A redox mechanism underlying nucleolar stress sensing by nucleophosmin

Yang, Kai; Wang, Ming; Zhao, Yuzheng; Sun, Xuxu; Yang, Yi; Li, Xie; Zhou, Aiwu; Chu, Huilin; Zhou, Hu; Xu, Jianrong; Wu, Mian; Yang, Jie; Yi, Jing

doi:10.1038/ncomms13599

Download PDF

Article
Open access
Published: 25 November 2016

A redox mechanism underlying nucleolar stress sensing by nucleophosmin

Kai Yang¹,
Ming Wang¹,
Yuzheng Zhao²,
Xuxu Sun¹,
Yi Yang ORCID: orcid.org/0000-0001-7896-1184²,
Xie Li²,
Aiwu Zhou³,
Huilin Chu¹,
Hu Zhou⁴,
Jianrong Xu⁵,
Mian Wu⁶,
Jie Yang¹ &
…
Jing Yi¹

Nature Communications volume 7, Article number: 13599 (2016) Cite this article

12k Accesses
79 Citations
11 Altmetric
Metrics details

Subjects

Abstract

The nucleolus has been recently described as a stress sensor. The nucleoplasmic translocation of nucleolar protein nucleophosmin (NPM1) is a hallmark of nucleolar stress; however, the causes of this translocation and its connection to p53 activation are unclear. Using single live-cell imaging and the redox biosensors, we demonstrate that nucleolar oxidation is a general response to various cellular stresses. During nucleolar oxidation, NPM1 undergoes S-glutathionylation on cysteine 275, which triggers the dissociation of NPM1 from nucleolar nucleic acids. The C275S mutant NPM1, unable to be glutathionylated, remains in the nucleolus under nucleolar stress. Compared with wild-type NPM1 that can disrupt the p53–HDM2 interaction, the C275S mutant greatly compromises the activation of p53, highlighting that nucleoplasmic translocation of NPM1 is a prerequisite for stress-induced activation of p53. This study elucidates a redox mechanism for the nucleolar stress sensing and may help the development of therapeutic strategies.

Three million images and morphological profiles of cells treated with matched chemical and genetic perturbations

Article Open access 09 April 2024

Srinivas Niranj Chandrasekaran, Beth A. Cimini, … Anne E. Carpenter

Bioorthogonal masked acylating agents for proximity-dependent RNA labelling

Article 09 April 2024

Shubhashree Pani, Tian Qiu, … Bryan C. Dickinson

DNA double-strand break–capturing nuclear envelope tubules drive DNA repair

Article 17 April 2024

Mitra Shokrollahi, Mia Stanic, … Karim Mekhail

Introduction

The nucleolus is a nuclear compartment for ribosome biogenesis where the transcription and processing of pre-ribosomal RNA and the assembly of ribosomal subunits take place. However, studies over the past two decades have revealed non-ribosomal functions for the nucleolus^1,2,3,4. Proteomic analysis has identified a large number of nucleolar proteins that are involved in diverse cellular processes, including ribosomal biogenesis, cell cycle control and stress signalling^5,6,7. Remarkably, many roles for the nucleolus are regulated by the sequestration or release of specific nucleolar proteins⁸ and various cellular stresses are often accompanied by dramatic changes in the structural organization and protein composition of the nucleoli⁷. However, it remains unknown how the various stress inducers lead to the release of nucleolar proteins into the nucleoplasm or the cytoplasm, a process that is often called ‘relocalization’ or ‘translocation.’ Thus, although recognizing the nucleolus as a stress sensor is a dramatic revision of our conception of the nucleolus, how the nucleolus senses stress has yet to be understood.

Nucleophosmin (NPM1) or B23 is an abundant nucleolar protein that can frequently shuttle between the nucleolus and the nucleoplasm or cytoplasm, interacting with a plethora of macromolecules to play a multifunctional role in cells⁹. Notably, NPM1 translocation from the nucleolus to the nucleoplasm is the most typical hallmark of nucleolar stress¹⁰. Nucleolar stress originally referred to the stressful events that impair the homeostasis of ribosomal biogenesis and activate the cellular stress response, which is also called ribosomal stress^11,12, and it is currently defined as alterations in the nucleolar structure and function that are induced by a variety of abnormal conditions¹³. NPM1 translocation appears to be a general event in nucleolar stress responses, even for stresses induced by different factors^14,15,16. Moreover, NPM1 might be an upstream signalling molecule that links various stresses with p53 accumulation^12,16,17. However, the mechanism underlying this well-known phenomenon of stress-induced NPM1 translocation remains unclear.

NPM1 has been identified as a phosphoprotein, as it is phosphorylated during specific cellular events¹⁸. The distinct phosphorylation patterns of NPM1 could be correlated with the regulation of NPM1 localization¹⁹. Many other types of posttranslational modifications (PTMs) of NPM1 have also been reported^20,21,22. These PTMs might presumably act as regulatory mechanisms underlying NPM1 translocation; however, none of the reported PTMs has been linked to the regulation of NPM1 translocation in response to nucleolar stress.

Using a combination of a number of innovative technologies, including live-cell imaging and a genetically encoded redox biosensor, we provide evidence addressing these questions and uncover a sensing mechanism for NPM1 in response to nucleolar and other stresses. We demonstrate that nucleolar oxidation is a general response to various cellular stresses. Accompanying the redox changes in the nucleolus, NPM1 undergoes S-glutathionylation on cysteine 275, which triggers the nucleoplasmic translocation of NPM1. This translocation can be attributed to the dissociation of NPM1 from nucleolar nucleic acids following S-glutathionylation, which is demonstrated by in vitro and in vivo NPM1–RNA/DNA interaction assays. The C275S mutant, which is unable to be glutathionylated and thus remained in the nucleolus, greatly compromises p53 stabilization under typical nucleolar stress conditions induced by actinomycin, regardless of the presence of the ribosomal protein L23 or ARF. Only wild-type (WT) NPM1 or a mutant localized to the nucleoplasm is able to lead to p53 accumulation under nucleolar stress, highlighting the fact that the nucleoplasmic translocation of NPM1 is a prerequisite for stress-induced p53 activation.

Results

Nucleolar oxidation is a general cellular stress response

Intracellular reactive oxygen species (ROS) can be increased by many stimuli^23,24. As NPM1 translocation has been observed under various cellular stress conditions in addition to typical nucleolar stress^14,16, we hypothesized that these stress inducers might produce a redox change in the nucleolar compartments. We constructed a nucleus-specific ratiometric redox probe based on reduction–oxidation-sensitive green fluorescent protein (roGFP1)²⁵. This nucleus-specific roGFP1 (NLS-roGFP1) fluoresced throughout the entire nucleus (Fig. 1a) and its distribution remained unchanged under oxidative stress conditions (Supplementary Fig. 1a). The cells expressing NLS-roGFP1 were then challenged by a series of stressors including hydrogen peroxide (H₂O₂), hypoxia, ultraviolet irradiation, heat shock, starvation (Earle’s balanced salt solution culture) and actinomycin D (Act.D). The nucleoli of all of the cells underwent a rapid oxidation to varying extents; however, these redox disturbances could be partially prevented by pretreatment with the anti-oxidant N-acetyl-cysteine (NAC) (Fig. 1b). These data indicated that an increase in ROS generation in the nucleolar compartment occurs universally under various cellular stresses. Given that the individual cells showed marked differences in the extent of the redox change, single live-cell imaging was used to monitor the real-time process of nucleolar oxidation. Two stress conditions were selected: Act.D (an inducer of nucleolar stress via the inhibition of RNA polymerase I¹¹) and H₂O₂ (a membrane-permeable ROS that is a general inducer of cellular oxidative stress). Nucleolar oxidation began 30 min after Act.D treatment and was aggravated at 1 h (Fig. 1c). Strikingly, H₂O₂ elicited the oxidation of the nucleolus within the first minute. This oxidation progressed and was maintained for at least 30 min but was attenuated by NAC pretreatment (Fig. 1d,e). Thus, nucleolar oxidation is indeed a general early event in response to various stresses, including typical nucleolar stress. In addition, we examined more cell types following their exposure to 500 μM H₂O₂ treatment, including the non-tumour cell line HEK293, the osteosarcoma cell line U2OS and the gastric carcinoma cell line MKN45. The oxidation of the nucleolus was recorded in live cells within 10 min (Supplementary Fig. 1b), suggesting that this is a general phenomenon.

