CK2-mediated TEL2 phosphorylation augments nonsense-mediated mRNA decay (NMD) by increase of SMG1 stability

Biochimica et Biophysica Acta 1829 (2013) 1047–1055

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbagrm

CK2-mediated TEL2 phosphorylation augments nonsense-mediated mRNA decay (NMD) by increase of SMG1 stability☆ Seyoung Ahn, Jinyoung Kim, Jungwook Hwang ⁎ Graduate School for Biomedical Science and Engineering, Hanyang University, Seoul 133-791, South Korea

a r t i c l e

i n f o

Article history: Received 6 March 2013 Received in revised form 13 June 2013 Accepted 21 June 2013 Available online 3 July 2013 Keywords: NMD UPF1 SMG1 TEL2 phosphorylation CK2 mRNP remodeling

a b s t r a c t Nonsense-mediated mRNA decay (NMD) is the best-characterized mRNA surveillance mechanism that degrades a premature-termination codon (PTC)-containing mRNA. During mammalian NMD, SMG1 and UPF1, key proteins in NMD, join at a PTC and form an SMG1–UPF1–eRF1–eRF3 (SURF) complex by binding UPF1 to eRF3 after PTC-recognition by the translating ribosome. Subsequently, UPF1 is phosphorylated after UPF1–SMG1 moves onto the downstream exon junction complex (EJC). However, the cellular events that induce UPF1 and SMG1 complex formation and increase NMD efficiency before PTC recognition remain unclear. Here, we show that telomere-maintenance 2 (TEL2) phosphorylation by casein-kinase 2 (CK2) increases SMG1 stability, which increases UPF1 phosphorylation and, ultimately, augments NMD. Inhibition of CK2 activity or downregulation of TEL2 impairs NMD. Intriguingly, loss of TEL2 phosphorylation reduces UPF1-bound PTC-containing mRNA and the formation of the SMG1–UPF1 complex. Thus, our results identify a new function of CK2-mediated TEL2 phosphorylation in a mammalian NMD. © 2013 The Authors. Published by Elsevier B.V. All rights reserved.

1. Introduction Nonsense-mediated mRNA decay (NMD) is a translation-dependent mRNA surveillance mechanism that eliminates not only eukaryotic mRNA harboring premature termination codons (PTC) that occur by frameshift, alternative splicing, and point mutations, but also mRNA that has a long 3′UTR, upstream ORF, or intron in the 3′UTR. This illustrates that NMD targets PTC-containing mRNA as well as natural mRNA (Refs. [1–5] and references therein). Bioinformatics studies have elucidated that around one-third of alternatively spliced mRNAs are subjected to NMD [6] and around one-third of disease-associated mutations are due to PTC-containing mRNAs [7]. Because NMD controls translation for a variety of genes, NMD malfunction or escape can cause several human diseases by production of C-terminally truncated proteins (Ref. [8] and references therein). Mammalian NMD activity is enhanced by the presence of an exon-junction complex (EJC) downstream of a PTC. EJC-dependent NMD normally requires: (1) cap-bound heterodimeric protein (CBP) 20-CBP80, (2) at least one EJC downstream of a PTC, and (3) a PTC more than ~ 55–50 nucleotides upstream of an exon-exon junction. ☆ This is an open-access article distributed under the terms of the Creative Commons Attribution–Noncommercial–No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited. ⁎ Corresponding author at: Graduate School for Biomedical Science and Engineering, Hanyang University, FTC 1209-15, Seoul 133-791, South Korea. Tel.: +82 2 2220 2427; fax: +82 2 2220 2422. E-mail address: [email protected] (J. Hwang).

When a translating ribosome stops at a PTC, up-frameshift factor (UPF) 1 and suppressor with morphological effect on genitalia (SMG) 1 join heterodimeric eukaryotic translation release factor (eRF) 1 and eRF3 to form the SMG1–eRF1–eRF3–UPF1 complex (SURF), by the help of CBP80–UPF1 interaction [1,9], presumably while UPF1 inhibits PABPC1–eRF3 interaction [10]. After SURF formation, the UPF1–SMG1 complex moves to the downstream EJC, resulting in UPF1 phosphorylation by SMG1 [9]. SMG1 belongs to the phosphatidylinositol 3-kinase (PI3K)-related protein kinases (PIKK) family and phosphorylates serine and threonine [11]. SMG1-mediated UPF1 phosphorylation occurs at two distinguished sites. One is N-terminal threonine 28 that is supposed to be bound by endonuclease SMG6, which replaces the UPF3–EJC interaction leading to cleavage of NMD targets [5,12,13]. The other site is at four C-terminal serines (1073, 1078, 1096 and 1116), which are supposed to be bound by the SMG5/7 complex, leading to UPF1 dephosphorylation [9,13]. A mammalian telomere-maintenance 2 (TEL2) (also known as TELO2 or hCLK2), the mammalian ortholog of the yeast TEL2 gene, has pleiotropic phenotypes including the response to replication stress [14–17]. Recent studies have revealed that casein kinase 2 (CK2) phosphorylates TEL2, which plays a role in the stability of the PIKK family proteins including SMG1, the mammalian target of rapamycin (mTOR), ataxia telangiectasia mutated (ATM) and ATM- and Rad3-related genes (ATR) [18,19]. In a mammalian cell, phosphorylated TEL2 forms a complex with SMG10, RuvB-like (RUVBL) 1 and RUVBL2 (also known as Pontin and Reptin, respectively), and SMG1, all of which are required for mammalian NMD [18,20,21], suggesting that the phosphorylated TEL2 complex may be involved in NMD. In addition, mutation of two constitutive

1874-9399/$ – see front matter © 2013 The Authors. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbagrm.2013.06.002

