Abstract
X-chromosome-linked inhibitor of apoptosis, XIAP, has been shown to contain a strong internal ribosome entry site (IRES) within its 5′ untranslated region (UTR) that promotes translation of XIAP mRNA under conditions of cellular stress. This claim came under scrutiny in a recent report demonstrating that the XIAP 5′ UTR undergoes splicing when inserted between the two reporter cistrons of the dual luciferase plasmid Rluc/Fluc. In this paper, we demonstrate that the splicing within the XIAP 5′ UTR specifically occurs only in the context of mRNA produced from the Rluc/Fluc but not the pβgal/CAT bicistronic reporter plasmid.
Keywords: IRES, bicistronic reporter, RT-PCR, mRNA
Van Eden et al. (2004a) recently proposed the use of stringent RNA test procedures to demonstrate internal ribosome entry sites (IRES) in eukaryotic mRNAs. The results of their work suggest that the 5′ UTR of the X-chromosome-linked inhibitor of apoptosis (XIAP), in addition to containing a weak IRES element, also contains a strong splice-acceptor site that may significantly contribute to the observed IRES activity of the XIAP 5′ UTR in bicistronic vectors. They observed that insertion of the XIAP 5′UTR into the inter-cistronic region of the dual luciferase plasmid pRL-FL significantly reduced expression of the first cistron, Renilla luciferase (RL) (Fig. 1E in Van Eden et al. 2004a), leading them to suspect that RL sequence was being removed by alternative splicing occurring between sequences in pRF-FL and sequences within the XIAP 5′ UTR. We re-examined this phenomenon using both dual luciferase bicistronic constructs (Fig. 1A) and pβgal/CAT bicistronic constructs (Fig. 2A) and discovered that spurious splicing within the 5′ UTR of XIAP occurs only in the context of Rluc/Fluc but not pβgal/CAT bicistronic reporter mRNA.
FIGURE 1.
Insertion of the XIAP 5′UTR into the linker region (LR) of dual luciferase bicistronic plasmid pR-F results in splicing and reduces levels of Renilla luciferase. (A) Schematic diagram depicting the dual luciferase bicistronic constructs. Plasmid pR-F was created by inserting the Renilla open reading frame (from plasmid pRL-null, Promega) and the Firefly open reading frame (from plasmid pSP-Luc+, Promega) into the expression plasmid pCI (Invitrogen). The black square represents the putative splice-acceptor site identified by Van Eden et al. (2004a). The regions amplified for the quantitative RT-PCR analysis are indicated with black bars beneath the Renilla and Firefly cistrons. The positions of the T7 and P2 primers that were used for RT-PCR analysis are indicated by gray arrows. (B) Renilla (black bars) and Firefly (gray bars) luciferase activities were determined 24 h post-transfection of 293A cells using the Dual-Luciferase Reporter System (Promega). The bars represent the average ± S.D. of three experiments performed in triplicate. (C) RT-PCR analysis of bicistronic mRNA was performed on total RNA (Absolutely RNA Miniprep kit, Stratagene) isolated from transfected 293A cells. The RNA was reverse transcribed using the First Strand cDNA Synthesis kit (Amersham) and included NotI-d(T)18 primer. PCR was then used to amplify the bicistronic mRNA using T7 forward primer and P2 reverse primer as described (Van Eden et al. 2004a). As a positive control, the PCR reaction was performed on plasmid DNA for each construct. As a negative control, reverse transcriptase was omitted and the samples were subjected to PCR amplification. (D) Quantitative RT-PCR was performed on DNAseI-treated total RNA isolated from 293A cells transfected with indicated plasmids. The RNA was isolated and reverse transcribed as in C. The quantitative real-time RT-PCR was performed using the QuantiTect SYBR green RT-PCR kit (QIAGEN) and analyzed on an ABI Prism 7000 sequence detection system using the ABI Prism 7000 SDS Software. qPCR reactions were carried out to detect the Renilla luciferase (5′-aacgcggcctcttcttattt-3′; 5′-atttgcctgatttgcccata-3′) and Firefly luciferase (5′-gaggttccatctgcaggta-3′; 5′-ccggtatccagatccacaac-3′) cistrons of the bicistronic mRNA. The Rluc/Fluc ratio was calculated as 2-[Ct(Rluc)-Ct(Fluc)]. The bars represent the mean ± S.E.M. of six experiments.
