Skip to main content
NCBI home page
RNA logoLink to RNA
. 2008 Mar;14(3):535–542. doi: 10.1261/rna.738508

Sequence-specific activation of TAK1-D by short double-stranded RNAs induces apoptosis in NCI-H460 cells

Reinhard Kodym 1, Elisabeth Kodym 1, Micheal D Story 1
PMCID: PMC2248254  PMID: 18230764

Abstract

Short double-stranded RNAs (dsRNA) are potent biological entities triggering a number of cellular effects. Most prominent among these is the post-transcriptional gene silencing of target genes by small interfering RNAs (siRNAs). In addition dsRNAs activate signal transduction processes through molecules like PKR or the Toll-like receptor important in viral defense and in explaining off target effects of siRNAs. Only a few of these dsRNA triggered pathways have been characterized yet. Here we show that the splicing variant D of the TAK1 gene is activated by short double-stranded RNAs in a sequence-specific manner. Activation of TAK1-D leads to the downstream activation of the p38 MAPK and of SAPK/JNK but not the NFκB pathway. In the human lung cancer cell line NCI-H460 the activation of these pathways leads to cell cycle arrest and apoptosis. Our results demonstrate that TAK1-D is activated by siRNAs of specific sequences, offering a new explanation for off target effects triggered by these molecules. In addition the dsRNA triggered activation of a cell death pathway in the human lung cancer cell line studied suggests that TAK1-D might be a new and promising therapeutic target for the treatment of nonsmall cell lung cancer.

Keywords: siRNA, TAK1, apoptosis, cell cycle

INTRODUCTION

Since the discovery of double-stranded RNA (dsRNA) mediated post-transcriptional gene silencing in Caenorhabditis elegans (Fire et al. 1998) and the description of 21-nucleotide (nt) dsRNAs effective in mammalian cells (Elbashir et al. 2001), small interfering RNAs (siRNAs) have been heavily used as a tool in biomedical research. In addition, the potential in vivo application of siRNAs as gene specific agents is now widely studied (Wall and Shi 2003). Both applications depend decisively on the specificity of the siRNA for its target mRNA. Apart from acting as micro-RNAs and down-regulating gene expression through low stringency binding to the 3′ untranslated region of mRNAs (Scacheri et al. 2004), siRNAs elicit cellular effects through binding and activating specific proteins. Currently three such target proteins have been identified: (1) The Toll-like receptor 3 (TLR3), a receptor involved in the immune response, which activates NFκB and triggers the production of type I interferons (Alexopoulou et al. 2001). (2) dsRNA-dependent protein kinase (PKR), a protein kinase activated preferentially by longer dsRNA molecules, which triggers the up-regulation of interferon inducible transcripts (Sledz et al. 2003) and inhibits protein synthesis through phosphorylation eIF-2α (de Haro et al. 1996). (3) 2′,5′-Oligoadenylate synthetase, which when activated catalyzes the formation of 2′,5′-oligoadenylates that activate RNAase L, leading to cleavage of cellular RNAs (Player and Torrence 1998). In addition to these pathways, which lead to the down-regulation of protein synthesis or activation of interferon inducible transcription, we have identified a new type of cellular stress response triggered by short dsRNAs in a sequence-specific manner.

RESULTS

While studying the effect of siRNAs on the expression of the protein kinase TLK1 we observed an unexpected activation of the cellular stress response pathways. After screening several mitogen activated protein kinase kinase kinase (MAPKKK) proteins we found an siRNA mediated activation of transforming growth factor beta activated kinase 1 (TAK1). Figure 1 shows in its upper panel that an antibody specific for TAK1 phosphorylated on T184 and T187 reacts with a protein slightly larger than 50 kDa after cells have been treated with siRNA si5. Most importantly, this phosphorylation does not occur after treatment with siRNA si6, which also targets the mRNA of TLK1. Autophosphorylation of T187 as well as of other residues in the activation loop of TAK1 has been shown to be essential for the activation of the kinase (Singhirunnusorn et al. 2005).

FIGURE 1.

FIGURE 1.

Open in a new tab

Sequence-specific activation of TAK1-D by short double-stranded RNAs. In intact NCI-H460 cells, treatment with siRNA si5 but not with si6 leads to autophosphorylation of TAK1 splice variant D on T184 and T187 within 3 h. TAK1-D is activated by these autophosphorylations and phosphorylates its downstream substrates MKK3/6 and MKK4. In contrast exposure of the cells to 10 ng/mL IL1β for 30 min activates exclusively the long form of TAK1 (TAK1-A), which in addition to MKK3/6 and MKK4 also activates the NFκB pathway through phosphorylation of IKKα/β.