1048

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

CK2-mediated phosphorylation sites leads to reduction of SMG1 with TEL2 and instability of SMG1 in both human cells and mouse models [18], suggesting that there may be a molecular mechanism by which TEL2 phosphorylation induces NMD via increase in SMG1 stability. In this study, we provide several lines of evidence demonstrating that TEL2 phosphorylation by CK2 upregulates SMG1 stability and increases UPF1 phosphorylation, which ultimately augments NMD. Inhibition of CK2 activity or downregulation of endogenous CK2 or TEL2 impairs exogenous as well as endogenous NMD targets. We also show that in a mammalian cell, phosphorylation of TEL2 upregulates complex formation with SMG1–UPF1 and increases joining of UPF1 to PTC-containing mRNA. All of these results suggest that CK2-mediated TEL2 phosphorylation coordinates NMD activity and may serve to direct mRNP remodeling during NMD. 2. Materials and methods 2.1. Cell culture, plasmid and siRNA transfection, cell lysis and drug treatments HeLa cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin/ streptomycin at 37 °C with 5% CO2. Transient transfections of DNA plasmid and siRNA were done using Lipofectamine 2000 (Invitrogen) and Oligofectamine (Invitrogen), respectively. Two days after siRNA transfection, cells were re-transfected with a pmCMV-Gl (β-globin) test plasmid [22], either nonsense-free (Norm) or PTC-harboring (Ter), a pmCMV-GPx1 (glutathione peroxidase 1) test plasmid [23], and a phCMV-MUP (mouse major urinary protein) reference plasmid [24]. When specified, pCI-Neo–FLAG–UPF1 [25] was also transfected. Total-cell lysates were obtained by hypotonic solution containing 10 mM Tris–Cl pH 7.5, 10 mM NaCl, 10 mM EDTA, and 0.5% Triton X-100 followed by a 10-min incubation on ice as done in the previous report [26]. Where specified, cells were treated with 100 μg/ml cycloheximide for 4 h, 10 μM Wortmannin for 4 h, or 75 μM 4,5,6,7tetrabromo-benzimidazole (TBB) for 12 h. 2.2. Plasmid constructions To construct N-terminal FLAG tagged pFLAG–TEL2-WT and pFLAG– TEL2-2A, pFLAG (Sigma) was digested with NotI and XbaI and the resulting vector fragment was ligated to a PCR-amplified product that had also been digested with NotI and XbaI. The PCR product was generated using pDEST–FLAG–TEL2-WT or 2A [18] and two primers: 5′-CG CGCGGCCGCGAGCCAGCACCCTCAGAGGTTCG-3′ (sense) and 5′-GCGTC TAGATTACTAGGGAGACGCGGGGGTGAGG-3′ (antisense), where the underlined nucleotides were recognized by NotI and XbaI, respectively. 2.3. siRNA For downregulation of endogenous UPF1, SMG1 and CK2, siRNA sequences were employed as previously described [9,27,28]. For control siRNA and downregulation of endogenous TEL2, the following commercial siRNAs were used: control siRNA (cat. no. SN-1003, Bioneer, Korea) and TEL2 siRNA (cat. no. 1078794, Bioneer, Korea), respectively. 2.4. Immunoprecipitation (IP) and Western blotting (WB) Cell lysates for the analysis of co-immunoprecipitated protein and mRNA were generated before and after IP as previously reported [26]. In order to remove nonspecific protein–protein interaction, 150 mM NaCl of cell lysate was used in IP using anti-UPF1, anti-FLAG or antiMYC antibody. Proteins were electrophoresed in 6–10% polyacrylamide and transferred to a PVDF membrane (Millipore). The following antibodies were used: FLAG (Sigma), UPF1 (Cell signaling), SMG1 (Cell signaling), MYC (Calbiochem), phosphorylated UPF1 (P-UPF1) [29],

TEL2 [20], phosphorylated TEL2 (P-TEL2) [18], CK2α (Santa Cruz Biotechnology) and β-actin (Santa Cruz Biotechnology). Antibodies were diluted 1:200–1:5000 in 5% non-fat dry milk, 0.1% Tween 20/TBS and incubated overnight at 4 °C with rocking. The protein–antibody complexes were visualized using an ECL reagent (Thermo). 2.5. Semiquantitative and real-time quantitative RT-PCR Total RNA was extracted using Trizol (Invitrogen) according to manufacturer's protocol. To eliminate possible exogenous or endogenous DNA contamination, RNA was further treated with RQ DNaseI (Promega) for 30 min at 37 °C followed by RNA extraction. The RNA was reverse-transcribed using random hexamers (Invitrogen) and reverse-transcribed cDNA was amplified by semiquantitative or quantitative RT-PCR. For semiquantitative RT-PCR (RT-sqPCR), β-Gl, GPx1, MUP, TBL2, GADD45B and GAPDH mRNAs were amplified as previously described [26,30,31]. PCR results were analyzed in 1% agarose gel using ethidium bromide staining. Real-time quantitative RT-PCR (RT-qPCR) was used to quantitate the level of Gl, GPx1, MUP, TBL2, GADD45B and GAPDH mRNA using CFX96 (Bio-Rad) and Fast SYBR Green Master Mix (Applied Biosystems). The primers used to amplify Gl, GPx1 mRNA were previously described [26]. The primers used to amplify MUP, TBL2, GADD45B and GAPDH mRNAs were the same as the primer pairs used in RT-sqPCR. 3. Results 3.1. Inhibition of CK2 activity reduces NMD efficiency CK2 phosphorylates two TEL2 serines, serine 487 and serine 491, which increases SMG1 stability [18]. To corroborate the theory that TEL2 phosphorylation by CK2 augments NMD through an increase in SMG1 stability and UPF1 phosphorylation, HeLa cells were transiently transfected with exogenous NMD test and reference plasmids (pmCMV-Gl and pmCMV-GPx1 test plasmids – either pmCMV-Gl Norm and pmCMV-GPx1 Norm, both of which does not contain a PTC, or pmCMV-Gl Ter and pmCMV-GPx1 Ter, both of which contain a PTC [22,23] – and the phCMV-MUP reference plasmid [24]) followed by treatment with the specified inhibitors, including cycloheximide (CHX, a protein synthesis inhibitor by arresting translating ribosomes), Wortmannin (Wort, canonical SMG1 inhibitor, preventing UPF1 phosphorylation) – CHX and Wort are known for NMD inhibitors [32–34] – or TBB (CK2 inhibitor [18]). Western blotting (WB) using an antiP-UPF1 antibody (which detects SMG1-mediated phosphoserine 1078 of UPF1) and anti-P-TEL2 antibody (which detects CK2-mediated phosphoserines 487 and 491 of TEL2), demonstrated that inhibition of CK2 activity by TBB significantly reduced phosphorylated UPF1 (P-UPF1/UPF1) and phosphorylated TEL2 (P-TEL2/TEL2) by ~3-fold each compared to DMSO treatment, while the expressions of UPF1, TEL2 and CK2 were comparable among cell lysates treated by inhibitors (Fig. 1A). Expectedly, TBB treatment slightly reduced the amount of SMG1 to ~70% in DMSO treated cells, suggesting that a decrease in phosphorylated UPF1 may result from the decreased amount of endogenous SMG1 by inhibition of CK2 (Fig. 1A). All of the WB results suggest that inhibition of CK2 activity may be involved in NMD by reducing two key NMD proteins, SMG1 and phosphorylated UPF1. To assess the role of CK2 activity in NMD, RT-sqPCR was employed. The results revealed that TBB treatments, like CHX and Wort, upregulate Gl and GPx1 Ter mRNA compared to DMSO treatment (Fig. 1B, upper). Moreover, TBB treatment increased endogenous NMD targets, TBL2 and GADD45B (Fig. 1B, bottom) [30,35]. To determine the quantitative level of exogenous and endogenous NMD reporters, RT-qPCR was performed. The results demonstrate that TBB indeed inhibits NMD like CHX and Wort did: the level of Gl Ter and GPx1 Ter mRNA was ~2% and ~8% of normal in DMSO treated cell lysates, respectively, but

A inhibitor SMG1 (α-SMG1) Mean (%) ±S.D.