FIGURE 2.
Insertion of the XIAP 5′UTR into the linker region (LR) of bicistronic plasmid pβgal/CAT does not result in splicing. (A) Schematic diagram depicting the bicistronic constructs. All constructs were described previously (Holcik et al. 1999, 2003). The black square represents the putative splice-acceptor site identified by Van Eden et al. (2004a). The regions amplified for the quantitative RT-PCR analysis are indicated by black bars beneath the βgal and CAT cistrons. (B) βgal activity (black bars) was determined as described (MacGregor et al. 1991) 24 h post-transfection of 293A cells and was normalized to Neomycin (NPTII ELISA, Agdia) produced from the same plasmid. The CAT activity (gray bars) was determine using CAT ELISA (Roche) and was normalized to Neomycin. The bars represent average ± S.D of three experiments performed in triplicate. (C) Quantitative RT-PCR was performed on total RNA isolated from 293A cells transfected with pβgal/CAT and pβgal/XIAP/CAT plasmids. The RNA was isolated and reverse transcribed as in Figure 1C. The quantitative real-time RT-PCR was performed using the QuantiTect SYBR green RT-PCR kit (QIAGEN) and analyzed on an ABI Prism 7000 sequence detection system using the ABI Prism 7000 SDS Software. qPCR reactions were carried out to detect the βGAL (5′-actatcccgaccgccttact-3′, 5′-ctgtagcggctgatgttgaa-3′) and CAT (5′-gcgtgttacggtgaaaacct-3′, 5′-gggcgaagaagttgtccata-3′) cistrons of the bicistronic mRNA. The βgal/CAT ratio was calculated as 2-[Ct(βgal)-Ct(CAT)]. The bars represent the mean ± S.E.M. of five experiments.
The pR-F dual luciferase plasmid used in this study (Fig. 1A) was constructed by inserting the Renilla and Firefly cistrons (from pRL-null and pSP-Luc+ plasmids [Promega], respectively) into the expression plasmid pCI (Promega). Similar to the plasmid pRL-FL used by Van Eden et al. (2004a), this construct also contains a chimeric intron (originating from the pCI vector) that precedes the Renilla luciferase cistron. When we tested the 265-nucleotide fragment of XIAP 5′ UTR in the Rluc/Fluc bicistronic plasmid, we observed that the presence of XIAP 5′ UTR resulted in a significant decrease in Renilla luciferase activity (Fig. 1B), as described by Van Eden et al. (2004a). To analyze the bicistronic mRNAs, the total RNA from transfected 293A cells was subjected to RT-PCR analysis as described by Van Eden et al. (2004a). We observed that the pR-XIAP-F construct failed to produce intact bicistronic mRNA, but instead produced several shorter transcripts indicative of multiple splice variants (Fig. 1C). Sequence analysis of these splice variants confirmed that they were generated by splicing between the chimeric intron of pR-F and the polypyrimidine tract of the XIAP 5′ UTR (smaller fragments) or a splice-donor site within the Renilla luciferase gene and the polypyrimidine tract of the XIAP 5′ UTR (large fragment) (data not shown). These splice variants were identical to those reported by Van Eden et al. (2004a). Surprisingly, the construct containing the XIAP 5′ UTR in the antisense orientation (pR-asXIAP-F) produced not only primarily full-length intact bicistronic mRNA but also several shorter fragments (Fig. 1C). The sequencing of these variants disclosed that they were generated by splicing between either the chimeric intron of pR-F or a splice-donor site within the Renilla luciferase gene and various sequences of the XIAP 5′ UTR (data not shown). We further employed quantitative RT-PCR