As shown in Figure 1 the TAK1 splicing variant activated by siRNA si5 is TAK1-D. In contrast, treatment of NCI-H460 cells with 10 ng/mL IL-1β leads to autophosphorylation and activation of the long splicing variant A of TAK1 in response to IL-1 receptor activation (Ninomiya-Tsuji et al. 1999). It is also evident from the upper panel of Figure 1 that IL-1 does not activate TAK1-D, likely due to the lack of the TAB2/3 interaction domain in TAK1-D. The long (TAK1-A) and short (TAK1-D) variants of TAK1 also differ in their ability to phosphorylate and activate downstream targets. Figure 1 shows that both TAK1-A and TAK1-D phosphorylate and activate mitogen activated protein kinase kinase (MAPKK) MKK3/6 and MKK4. In contrast only the long form (TAK1-A) is able to activate the NFκB pathway through phosphorylation of IKKα/β.

While the lack of the C-terminal protein interaction domain in TAK1-D makes it unlikely that siRNA si5 activates the kinase through interaction with upstream signaling components in intact NCI-H460 cells, it does not rule out that possibility. To verify if siRNA si5 indeed interacts with TAK1-D directly, the activation of the kinase was studied in vitro using purified proteins. As TAK1 coexpressed with TAB1 is constitutively active (Sakurai et al. 2000), the autophosphorylation shown in Figure 2A after the addition of dsRNA si6 is the result of this activity. When siRNA si5 was added to the kinase reaction, an increase in autophosphorylation activity of the TAK1-D TAB1 complex was observed and is shown in Figure 2A. To rule out an effect of si5 or si6 on another kinase coprecipitated with TAK1-D, an inactive FLAG tagged TAK1-D was constructed by mutating lysine at position 63 to tryptophan (K63W) (Sakurai et al. 2000). Figure 2B demonstrates that TAK1-D(K63W) does not show any detectable autophosphorylation in the kinase assay, indicating that the kinase assay shown in Figure 2A specifically measures TAK1-D kinase activity. To test whether the increased activation of the TAK1-D/TAB1 complex by siRNA si5 is caused by an increased affinity of the RNA molecule to the kinase, binding of radiolabeled siRNAs to immunoprecipitated FLAG tagged TAK1-D was measured. In Figure 2C the amount of RNA bound to immobilized FLAG-TAK1-D is given. It can be seen that the number of si5 molecules bound to TAK1-D under buffer conditions used for the kinase assay is significantly greater than the number of si6 or siNT RNAs. These data demonstrate that there is a moderately increased affinity of siRNA si5 to TAK1-D, which at least partially explains the observed modulation of TAK1-D activity by siRNA si5.

FIGURE 2.

FIGURE 2.

Open in a new tab

Effect of siRNAs on the activity of purified recombinant TAK1-D protein. (A) In vitro kinase assay showing the activation of TAK1-D autophosphorylation by siRNA si5. FLAG-tagged TAK1-D and His-tagged TAB1 were coexpressed in Hela cells and purified by immunoprecipitation with anti FLAG antibody. The autoradiograph in the top panel and the phosphorylation kinetics (▼: si5, ●: si6) show that constitutive autophosphorylation activity of the TAK1-D/TAB1 complex is increased by siRNA si5. The lower panel shows the Ponceau S stained proteins as a loading control. (B) Autoradiograph (top) and Ponceau S stained loading control (bottom) of a kinase reaction performed with wild-type and kinase inactive (K63W) TAK1-D indicating that the signal detected in the kinase assay is specific for TAK1-D. (C) Binding of radiolabeled siRNAs on FLAG tagged TAK1-D immobilized on magnetic beads. siRNA si5 shows increased binding. Error bars indicate the standard error of three experiments.

To address the question of which features of the dsRNA sequence are important for TAK1-D activation, NCI-H460 cells were treated with a panel of randomly chosen siRNA molecules. Figure 3 demonstrates that siRNA si5 and si10 are most effective in activating the TAK1-D p38 pathway. Sequence similarities between the 5′ (C followed by four complementary A or U bases) and the 3′ (the AUGA sequence) regions of these two RNAs suggest that these sequence motifs might play a role in the activation of TAK1-D. To gain additional insight into the sequence and structural requirements for a short dsRNA molecule to activate TAK1-D, dsRNAs with modifications of the si5 sequence were synthesized, and their ability to activate TAK1-D was tested in NCI-H460 cells. Figure 4 shows the sequence of these molecules and their effect on the TAK1-D p38 pathway. In the leftmost lanes the effect of si6 and si5 are shown as a control. Transfection of NCI-H460 cells with single-stranded RNA molecules with the sequence of the si5 sense strand (si5se) and the si5 antisense strand (si5as) fails to activate TAK1-D. As the stability of the single-stranded molecules is lower in the cell culture system than the one of dsRNA, these findings indicate but do not prove the requirement of dsRNA to activate TAK1-D. In addition, whether or not the two base dT overhang on the 3′ end of both strands is required for TAK1-D activation (si5w) was tested. The strong effect of the molecule with both overhangs deleted proved that these structures are not required. Having demonstrated that the terminal two bases are dispensible for TAK1-D activation, the influence of the length of the dsRNA was tested. While a dsRNA two bases shorter than si5 (si5s) completely failed to activate TAK1-D, a two bases longer molecule (si5l) shows only a slight TAK1-D activation. To investigate the role of the homopolymeric stretch of four A or U found in the activating RNAs si5 and si10 (Fig. 3), one U was changed to a C in dsRNA si5d. This single base change significantly reduced the capacity of the molecule to activate TAK1-D, indicating the importance of this motif. When one compares the sequence of the activating siRNAs si5 and si10 in Figure 3 with the other sequences, it can be noted that they contain a high proportion of A and U bases. To test whether this sequence feature is important for TAK1-D activation, a dsRNA with alternating A and U bases was constructed (AU). It can be seen in Figure 4 that this sequence does not activate the TAK1-D pathway. Sequence specificity has also been reported for the activation of the innate immune response (Judge et al. 2005) by siRNAs. In addition it has been shown that incorporation of 2′-O-methyl (2′-O-Me)uridine or guanosine completely inhibits this response (Judge et al. 2006). Therefore, we tested whether selective substitution of 2′-O-Me ribonucleosides in the si5 sequence influences its ability to activate TAK1-D. As shown, partial substitution in both strands (OMeD) or in the sense strand (OMeS) resulted in molecules unable to activate TAK1-D. Interestingly partial incorporation of 2′-O-Me nucleosides in the antisense strand did not interfere with the ability of the molecule to trigger the stress activate protein kinase pathway.