100 75 89 71 0 10 4 5

SMG1/ β-Actin UPF1 (α-UPF1)

–

inhibitor pmCMV-Gl + pmCMV-GPx1

RT-sqPCR

–

CHX Wort TBB

B

1049

Norm Ter Norm CHX Ter Norm Wort Ter Norm TBB Ter

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

Gl mRNA GPx1 mRNA MUP mRNA

TEL2 (α-TEL2) P-TEL2 (α-P-TEL2) Mean (%) ±S.D.

–

P-UPF1/UPF1

100 19 34 35 0 2 4 5

P-TEL2/TEL2

RT-sqPCR

WB

100 141 35 30 0 20 4 5

CHX Wort TBB

P-UPF1 (α-P-UPF1) Mean (%) ±S.D.

inhibitor TBL2 mRNA GADD45B mRNA GAPDH mRNA

CK2 (α-CK2) β-Actin (α-β-Actin)

C

RT-qPCR

** GPx1 mRNA/MUP mRNA (% of Norm)

Gl mRNA/MUP mRNA (% of Norm)

** 100 90 30 19

20

16

18

10 2 0 Norm Ter NormTer Norm Ter Norm Ter –

CHX

Wort

TBB

100 90 30

25

17

20

8

10 0

Norm Ter NormTer NormTer Norm Ter

inhibitor

–

CHX

* 21

20 15 10

4

5

3

1

Wort

inhibitor

58

60 50 15

11

10 5

TBB

**

70

Relative level of GADD45B mRNA

Relative level of TBL2 mRNA

25

24

6 1

0

0 –

CHX

Wort

TBB

inhibitor

–

CHX

Wort

TBB

inhibitor

Fig. 1. Inhibition of CK2 kinase activity impairs NMD. (A) HeLa cells were treated with the indicated inhibitors and cell lysates were subjected to Western blotting (WB) using the specified antibody. The four leftmost lanes analyzed 3-fold dilutions of lysates and demonstrate that WB conditions are semiquantitative. The normalized level in the presence of DMSO was set to 100%. Mean values with standard deviations (S.D.) were calculated from four independent experiments. (B) HeLa cells were transiently transfected with NMD test and reference plasmids. Semiquantitative RT-PCR (RT-sqPCR) was performed to compare relative amounts of exogenous (Top, Gl and GPx1, where MUP was employed for transfection efficiency and a loading control) and endogenous NMD substrates (TBL2 and GADD45B, where GAPDH was employed for a loading control). (C) As in (B). However, quantitative RT-PCR (RT-qPCR) was performed. The levels of TBL2 and GADD45B mRNA were normalized to GAPDH mRNA. The columns and error bars represent the mean and standard deviation of at least three independently performed experiments. Numbers above the bars indicate the mean value from the experiments. Asterisks denote statistically significant differences; *: p b 0.05, **: p b 0.01 (paired Student's t test).

~18% and ~17% of normal in the TBB treated cells. This suggests that inhibition of CK2 stabilized Gl Ter mRNA 9-fold and GPx1 Ter mRNA 2-fold (Fig. 1C). Similar to exogenous NMD targets, we found that TBB treatment upregulated TBL2 and GADD45B mRNA expression approximately 3- and 6-fold, respectively, indicating that inhibition of CK2 activity also regulates endogenous NMD targets (Fig. 1C). All together, our findings suggest that CK2 activity enhances NMD efficiency by increasing phosphorylated TEL2 and phosphorylated UPF1.

3.2. Downregulation of endogenous TEL2 and CK2 reduces NMD efficiency To rule out the possibility that the effect of TBB treatment could be due to pleiotropic effects rather than specific CK2 activity inhibition, we examined the effects of endogenous TEL2 and CK2 on NMD by employing siRNA-mediated knockdown. We transiently transfected HeLa cells with siRNA that downregulates the expression of SMG1

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

Control SMG1 UPF1 TEL2 CK2

SMG1 ( -SMG1) UPF1 ( -UPF1) P-UPF1 ( -P-UPF1) Mean (%) 100 27 72 22 28 ±S.D. 0 5 8 5 6

Gl mRNA GPx1 mRNA MUP mRNA

UPF1 TEL2 CK2

Control SMG1

-Actin ( - -Actin)

siRNA

RT-sqPCR

CK2 ( -CK2)

TBL2 mRNA GADD45B mRNA GAPDH mRNA

RT-qPCR

** GPx1 mRNA/MUP mRNA (% of Norm)

Gl mRNA/MUP mRNA (% of Norm)

** 100 90 30

27 18

20

14

13 7

10 0

100 90 60

Control

UPF1

TEL2

CK2

0 Norm Ter Norm Ter Norm Ter Norm Ter Norm Ter Control

Relative level of GADD45B mRNA

1

0.5 0

43

25

SMG1

UPF1

TEL2

CK2

siRNA

*

4

1.4

1.5

50

20

siRNA

**

2

1

SMG1

54

47

40

NormTer NormTer NormTer Norm Ter NormTer

Relative level of TBL2 mRNA

pmCMV-Gl + pmCMV-GPx1

P-UPF1/UPF1 TEL2 ( -TEL2)

C

siRNA

RT-sqPCR

siRNA

Norm CK2 Ter

B

A

WB

each (Fig. 2A). Interestingly, downregulation of TEL2 and CK2 reduced SMG1 expression, indicating that TEL2 and CK2 are necessary for SMG1 stability as previously reported [18] (Fig. 2A). These WB results suggest that NMD may be repressed by reduced amount of endogenous TEL2 and CK2. To determine the role of endogenous TEL2 and CK2 on NMD, NMD target mRNAs were analyzed by RT-PCR. RT-sqPCR (Fig. 2B) and RT-qPCR (Fig. 2C) demonstrated that downregulating TEL2 reduced NMD of PTC-containing Gl and GPx1 mRNA by approximately 2-fold. Consistent with what was found in exogenous NMD test plasmids,