to precisely determine the ratio of Rluc and Fluc cistrons as described (Teshima-Kondo et al. 2004; Miura et al. 2005). The advantage of this approach over RT-PCR is that it can accurately measure the amount of Rluc and Fluc portions of the same mRNA and hence estimate the degree of splicing. Thus, in the absence of splicing, the ratio of Rluc and Fluc cistrons should be the same for both pR-F and pR-XIAP-F bicistronic mRNAs. Conversely, splicing of the bicistronic mRNA would result in a decrease of the Rluc cistron RNA relative to Fluc cistron RNA. Indeed, we observed that the insertion of XIAP 5′ UTR into the dual luciferase reporter vector resulted in a threefold decrease in Rluc cistron RNA (P=0.0043, Mann-Whitney test). This decrease in the Rluc RNA levels in pR-XIAP-F strongly correlated with the ~threefold decrease in the Renilla luciferase activity generated from the same plasmid (Fig. 1B) and the appearance of alternative splice products as detected by RT-PCR (Fig. 1C). The insertion of the antisense XIAP 5′ UTR fragment resulted in a 1.4-fold decrease in Rluc cistron RNA (P = 0.0931, Mann-Whitney test; Fig. 1D). Again, this decrease in Rluc RNA levels in pR-asXIAP-F correlated with the appearance and relative abundance of alternative splice products detected by RT-PCR (Fig. 1C). However, we did not observe any reduction in Renilla luciferase activity (Fig. 1B). It should be noted that the data in Figure 1B represent the raw data obtained from the dual luciferase assays and therefore do not reflect possible differences in transfection efficiency of individual constructs.
Taken together, these data are consistent with the data reported by Van Eden et al. (2004a) demonstrating that the XIAP 5′ UTR undergoes splicing within the dual luciferase bicistronic mRNA and that this splicing generates monocistronic Fluc transcripts. The increase in Fluc activity in pR-XIAP-F (Fig. 1B) is therefore largely due to the presence of monocistronic Fluc transcripts and not XIAP IRES activity.
In contrast, similar analyses of the pβgal/CAT bicistronic constructs failed to show any decrease in βgal values regardless of the length of the XIAP 5′ UTR used (Fig. 2B). Unlike pR-F, the pβgal/CAT construct is based on the pcDNA3 cloning vector (Invitrogen) that also contains the Neomycin gene driven by the SV40 promoter. We therefore normalized the pβgal and CAT values to Neomycin to correct for transfection efficiencies of different constructs. We observed that the levels of βgal produced from empty vector or from constructs containing 1 kb of XIAP 5 UTR in either the sense or antisense orientation, as well as the 265-nucleotide XIAP 5′ UTR fragment (identical to that used in pR-XIAP-F, see above) were virtually identical. Because the βgal/XIAP/CAT bicistronic mRNA is significantly longer than R-XIAP-F mRNA, we were unable to perform RT-PCR analysis to visualize the intact bicistronic mRNA. However, the quantitative RT-PCR analysis of the total RNA isolated from 293A cells transfected with pβgal/CAT or pβgal/XIAP/CAT plasmids confirmed that the insertion of the XIAP 5′ UTR did not affect the ratio of βgal and CAT cistron mRNAs (P<0.9999, Mann-Whitney test) (Fig. 2C). Taken together, these data strongly indicate the lack of spurious splicing in pβgal/XIAP/CAT bicistronic mRNA. Therefore, the significant increase in CAT levels observed with pβgal/XIAP/CAT and pβgal/5′ (265)/CAT plasmids (Fig. 2B) is due to the strong IRES activity of the XIAP 5′ UTR.