NPM1 translocation is elicited by nucleolar oxidation

We used immunofluorescence to observe endogenous NPM1 translocation in cells challenged by the aforementioned stress conditions. The relative fluorescence intensity (RFI; see ‘Methods’) in nucleoli showed that NPM1 was largely dispersed to the nucleoplasm under all of the stress conditions tested. The antioxidant NAC was able to attenuate or even completely prevent this translocation (Fig. 2a and Supplementary Fig. 2a), suggesting that the nucleoplasmic translocation of NPM1 was induced by nucleolar redox changes. We next visualized the dynamics of EGFP-NPM1 translocation in response to H₂O₂ treatment in single live cells. The histone subunit H2B was labelled with mCherry, to show the shape of the nucleus and to control the focal plane in that cell. Treatment with 500 μM H₂O₂ produced a dispersion of nucleolar EGFP-NPM1 to the nucleoplasm that lasted for at least 30 min, resulting in a decrease in nucleolar NPM1 (Fig. 2b). Treatments with increasing doses of H₂O₂ (250 and 500 μM) caused a sustained loss of nucleolar NPM1 intensity in a dose-dependent manner. The RFI following 500 μM H₂O₂ treatment was decreased to 60%; in contrast, a lower concentration of H₂O₂ (125 μM) produced only a slight, transient loss within the first few minutes (Fig. 2c). Pretreatment with NAC restored the nucleolar intensity (Fig. 2d), whereas the protein disulfide reductant dithiothreitol (DTT) produced a more powerful effect in blocking NPM1 translocation compared with NAC; a complete nucleolar restoration of NPM1 was achieved by DTT (Fig. 2e). To ensure that the GFP tag did not significantly impair translocation efficiency, we examined the localization of a FLAG-tagged NPM1 by immunofluorescence. The RFI of nucleolar FLAG-tagged NPM1 was decreased following 500 μM H₂O₂ treatment slightly more effectively (decreased to 40%; Supplementary Fig. 2b). Nevertheless, both tagged proteins produced similar results as the endogenous setting shown in Fig. 2a. The nucleoplasmic translocation of NPM1 was also examined in additional types of live cells; decreases in nucleolar RFI, which ranged from ∼25 to 55%, were observed within 20 min (Supplementary Fig. 2c), suggesting that this is a general phenomenon.

To determine the causal relationship between nucleolar redox change and NPM1 translocation, the NLS-roGFP1 probe and mCherry-NPM1 were co-transfected into the same cell. The addition of H₂O₂ caused an immediate and stepwise nucleolar oxidation that was accompanied by NPM1 translocation. However, when DTT was co-administered, no translocation of NPM1 occurred, although the redox probe indicated a slight transient oxidation of the nucleolus (Fig. 2f). These data confirmed that the shuttling of NPM1 between the nucleolus and nucleoplasm might be controlled by the redox states of the two compartments, and that its nucleoplasmic translocation following stress is elicited by nucleolar oxidation.

Cys²⁷⁵ oxidation is responsible for NPM1 translocation

The DTT reversal experiments strongly implied a correlation between the reversible oxidative modification of proteins and stress-induced NPM1 translocation. To validate the association of cysteine thiol modification with translocation and to map the potential modification site(s), we constructed a 3C/S mutant EGFP-NPM1 that eliminated thiol modification sites by substituting all three cysteine residues (Cys²¹, Cys¹⁰⁴ and Cys²⁷⁵) with serines. A WT mCherry-NPM1 was co-transfected into the same cells to compare the translocation dynamics. Unlike the WT protein, mutant 3C/S was resistant to translocation (Fig. 3a). Cysteine mutagenesis was then performed in pairs and only the mutants containing C275S failed to respond to H₂O₂-induced translocation (Fig. 3b), revealing that Cys²¹ or Cys¹⁰⁴ was not responsible for this process. Next, a C275S single mutant was examined, which indeed lost the ability to translocate following stress in a way that was similar to the 3C/S mutant (Fig. 3c). This finding was also verified in the EGFP-C275S NPM1 and mCherry-H2B co-transfected living cell system (Supplementary Fig. 3a), and in a FLAG-C275S NPM1 immunofluorescence assay (Supplementary Fig. 3b). These results support the hypothesis that oxidative modification on Cys²⁷⁵ is essential for redox-regulated NPM1 nucleoplasmic translocation.

Cys²⁷⁵ is within the C-terminal domain that is important for nucleolar localization²⁶. The mutation of two tryptophans in this domain, Trp²⁸⁸ and Trp²⁹⁰, into alanines led to nucleoplasmic localization^26,27,28. We therefore speculated that the oxidative modification on Cys²⁷⁵ altered NPM1 localization by interfering with the nucleolar localization function of the carboxy-terminal domain. If this were the case, the explanation for the nucleolar retention of C275S under stress might be the absence of the thiol modification and the consequent interference, rather than the acquisition of nucleolar retention. Therefore, we predicted that the additional mutagenesis of two tryptophans (leading to the loss of nucleolar localization) would abolish the nucleolar retention of C275S. To test this, we constructed a W288/290A-C275S mutant that was tagged with FLAG and the localization of this NPM1 triple mutant was examined by immunofluorescence using an anti-FLAG antibody. We found that the W288/290A-C275S triple mutant remained localized to the nucleoplasm, behaving similarly to the W288/290A double mutant (Fig. 3d). These data strongly indicated that the stress-induced NPM1 translocation might be a consequence of the disruption of the C-terminal function of nucleolar localization, similar to what was caused by the W288/290A mutant.

NPM1 is S-glutathionylated at Cys²⁷⁵ under stress

Reversible thiol oxidative modifications to cysteine include the formation of disulfide, sulfenic and sulfinic acids, and S-glutathionylation, which can change protein conformation or alter interactions with other macromolecules. Therefore, we investigated the type of oxidative modification occurring on Cys²⁷⁵. As the responsible cysteine residue was single, the possibility of a formation of an intramolecular disulfide bond was excluded. The oligomerization (to form pentamers) of NPM1 was unaffected by redox changes²⁹; therefore, the formation of intermolecular disulfides between two or more NPM1 was also unlikely. To mimic the negative charge of sulfenic acid, Cys²⁷⁵ was replaced with aspartic acid. This C275D mutant remained localized in the nucleolus under non-stress conditions and was resistant to being translocated following stress (Supplementary Fig. 4a). Thus, modification with sulfenic acid on Cys²⁷⁵ was not possible either. Although a disulfide based on Cys²⁷⁵ of NPM1 and another cysteine of other proteins could not be excluded, we first turned to consider S-glutathionylation on this residue.

S-glutathionylation is a reversible but relatively stable oxidative modification containing a disulfide bond between the cysteine thiol of a protein and glutathione (GSH). We performed a co-immunoprecipitation (co-IP) assay and detected a band slightly larger than 38 kDa (NPM1) in the precipitates using antibodies against either GSH or NPM1. The expression of these bands was enhanced by H₂O₂ in a dose-dependent manner and NAC abolished the enhancement of this modification (Fig. 4a). These modification bands were similarly detectable in other cell types exposed to the treatments (Supplementary Fig. 4b). These data indicate that NPM1 is glutathionylated under conditions of oxidative stress. Next, we used mass spectrometry to analyse the samples exposed to H₂O₂, verifying the existence of the S-glutathionylation of NPM1 at Cys²⁷⁵ (Fig. 4b and Supplementary Fig. 4c).

**Figure 4: NPM1 is S-glutathionylated at Cys²⁷⁵ under stress.**

Glutathione S-transferase Pi (GSTP) catalyses the forward reaction of S-glutathionylation³⁰. We knocked down GSTP expression using small interfering RNA (siRNA) with varying efficiencies. Based on the level of remaining GSTP messenger RNA, the H₂O₂-induced NPM1 translocation to the nucleoplasm could be completely or minimally blocked (Fig. 4c). These data verified that S-glutathionylation is a cause of NPM1 translocation. Glutaredoxin 1 (Grx1) is a thiol oxidoreductase that selectively interacts with GSH³¹. In previous reports, a Grx1-roGFP2 fusion protein was constructed to monitor the dynamics of reduced GSH versus oxidized GSH (GSSG)^32,33,34. We therefore constructed an EGFP-Grx1-NPM1 fusion protein and observed its translocation dynamics in a live-cell imaging system. Compared with the control mCherry-NPM1 that was co-expressed in the same cells, nucleolar EGFP-Grx1-NPM1 showed an earlier and more thorough dispersion following H₂O₂ treatment with a short recovery period (Fig. 4d, top). In contrast, catalytically inactive Grx1-fused NPM1 (ref. 34), Grx1(ss)-NPM1, showed translocation dynamics that were indistinguishable from those of the control mCherry-NPM1 (Fig. 4d, middle), confirming that the drastic and quick translocation of Grx1-NPM1 was indeed caused by the Grx1-catalysed NPM1 S-glutathionylation cycle. The Grx1-NPM1 C275S mutant further verified that Grx1-catalysed NPM1 S-glutathionylation occurred on Cys²⁷⁵ (Fig. 4d, bottom). Using Grx1-roGFP2 as a biosensor for the GSH/GSSG ratio³⁴, we found a drastic increase in the levels of GSSG following H₂O₂ treatment (Supplementary Fig. 4d). These results, together with the preceding data, strongly suggest that S-glutathionylation on Cys²⁷⁵ impedes NPM1 nucleolar localization under conditions of oxidative stress.

The C-terminal domain of NPM1, which encompasses the last 70 residues of NPM1 (NPM1-C70), residues 225–294, is critical for its interaction with nucleolar nucleic acids³⁵. The solution structure of NPM1-C70 has been previously solved, showing a surface through which NPM1 might associate with nucleolar binding partners²⁶. Interestingly, Cys²⁷⁵ is located near the centre of this interface²⁶. Based on this structure, we simulated the covalent modification of Cys²⁷⁵ by a bulky GSH molecule. The simulated docking indicated that the GSH molecule, apart from forming a disulfide linkage with Cys²⁷⁵, could also form stabilizing interactions with residues around Cys²⁷⁵. This GSH modification might impede the association of the interface with the potential nucleolar macromolecules (Fig. 4e), which could explain the translocation of NPM1 following stress.