Norm UPF1 Ter Norm TEL2 Ter

and UPF1 (each of which are essential for mammalian NMD), TEL2 or CK2, or a nonspecific control siRNA. This was followed by transfection of exogenous NMD test and reference plasmids. The amount of protein detected by WB (normalized to the abundance of β-actin to control for variations in protein loading) demonstrated that specific siRNAs downregulated SMG1, UPF1, TEL2 and CK2 to approximately 20%, 20%, 30% and 30%, respectively, of the abundance found in the presence of control siRNA (Fig. 2A). Similar to reduced phosphorylated UPF1 by TBB treatment in Fig. 1A, downregulation of TEL2 and CK2 significantly reduced phosphorylated UPF1 by approximately 4-fold

Norm Control Ter Norm SMG1 Ter

1050

3.2 3 2 1

1

0 Control

TEL2

siRNA

Control

TEL2

siRNA

Fig. 2. Knockdown of TEL2 and CK2 abrogates NMD. (A) HeLa cells were transiently transfected with the indicated siRNAs. The normalized level in the presence of control siRNA was set to 100%. Mean values with standard deviations (S.D.) were calculated from four independent experiments. (B) As in (A). However, 2 days after siRNA transfection, cells were transiently transfected with NMD test and reference plasmids. Semiquantitative RT-PCR (RT-sqPCR) was performed. (C) As in (B). The levels of TBL2 and GADD45B mRNA were normalized to GAPDH mRNA. However, quantitative RT-PCR (RT-qPCR) was performed. The columns and error bars represent the mean and standard deviation of at least three independently performed experiments. Numbers above the bars indicate the mean value from the experiments. Asterisks denote statistically significant differences; *: p b 0.05, **: p b 0.01 (paired Student's t test).

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

3.3. NMD targets are upregulated by expression of unphosphorylated TEL2

pFLAG-

–

TEL2-WT TEL2-2A TEL2-WT TEL2-WT TEL2-2A

FLAG-TEL2 ( -FLAG)

-UPF1

Since downregulation of TEL2 or CK2 inhibits TEL2 phosphorylation and reduces SMG1-mediated UPF1 phosphorylation, expressing unphosphorylated TEL2 was expected to inhibit NMD. To test this, HeLa cells were transiently co-transfected with FLAG–TEL2-WT, FLAG–TEL2-2A (where two constitutive CK2-mediated phosphosites of TEL2 were exchanged with two alanines [18]) or, as a negative control, FLAG alone with NMD test and reference plasmids. FLAG– TEL2-WT and 2A proteins were expressed at similar levels that were approximately 3-fold above the level of endogenous TEL2 (Fig. 3A, left panel). As published in a previous report [18], the expression of unphosphorylated TEL2 (TEL2-2A) reduced endogenous SMG1 by

rIgG

–

TEL2-WT TEL2-2A

downregulating TEL2 upregulated endogenous NMD targets, TBL2 and GADD45B, by approximately 1.4- and 3-fold (Fig. 2C). We conclude that NMD probably involves CK2 and TEL2, and TEL2 phosphorylation. However, it is possible that downregulation of TEL2 indirectly inhibits NMD. To address this, we compared the stability of PTC-free and PTC-containing mRNAs when TEL2 was downregulated by siRNA and, after cells were incubated for 2 days, cells were treated with RNA polymerase inhibitor, 5,6-dichloro-1-β-D-ribofuranosylbenzimidazole (DRB). When expressed as a ratio of TEL2 siRNA to control siRNA, PTCcontaining Gl and GPx1 mRNA stability approximately doubled in TEL2-depleted cells (Supplemental Fig. 1A). More importantly, the halflives of TBL2 and GADD45B were significantly increased upon downregulation of TEL2 siRNA, but not control siRNA (Supplemental Fig. 1B), suggesting that TEL2 is a regulator of NMD.

A

1051

TEL2 ( -TEL2) P-TEL2 ( -P-TEL2)

SMG1/ -Actin UPF1 ( -UPF1)

P-UPF1 ( -p(S/T)Q)

WB

WB

100109 57 0 18 3

Mean (%) - 100 47 ±S.D. - 0 5

P-UPF1 ( -P-UPF1) Mean (%) ±S.D.

100105 52 0 9 8

pFLAG-

UPF1 ( -UPF1)

SMG1 ( -SMG1) Mean (%) ±S.D.

IP

P-UPF1/UPF1 -Actin ( - -Actin)

P-UPF1/UPF1

100

*

*

*

90 10

6 5

3.3

3

0 Norm Ter Norm Ter Norm Ter – TE L2- W T TEL2-2A pFLAG-

** Relative level of TBL2 mRNA

2.0

1.0

1

0.9

–

21.6 20

12.3 9.1

10 0

Norm Ter Norm Ter Norm Ter – TE L2- W T TEL2-2A pFLAG-

*

1.7

1.5 1.0

1

1.1

0.5 0

TE L2- W T TEL2-2A pFLAG-

**

90 30

2.0

0.5 0

100

*

* 1.3

1.5

GPx1 mRNA/MUP mRNA (% of Norm)

RT-qPCR

Relative level of GADD45B mRNA

B Gl mRNA/MUP mRNA (% of Norm)

-Actin ( - -Actin)

–

TE L2- W T TEL2-2A pFLAG-

Fig. 3. Phosphorylation of TEL2 by CK2 augments NMD. (A) HeLa cells were transiently transfected with the specified pFLAG–TEL2 variants, or empty vector, pFLAG as a negative control. Western blotting (WB) was performed using the specified antibody (left panel). The four leftmost lanes analyzed 3-fold dilutions of lysates. Cell lysates were employed for immunoprecipitation (IP) using anti (α)-UPF1 antibody, not (−), or, to control for nonspecific IP, rabbit IgG (rIgG) (right panel). The normalized levels in the presence of pFLAG expression (left panel) or pFLAG–TEL2-WT overexpression (right panel) were set to 100%. Mean values with standard deviations (S.D.) were calculated from three independent experiments. (B) As in (A). However, NMD test and reference plasmids were co-transfected with the specified pFLAG–TEL2 variants. Quantitative RT-PCR (RT-qPCR) was performed. The columns and error bars represent the mean and standard deviation of at least three independently performed experiments. Numbers above the bars indicate the mean value from the experiments. Asterisks denote statistically significant differences; *: p b 0.05, **: p b 0.01 (paired Student's t test).