The data presented here and in two additional recent papers from the Lloyd laboratory (Sherrill et al. 2004; Van Eden et al. 2004b) indicate that splicing may be a common problem associated with the use of the pRL-FL plasmid. The insertion of the 5′ UTR of BCL-2 (which contains an IRES) into pRL-FL also resulted in the generation of two splice variants, one lacking the first cistron (Sherrill et al. 2004). Similarly, the insertion of the HIAP2/cIAP1 5′ UTR (which contains an IRES) into the same plasmid resulted in the generation of an alternative splice variant as well (Van Eden et al. 2004b). Significantly, the splicing of XIAP and HIAP2/cIAP1 5′ UTRs occurred using a splice-donor site that is located in the Renilla luciferase sequence. In the case of BCL-2 and the minor XIAP transcript, the splicing occurred with the donor splice site of the chimeric intron that is present in the pRL-FL plasmid and precedes the RL coding sequence. We have cloned the HIAP2/cIAP1 5′ UTR into the pβgal/CAT reporter vector and, again, were unable to detect splicing within the 5′ UTR of HIAP2/cIAP1 (using either ribonuclease protection assay) (Fig. 2B in Warnakulasuriyarachchi et al. 2004) or Northern blot (T. Graber and M. Holcik, unpubl.).
Although it is not unusual to obtain somewhat different data using different reporter systems, our data suggest that the pRL-FL bicistronic reporter system is prone to detecting spurious splicing events, even if these do not occur naturally, and therefore pRL-FL may not be the most appropriate system for the detection of IRES activity. The presence of a chimeric intron in pRL-FL provides a strong splice-donor site that may trans-splice with splice-acceptor-like sequences within the 5′ UTRs tested. Furthermore, because the Renilla luciferase sequence contains an additional splice-donor site, the insertion of sequences downstream of RL can result in nonphysiological splicing that is not seen in the context of endogenous mRNAs. For example, numerous Northern blot analyses of endogenous XIAP and HIAP2/cIAP1 mRNAs in several cell lines and primary tissues from mouse, rat, and human failed to detect any alternative splice products (e.g., Duckett et al. 1996; Liston et al. 1996, 1997; Farahani et al. 1997; Lagace et al. 2001; Holcik et al. 2002). Clearly, pRL-FL does not produce alternatively spliced mRNAs in all cases, as the insertion of a HCV RNA segment did not yield any detectable splicing while providing IRES activity (Van Eden et al. 2004a). Because many laboratories are currently using the pRL-FL system due to its convenience, we would caution that stringent RNA test procedures are indeed necessary to separate splicing from translation using the pRL-FL system. Alternatively, the mutation of the splice-donor site within the Renilla luciferase gene and the removal of the chimeric intron could improve the usability of the pRL-FL system. It should be noted, however, that previous work by Hennecke et al. (2001) identified additional problems associated with the arrangement of Renilla and Firefly luciferase reporter genes in the bicistronic constructs. Thus, the use of an alternative reporter system, such as βgal and CAT, may be a more appropriate tool to avoid complications caused by spurious confounding events when testing for IRES activity.
Article published online ahead of print. Article and publication date are at http://www.rnajournal.org/cgi/doi/10.1261/rna.2158605.