S-glutathionylated NPM1 dissociates from rRNA/rDNA

How the C-terminal domain determines the nucleolar localization of NPM1 remains unclear, but binding with rRNA is widely believed to be critical^36,37. Two studies have reported that NPM1 could also bind to nucleolar rDNA chromatin^38,39. We hypothesized that the nucleoplasmic translocation of NPM1 under stress might be due to its dissociation from these nucleic acids. The cells were incubated with RNase or Dnase, to digest rRNA or rDNA, respectively. Both of the treatments caused a dramatic dispersion of NPM1 into the entire nucleoplasm (Fig. 5a, left). These data suggested that the normal NPM1 nucleolar localization relies on the direct or indirect anchoring of NPM1 to both rRNA and rDNA, although it remains possible that RNase treatment could cause the destruction of pre-ribosomal particles, which could also account for the diffusion of NPM1 into the nucleoplasm. Digestion experiments were then performed using the C275S mutant and H₂O₂ treatments. Following stress, the C275S mutant remained in the nucleoli but dispersed into the nucleoplasm after RNase or DNase digestions, indicating that the mutant protein moved out of the nucleoli whenever the nucleic acids disappeared (Fig. 5a, right). We used RNA immunoprecipitation (RIP) and chromatin immunoprecipitation (ChIP) assays to quantitatively compare the ability of NPM1 to bind with rRNA and rDNA before and after H₂O₂ treatment. Primers were designed for three rRNA sequences targeting 47S, 18S and 28S transcripts, and for six rDNA sequences covering the rRNA gene promoter and coding region (Supplementary Fig. 5a). Equal quantities of FLAG-NPM1 were efficiently precipitated under the various conditions (Fig. 5b–e) and the quantities of rRNA and rDNA in the precipitates were assessed by real-time reverse transcriptase–PCR and PCR, respectively, along with several internal controls (Supplementary Fig. 5b–d). All three rRNA products bound to NPM1, especially products I and III, were markedly decreased under stress conditions compared with controls and these decreases were largely prevented by NAC (Fig. 5b). Likewise, all six rDNA products bound with NPM1 were decreased under stress and these decreases were also reversed by NAC (Fig. 5c). These results demonstrated that NPM1 becomes dissociated from rRNA and rDNA under conditions of stress. Unlike WT NPM1, the binding capacity to either rRNA or rDNA of the C275S mutant was not decreased under oxidative stress (Fig. 5d,e). These data indicate that the nucleolar retention of C275S under stress is due to the fact that the mutant protein has lost its ability to undergo thiol modification, thereby becoming unable to dissociate from the nucleic acids.

**Figure 5: NPM1 dissociates from rRNA/rDNA following S-glutathionylation.**

RIP assays were also performed to assess whether GSSG could promote the dissociation of NPM1 from nucleic acids. The amount of rRNA bound to NPM1 was less in the GSSG-containing sample than in the control (Supplementary Fig. 5e).

Next, we tested whether S-glutathionylation could block the interactions between NPM1, and both rDNA and rRNA in vitro. The amino and C termini of NPM1 truncates N224 (1–224) and C70 (225–294) were constructed and their subcellular localization was compared. Immunostaining revealed that N224 showed no nucleolar localization, whereas C70 was normally distributed in the nucleoli, nucleoplasm and cytoplasm (Supplementary Fig. 5f). This finding verified that NPM1-C70, which, unlike NPM1-N224, mimics the nucleolar localization of full-length NPM1, should be used to observe the in vitro interaction with nucleolar nucleic acids. NPM1-C70 and the mutant C275S were expressed in prokaryotic cells and purified. Previously, Chiarella et al.³⁵ found a significant number of putative G-quadruplex-forming sequences on the non-template of the rDNA gene that encodes the 47S rRNA precursor and verified that the G-quadruplexes formed by these sequences had a greater affinity for NPM1-C70. We therefore selected one of the putative G-quadruplex-forming sequences in the 47S rRNA precursor to represent the rDNA nucleic acid and its rRNA counterpart to represent the rRNA nucleic acid. We designed the surface plasmon resonance (SPR) assay similar to the experiments previously conducted by Chiarelle et al.³⁵ and tested whether the interaction of NPM1-C70 and these G-quadruplexes could be interrupted following the addition of GSSG in vitro. Our results showed that GSSG blocked the interaction of NPM1-C70 with rDNA G-quadruplex (Fig. 6, G4-DNA, WT versus WT+GSSG; Supplementary Fig. 5g and Table 1 for the K_D values). Importantly, the interaction of the translocation-disabled C275S mutant with the rDNA G-quadruplex was unaffected by GSSG (Fig. 6, G4-DNA, C275S versus C275S+GSSG; Supplementary Fig. 5g and Table 1). The measurement of the interactions of NPM1-C70 WT or C275S with rRNA G-quadruplex indicated similar effects (Fig. 6, G4-RNA; Supplementary Fig. 5g and Table 1).

**Figure 6: NPM1 dissociates from rRNA/rDNA after exposed to oxidized GSH *in vitro*.**

Table 1 The K_D values of the interactions of rRNA or rDNA G-quadruplexes with NPM1-C70 WT or C275S mutants respectively.

Full size table

Supporting the model proposed in Fig. 4e, these data provide evidence that the S-glutathionylation on Cys²⁷⁵ forms a steric obstacle between NPM1 molecules and nucleolar DNA and/or RNA. This obstacle may be sufficient to disassociate NPM1 from nucleolar nucleic acids and thereby trigger its nucleoplasmic translocation.

NPM1 translocation is required for p53 activation

The NPM1 translocation and activation of the master stress-response protein p53 have often been described concomitantly under nucleolar and other stresses^12,16,17. We next addressed the question of whether and how p53 stabilization is dependent on the translocation of NPM1. First, we determined the endogenous levels and localization of NPM1 and p53 using immunofluorescence in U2OS cells challenged with the most commonly used inducer for nucleolar stress, Act.D. As expected, Act.D elicited concurrent NPM1 translocation and p53 accumulation in the nucleoplasm, which could be attenuated by NAC or completely reversed by DTT (Fig. 7a and Supplementary Fig. 6a). siRNA for the glutathionylation enzyme GSTP compromised Act.D-induced p53 accumulation (Supplementary Fig. 6b). Consistent with previous reports¹⁶, the knockdown of NPM1 mRNA levels greatly reduced the accumulation of p53 following the same Act.D treatment or a short-term H₂O₂ treatment (Fig. 7b). We then added back the WT and the mutant C275S to these NPM1 knockdown cells and found that WT, but not translocation-disabled mutant C275S, could rescue the lost p53 accumulation under stress (Fig. 7c and Supplementary Fig. 6c). The levels of p53 accumulation were also demonstrated by western blotting in NPM1 knockdown and restored cells that were exposed to Act.D and H₂O₂. Nutlin-3, an agent that directly disrupts the p53–HDM2 interaction⁴⁰, was used here as a positive control. Interestingly, Act.D or H₂O₂ did not produce p53 accumulation in cells bearing C275S NPM1, whereas Nutlin-3 did (Fig. 7d), indicating that unlike Nutlin-3, Act.D or H₂O₂-induced p53 accumulation was dependent on NPM1 translocation. Furthermore, the transcriptional activity of p53 in these cells was determined by measuring the mRNA levels of the p53 target genes p21 and PUMA. The results similarly showed that p53 transactivation was impaired in the cells with C275S NPM1 (Fig. 7e). These results strongly argue that the translocation of NPM1, mediated by S-glutathionylation on Cys²⁷⁵, is required for the activation of p53 under stress.

To clarify whether merely the nucleoplasmic localization of NPM1 could induce p53 accumulation, we repeated the rescue assays in cells with NPM1 knocked down and then with the adding back of the FLAG-tagged W288/290A mutant. Surprisingly, the W288/290A mutant, although automatically localized to the nucleoplasm, could stabilize p53 only under Act.D or H₂O₂ exposure conditions (Fig. 7c and Supplementary Fig. 6d). These data indicate that the nucleoplasmic localization of NPM1 is necessary but not sufficient to induce p53 activation; some putative stress-mediated modifications to p53 might be simultaneously required, which would inevitably occur under stress conditions.

Human double minute 2 homologue (HDM2) is a negative regulator of p53. Studies have revealed that NPM1 binding to HDM2 could induce p53 accumulation by blocking the activity of HDM2 E3 ligase¹⁶. Other papers have reported that ribosomal proteins, including L5, L11 and L23, are inhibitors of the HDM2–p53 interaction and inducers of p53 stabilization^41,42,43. We performed co-IP using an antibody against HDM2 to assess the interaction of endogenous HDM2 with exogenous FLAG-NPM1 in cells with endogenous NPM1 knocked down and the adding back of FLAG-NPM1 WT, W288/290A and C275S mutants following Act.D treatment. We found that the WT and W288/290A mutant, but not the C275S mutant, could bind with HDM2 and that the p53–HDM2 interaction was prevented only by the WT and WW mutant. In all three circumstances, ribosomal protein L23 remained bound to HDM2 (Fig. 8a). This means that although L23 is required for p53 stabilization under stress, as previous studies have shown, only NPM1 translocation to the nucleoplasm dictates the final outcome of p53 accumulation, which occurs through NPM1’s binding with HDM2 and thus the disruption of the HDM2–p53 interaction.