1052

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

GPx1 stability approximately tripled when TEL2-2A was overexpressed (Supplemental Fig. 2A). Similarly, the half-lives of TBL2 and GADD45B were increased upon overexpression of TEL2-2A, but not TEL2-WT (Supplemental Fig. 2B). All together, our findings suggest that CK2mediated TEL2 phosphorylation is a bona fide regulator of NMD.

-UPF1

Norm Norm TEL2-WT Norm TEL2-2A Norm Norm TEL2-WT Ter

– –

–

Our data indicate NMD is regulated by TEL2 phosphorylation. Since UPF1 preferentially associates with PTC-containing mRNA in a translation-dependent manner [26,36], UPF1 was expected to inefficiently associate PTC-containing mRNA when unphosphorylated TEL2 was expressed. To determine how much unphosphorylated TEL2 expression inhibits UPF1 binding to PTC-containing mRNA, HeLa cells were transiently transfected with NMD test and reference plasmids. These transfections also included either pFLAG–TEL2-WT or pFLAG–TEL2-2A. The lysates of transfected cells were obtained and immunoprecipitated using anti-UPF1 antibody or, to control for nonspecific IP, rabbit IgG (rIgG) (Fig. 4A). In addition, to exclude the possibility that the RNA IP was not specific, RT-sqPCR was performed. No Gl mRNA was detected on the rIgG IP, showing that the RNA IP was specific (Fig. 4B, left panel). RT-qPCR revealed that the level of Gl Ter mRNA before IP was approximately 5% and 14% of normal in the expression of FLAG–TEL2-WT and FLAG–TEL2-2A, respectively (Fig. 4B, right panel). Consistent with the previous reports [26], RT-qPCR demonstrated that UPF1-bound Gl Ter mRNA after IP was 198% of normal, which means UPF1 binds PTC-containing mRNA about 40 times more efficiently than does its PTC-free counterpart (Fig. 4B, right panel, compare 5% with 198%). Surprisingly, when unphosphorylated TEL2 was expressed, RT-qPCR demonstrated that FLAG–TEL2-2A expression decreased the UPF1bound PTC-containing mRNA, Gl Ter mRNA, to about 15% of normal, indicating that unphosphorylated TEL2 prevents the binding of UPF1 to PTC containing mRNA (Fig. 4B, right panel). These observations are

pFLAGpmCMV-Gl

WB

UPF1 ( -UPF1)

MUP mRNA

10

5

0 Norm Ter Norm Ter Norm Ter Norm Ter

–

TEL2-2A

Gl mRNA

15

14

TEL2-WT

pmCMV-Gl

198

TEL2-2A

pFLAG

**

*

250 200 150 100 20

TEL2-WT

rIgG

RT-qPCR IP Gl mRNA/MUP mRNA (% of Norm)

TEL2-WT TEL2-2A

–

Norm Norm Norm Norm Norm

TEL2-WT TEL2-2A

–

-UPF1

-Actin ( - -Actin)

B

RT-sqPCR

3.4. Unphosphorylated TEL2 inhibits UPF1 binding to Gl Ter mRNA

IP

Norm TEL2-2A Ter

A

rIgG

approximately 2-fold (Fig. 3A, left panel). In addition, phosphorylated UPF1 was decreased by approximately 2-fold, presumably, resulting from the decrease of SMG1 (Fig. 3A), which, however, was less than the decrease of phosphorylated UPF1 by TEL2 siRNA (compare Fig. 3A with 2A). To confirm the decrease of phosphorylated UPF1 by overexpression of FLAG–TEL2-2A, endogenous UPF1 immunoprecipitation (IP) in HeLa cell variants transiently expressing TEL2 was performed (Fig. 3A, right panel). The lysates of transfected cells were obtained and immunoprecipitated using anti-UPF1 antibody or, to control for nonspecific IP, rabbit IgG (rIgG). Lysates subjected to WB before or after IP demonstrated that none of the UPF1 proteins or β-actin (negative-control protein) were immunoprecipitated using rIgG and UPF1 antibody, respectively, indicating that the IPs were specific (Fig. 3A, right panel). The results of phosphorylated UPF1 using antip(S/T)Q antibody showed that overexpression of unphosphorylated TEL2 reduced serine/threonine phosphorylated UPF1 by approximately 2-fold. These observations suggest that phosphorylation of TEL2 by CK2 increases SMG1 and phosphorylated UPF1. Next, the effect of phosphorylation of TEL2 on NMD was examined. RT-qPCR revealed that the expression of unphosphorylated TEL2 reduced NMD by approximately 2-fold compared to the expression of normal TEL2 (Fig. 3B). Consistent with exogenous NMD test plasmids, unphosphorylated TEL2 upregulated TBL2 and GADD45B by approximately 1.3- and 1.7-fold, respectively. Unexpectedly, repression of NMD in natural targets by the expression of unphosphorylated TEL2 was not as efficient as repression by TBB treatment and downregulation of TEL2. One possible explanation is that endogenous phosphorylated TEL2 may be sufficient to enhance NMD or other TEL2 phosphorylation sites by CK2 play a role in SMG1 stability. The collective data suggest that CK2mediated TEL2 phosphorylation augments NMD. In a same way with downregulation of TEL2 by siRNA, mRNA stability was tested by employing DRB. When expressed as a ratio of TEL2-2A overexpression to TEL2-WT overexpression, PTC-containing Gl and

-UPF1

pFLAG-

IP

Fig. 4. TEL2 phosphorylation increases UPF1 binding to NMD targets. (A) HeLa cell lysates that were transiently co-transfected with the specified pFLAG–TEL2 variants and NMD test and reference plasmids were subjected to immunoprecipitation (IP) as done in Fig. 3A. (B) Semiquantitative RT-PCR (RT-sqPCR, left panel) and quantitative RT-PCR (RT-qPCR, right panel) were employed to quantitate Gl and MUP mRNAs. The columns and error bars represent the mean and standard deviation of at least three independently performed experiments. Numbers above the bars indicate the mean value from the experiments. Asterisks denote statistically significant differences; *: p b 0.05, **: p b 0.01 (paired Student's t test).