FIGURE 3.

FIGURE 3.

Open in a new tab

Activation of TAK1-D by siRNAs of specific sequence in intact cells. Transfection of NIH-H460 cells with randomly chosen siRNAs demonstrates that the activation of the TAK1-D p38 pathway depends on the siRNA sequence, and siRNAs si5 and si10 activate the pathway most effectively.

FIGURE 4.

FIGURE 4.

Open in a new tab

Activation of TAK1-D by siRNAs of specific sequence in intact cells. Consequences of modifications of the sequence of siRNA si5 on the activation of TAK1-D and its downstream target p38 MAPK in NCI-H460 cells. si5, si5w, si5s, si5l, and si5d are double-stranded RNAs, and the aligned sequences of these molecules are shown. si5se and si5as are single-stranded RNAs with the si5 sense and si5 antisense sequence. Molecules OMeD, OMeS, and OMeA are dsRNAs modified with a 2′-O-Me group at the dot marked bases.

While there is a considerable body of knowledge about the cell biologic functions of the long splicing variants of TAK1, mainly about TAK1-A, the role of the short splicing variants has not yet been characterized. Having demonstrated a specific activation of the D variant of TAK1 by short dsRNAs of defined sequence, it is interesting to ask which downstream signaling events are triggered by this activation.

The upper panel of Figure 5A shows that the activating autophosphorylation of TAK1-D on T184 and T187 occurs within 2 h after the addition of siRNA si5 to the cell culture. This Western blot demonstrates again that the activation does not occur after exposure of NCI-H460 cells to siRNA of different nucleotide sequence, like si6 or siNT. The activation of TAK1-D leads to a phosphorylation of the MAPKKs of both stress activated protein kinase signaling pathways, indicated by the occurrence of activation phosphorylation on MKK3/6 and MKK4 (Fig. 5A). Consecutively, activation of MKK3/6 leads to phosphorylation and activation of the p38 MAPK, while activation of MKK4 phosphorylates and activates SAPK/JNK. The bottom panel of Figure 5A shows that the protein levels of Tlk1, the gene product siRNA si5 and siRNA si6 are directed against, remain unchanged during the first 6 h after addition of the siRNAs. Figure 5B shows that after 48 h both siRNAs si5 and si6 drastically reduce the expression level of their target gene TLK1 in HeLa cells. This provided further confirmation, apart from the rapid occurrence of TAK1-D phosphorylation, that gene silencing effects of the siRNAs play no role in the phenomena described.

FIGURE 5.

FIGURE 5.

Open in a new tab

Downstream signaling events triggered by the dsRNA mediated activation of TAK1-D (A) Transfection of NCI-H460 cells with siRNA si5 leads to autophosphorylation and activation of TAK1-D within 2 h. Phosphorylation of MKK3/6 leads to activation of p38 MAPK, while phosphorylation of MKK4 activates SAPK/JNK. No activation of the two MAPK cascades can be observed after exposure of cells to siRNA si6 or siNT. The bottom panel demonstrates that the intracellular concentrations of TLK1, the mRNA of which is targeted by si5 and si6, remain constant and therefore do not contribute to the activation of p38 MAPK and SAPK/JNK. (B) TLK1 protein levels in HeLa cells are reduced to almost undetectable levels by si5 and si6 after 48 h. (C) Mouse peritoneal macrophages were treated with siRNA si5 and si6 for 4 h. The Western blot demonstrates that the activation of the TAK1-D p38 pathway by siRNAs of specific sequence is not limited to the human small cell lung cancer cell line NCI-H460. (D) Response of the cell cycle control phosphatases cdc25 to activation of the p38 MAPK and SAPK/JNK pathway. Nuclear cdc25A shows an increased phosphorylation as early as 2 h after exposure of NCI-H460 cells to siRNA si5, as visible by the increased intensity of the upper band in the top panel. For cdc25B the nuclear levels of the phophatase decrease in response to activation of TAK1-D. Cdc25C does not seem to participate in these signaling events, as all cdc25C protein can be found in the cytoplasm. The lower two panels show immunoblots of control protein for the subcellular fractionation.