**Figure 8: The presence of nucleoplasmic NPM1 is a prerequisite for stress-induced p53 activation.**

The tumour suppressor protein ARF is a molecule that has a key function in the p53-HDM2-NPM1 interrelationship^44,45. We next tried to assess the correlation of NPM1 translocation and ARF in respect of the activation of p53. The vector HA-p14/ARF⁴⁶ was ectopically expressed in U2OS cells that are negative in ARF expression. Examining the localization of the overexpressed ARF and endogenous NPM1 by immunofluorescent staining, we found that exogenous ARF was distributed both in the nucleolus and nucleoplasm (Supplementary Fig. 6e); the nucleolar ARF did not move out following Act.D treatment and, in contrast, the nucleolus/nucleoplasm ratio of ARF was slightly increased, whereas NPM1 translocate to the nucleoplasm under the same condition (Fig. 8b). We determined the levels of p53 by western blottings and found that, compared with the mock DNA, overexpression of ARF in U2OS cells with either normal or silenced NPM1 led to an increased accumulation of p53 under non-stress condition. However, after cells were exposed to Act.D, overexpression of ARF alone did not produce further accumulation of p53 in NPM1 knockdown cells. An enhanced p53 accumulation was observed when WT NPM1 was added back, but adding back of C275S mutant could not causes the increase of p53 (Fig. 8c). These data suggest that the nucleoplasmic fraction of ARF alone is able to induce p53 accumulation under basal conditions, yet the further accumulation under stress conditions is determined by the presence of the nucleoplasmic NPM1, independent on ARF. This idea was substantiated by the results of co-IP, in which we investigated the interactions between HDM2 and p53, the added-back WT or mutant NPM1 and exogenous ARF, respectively, following Act.D treatment. We found that the disruption of the HDM2–p53 interaction occurred in the cells bearing the WT but not C275S NPM1 and amounts of ARF bond to HDM2 appeared similar in both cell groups (Fig. 8d). This means that although ARF expression can promote p53 stabilization, in line with previous studies^47,48,49, the enhancement of p53 accumulation under nucleolar stress condition requires the presence of the nucleoplasmic NPM1.

Collectively, it is concluded that the nucleoplasmic translocation of NPM1 is a prerequisite for p53 stabilization under nucleolar stress condition.

In summary, redox changes in the nucleolar compartment and the consequent S-glutathionylation at Cys²⁷⁵ serve as a fundamental mechanism for NPM1 sensing nucleolar and other stresses. The translocation of NPM1 to the nucleoplasm, which is due to its dissociation from rRNA and rDNA, accounts for the NPM1 regulation of p53 activation under stress (Fig. 9).

**Figure 9: NPM1 sensing for nucleolar stress.**

Discussion

To our knowledge, the redox state of the nucleolus has not previously been addressed, although methods for analysing the redox conditions in specific subcellular compartments in living cells have been rapidly developed^50,51. We constructed a nuclear-specific redox sensor to observe the process of nucleolar oxidation on different cellular insults. The nucleoli became prone to oxidation after cells were exposed to various insults that had been previously reported to induce nucleolar stress⁷. Thus, nucleolar stress can be signified by oxidative stress in the nucleoli.

Translocation of the protein NPM1 from nucleoli to the nucleoplasm was initially observed when ribosome biogenesis was halted by Act.D⁵², and soon thereafter was also found in cells exposed to cytotoxic agents¹⁴, viral proteins⁵³, ultraviolet radiation¹⁷, heat shock¹⁵ and agents inducing DNA damage⁵⁴, apoptosis or senescence^55,56. NPM1 is thus considered to be a sensing molecule for stress^10,57. Why cellular stressors that differ in nature all lead to NPM1 translocation is puzzling. Here, our findings provide an explanation by showing that a redox change in the nucleolar compartment is a general response to diverse stress conditions.

NPM1 is believed to bind to rRNA^21,58,59. The capability of NPM1 to bind to DNA in the nucleoplasm has also been reported⁶⁰; nevertheless, only a few documents have suggested that NPM1 binds to rDNA^35,38,39. Our RIP and ChIP results demonstrated that the associations between NPM1 and rRNA or rDNA, whether direct or indirect, were weakened by oxidative modifications. In parallel, NPM1 could be driven out of the nucleoli by digestion with RNase or, to a lesser extent, with DNase, probably dissociating from the fibrillar centre (rDNA), fibrillar components (rRNA) and granular components (pre-ribosomal particles). Most importantly, the S-glutathionylation of NPM1 interfered with its binding to G-quadruplexes of rDNA or rRNA, as shown by our in vitro interaction assays. These findings indicate that the anchoring of NPM1 to rRNA and rDNA could be the force holding NPM1 within an unstressed nucleolus; the translocation of NPM1 under stress is due to its dissociation from rRNA and rDNA.

How NPM1 and other nucleolar proteins shuttle rapidly between the nucleolar-bound and unbound states remains an open question³. Based on the findings in this study, we assume that a rapid, reversible S-glutathionylation at Cys²⁷⁵ could determine the localization of NPM1, which may also hold true in its shuttling under conditions that are considered to be unstressed. Three lines of evidence support this hypothesis. First, the nucleolar localization of NPM1 that was disrupted by a low dose of H₂O₂ could be rapidly auto-restored (Fig. 2c), meaning that two modification states could be switched along with physiological ROS fluctuations. Second, the nucleoplasmic translocalization of NPM1 can be mildly prevented by NAC, a GSH synthesis precursor, and be thoroughly stopped by DTT, a reducing agent that reduces the disulfide bonds of proteins, including those formed with GSSG. Third, GSTP and Grx1 could modulate the translocation dynamics of NPM1, indicating that the presence or absence of S-glutathionylation dictates the localization of NPM1 in the nucleoplasm or the nucleolus, respectively.

S-glutathionylation of NPM1 was initially reported following exposure of alkylating agent⁶¹. In our study, S-glutathionylation is eventually verified as the type of thiol modifications on Cys²⁷⁵ of NPM1 following nucleolar oxidation, which is based on applications of the enzymes that specifically catalyse S-glutathionylation. It is worth noting that Grx is traditionally considered an enzyme catalysing the reduction of S-glutathionylated proteins, but is now also considered an efficient catalyst for protein oxidation, as demonstrated by assays using roGFP/rxYFP probes⁶². EGFP-Grx1-NPM1 we constructed shows an earlier and more thorough dispersion with a short recovery, which is dependent on the catalytical activity of Grx1 (Fig. 4d), confirming that Grx is responsible for a reversible S-glutathionylation of NPM1.

The nucleolus is considered to be a central hub coordinating the stress response⁸. How the signals of nucleolar stress can be connected to p53 stabilization, the initial step of activation of p53 pathway, is emerging as a research focus^7,11,12,13. The translocation of NPM1 to the nucleoplasm occurs before p53 accumulation under ultraviolet radiation stress¹². Notably, it was demonstrated a decade ago that NPM1 can, via its C-terminal domain, interact with p53 (ref. 17). NPM1 is also found to bind with HDM2, competing with p53 and blocking the HDM2-mediated ubiquitination of p53 (ref. 16). The interaction of ribosomal proteins with HDM2 is required for the activation of p53 induced by nucleolar stress^41,42,43. ARF can also bind to HDM2, to stabilize p53 in diverse cellular events^48,63,64,65. Our findings emphasize that the nucleoplasmic localization of NPM1 alone allows it to induce p53 stabilization under stress. The binding of ribosomal proteins or ARF with HDM2, which had been thought to be sufficient for p53 stabilization^41,42,43,48, is actually insufficient when NPM1 stays within the nucleoli.

There existed complicated reciprocal regulations between ARF and NPM1 (refs 66, 67, 68, 69, 70). Moreover, ARF-NPM1-p53 has often been considered as an axis that functions under varying physiological and pathological scenarios^66,71. However, it has not been adequately studied whether ARF translocate from the nucleolus to the nucleoplasm under stress conditions and whether this translocation is required for NPM1 action on HDM2–p53 (refs 72, 73). Moulin et al.⁷⁰ showed that depletion of NPM had little impact on the amount of ARF that associated with MDM2 and p53, which, together with their earlier report⁴⁹, indicate that only a nucleoplasmic fraction of the total ARF is involved in the interaction with MDM2. In the present study we further clarify that, unlike NPM1, ARF does not translocate to the nucleoplasm following a typical nucleolar stress. Moreover, under stress condition, the amounts of ARF bound to HDM2 in cells bearing the WT or mutant NPM1 remain similar. Therefore, irrespective of ARF, nucleoplasmic translocalization of NPM1 plays an indispensable role that connects nucleolar stress to p53 activation.

The findings in this and future studies could help with the development of therapeutic strategies^74,75,76, because NPM1 mutation is a founder genetic lesion in acute myeloid leukaemia (AML) and an attractive target for intervention^{76,77,78,79,80,81}. It has been found that the p53-dependent death pathway is activated in AML cells carrying WT NPM1 but not in cells carrying aberrant cytoplasmic NPM1 (refs 28, 82). We speculate that the abovementioned mechanism could account for the inactivation of p53 in these cells; the mislocalization of NPM1 resulted in the loss of the ability to sense nucleolar stress and to access to HDM2 in the nucleoplasm. NPM1 mutations are always heterozygous⁷⁷ and it is hypothesized that the NPM1-mutated AML cells might be vulnerable to drugs that trigger a nucleolar stress response, because it contains a low level of non-mutant NPM1 and because p53 in these cells is not mutated⁷⁸. A few NPM1-interacting anti-proliferative compounds have been identified^76,79,80,81, among which the natural product avrainvillamide binds to NPM1 specifically at Cys²⁷⁵. Interestingly, avrainvillamide at a low dose disperses NPM1 into the nucleoplasm in the NPM1-WT AML cells but re-localizes the mutated NPM1 to the nucleoli and, at higher doses, induces apoptosis in NPM1-mutated AML cells^76,83. More excitingly, a recent clinical application of Act.D at a concentration higher than the one used in the present in vitro study led to a morphologically and molecularly complete remission of an AML patient⁷⁸. In this case, the disruption of nucleolar localization of the WT NPM1 of the patient by Act.D, probably through S-glutathionylation on Cys²⁷⁵, might be one of the relevant mechanisms.