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

––

TEL2

mIgG

–

siRNA

TEL2

––

–

pCI-Neo-FLAG-

UPF1 TEL2

–

-FLAG

A

IP

siRNA FLAG-UPF1 ( -FLAG)

FLAG-UPF1 ( -FLAG) WB

1053

SMG1 ( -SMG1)

UPF1 ( -UPF1) Mean (%) - 100 45 ±S.D. - 0 6

WB

-Actin ( - -Actin)

SMG1/FLAG-UPF1 UPF2 ( -UPF2) TEL2 ( -TEL2) IgG heavy chain

( - -Actin)

–

pFLAG-

–

-MYC

pCMV-MYC-

–

TEL2-WT TEL2-2A

UPF1

mIgG

–

TEL2-WT TEL2-2A

B

TEL2-WT TEL2-2A

-Actin

SMG1 ( -SMG1) Darker exposure Mean (%) - 100 32 ±S.D. - 0 4

WB

WB

-Actin ( - -Actin)

pFLAG-

MYC-UPF1 ( -MYC)

MYC-UPF1 ( -MYC) MYC-UPF1 ( -UPF1) UPF1 FLAG-TEL2 ( -FLAG) TEL2 ( -TEL2)

IP

SMG1/MYC-UPF1 UPF2 ( -UPF2) FLAG-TEL2 ( -FLAG) IgG heavy chain -Actin

– –

–

TEL2-WT TEL2-2A

-FLAG TEL2-WT TEL2-2A

C

( - -Actin)

IP

pFLAGFLAG-TEL2 ( -FLAG) TEL2 ( -TEL2)

WB

SMG1 ( -SMG1) Darker exposure Mean (%) - 100 29 - 0 4 ±S.D.

SMG1/FLAG-TEL2 UPF1 ( -UPF1) UPF2 ( -UPF2) IgG heavy chain -Actin

( - -Actin)

Fig. 5. Phosphorylated TEL2 forms a complex with SMG1–UPF1. (A) (Left panel) HeLa cells were transiently transfected with TEL2 or control siRNA. Two days after transfection, cells were retransfected with FLAG–UPF1 followed by 2 days incubation. The relative expression of FLAG–UPF1 compared with endogenous UPF1 was observed by Western blotting (WB) (left panel). Cells lysates were employed for immunoprecipitation (IP) using anti(α)-FLAG antibody or mouse IgG (mIgG) (right panel). (B) HeLa cell lysates that were cotransfected with pFLAG, pFLAG–TEL2-WT or 2A and pCMV–MYC–UPF1. The relative expressions of MYC–UPF1 and FLAG–TEL2 variants compared to endogenous UPF1 and TEL2 were obtained by WB (left panel). IPs were performed using anti-MYC antibody or mIgG (right panel). (C) HeLa cell lysates that were transiently transfected with the specific pFLAG–TEL2-WT or 2A were employed for IPs using anti-FLAG antibody. The normalized levels in the presence of control siRNA (A) and pFLAG–TEL2-WT overexpression (B and C) were set to 100%. Mean values with standard deviations (S.D.) were calculated from three independent experiments.

1054

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055

consistent with the idea that unphosphorylated TEL2 reduces NMD efficiency by blocking UPF1 from binding to PTC-containing mRNA. 3.5. Phosphorylated TEL2 regulates the formation of UPF1–SMG1 complex Since UPF1 directly binds SMG1, preceding UPF1 phosphorylation [9], we expected that phosphorylated TEL2 might upregulate the complex with SMG1 and UPF1 by increasing SMG1 stability, which is expected to join NMD targets. To address this hypothesis, HeLa cells were transiently transfected with TEL2 or control siRNA. TEL2 siRNA reduced endogenous TEL2 to 20% of normal, and was subsequently transfected with pCI-Neo–FLAG–UPF1, where the expression of FLAG–UPF1 proteins was nearly double the level of endogenous UPF1 (Fig. 5A, left panel). Cell lysates were then subjected to before or after IP using anti-FLAG antibody. Consistent with Fig. 2A, endogenous SMG1 was reduced upon depletion of TEL2. As a result of reduced endogenous SMG1, the levels of co-immunoprecipitated SMG1 with FLAG–UPF1 decreased by approximately 2-fold (Fig. 5A, right panel). Moreover, endogenous TEL2 was co-immunoprecipitated with FLAG– UPF1, which was reduced by downregulation of TEL2 (Fig. 5A, right panel). To test whether phosphorylation of TEL2 is also involved in UPF1– SMG1 complex remodeling, HeLa cell lysates that were transiently transfected with pCMV–MYC–UPF1 and either pFLAG–TEL2-WT or 2A were employed for IP using anti-MYC antibody (Fig. 5B, right panel), where the MYC–UPF1 and FLAG–TEL2 variant proteins were expressed at similar level of endogenous UPF1 and TEL2, respectively (Fig. 5B, left panel). The inhibition of UPF1–SMG1 complex formation was also observed with a unphosphorylated TEL2-mediated decrease in the coIP of MYC–UPF1 with SMG1, but not with UPF2, suggesting that inhibition of NMD by overexpression of unphosphorylated TEL2 may result from the reduction of the SMG1–UPF1 complex (Fig. 5B, right panel). Next, the reciprocal experiments were performed. HeLa cell lysates that were transiently transfected with pFLAG, pFLAG–TEL2-WT or pFLAG–TEL2-2A were analyzed before or after IP using anti-FLAG (Fig. 5C). Results have shown that SMG1 and UPF1 was more co-immunoprecipitated with FLAG–TEL2-WT than FLAG–TEL2-2A, suggesting that phosphorylated TEL2 upregulates, presumably, the TEL2–SMG1–UPF1 complex (Fig. 5C). Furthermore, endogenous UPF2 was not observed in co-immunoprecipitates, suggesting that the TEL2–SMG1–UPF1 complex is disassembled before the SMG1–UPF1 complex moves to the UPF2-bound EJC (Fig. 5C).