Having demonstrated the sequence-specific activation of TAK1-D in a human nonsmall cell lung cancer cell line it is interesting to ask if this activation also occurs in other cell types. It has been reported (Dempsey et al. 2000) that the splicing variant D of TAK1 could not be detected in many cell lines despite its expression in the parental tissue, most probably due to down-regulation of the transcript during in vitro culture. Therefore, we used macrophages 24 h after obtaining them from athymic NCR-NU/NU female mice by peritoneal lavage. Figure 5C shows that p38 is activated by phosphorylation 4 h after treatment with siRNA si5 but not siRNA si6. This indicates that the sequence-specific activation of TAK1-D by short dsRNAs appears to be a general phenomenon occurring in various mammalian cell types.

P38 MAPK as well as SPK/JNK have a large number of downstream targets. Among these, the phosphorylation of the cell cycle control phosphatase cdc25 (Bulavin et al. 2001; Goss et al. 2003) is of particular interest, as these phosphorylations have the potential to transduce an immediate and direct effect of TAK1-D activation to the cell cycle. We looked at the subcellular distributions of cdc25 isoforms in NCI-H460 cells, which are shown in Figure 5D. Cdc25C was found to be exclusively cytoplasmic and phosphorylated on S216. This indicates that cdc25C does not take part in the classic activation of nuclear cyclin dependent kinases through removal of the inhibitory phosphorylation. In contrast, both cdc25A and cdc25B were found in the nucleus and therefore are capable of cell cycle control. The upper panel of Figure 5D shows that the intensity of the upper band recognized by the cdc25A antibody increases in intensity within 2 h after the addition of siRNA si5. The occurrence of this shifted form of cdc25A is indicative of phosphorylation and inactivation of cdc25A (Zhao et al. 2002). The amount of cdc25B in the nucleus decreases with kinetics similar to the phosphorylation of cdc25A, indicative for a phosphorylation dependent translocation and proteolytic digestion of cdc25B (Cans et al. 1999). Therefore, it can be concluded that activation of TAK1-D in NCI-H460 cells leads to an inactivation of cdc25A and cdc25B through p38 MAPK and SAPK/JNK mediated phosphorylation and proteolytic destruction.

A pronounced effect on the cell cycle can be expected from the inactivation of cdc25A and cdc25B. BrdU pulse labeling and flow cytometric analysis were used to quantify the effect of TAK1-D induced inactivation of cdc25A and cdc25B on the cell cycle. Figure 6A shows that within a few hours after transfection of NCI-H460 cells with siRNA si5 the number of cells in the S phase of the cell cycle drops to about half of the number found prior to the addition of the si5 siRNA. After 24 h DNA synthesis was found to have ceased in almost every cell. Because the fraction of cells containing a nuclear DNA amount specific for G1, S, or G2 cells did not change during the first 24 h after addition of siRNA si5, it can be concluded that activation of TAK1-D leads to an arrest of NCI-H460 cells in the G1, S, and G2 phase of the cell cycle.

FIGURE 6.

FIGURE 6.

Open in a new tab

Reaction of NCI-H460 cells to activation of p38 MAPK and SAPK/JNK. (A) BrdU labeling was used to determine the amount of NCI-H460 cells in the S phase of the cell cycle. Cells were pulse labeled with 100 μM BrdU for 15 min (0 and 4 h) or 10 μM for 1 h (24 h). Dot blots show DNA content versus BrdU signal. The number of S phase cells decreases within a few hours after treatment of cells with siRNA si5. (B) Flow cytometry dot blots, showing the cellular DNA content versus the amount of cleaved Caspase 3. Within 24 h after exposure to siRNA si5 cells in G1 as well as in the S phase of the cell cycle show Caspase 3 activation. (C) Activation of Caspase 3 leads to PARP cleavage and to (D) nuclear chromatin condensation visible in Hoechst 33342 stained NCI-H460 cells, typical of apoptosis.

Cell cycle arrest triggers programmed cell death in many tumor cell lines. As shown in Figure 6B this also applies for TAK1-D mediated cell cycle arrest in NCI-H460 cells. Twenty-four hours after treatment with siRNA, si5 cells from the G1 and the S phase of the cell cycle show caspase 3 cleavage, indicative for the activation of the apoptotic pathway. After 48 h the fraction of G1 and S phase cells staining positive for cleaved caspase 3 is higher; still no Caspase 3 cleavage could be observed in G2 phase NCI-H460 cells. The fact that NCI-H460 cells undergo classical apoptosis after TAK1-D induced cell cycle arrest is further demonstrated by the cleavage of the Caspase 3 substrate PARP shown in Figure 6C and by the occurrence of nuclear condensation and fragmentation shown in the fluorescent micrographs of Figure 6D.