Methods

Antibodies and reagents

Antibodies against NPM1 (60096-1-Ig) were obtained from ProteinTech Group (USA). Antibody against IgG1 (5415) was from Cell Signaling Technology. Anti-Flag M2 (F1804) and anti-tubulin (D66; T0198) mouse antibodies were from Sigma. Antibodies against vinculin (ab129002), GFP (ab290), L23 (ab112587), GSH (D8; ab19534), Fibrillarin (ab5821), p53 (E47; ab32509) and HA (ab137838) were from Abcam. Anti-p53 (Pantropic) (DO-1; OP43L) and anti-MDM2 (5B10C; OP146) were purchased from Calbiochem. GSSG (G2140), H₂O₂ (H1009), NAC (A9165), DTT (D0632) and Act.D (A4262) were purchased from Sigma-Aldrich. Nutlin-3 (3984) was obtained from Tocris Bioscience.

Cell culture and treatments

HEK293 (derived from ATCC), HEK293T, 293FT (derived from Invitrogen), HeLa and MKN45 (or U2OS (derived from ATCC)) cells were maintained at 37 °C in a humidified atmosphere of 5% CO₂ with DMEM high-glucose (or α-MEM) supplemented with 10% fetal bovine serum and penicillin/streptomycin. All cell lines were tested for mycoplasma contamination and experiments were conducted under mycoplasma-negative conditions. The Nutlin3 and Act.D solution was prepared in dimethyl sulfoxide. The NAC was prepared with 2 M Tris-base solution to produce a final pH of 7.4. Ultraviolet treatments were performed by applying UVC radiation using a CL-1000 UV cross-linker (UVP Inc., USA). Hypoxia treatment was performed in a triple-gas incubator (Huaxi Electronic Tec., China) with an atmosphere containing 1% O₂ and 5% CO₂. Heat-shock experiments were performed by heating the cells in a 42 °C water bath for 30 min. Cell starvation was induced by changing the medium into a nutrient-free Earle’s balanced salt solution (Gibco).

Plasmids

The pcDNA3.1-FLAG-NPM1 was a gift of Dr Benjamin Y.M. Yung (The Hong Kong Polytechnic University, Hong Kong). The mCherry-C1-H2B was a gift of Dr Chuanmao Zhang (College of Life Sciences, Peking University, China). To construct a nucleus-targeted roGFP1, NLS-roGFP1, the threefold nuclear localization signal (3 × NLS) DPKKKRKVDPKKKRKVDPKKEKRKV was added to the C terminus and expressed in HeLa cells using a modified pEGFP-N1 as the expression vector. For tracking NPM1 proteins in living cell, full coding sequences of NPM1 were subcloned from the pcDNA3.1-FLAG-NPM1 plasmid into pEGFP-N1 and mCherry-C1 (Clontech Laboratories Inc.). For Grx1-NPM1 fusion protein, full coding sequences of Grx1 were subcloned into the upstream of the NPM1 transcription initiation site in pEGFP-N1-NPM1. Grx1(ss), as a catalytically inactive Grx1, was generated by replacing Cys²³ and Cys²⁶ residues with serines.

Mutagenesis was performed using the QuickChange Lightning Site-Directed Mutagenesis Kit (Agilent Technologies) following the manufacturer’s instructions and using primers listed in the Supplementary Table 1.

Plasmids DNA and siRNA oligo were transfected into cells with Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions.

For GSTP siRNA sequences, siGSTP-1 (sense): 5′-GGAGGACCUCCGCUGCAAAdTdT-3′; siGSTP-2 (sense): 5′-CUCCGCUGCAAAUACAUCUdTdT-3′.

For NPM1-shRNA (5′-untranslated region) oligonucleotides, the following sequences were annealed and subcloned into the pLVX-shRNA2 lentiviral expression vector:

5′-gatccGCCTACCGTGTTTGATAAATTTCAAGAGAATTTATCAAACACGGTAGGTTTTTTACGCGTg-3′, 5′-aattcACGCGTAAAAAACCTACCGTGTTTGATAAATTCTCTTGAAATTTATCAAACACGGTAGGCg-3′.

To establish a stable NPM1-shRNA (5′-untranslated region) U2OS cell line, ZsGreen1 co-expressing lentiviral expression vectors were transiently transfected into 293FT cells. After 72 h, the supernatants were harvested to infect U2OS cells with 10 μg ml⁻¹ polybrene at final concentration. Lastly, the ZsGreen1-positive cells were sorted on a FACSAria II flow cytometer (BD Biosciences).

Immunofluorescence

Cells grown on coverslips were fixed in 4% paraformaldehyde for 10 min before permeabilization with 0.3% Triton X-100, washed with PBS and incubated with 2% BSA containing primary antibodies against NPM1 (1: 100, PTG-lab), p53 (1: 100, Abcam), HA (1: 100, Abcam), Fibrillarin (1: 100, Abcam) and FLAG (1: 100, Sigma) at 4 °C overnight. The monolayers were then washed with PBS and incubated with appropriate secondary antibodies in PBS at 37 °C for 2 h. Alexa Fluor 488- and 555-labelled goat anti-mouse antibodies, and Alexa Fluor 555- and 633-labelled goat anti-rabbit antibodies were obtained from Molecular Probes (Invitrogen). Cells were washed again with PBS, counterstained with 4,6-diamidino-2-phenylindole (Beyotime, China) and then used for observation. Images were taken on random, but tissue areas and cells with extraordinarily strong staining were excluded for examination.

Microscope image acquisition

Images were acquired using the Zeiss 710 laser scanning confocal microscopy system on a Zeiss Axio Observer Z1 inverted microscope, equipped with a Plan Apochromat × 63.0, 1.4 numerical aperture oil-immersion, differential interference contrast objective and a 488 nm argon laser. For immunofluorescence assays, scanning was performed using 1,024 × 1,024 format, 1.58 μs pixel dwell, 12 bit depth, 4 × line average and 3.0 μm pinhole. For live-cell imaging, images were recorded every minute. Cells were plated on a 35 mm glass-bottom dish (Nest Biotechnology) and observed at 12–24 h post transfection in a humidified atmosphere using a CO₂ incubator (PeCon) at 37 °C, scanned in 512 × 512 format, 0.5 μs pixel dwell, 12-bit depth, 2 × line average and 8.4 μm pinhole. For redox probe observation, NLS-roGFP1 were excited with the 405 nm and 488 nm lasers, and emissions were detected at 500–550 nm, scanning with 512 × 512 format, 0.5 μs pixel dwell, 12 bit depth, 2 × line average and 8.4 μm pinhole. Images were taken on random, but cells with extraordinarily strong or low expression levels were excluded for examination.

Image processing and fluorescence signal quantifications

Fluorescence images obtained from immunofluorescence and live-cell imaging were processed in ImageJ software (version 1.49 v with 64 bit Java Platform; NIH) and exported as TIFF mode files in red/green/blue (RGB) channels. To normalize individual cell differences, in general 30 cells or more in each group were examined for quantification of average fluorescent intensity. The absolute fluorescence intensity and line plots were obtained from bit channel files before the images were transferred to RGB channel files. RFI mean absolute fluorescence intensity changes relative to control groups in immunofluorescence or initiation time (0 s) in live-cell imaging. For NLS-roGFP1 images, the pixel-by-pixel ratio of the 488 nm excitation image to the 405 nm excitation image of the same cell was used to pseudocolour the images with ImageJ as previously described⁸⁴. Briefly, the colour brightness was proportional to the fluorescence intensities in both channels. Shown as a calibration bar, the RGB values 255, 0, 255 (purple) represented the lowest 405/488 nm ratio, and 255, 0, 0 (red) represented the highest ratio. A higher ratio of 405/488 indicated a more oxidative state.

Co-immunoprecipitation

Cells grown at 10⁷ were washed with PBS and scraped before being suspended in 1 ml PBS on ice. After centrifugation, the cell pellet was resuspended in 1 ml lysis buffer (50 mM Tris-HCl pH 7.4, 1% Triton X-100, 150 mM NaCl, 1 mM EDTA, plus cocktail inhibitor (Roche)) for 30 min gentle agitation at 4 °C, followed by sonication for two bursts of 10 s each at half power with a 2 mm stepped microtip (Sonics VCX-130, USA). After centrifugation, the supernatant was mixed with anti-NPM1 antibody (1: 200, PTG-lab), anti-GSH antibody (1: 100, Abcam) and anti-HDM2 antibody (1: 100, Calbiochem) overnight at 4 °C. Next, 30 μl of Protein A/G-Agarose (EMD Millipore) was added for a further 4 h agitation at 4 °C. Lastly, the resins were washed six times with 1 ml lysis buffer and then mixed with 2 × SDS sample buffer, boiled 10 min and analysed by western blotting. Specifically for glutathionylated NPM1 samples, β-mercaptoethanol was not included in the 2 × SDS sample buffer.