mostly localizes in the cytoplasm and nuclear periphery [18], where a majority of NMD occurs, suggesting that phosphorylated TEL2 may have NMD-related function in cytoplasm besides replication. Together, all these observations have driven us to determine the function of TEL2 phosphorylation on NMD and to describe the events surrounding SMG1–UPF1 binding to PTC-containing mRNA. All of our observations in this study provide insights into the dynamics of UPF1 binding to PTC-containing mRNA, induced by CK2-mediated TEL2 phosphorylation, which stabilizes endogenous SMG1. This finding highlights the importance of TEL2 as a key regulator of SMG1–UPF1 joining to a target of PTC-harboring mRNA. The following observations support this notion: NMD is inhibited upon (1) the inhibition of CK2 activity (Fig. 1), (2) downregulation of endogenous TEL2 and CK2 (Fig. 2), and (3) overexpression of unphosphorylated TEL2 (Fig. 3). Furthermore, we elucidate that UPF1 binds PTC-containing mRNA more efficiently than its PTC-free counterpart in the presence of phosphorylated TEL2, while expression of unphosphorylated TEL2 inhibits UPF1 binding to PTCcontaining mRNA (Fig. 4.). In addition, we elucidate that phosphorylated TEL2 remodels the UPF1–SMG1 complex, presumably, which are involved in NMD. In summary, before the SMG1 and UPF1 complex join NMD targets, SMG1 stability is upregulated by the formation of the phosphorylated TEL2 complex with PIHD1 and RUVBL1 and 2. Increased endogenous SMG1 augments UPF1 phosphorylation, which ultimately increases NMD efficiency. Future studies are expected to unlock more about the upstream events triggering TEL2 complex formation, which may answer why NMD efficiencies are different among various cell lines and tissues. Acknowledgement We are grateful to Lynne Maquat for pCMV–MYC–UPF1, pCI-Neo– FLAG–UPF1, β-globin, GPx1 and MUP plasmids, Yoon Ki Kim for phosphorylated UPF1 antibody, Simon Boulton for phosphorylated TEL2 antibody, and Akio Yamashita for TEL2 antibody. This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012R1A1A1009809 to J.H.) and the grant for Medical Research Center (2011-0028261) funded by the National Research Foundation of Korea (NRF) of the Korea government (MSIP). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi. org/10.1016/j.bbagrm.2013.06.002.

4. Discussion Mammalian NMD involves many proteins: CBP80, CTIF, PNRC2, UPF1, SMG1, eRF1, eRF3, UPF2, UPF3 or UPF3X, SMG8, SMG9, RUVBL1, RUVBL2 etc. A broad spectrum of studies have reported on the molecular mechanisms after PTC recognition by a ribosome; SURF complex formation, UPF1 phosphorylation by SMG1, dephosphorylation of phosphorylated UPF1, endonuclease activity of SMG6 that binds phosphorylated UPF1, disassembly of UPF1-bound complex on the EJC and exonuclease activity degrading NMD targets (see recent reviews [1,3,37,38]). In contrast to various mechanistic studies about NMD after PTC recognition, few lines of evidence about the events before PTC recognition have shown that SMG1 directly interacts with SMG8 and SMG9, which suppress SMG1 kinase activity and are expected to form the complex before SMG1 forms the SURF complex [39]. In addition, RUVBL1 and 2 stabilize SMG1 and augments NMD through induction of the mRNA surveillance complex before a ribosome recognizes PTC [21]. Interestingly, the CK2 phosphosite of TEL2 is immunoprecipitated with both RUVBL1 and 2 via direct binding to PIH1D1 [18], suggesting that RUVBL1 and 2 may be able to regulate NMD through TEL2 phosphorylation. Furthermore, phosphorylated TEL2

References [1] J. Hwang, L.E. Maquat, Nonsense-mediated mRNA decay (NMD) in animal embryogenesis: to die or not to die, that is the question, Curr. Opin. Genet. Dev. 21 (2011) 422–430. [2] J.T. Mendell, N.A. Sharifi, J.L. Meyers, F. Martinez-Murillo, H.C. Dietz, Nonsense surveillance regulates expression of diverse classes of mammalian transcripts and mutes genomic noise, Nat. Genet. 36 (2004) 1073–1078. [3] D.R. Schoenberg, L.E. Maquat, Regulation of cytoplasmic mRNA decay, Nat. Rev. Genet. 13 (2012) 246–259. [4] C.J. Shoemaker, R. Green, Translation drives mRNA quality control, Nat. Struct. Mol. Biol. 19 (2012) 594–601. [5] J. Hwang, Y.K. Kim, When a ribosome encounters a premature termination codon, BMB Rep. 46 (2013) 9–16. [6] B.P. Lewis, R.E. Green, S.E. Brenner, Evidence for the widespread coupling of alternative splicing and nonsense-mediated mRNA decay in humans, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 189–192. [7] M. Khajavi, K. Inoue, J.R. Lupski, Nonsense-mediated mRNA decay modulates clinical outcome of genetic disease, Eur. J. Hum. Genet. 14 (2006) 1074–1081. [8] L.B. Gardner, Nonsense-mediated RNA decay regulation by cellular stress: implications for tumorigenesis, Mol. Cancer Res. 8 (2010) 295–308. [9] I. Kashima, A. Yamashita, N. Izumi, N. Kataoka, R. Morishita, S. Hoshino, M. Ohno, G. Dreyfuss, S. Ohno, Binding of a novel SMG-1–Upf1–eRF1–eRF3 complex (SURF) to the exon junction complex triggers Upf1 phosphorylation and nonsense-mediated mRNA decay, Genes Dev. 20 (2006) 355–367.