DISCUSSION

Many RNA binding proteins that are currently known to activate signal transduction cascades after binding to small dsRNAs exhibit little or no specificity for the RNA sequence. For instance no direct dependence of the activation of PKR on the siRNA sequence has been established although it has been shown that modification of the ribonucleotides, which participate in binding to PKR, can modulate the activation of PKR by siRNAs (Puthenveetil et al. 2006). A significantly higher degree of sequence specificity was found for dsRNA mediated activation of the mammalian innate immune system. While immune stimulation through activation of TLR3 was found to be triggered by the unphysiologic dsRNA poly(I:C) (Alexopoulou et al. 2001), it was later discovered that activation of TLR7 is preferentially triggered by siRNA containing the sequence motif GUGU (Judge et al. 2005) or by even longer defined RNA sequences (Hornung et al. 2005).

If one considers the data presented in Figures 3 and 4 it can be concluded that activation of TAK1-D depends strongly on the sequence of the dsRNA. Nevertheless, the lack of TAK1-D activation by single-stranded RNA does not rule out an effect of these molecules due to the lower stability of single-stranded RNA under cell culture conditions. While the sequence in the center of the molecule seems to be less important when si5 is compared to si10, it is interesting to note that insertions or deletions in that region of the molecule abolish its ability to activate TAK1-D. As the length of the molecule does not seem to be of special importance, which is demonstrated by the fact that omission of the dT tails does not reduce TAK1-D activation, it can be assumed that the distance between the 5′ and 3′ sequence motifs of si5 is important. The strong dependence of the effect on the three-dimensional structure of the molecule is further highlighted by the fact that 2′-O-Me modifications in the sense strand, but not in the antisense strand, inhibit dsRNA mediated activation of the stress induced protein kinase pathway.

Four splicing variants of TAK1 have been described (Dempsey et al. 2000). Publications dealing with the function of TAK1 describe almost exclusively the properties of the long splicing variants A and B. B differs from A by a deletion of 28 amino acids due to alternative splicing of exon 12 (Dempsey et al. 2000). The short splicing variants C and D lack 99 amino acids of the C-terminal sequence because a frame shift caused by alternative splicing of exon 16 leads to truncation of the protein (Dempsey et al. 2000). Most importantly, the deleted C-terminal sequence in splice variants C and D has been demonstrated to be required for the activation of the kinase through upstream signaling molecules like TAB2 or TAB3 (Besse et al. 2007). Therefore, it can be assumed that the short splicing variants of TAK1 are activated by mechanisms different from the well-established upstream pathways of TAK1-A and TAK1-B reviewed recently (Delaney and Mlodzik 2006). To the best of our knowledge the siRNA triggered activation of TAK1-D described here is the first function demonstrated for the short splicing variants of TAK1. Given the fact that TAK1-D has been reported to be expressed in many human tissues like brain, heart, liver, lung, skeletal muscle, prostate, and peripheral blood mononuclear cells (Dempsey et al. 2000), the biologic functions TAK1-D participates in might be important.

Interestingly, we did not observe an activation of TAK1-C, which differs from TAK1-D by the presence of 28 additional amino acids coded by exon 12 (Dempsey et al. 2000), although we found this splicing variant to be expressed on the mRNA level in NCI-H460 cells. This indicates that the activation of TAK1 by small dsRNAs depends on structural features of the C-terminal part of the molecule specific for splice variant D.

In vitro studies of TAK1 activation, like the ones presented here, are complicated by the fact that recombinant TAK1 expressed in mammalian cells is catalytically inactive (Sakurai et al. 2000). To gain kinase activity TAK1 has to be coexpressed with its binding protein TAB1 (Sakurai et al. 2000), or a fusion protein consisting of the N-terminal part of TAK1 and the C-terminal sequence of TAB1 (Sakurai et al. 2002) has to be used. The crystal structure of such a fusion protein explained the requirement of TAK1 for TAB1 to gain catalytic activity (Brown et al. 2005). Unfortunately coexpression with TAB1 is sufficient to activate TAK1, requiring no other cofactors (Sakurai et al. 2000). Therefore, the observed increased autophosphorylation activity induced by siRNA si5 in our in vitro system indicates a very significant activation of the enzyme, comparable to the one found in intact cells.

Having demonstrated that short dsRNAs of a specific nucleotide sequence activate splicing variant D of TAK1 and that this activation leads to cell cycle arrest and apoptosis in a nonsmall cell lung cancer cell line two consequences of these findings can be envisioned:

  • The activation of TAK1-D offers an explanation for off-target effects of siRNAs, especially when mixed populations of molecules are used (Kettner-Buhrow et al. 2006), and suggests the requirement for a very careful interpretation of experiments using siRNA.

  • If the effect of siRNA si5 on NCI-H460 cells is considered, it is tempting to assume it offers a treatment option, at least for a subset of human cancers.

MATERIALS AND METHODS

siRNA molecules

The 2′-O-Me modified RNAs were synthesized by Sigma-Proligo. All other RNA molecules were purchased from Ambion. The sequences are shown in Figures 3 and 4. For transfection the siRNAs were diluted in serum free cell culture medium and mixed with siLentFect (Bio-Rad) to give a siRNA concentration of 118 nM and a siLentFect concentration of 0.8% (v/v). After a 20-min incubation at room temperature the mixture was added to the cell culture, giving a final siRNA concentration of 20 nM.