Western blotting

Cells were lysed in sample solution. Proteins were separated with 10% SDS–PAGE gels and transferred to nitrocellulose membranes. To reduce nonspecific background, membranes were blocked with 5% milk in TBS-T buffer. Bands were detected using various antibodies against NPM1 (1: 2000, PTG-lab), GSH (1:1,000, Abcam), L23 (1:1,000, Abcam), p53 (1:1,000, Abcam or Calbiochem), vinculin (1:2,000, Abcam), HDM2 (1:1,000, Calbiochem) and FLAG (1:1,000, Sigma) at 4 °C overnight. The membranes were incubated with peroxidase-conjugated secondary antibodies (1:5,000, Jackson) for 2 h at 37 °C before detection using ECL system (EMD Millipore). Images scanned with ImageQuant LAS 4000 mini (GE) were exported as TIFF images in RGB mode. All the uncropped versions of images were showed in Supplementary Fig. 7.

Mass spectrometry analysis

HEK293T cells transfected with FLAG-NPM1 were treated with 500 μM H₂O₂ for 30 min. FLAG-NPM1 was then immunoprecipitated with anti-FLAG M2 gel (Sigma), mixed with 2 × SDS sample buffer without β-mercaptoethanol and loaded for SDS–PAGE. After examination of the Coomassie Brilliant Blue staining, the gel bands containing FLAG-NPM1 protein were clipped out and cut into small pieces in a 1.5 ml microtube. The chopped gels were washed three times with 50 mM NH₄HCO₃ in 30% acetonitrile by shaking at room temperature (RT) for 20 min. The gels were further incubated with 300 μl of acetonitrile at RT for 10 min. After removing the acetonitrile, the gels were dried and trypsin-digested following the manufacturer’s protocol. Peptides were extracted with 85% acetonitrile and 0.1% trifluoroacetic acid for twice. After removing the pieces of gels, the remained solution was concentrated with a SpeedVac concentrator. Then, the peptides were re-dissolved in 30 μl of 0.1% formic acid for liquid chromatography–tandem mass spectrometry analysis (LTQ Orbitrap Elite, Thermo Scientific). The fragment ions observed in the mass spectra were analysed with Mascot.

Structure model

NPM1-GSH complex was modelled in COOT software⁸⁵ and images were prepared in PyMOL software (DeLano Scientific, http://www.pymol.org).

RNase A and DNase I digestion

HeLa cells grown on coverslips were permeabilized with 0.1% Triton X-100 in PBS for 2 min, washed immediately and treated with RNase A (EN0531, 1 mg ml⁻¹, Fermentas, Thermo) or DNase I (EN0523, 0.5 U μl⁻¹, Fermentas, Thermo) in solution buffer for 10 min at 37 °C and cells were then fixed with 4% paraformaldehyde for 10 min and analysed by immunofluorescence.

RNA immunoprecipitation

HEK293T cells were scraped into 1 ml PBS. After centrifugation at 600 g for 2 min, the cell pellet was resuspended in 1 ml diethyl pyrocarbonate (DEPC)-treated lysis buffer (50 mM Tris-HCl pH 7.4, 1% Triton X-100, 150 mM NaCl, 1 mM EDTA, plus cocktail inhibitor (Roche)) containing 40 U ml⁻¹ Ribonuclease Inhibitor (TaKaRa, China) for 30 min agitation. Then, the samples were sonicated for two bursts of 10 s each at half power and centrifuged at 12,000 r.p.m. for 20 min to remove the debris. For the INPUT sample, 10% of the supernatant was used for western blot analysis and 10% was used for RNA extraction with Trizol reagent (Invitrogen). The rest of the supernatant was incubated with 20 μl anti-FLAG M2 gel for overnight rotation. After being washed with 1 ml diethyl pyrocarbonate-treated lysis buffer six times, immunoprecipitates were then separately subjected to western blotting or RNA extraction as described above. Samples comprising 10% of the resins were mixed with 2 × SDS sample buffer and analysed by western blotting. Trizol (1 ml) was directly added to the rest of the resins for isolating the RNA. All steps were performed at 2–8 °C.

Nuclear preparation and ChIP analysis

HEK293T cells were cross-linked with formaldehyde (0.25%, final concentration) for 10 min at RT in dishes, then washed with PBS before being scraped into 1 ml PBS. After centrifugation, cell pellet was resuspended in 1 ml of buffer A (10 mM Hepes-KOH pH 7.4, 10 mM KCl, 1.5 mM MgCl₂, 0.5 mM EDTA, 0.5 mM EGTA, plus cocktail inhibitor (Roche)) and flushed through a 23 G needle syringe 27 times. The released nuclei were monitored microscopically and washed once with buffer A with centrifugation. Then, the nuclei were resuspended in 0.1 ml of TE buffer (20 mM Tris-HCl pH 7.4 and 2 mM EDTA) containing 2% SDS and incubated at 37 °C for 15 min to disrupt the nucleolar structure. An additional 0.9 ml lysis buffer (50 mM Tris-HCl pH 7.4, 1% Triton X-100, 150 mM NaCl, 1 mM EDTA, plus cocktail inhibitor (Roche)) was added to each sample before sonication for four bursts of 15 s each at 80% power. After centrifugation at 12,000 r.p.m. for 20 min to remove the debris, 10% of the nuclear chromatin supernatant was used as INPUT for western blotting and 10% was used for genomic DNA extraction. Extraction was performed using the TIANamp Genomic DNA Kit (TIANGEN BIOTECH, China). Lastly, the rest of the supernatant was used immediately in ChIP assays. The nuclear chromatin supernatant was then incubated with 20 μl anti-FLAG M2 gel and rotated at 4 °C overnight. The resins were washed as follows: twice with 1 ml of lysis buffer containing 0.2% SDS, twice with 1 ml of lysis buffer and twice with 1 ml of TE buffer. Then, 10% of each sample was mixed with 2 × SDS sample buffer, boiled for 10 min and analysed by western blotting, and the rest of the resins were subjected to genomic DNA extraction as described above. The purified DNA was resuspended in ddH₂O for real-time quantitative PCR analysis.

Real-time quantitative PCR analysis

Real-time quantitative PCR for RIP and ChIP assays was performed using FastStart SYBR-Green/Rox (Roche) on ABI-7500 Fast equipment (Applied Biosystems). Samples were analysed in triplicate and normalized against vehicle group as an internal control. Relative changes in gene expression were calculated using the ΔΔCt method. For rRNA samples, an equal volume of isolated RNA was reverse-transcribed using AMV Reverse Transcriptase (Promega). The amplification procedure was performed as follows: 95 °C 10 s, 60 °C 30 s and 72 °C 30 s for 40 cycles, with a melting-curve analysis to verify the specificity of the amplification. The raw data were exported and analysed with GraphPad Prism software. The sequences of primers used in RIP and ChIP assay are shown in the Supplementary table 1.

Prokaryotic protein expression and purification

The gene fragments encoding the C terminus of NPM1 (NPM1-C70) WT and C275S were fused with SUMO tag and inserted into pCDFDuet vector (Novagen) at the NcoI/XhoI restriction sites. Plasmids were transformed into Escherichia coli BL21 (DE3) stain. Cells were grown in Luria broth supplemented with 50 μg ml⁻¹ streptomycin, induced with 1 mM isopropyl-D-thiogalactopyranoside and then transferred to 18 °C for 24 h before harvesting. The cell pellets were resuspended in Buffer A (500 mM NaCl, 20 mM phosphate and 20 mM imidazole pH 7.4) and sonicated. After separating the debris by centrifugation, the supernatant was loaded on a Ni column pre-equilibrated with Buffer A. After washing with five column volumes of wash buffer (Buffer A contain 50 mM imidazole), the His-SUMO-tagged protein was eluted with Buffer B (500 mM NaCl, 20 mM phosphate and 500 mM imidazole pH 7.4). Fractions containing the His-SUMO-tagged protein were then desalted and exchanged to digestion buffer (20 mM Tris-HCl pH 8.0, 150 mM NaCl and 1 mM DTT) and digested with SUMO protease at 37 °C for 1 h. The cleavage mixture obtained was then buffer changed to Buffer A (500 mM NaCl, 20 mM phosphate and 10 mM imidazole pH 7.4) and loaded onto Ni column to separate WT or C275S from the His-SUMO and the uncleaved fusion protein. At last, WT or C275S was recovered from the flow through and buffer exchanged to 50 mM Tris-HCl pH 7.4 by ultrafiltration, concentrated and ready to use.

Surface plasmon resonance

Oligonucleotides (oliogos) for SPR analysis were prepared following the study of Chiarella et al.³⁵. One of a G-quadruplex-forming sequences in 47S rRNA precursor, 5′-GGGGTGGGGGGGAGGG-3′, namely 13079NT, was selected to represent the rDNA nucleic acid and its rRNA counterpart was selected to represent the rRNA nucleic acid. Lyophilized HPLC-grade DNA or RNA oligos were obtained from Invitrogen. The two oligos were both biotinylated at their 5′-end and dissolved in 20 mM Hepes pH 7.4 with 150 mM KCl. For annealing to form G-quadruplex structure, both oligos were heated at 95 °C for 15 min and were let to gently cool down overnight at room temperature.