S. Ahn et al. / Biochimica et Biophysica Acta 1829 (2013) 1047–1055 [10] G. Singh, I. Rebbapragada, J. Lykke-Andersen, A competition between stimulators and antagonists of Upf complex recruitment governs human nonsense-mediated mRNA decay, PLoS Biol. 6 (2008) e111. [11] N. Izumi, A. Yamashita, S. Ohno, Integrated regulation of PIKK-mediated stress responses by AAA+ proteins RUVBL1 and RUVBL2, Nucleus 3 (2012) 29–43. [12] I. Kashima, S. Jonas, U. Jayachandran, G. Buchwald, E. Conti, A.N. Lupas, E. Izaurralde, SMG6 interacts with the exon junction complex via two conserved EJC-binding motifs (EBMs) required for nonsense-mediated mRNA decay, Genes Dev. 24 (2010) 2440–2450. [13] Y. Okada-Katsuhata, A. Yamashita, K. Kutsuzawa, N. Izumi, F. Hirahara, S. Ohno, N- and C-terminal Upf1 phosphorylations create binding platforms for SMG-6 and SMG-5:SMG-7 during NMD, Nucleic Acids Res. 40 (2012) 1251–1266. [14] S.J. Collis, L.J. Barber, A.J. Clark, J.S. Martin, J.D. Ward, S.J. Boulton, HCLK2 is essential for the mammalian S-phase checkpoint and impacts on Chk1 stability, Nat. Cell Biol. 9 (2007) 391–401. [15] N. Jiang, C.Y. Benard, H. Kebir, E.A. Shoubridge, S. Hekimi, Human CLK2 links cell cycle progression, apoptosis, and telomere length regulation, J. Biol. Chem. 278 (2003) 21678–21684. [16] M. Shikata, F. Ishikawa, J. Kanoh, Tel2 is required for activation of the Mrc1mediated replication checkpoint, J. Biol. Chem. 282 (2007) 5346–5355. [17] T. Trcek, H. Sato, R.H. Singer, L.E. Maquat, Temporal and spatial characterization of nonsense-mediated mRNA decay, Genes Dev., 2013. [18] Z. Horejsi, H. Takai, C.A. Adelman, S.J. Collis, H. Flynn, S. Maslen, J.M. Skehel, T. de Lange, S.J. Boulton, CK2 phospho-dependent binding of R2TP complex to TEL2 is essential for mTOR and SMG1 stability, Mol. Cell 39 (2010) 839–850. [19] H. Takai, R.C. Wang, K.K. Takai, H. Yang, T. de Lange, Tel2 regulates the stability of PI3K-related protein kinases, Cell 131 (2007) 1248–1259. [20] N. Izumi, A. Yamashita, H. Hirano, S. Ohno, Heat shock protein 90 regulates phosphatidylinositol 3-kinase-related protein kinase family proteins together with the RUVBL1/2 and Tel2-containing co-factor complex, Cancer Sci. 103 (2012) 50–57. [21] N. Izumi, A. Yamashita, A. Iwamatsu, R. Kurata, H. Nakamura, B. Saari, H. Hirano, P. Anderson, S. Ohno, AAA+ proteins RUVBL1 and RUVBL2 coordinate PIKK activity and function in nonsense-mediated mRNA decay, Sci. Signal. 3 (2010) ra27. [22] J. Zhang, X. Sun, Y. Qian, L.E. Maquat, Intron function in the nonsense-mediated decay of beta-globin mRNA: indications that pre-mRNA splicing in the nucleus can influence mRNA translation in the cytoplasm, RNA 4 (1998) 801–815. [23] P.M. Moriarty, C.C. Reddy, L.E. Maquat, Selenium deficiency reduces the abundance of mRNA for Se-dependent glutathione peroxidase 1 by a UGA-dependent mechanism likely to be nonsense codon-mediated decay of cytoplasmic mRNA, Mol. Cell Biol. 18 (1998) 2932–2939. [24] P. Belgrader, L.E. Maquat, Nonsense but not missense mutations can decrease the abundance of nuclear mRNA for the mouse major urinary protein, while both types of mutations can facilitate exon skipping, Mol. Cell Biol. 14 (1994) 6326–6336.

1055

[25] E. Park, M.L. Gleghorn, L.E. Maquat, Staufen2 functions in Staufen1-mediated mRNA decay by binding to itself and its paralog and promoting UPF1 helicase but not ATPase activity, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 405–412. [26] J. Hwang, H. Sato, Y. Tang, D. Matsuda, L.E. Maquat, UPF1 association with the cap-binding protein, CBP80, promotes nonsense-mediated mRNA decay at two distinct steps, Mol. Cell 39 (2010) 396–409. [27] T.M. Franks, G. Singh, J. Lykke-Andersen, Upf1 ATPase-dependent mRNP disassembly is required for completion of nonsense-mediated mRNA decay, Cell 143 (2010) 938–950. [28] J.C. Tapia, V.A. Torres, D.A. Rodriguez, L. Leyton, A.F. Quest, Casein kinase 2 (CK2) increases survivin expression via enhanced beta-catenin-T cell factor/lymphoid enhancer binding factor-dependent transcription, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 15079–15084. [29] H. Cho, K.M. Kim, S. Han, J. Choe, S.G. Park, S.S. Choi, Y.K. Kim, Staufen1-mediated mRNA decay functions in adipogenesis, Mol. Cell 46 (2012) 495–506. [30] I.G. Bruno, R. Karam, L. Huang, A. Bhardwaj, C.H. Lou, E.Y. Shum, H.W. Song, M.A. Corbett, W.D. Gifford, J. Gecz, S.L. Pfaff, M.F. Wilkinson, Identification of a microRNA that activates gene expression by repressing nonsense-mediated RNA decay, Mol. Cell 42 (2011) 500–510. [31] N. Hosoda, Y.K. Kim, F. Lejeune, L.E. Maquat, CBP80 promotes interaction of Upf1 with Upf2 during nonsense-mediated mRNA decay in mammalian cells, Nat. Struct. Mol. Biol. 12 (2005) 893–901. [32] J.R. Hogg, S.P. Goff, Upf1 senses 3′UTR length to potentiate mRNA decay, Cell 143 (2010) 379–389. [33] M. Pal, Y. Ishigaki, E. Nagy, L.E. Maquat, Evidence that phosphorylation of human Upfl protein varies with intracellular location and is mediated by a wortmanninsensitive and rapamycin-sensitive PI 3-kinase-related kinase signaling pathway, RNA 7 (2001) 5–15. [34] A. Tinguely, G. Chemin, S. Peron, C. Sirac, S. Reynaud, M. Cogne, L. Delpy, Cross talk between immunoglobulin heavy-chain transcription and RNA surveillance during B cell development, Mol. Cell Biol. 32 (2012) 107–117. [35] I. Rebbapragada, J. Lykke-Andersen, Execution of nonsense-mediated mRNA decay: what defines a substrate? Curr. Opin. Cell Biol. 21 (2009) 394–402. [36] T. Kurosaki, L.E. Maquat, Rules that govern UPF1 binding to mRNA 3′ UTRs, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 3357–3362. [37] L.E. Maquat, J. Hwang, H. Sato, Y. Tang, CBP80-promoted mRNP rearrangements during the pioneer round of translation, nonsense-mediated mRNA decay, and thereafter, Cold Spring Harb. Symp. Quant. Biol. 75 (2010) 127–134. [38] P. Nicholson, O. Muhlemann, Cutting the nonsense: the degradation of PTC-containing mRNAs, Biochem. Soc. Trans. 38 (2010) 1615–1620. [39] A. Yamashita, N. Izumi, I. Kashima, T. Ohnishi, B. Saari, Y. Katsuhata, R. Muramatsu, T. Morita, A. Iwamatsu, T. Hachiya, R. Kurata, H. Hirano, P. Anderson, S. Ohno, SMG-8 and SMG-9, two novel subunits of the SMG-1 complex, regulate remodeling of the mRNA surveillance complex during nonsense-mediated mRNA decay, Genes Dev. 23 (2009) 1091–1105.