Cell culture

NCI-H460 cells were cultured in RPMI1640 medium, HeLa cells and mouse peritoneal macrophages were grown in Alpha-MEM. The medium was supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Cells were cultured at 37°C in a humidified atmosphere containing 5% CO2.

Western blotting

Total cellular proteins were solubilized by boiling the cells in nonreducing sample buffer (100 mM Tris at pH6.8, 2% SDS, 10% glycerol). Cytoplasmic and nuclear fractions were prepared as described previously (Dyer and Herzog 1995). After we determined the protein concentration using the DC Protein Assay kit (Bio-Rad), samples were reduced by adding DDT to a final concentration of 100 mM. For each sample 25 μg of protein were loaded into a SDS-PAGE gel. After the run proteins were electro-transfered to a PVDF membrane. After blocking unspecific binding sites of the membrane with 5% nonfat dry milk in TBST, membranes were incubated with the primary antibody. After three washes in TBST the membranes were incubated with the peroxidase conjugated secondary antibody, washed again three times in TBST, and the signal detected using an ECL Plus Western Blotting Detection System (Amersham). The following primary antibodies were used: pTAK1(T184/T187) (#4508), pMKK3/6(S189/S207) (#9236), pMKK4 (#9156), pIKKα/β(S176/S180) (#2697), pp38(T180/Y182) (#9211), p38 (#2912), pSAPK/JNK(T183/Y185) (#9251), SAPK/JNP (#9252), Cleaved Caspase-3 (#9664), and PARP (#9542) (Cell Signal Technology); TAK1(A and B) (#sc-7967), CDC25C (#sc-13138), and Lamin B (#sc-20682) (Santa Cruz Biotechnology); CDC25A (#05-743) (Upstate); TAK1-D (#AHP969) (AbD Serotec); βTubulin (#T-4026) (Sigma).

Production of recombinant TAK1-D and TAB1

The coding sequence of TAK1-D cDNA was cloned into the expression vector pFLAG-CMV-4. From this vector TAK1-D fused to the FLAG tag at its N terminus could be expressed under the control of the CMV promoter. For the TAB1 expression construct the coding sequence of TAB1 was cloned into the vector pcDNA3.1/His. This construct coded for TAB1 fused to a 6×His tag at the N terminus of TAB1. For protein expression Hela cells in a 10-cm petri dish were transfected with 12 μg pFLAG-CMV4-TAK1D and 12 μg pcDNA1.1/His/TAB1. Twenty-four hours after transfection the cells were lyzed in 1 mL of a buffer containing 20 mM Tris (pH 7.5), 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM sodium orthovanadate, 1 mM NaF, and 1 mM phenylmethanesulfonylfluoride. After removal of the insoluble material by centrifugation (10 min, 16,000g) FLAG-TAK1-D and the bound His-TAB1 were immunoprecipitated with agarose conjugated anti-FLAG M2 antibody (Sigma) at 4°C for 2 h. After washing the resin three times in cell lysis buffer, proteins were eluted by incubating the resin for 30 min at 4°C with Triton free lysis buffer containing 150 μg/mL 3×FLAG peptide (Sigma).

Kinase assay

TAK1-D kinase assays were performed similarly to previously published protocols (Kishimoto et al. 2000). Briefly 10 μL reactions were set up containing about 70 ng FLAG-TAK1-D, 70 ng His-TAB1, and 1 μg bacterially expressed Mal-E tagged MKK6 (Upstate) in kinase buffer (20 mM Tris at pH 7.5, 10 mM MgCl2, 1 mM DTT). siRNAs were added to the reaction mixture at a final concentration of 2.5 μM. The kinase reaction was started by adding 185 kBq of [γ-32P]ATP. The reaction was stopped by adding SDS-PAGE sample buffer, and the proteins were separated by SDS-PAGE and transfered to a PVDF membrane as described above. Radioactivity incorporated into the proteins was determined by using a phosphorimager (Typhoon 9410, Amersham), and the proteins were stained with Ponceau S (Sigma).

RNA binding assay

We radiolabeled 150 pmol of siRNA with 370 kBq [γ-32P]ATP using T4 polynucleotide kinase. Unincorporated ATP was removed by purification of the RNA through NucAway spin columns (Ambion). FLAG tagged TAK1-D was expressed in HeLa cells and purified as described above with the only difference that Dynabead protein A (Invitrogen) conjugated FLAG antibody was used instead of the agarose conjugate. The antibody bound FLAG-TAK1-D was washed once with the buffer used for the kinase assay and then incubated with ∼18 pmol radiolabeled RNA for 10 min at room temperature in the same buffer. After two washes in kinase buffer the bound radioactivity was determined by liquid scintillation counting. The measured activity was corrected for unspecific binding by subtracting the activity bound to antibody conjugated beads.