REFERENCES
- Duckett, C.S., Nava, V.E., Gedrich, R.W., Clem, R.J., Van Dongen, J.L., Gilfillan, M.C., Shiels, H., Hardwick, J.M., and Thompson, C.B. 1996. A conserved family of cellular genes related to the baculovirus iap gene and encoding apoptosis inhibitors. EMBO J. 15: 2685–2694. [PMC free article] [PubMed] [Google Scholar]
- Farahani, R., Fong, W.G., Korneluk, R.G., and MacKenzie, A.E. 1997. Genomic organization and primary characterization of miap-3: The murine homologue of human X-linked IAP. Genomics 42: 514–518. [DOI] [PubMed] [Google Scholar]
- Hennecke, M., Kwissa, M., Metzger, K., Oumard, A., Kroger, A., Schirmbeck, R., Reimann, J., and Hauser, H. 2001. Composition and arrangement of genes define the strength of IRES-driven translation in bicistronic mRNAs. Nucleic Acids Res. 29: 3327–3334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Holcik, M., Lefebvre, C.A., Yeh, C., Chow, T., and Korneluk, R.G. 1999. A new internal-ribosome-entry-site motif potentiates XIAP-mediated cytoprotection. Nat. Cell Biol. 1: 190–192. [DOI] [PubMed] [Google Scholar]
- Holcik, M., Lefebvre, C.A., Hicks, K., and Korneluk, R.G. 2002. Cloning and characterization of the rat homologues of the inhibitor of apoptosis protein 1, 2, and 3 genes. BMC Genomics 3: 5–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Holcik, M., Gordon, B.W., and Korneluk, R.G. 2003. The internal ribosome entry site-mediated translation of antiapoptotic protein XIAP is modulated by the heterogeneous nuclear ribonucleoproteins C1 and C2. Mol. Cell. Biol. 23: 280–288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lagace, M., Xuan, J.Y., Young, S.S., McRoberts, C., Maier, J., Rajcan-Separovic, E., and Korneluk, R.G. 2001. Genomic organization of the x-linked inhibitor of apoptosis and identification of a novel testis-specific transcript. Genomics 77: 181–188. [DOI] [PubMed] [Google Scholar]
- Liston, P., Roy, N., Tamai, K., Lefebvre, C., Baird, S., Cherton-Horvat, G., Farahani, R., McLean, M., Ikeda, J.E., MacKenzie, A., et al. 1996. Suppression of apoptosis in mammalian cells by NAIP and a related family of IAP genes. Nature 379: 349–353. [DOI] [PubMed] [Google Scholar]
- Liston, P., Lefebvre, C., Fong, W.G., Xuan, J.Y., and Korneluk, R.G. 1997. Genomic characterization of the mouse inhibitor of apoptosis protein-1 and 2 genes. Genomics 46: 495–503. [DOI] [PubMed] [Google Scholar]
- MacGregor, G.R., Nolan, G.P., Fiering, S., Roederer, M., and Herzenberg, L.A. 1991. Use of E. coli lacZ (β-Galactosidase) as a reporter gene. In Methods in molecular biology (eds. E.J. Murray and J.M. Walker), Vol. 7, pp. 217–235. Humana, Clifton, NJ. [DOI] [PubMed]
- Miura, P., Thompson, J., Chakkalakal, J.V., Holcik, M., and Jasmin, B.J. 2005. The utrophin A 5′ UTR confers IRES-mediated translation control during regeneration of skeletal muscle fibers. J. Biol. Chem. http://www.jbc.org/cgi/reprint/M503994200v1. [DOI] [PubMed]
- Sherrill, K.W., Byrd, M.P., Van Eden, M.E., and Lloyd, R.E. 2004. BCL-2 translation is mediated via internal ribosome entry during cell stress. J. Biol. Chem. 279: 29066–29074. [DOI] [PubMed] [Google Scholar]
- Teshima-Kondo, S., Kondo, K., Prado-Lourenco, L., Gonzalez-Herrera, I.G., Rokutan, K., Bayard, F., Arnal, J.F., and Prats, A.C. 2004. Hyperglycemia upregulates translation of the fibroblast growth factor 2 mRNA in mouse aorta via internal ribosome entry site. FASEB J. 18: 1583–1585. [DOI] [PubMed] [Google Scholar]
- Van Eden, M.E., Byrd, M.P., Sherrill, K.W., and Lloyd, R.E. 2004a. Demonstrating internal ribosome entry sites in eukaryotic mRNAs using stringent RNA test procedures. RNA 10: 720–730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- ———. 2004b. Translation of cellular inhibitor of apoptosis protein 1 (c-IAP1) mRNA is IRES mediated and regulated during cell stress. RNA 10: 469–481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Warnakulasuriyarachchi, D., Cerquozzi, S., Cheung, H.H., and Holcik, M. 2004. Translational induction of the inhibitor of apoptosis protein HIAP2 during endoplasmic reticulum stress attenuates cell death and is mediated via an inducible internal ribosome entry site element. J. Biol. Chem. 279: 17148–17157. [DOI] [PubMed] [Google Scholar]