) are indicated as mean values ± standard deviations. IP, immunoprecipitate; IgG, immunoglobulin G; Dox, doxycycline.

FIG. 6.

Tethering of KSRP targets HIV-1 mRNAs for degradation and dramatically inhibits viral replication. (A) 293T cells were transfected with an HIV-1 proviral DNA, a construct expressing a truncated HIV-1 mRNA (in which most of the sequences [nucleotides 715 to 8887], including the RRE, are deleted), and constructs expressing Rev, KSRP, or Rev-KSRP. Cytoplasmic levels of HIV-1 mRNAs (labeled 9-kb, 4-kb, and 2-kb) and control mRNA (labeled control) were analyzed by Northern blotting with a probe against the U3 sequences. (B) Levels of cytoplasmic 9-, 4-, and 2-kb mRNAs were quantitated and normalized to levels of control mRNA. (C) Nuclear RNA was isolated and analyzed by Northern blotting as described for panel A. (D) Levels of nuclear 9-, 4-, and 2-kb mRNAs were quantitated. (E) Cytoplasmic levels of HIV-1 mRNAs and control mRNA were analyzed by Northern blotting with cells cotransfected with pcDNA or constructs expressing Rev or Rev-GFP. (F) Levels of 9-, 4-, and 2-kb mRNAs were quantitated and normalized to levels of control mRNA. (G) Cytoplasmic levels of HIV-1 mRNAs and control mRNA were analyzed by Northern blotting with cells cotransfected with pcDNA or a construct expressing Rev-KH_1-4. (H) Levels of 9-, 4-, and 2-kb mRNAs were quantitated and normalized to levels of control mRNA. (I) Expression of transfected proteins was analyzed by anti-Xpress immunoblotting. (J) Supernatants withdrawn from transfected cells were analyzed for viral infectivity and p24 antigen levels. For all bar graphs, the levels of each class of HIV-1 mRNA in cells transfected with pcDNA are set at 100 and mean values with standard deviations (error bars) are shown.