Flow cytometry

For the detection of cleaved Caspase 3, cells were trypsinized, combined with already detached cells, and washed once in PBS. Thereafter cells were fixed in 3.7% formaldehyde for 10 min at 37°C. After permeabilization in 90% methanol at −20°C, cells were incubated for 10 min. at room temperature in PBS containing 0.5% BSA and 0.5 mg/mL ribonuclease A. Cells were incubated with primary antibody directed against cleaved Caspase 3 in a 1:100 dilution in PBS containing 0.5% BSA for 1 h. After one wash the cells were incubated with FITC conjugated secondary antibody (Jackson ImmunoResearch) for 30 min at room temperature in PBS containing 0.5% BSA. After one wash cells were resuspended in 0.5 mL PBS containing 50 μg/mL Probidium iodide and analyzed on a flow cytometer.

For the BrdU labeling experiments cells were processed using a FITC BrdU Flow Kit (BD) according to the instructions given by the manufacturer.

ACKNOWLEDGMENTS

We thank D. Saha for the mouse peritoneal macrophages and C. Chat for her support. Part of this work was supported by grant number P17200-B09 from the Austrian Science Fund (FWF) to R.K. and with New Investigator funding from the Department of Radiation Oncology to M.D.S.

Footnotes

Article published online ahead of print. Article and publication date are at http://www.rnajournal.org/cgi/doi/10.1261/rna.738508.