SPR measurement was performed on a BIAcore T200 instrument with four flow channels and a sensor chip with dextran matrix and pre-coated streptavidin (GE Healthcare). The chip was activated with three consecutive 1 min injections of 50 mM NaOH and 1 M NaCl. For immobilization of G-quadruplex oligos, the flow rate was 5 μl min⁻¹ to minimize sample consumption. Biotinylated DNA (10 μM) and RNA oligos diluted with the running buffer was injected and captured onto the channel surface, to achieve the immobilization level at about 300 RU (response unit) of both. For the binding assay, the reaction temperature was controlled at 25±0.1 °C and the flow rate was set at 30 μl min⁻¹. The response obtained from the detection channel (FcDNA or FcRNA) was normalized by subtracting the signal simultaneously acquired from the control channel (Fcvehicle), which could eliminate nonspecific binding and buffer-induced bulk refractive index changes. Tris-HCl (50 mM, pH 7.4) was used as running buffer. Different concentrations (0.6, 1.2, 3, 6, 15 and 30 μg ml⁻¹) of WT and C275S of NPM1-C70 were prepared and their bindings with rDNA or rRNA were measured in SPR assay. To specify the effect of GSSG on NPM1-C70/G-quadruplex binding, some of NPM1-C70 samples were mixed with 10 mM GSSG. The sample solutions were injected onto the chip surface for 120 s. After each binding reaction, a further dissociation time of 180 s was applied and as the regeneration buffer, 10 mM NaOH solution was injected for 60 s to allow the signal back to the baseline.

Owing to the fast K_on and K_off, the K_D values were calculated with a one-site steady-state binding model⁸⁶, with equation Y=B_max*X/(K_D+X)+background, where X is different analyte concentration and Y is the corresponding maximal response unit.

Statistical analysis

GraphPad Prism software (version 6.0c) was used to analyse and plot all data. Statistical analysis was performed with two-tailed unpaired t-test with 95% confidence interval. In box and whisker plots (Figs 1b and 2a), the box showed the top and bottom quartiles (25–75%) with a line at the median and the whiskers showed the minimum to the maximum values of all data. P-values indicated the significances of differences.

Data availability

The data that supports the conclusions of this study are available from the corresponding author upon request.

Additional information

How to cite this article: Yang, K. et al. A redox mechanism underlying nucleolar stress sensing by nucleophosmin. Nat. Commun. 7, 13599 doi: 10.1038/ncomms13599 (2016).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