REFERENCES

  1. Alexopoulou, L., Holt, A., Medzhitov, R., Flavell, R. Recognition of double-stranded RNA and activation of NF-κB by Toll-like receptor 3. Nature. 2001;413:732–738. doi: 10.1038/35099560. [DOI] [PubMed] [Google Scholar]
  2. Besse, A., Lamothe, B., Campos, A., Webster, W., Maddineni, U., Lin, S., Wu, H., Darnay, B. TAK1-dependent signaling requires functional interaction with TAB2/TAB3. J. Biol. Chem. 2007;282:3918–3928. doi: 10.1074/jbc.M608867200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brown, K., Vial, S., Dedi, N., Long, J., Dunster, N., Cheetham, G. Structural basis for the interaction of TAK1 kinase with its activating protein TAB1. J. Mol. Biol. 2005;354:1013–1020. doi: 10.1016/j.jmb.2005.09.098. [DOI] [PubMed] [Google Scholar]
  4. Bulavin, D., Higashimoto, Y., Popoff, I., Gaarde, W., Basrur, V., Potapova, O., Appella, E., Fornace, A.J. Initiation of a G2/M checkpoint after ultraviolet radiation requires p38 kinase. Nature. 2001;411:102–107. doi: 10.1038/35075107. [DOI] [PubMed] [Google Scholar]
  5. Cans, C., Ducommun, B., Baldin, V. Proteasome-dependent degradation of human CDC25B phosphatase. Mol. Biol. Rep. 1999;26:53–57. doi: 10.1023/a:1006912105352. [DOI] [PubMed] [Google Scholar]
  6. de Haro, C., Mendez, R., Santoyo, J. The eIF-2α kinases and the control of protein synthesis. FASEB J. 1996;10:1378–1387. doi: 10.1096/fasebj.10.12.8903508. [DOI] [PubMed] [Google Scholar]
  7. Delaney, J., Mlodzik, M. TGF-β activated kinase-1: New insights into the diverse roles of TAK1 in development and immunity. Cell Cycle. 2006;5:2852–2855. doi: 10.4161/cc.5.24.3558. [DOI] [PubMed] [Google Scholar]
  8. Dempsey, C., Sakurai, H., Sugita, T., Guesdon, F. Alternative splicing and gene structure of the transforming growth factor β-activated kinase 1. Biochim. Biophys. Acta. 2000;1517:46–52. doi: 10.1016/s0167-4781(00)00258-x. [DOI] [PubMed] [Google Scholar]
  9. Dyer, R., Herzog, N. Isolation of intact nuclei for nuclear extract preparation from a fragile B-lymphocyte cell line. Biotechniques. 1995;19:192–195. [PubMed] [Google Scholar]
  10. Elbashir, S., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl, T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature. 2001;411:494–498. doi: 10.1038/35078107. [DOI] [PubMed] [Google Scholar]
  11. Fire, A., Xu, S., Montgomery, M., Kostas, S., Driver, S., Mello, C. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans . Nature. 1998;391:806–811. doi: 10.1038/35888. [DOI] [PubMed] [Google Scholar]
  12. Goss, V., Cross, J., Ma, K., Qian, Y., Mola, P., Templeton, D. SAPK/JNK regulates cdc2/cyclin B kinase through phosphorylation and inhibition of cdc25c. Cell. Signal. 2003;15:709–718. doi: 10.1016/s0898-6568(03)00009-3. [DOI] [PubMed] [Google Scholar]
  13. Hornung, V., Guenthner-Biller, M., Bourquin, C., Ablasser, A., Schlee, M., Uematsu, S., Noronha, A., Manoharan, M., Akira, S., de Fougerolles, A., et al. Sequence-specific potent induction of IFN-α by short interfering RNA in plasmacytoid dendritic cells through TLR7. Nat. Med. 2005;11:263–270. doi: 10.1038/nm1191. [DOI] [PubMed] [Google Scholar]
  14. Judge, A., Sood, V., Shaw, J., Fang, D., McClintock, K., MacLachlan, I. Sequence-dependent stimulation of the mammalian innate immune response by synthetic siRNA. Nat. Biotechnol. 2005;23:457–462. doi: 10.1038/nbt1081. [DOI] [PubMed] [Google Scholar]
  15. Judge, A., Bola, G., Lee, A., MacLachlan, I. Design of noninflammatory synthetic siRNA mediating potent gene silencing in vivo. Mol. Ther. 2006;13:494–505. doi: 10.1016/j.ymthe.2005.11.002. [DOI] [PubMed] [Google Scholar]
  16. Kettner-Buhrow, D., Dittrich-Breiholz, O., Schneider, H., Wolter, S., Resch, K., Kracht, M. Small interfering RNAs generated by recombinant dicer induce inflammatory gene expression independent from the TAK1-NFκB-MAPK signaling pathways. Biochem. Biophys. Res. Commun. 2006;347:566–573. doi: 10.1016/j.bbrc.2006.06.121. [DOI] [PubMed] [Google Scholar]
  17. Kishimoto, K., Matsumoto, K., Ninomiya-Tsuji, J. TAK1 mitogen-activated protein kinase kinase kinase is activated by autophosphorylation within its activation loop. J. Biol. Chem. 2000;275:7359–7364. doi: 10.1074/jbc.275.10.7359. [DOI] [PubMed] [Google Scholar]
  18. Ninomiya-Tsuji, J., Kishimoto, K., Hiyama, A., Inoue, J., Cao, Z., Matsumoto, K. The kinase TAK1 can activate the NIK-I κB as well as the MAP kinase cascade in the IL-1 signalling pathway. Nature. 1999;398:252–256. doi: 10.1038/18465. [DOI] [PubMed] [Google Scholar]
  19. Player, M., Torrence, P. The 2-5A system: Modulation of viral and cellular processes through acceleration of RNA degradation. Pharmacol. Ther. 1998;78:55–113. doi: 10.1016/S0163-7258(97)00167-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Puthenveetil, S., Whitby, L., Ren, J., Kelnar, K., Krebs, J., Beal, P. Controlling activation of the RNA-dependent protein kinase by siRNAs using site-specific chemical modification. Nucleic Acids Res. 2006;34:4900–4911. doi: 10.1093/nar/gkl464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Sakurai, H., Miyoshi, H., Mizukami, J., Sugita, T. Phosphorylation-dependent activation of TAK1 mitogen-activated protein kinase kinase kinase by TAB1. FEBS Lett. 2000;474:141–145. doi: 10.1016/s0014-5793(00)01588-x. [DOI] [PubMed] [Google Scholar]
  22. Sakurai, H., Nishi, A., Sato, N., Mizukami, J., Miyoshi, H., Sugita, T. TAK1-TAB1 fusion protein: A novel constitutively active mitogen-activated protein kinase kinase kinase that stimulates AP-1 and NF-κB signaling pathways. Biochem. Biophys. Res. Commun. 2002;297:1277–1281. doi: 10.1016/s0006-291x(02)02379-3. [DOI] [PubMed] [Google Scholar]
  23. Scacheri, P., Rozenblatt-Rosen, O., Caplen, N., Wolfsberg, T., Umayam, L., Lee, J., Hughes, C., Shanmugam, K., Bhattacharjee, A., Meyerson, M., et al. Short interfering RNAs can induce unexpected and divergent changes in the levels of untargeted proteins in mammalian cells. Proc. Natl. Acad. Sci. 2004;101:1892–1897. doi: 10.1073/pnas.0308698100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Singhirunnusorn, P., Suzuki, S., Kawasaki, N., Saiki, I., Sakurai, H. Critical roles of threonine 187 phosphorylation in cellular stress-induced rapid and transient activation of transforming growth factor-β-activated kinase 1 (TAK1) in a signaling complex containing TAK1-binding protein TAB1 and TAB2. J. Biol. Chem. 2005;280:7359–7368. doi: 10.1074/jbc.M407537200. [DOI] [PubMed] [Google Scholar]
  25. Sledz, C., Holko, M., de Veer, M., Silverman, R., Williams, B. Activation of the interferon system by short-interfering RNAs. Nat. Cell Biol. 2003;5:834–839. doi: 10.1038/ncb1038. [DOI] [PubMed] [Google Scholar]
  26. Wall, N., Shi, Y. Small RNA: Can RNA interference be exploited for therapy? Lancet. 2003;362:1401–1403. doi: 10.1016/S0140-6736(03)14637-5. [DOI] [PubMed] [Google Scholar]
  27. Zhao, H., Watkins, J., Piwnica-Worms, H. Disruption of the checkpoint kinase 1/cell division cycle 25A pathway abrogates ionizing radiation-induced S and G2 checkpoints. Proc. Natl. Acad. Sci. 2002;99:14795–14800. doi: 10.1073/pnas.182557299. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from RNA are provided here courtesy of The RNA Society

RESOURCES