Andersen, J. S. et al. Directed proteomic analysis of the human nucleolus. Curr. Biol. 12, 1–11 (2002).
PubMed Google Scholar
Olson, M. O., Dundr, M. & Szebeni, A. The nucleolus: an old factory with unexpected capabilities. Trends Cell Biol. 10, 189–196 (2000).
CAS PubMed Google Scholar
Pederson, T. & Tsai, R. Y. In search of nonribosomal nucleolar protein function and regulation. J. Cell Biol. 184, 771–776 (2009).
CAS PubMed PubMed Central Google Scholar
Warner, J. R. & McIntosh, K. B. How common are extraribosomal functions of ribosomal proteins? Mol. Cell 34, 3–11 (2009).
CAS PubMed PubMed Central Google Scholar
Andersen, J. S. et al. Nucleolar proteome dynamics. Nature 433, 77–83 (2005).
ADS CAS PubMed Google Scholar
Boisvert, F. M., van Koningsbruggen, S., Navascues, J. & Lamond, A. I. The multifunctional nucleolus. Nat. Rev. Mol. Cell Biol. 8, 574–585 (2007).
CAS PubMed Google Scholar
Boulon, S., Westman, B. J., Hutten, S., Boisvert, F. M. & Lamond, A. I. The nucleolus under stress. Mol. Cell 40, 216–227 (2010).
CAS PubMed PubMed Central Google Scholar
Emmott, E. & Hiscox, J. A. Nucleolar targeting: the hub of the matter. EMBO Rep. 10, 231–238 (2009).
CAS PubMed PubMed Central Google Scholar
Lindstrom, M. S. NPM1/B23: a multifunctional chaperone in ribosome biogenesis and chromatin remodeling. Biochem. Res. Int. 2011, 195209 (2011).
PubMed Google Scholar
Yao, Z. et al. B23 acts as a nucleolar stress sensor and promotes cell survival through its dynamic interaction with hnRNPU and hnRNPA1. Oncogene 29, 1821–1834 (2010).
CAS PubMed Google Scholar
Zhang, Y. & Lu, H. Signaling to p53: ribosomal proteins find their way. Cancer Cell 16, 369–377 (2009).
CAS PubMed PubMed Central Google Scholar
Rubbi, C. P. & Milner, J. Disruption of the nucleolus mediates stabilization of p53 in response to DNA damage and other stresses. EMBO J. 22, 6068–6077 (2003).
CAS PubMed PubMed Central Google Scholar
Holmberg Olausson, K., Nister, M. & Lindstrom, M. S. p53 -dependent and independent nucleolar stress responses. Cells 1, 774–798 (2012).
CAS PubMed Google Scholar
Chan, P. K., Bloom, D. A. & Hoang, T. T. The N-terminal half of NPM dissociates from nucleoli of HeLa cells after anticancer drug treatments. Biochem. Biophys. Res. Commun. 264, 305–309 (1999).
CAS PubMed Google Scholar
Kodiha, M., Banski, P. & Stochaj, U. Computer-based fluorescence quantification: a novel approach to study nucleolar biology. BMC Cell Biol. 12, 25 (2011).
CAS PubMed PubMed Central Google Scholar
Kurki, S. et al. Nucleolar protein NPM interacts with HDM2 and protects tumor suppressor protein p53 from HDM2-mediated degradation. Cancer Cell 5, 465–475 (2004).
CAS PubMed Google Scholar
Colombo, E., Marine, J. C., Danovi, D., Falini, B. & Pelicci, P. G. Nucleophosmin regulates the stability and transcriptional activity of p53. Nat. Cell Biol. 4, 529–533 (2002).
CAS PubMed Google Scholar
Schmidt-Zachmann, M. S., Hugle-Dorr, B. & Franke, W. W. A constitutive nucleolar protein identified as a member of the nucleoplasmin family. EMBO J. 6, 1881–1890 (1987).
CAS PubMed PubMed Central Google Scholar
Negi, S. S. & Olson, M. O. Effects of interphase and mitotic phosphorylation on the mobility and location of nucleolar protein B23. J. Cell Sci. 119, 3676–3685 (2006).
CAS PubMed Google Scholar
Liu, X. et al. Sumoylation of nucleophosmin/B23 regulates its subcellular localization, mediating cell proliferation and survival. Proc. Natl Acad. Sci. USA 104, 9679–9684 (2007).
ADS CAS PubMed Google Scholar
Okuwaki, M. The structure and functions of NPM1/Nucleophsmin/B23, a multifunctional nucleolar acidic protein. J. Biochem. 143, 441–448 (2008).
CAS PubMed Google Scholar
Guillonneau, M. et al. Oxidative stress disassembles the p38/NPM/PP2A complex, which leads to modulation of nucleophosmin-mediated signaling to DNA damage response. FASEB J. 30, 2899–2914 (2016).
CAS PubMed Google Scholar
Wang, Y., Yang, J. & Yi, J. Redox sensing by proteins: oxidative modifications on cysteines and the consequent events. Antioxid. Redox Signal. 16, 649–657 (2012).
PubMed Google Scholar
Schieber, M. & Chandel, N. S. ROS function in redox signaling and oxidative stress. Curr. Biol. 24, R453–R462 (2014).
CAS PubMed PubMed Central Google Scholar
Dooley, C. T. et al. Imaging dynamic redox changes in mammalian cells with green fluorescent protein indicators. J. Biol. Chem. 279, 22284–22293 (2004).
CAS PubMed Google Scholar
Grummitt, C. G., Townsley, F. M., Johnson, C. M., Warren, A. J. & Bycroft, M. Structural consequences of nucleophosmin mutations in acute myeloid leukemia. J. Biol. Chem. 283, 23326–23332 (2008).
CAS PubMed PubMed Central Google Scholar
Nishimura, Y., Ohkubo, T., Furuichi, Y. & Umekawa, H. Tryptophans 286 and 288 in the C-terminal region of protein B23.1 are important for its nucleolar localization. Biosci. Biotechnol. Biochem. 66, 2239–2242 (2002).
CAS PubMed Google Scholar
Falini, B. et al. Cytoplasmic nucleophosmin in acute myelogenous leukemia with a normal karyotype. N. Engl. J. Med. 352, 254–266 (2005).
CAS PubMed Google Scholar
Chan, P. K. & Chan, F. Y. Nucleophosmin/B23 (NPM) oligomer is a major and stable entity in HeLa cells. Biochim. Biophys. Acta 1262, 37–42 (1995).
PubMed Google Scholar
Townsend, D. M. et al. Novel role for glutathione S-transferase pi. Regulator of protein S-Glutathionylation following oxidative and nitrosative stress. J. Biol. Chem. 284, 436–445 (2009).
CAS PubMed PubMed Central Google Scholar
Gravina, S. A. & Mieyal, J. J. Thioltransferase is a specific glutathionyl mixed disulfide oxidoreductase. Biochemistry 32, 3368–3376 (1993).
CAS PubMed Google Scholar
Bhaskar, A. et al. Reengineering redox sensitive GFP to measure mycothiol redox potential of Mycobacterium tuberculosis during infection. PLoS Pathog. 10, e1003902 (2014).
PubMed PubMed Central Google Scholar
Gibbs-Flournoy, E. A., Simmons, S. O., Bromberg, P. A., Dick, T. P. & Samet, J. M. Monitoring intracellular redox changes in ozone-exposed airway epithelial cells. Environ. Health Perspect. 121, 312–317 (2013).
PubMed Google Scholar
Gutscher, M. et al. Real-time imaging of the intracellular glutathione redox potential. Nat. Methods 5, 553–559 (2008).
CAS PubMed Google Scholar
Chiarella, S. et al. Nucleophosmin mutations alter its nucleolar localization by impairing G-quadruplex binding at ribosomal DNA. Nucleic Acids Res. 41, 3228–3239 (2013).
CAS PubMed PubMed Central Google Scholar
Kuroda, T. et al. RNA content in the nucleolus alters p53 acetylation via MYBBP1A. EMBO J. 30, 1054–1066 (2011).
CAS PubMed PubMed Central Google Scholar
Savkur, R. S. & Olson, M. O. Preferential cleavage in pre-ribosomal RNA byprotein B23 endoribonuclease. Nucleic Acids Res. 26, 4508–4515 (1998).
CAS PubMed PubMed Central Google Scholar
Grandori, C. et al. c-Myc binds to human ribosomal DNA and stimulates transcription of rRNA genes by RNA polymerase I. Nat. Cell Biol. 7, 311–318 (2005).
CAS PubMed Google Scholar
Murano, K., Okuwaki, M., Hisaoka, M. & Nagata, K. Transcription regulation of the rRNA gene by a multifunctional nucleolar protein, B23/nucleophosmin, through its histone chaperone activity. Mol. Cell Biol. 28, 3114–3126 (2008).
CAS PubMed PubMed Central Google Scholar
Anil, B., Riedinger, C., Endicott, J. A. & Noble, M. E. The structure of an MDM2-Nutlin-3a complex solved by the use of a validated MDM2 surface-entropy reduction mutant. Acta Crystallogr. D Biol. Crystallogr. 69, 1358–1366 (2013).
CAS PubMed Google Scholar
Jin, A., Itahana, K., O'Keefe, K. & Zhang, Y. Inhibition of HDM2 and activation of p53 by ribosomal protein L23. Mol. Cell Biol. 24, 7669–7680 (2004).
CAS PubMed PubMed Central Google Scholar
Dai, M. S. & Lu, H. Inhibition of MDM2-mediated p53 ubiquitination and degradation by ribosomal protein L5. J. Biol. Chem. 279, 44475–44482 (2004).
CAS PubMed Google Scholar
Sasaki, M. et al. Regulation of the MDM2-P53 pathway and tumor growth by PICT1 via nucleolar RPL11. Nat. Med. 17, 944–951 (2011).
CAS PubMed PubMed Central Google Scholar
Lindstrom, M. S. & Zhang, Y. B23 and ARF: friends or foes? Cell Biochem. Biophys. 46, 79–90 (2006).
CAS PubMed Google Scholar
Zhang, Y. The ARF-B23 connection: implications for growth control and cancer treatment. Cell Cycle 3, 259–262 (2004).
CAS PubMed Google Scholar
Zhu, C. et al. K259-SUMOylation of DGCR8 promoted by p14ARF exerts a tumor-suppressive function. J. Mol. Cell Biol. 8, 456–458 (2016).
Google Scholar
Clark, P. A., Llanos, S. & Peters, G. Multiple interacting domains contribute to p14ARF mediated inhibition of MDM2. Oncogene 21, 4498–4507 (2002).
CAS PubMed Google Scholar
Zhang, Y., Xiong, Y. & Yarbrough, W. G. ARF promotes MDM2 degradation and stabilizes p53: ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways. Cell 92, 725–734 (1998).
CAS PubMed Google Scholar
Llanos, S., Clark, P. A., Rowe, J. & Peters, G. Stabilization of p53 by p14ARF without relocation of MDM2 to the nucleolus. Nat. Cell Biol. 3, 445–452 (2001).
CAS PubMed Google Scholar
Albrecht, S. C., Barata, A. G., Grosshans, J., Teleman, A. A. & Dick, T. P. In vivo mapping of hydrogen peroxide and oxidized glutathione reveals chemical and regional specificity of redox homeostasis. Cell Metab. 14, 819–829 (2011).
CAS PubMed Google Scholar
Belousov, V. V. et al. Genetically encoded fluorescent indicator for intracellular hydrogen peroxide. Nat. Methods 3, 281–286 (2006).
CAS PubMed Google Scholar
Yung, B. Y., Busch, R. K., Busch, H., Mauger, A. B. & Chan, P. K. Effects of actinomycin D analogs on nucleolar phosphoprotein B23 (37,000 daltons/pI 5.1). Biochem. Pharmacol. 34, 4059–4063 (1985).
CAS PubMed Google Scholar
Matthews, D. A. Adenovirus protein V induces redistribution of nucleolin and B23 from nucleolus to cytoplasm. J. Virol. 75, 1031–1038 (2001).
CAS PubMed PubMed Central Google Scholar
Yogev, O., Saadon, K., Anzi, S., Inoue, K. & Shaulian, E. DNA damage-dependent translocation of B23 and p19 ARF is regulated by the Jun N-terminal kinase pathway. Cancer Res. 68, 1398–1406 (2008).
CAS PubMed Google Scholar
Duangmano, S., Sae-Lim, P., Suksamrarn, A., Domann, F. E. & Patmasiriwat, P. Cucurbitacin B inhibits human breast cancer cell proliferation through disruption of microtubule polymerization and nucleophosmin/B23 translocation. BMC Complement. Altern. Med. 12, 185 (2012).
CAS PubMed PubMed Central Google Scholar
Kar, B., Liu, B., Zhou, Z. & Lam, Y. W. Quantitative nucleolar proteomics reveals nuclear re-organization during stress- induced senescence in mouse fibroblast. BMC Cell Biol. 12, 33 (2011).
CAS PubMed PubMed Central Google Scholar
Kurki, S., Peltonen, K. & Laiho, M. Nucleophosmin, HDM2 and p53: players in UV damage incited nucleolar stress response. Cell Cycle 3, 976–979 (2004).
CAS PubMed Google Scholar
Hingorani, K., Szebeni, A. & Olson, M. O. Mapping the functional domains of nucleolar protein B23. J. Biol. Chem. 275, 24451–24457 (2000).
CAS PubMed Google Scholar
Dumbar, T. S., Gentry, G. A. & Olson, M. O. Interaction of nucleolar phosphoprotein B23 with nucleic acids. Biochemistry 28, 9495–9501 (1989).
CAS PubMed Google Scholar
Federici, L. et al. Nucleophosmin C-terminal leukemia-associated domain interacts with G-rich quadruplex forming DNA. J. Biol. Chem. 285, 37138–37149 (2010).
CAS PubMed PubMed Central Google Scholar
Townsend, D. M. et al. A glutathione S-transferase pi-activated prodrug causes kinase activation concurrent with S-glutathionylation of proteins. Mol. Pharmacol. 69, 501–508 (2006).
CAS PubMed Google Scholar
Schwarzlander, M., Dick, T. P., Meyer, A. J. & Morgan, B. Dissecting redox biology using fluorescent protein sensors. Antioxid. Redox Signal. 24, 680–712 (2016).
PubMed Google Scholar
Honda, R. & Yasuda, H. Association of p19(ARF) with Mdm2 inhibits ubiquitin ligase activity of Mdm2 for tumor suppressor p53. EMBO J. 18, 22–27 (1999).
CAS PubMed PubMed Central Google Scholar
Hesse, R. G., Kouklis, G. K., Ahituv, N. & Pomerantz, J. H. The human ARF tumor suppressor senses blastema activity and suppresses epimorphic tissue regeneration. eLife 4, 07702 (2015).
Google Scholar
Pomerantz, J. et al. The Ink4a tumor suppressor gene product, p19Arf, interacts with MDM2 and neutralizes MDM2's inhibition of p53. Cell 92, 713–723 (1998).
CAS PubMed Google Scholar
Chen, D., Shan, J., Zhu, W. G., Qin, J. & Gu, W. Transcription-independent ARF regulation in oncogenic stress-mediated p53 responses. Nature 464, 624–627 (2010).
ADS CAS PubMed PubMed Central Google Scholar
Korgaonkar, C. et al. Nucleophosmin (B23) targets ARF to nucleoli and inhibits its function. Mol. Cell Biol. 25, 1258–1271 (2005).
CAS PubMed PubMed Central Google Scholar
Colombo, E. et al. Nucleophosmin is required for DNA integrity and p19 Arf protein stability. Mol. Cell Biol. 25, 8874–8886 (2005).
CAS PubMed PubMed Central Google Scholar
Itahana, K. et al. Tumor suppressor ARF degrades B23, a nucleolar protein involved in ribosome biogenesis and cell proliferation. Mol. Cell 12, 1151–1164 (2003).
CAS PubMed Google Scholar
Moulin, S., Llanos, S., Kim, S. H. & Peters, G. Binding to nucleophosmin determines the localization of human and chicken ARF but not its impact on p53. Oncogene 27, 2382–2389 (2008).
CAS PubMed Google Scholar
Wang, L. et al. Intracellular CD24 disrupts the ARF-NPM interaction and enables mutational and viral oncogene-mediated p53 inactivation. Nat. Commun. 6, 5909 (2015).
CAS PubMed PubMed Central Google Scholar
Lee, C., Smith, B. A., Bandyopadhyay, K. & Gjerset, R. A. DNA damage disrupts the p14ARF-B23(nucleophosmin) interaction and triggers a transient subnuclear redistribution of p14ARF. Cancer Res. 65, 9834–9842 (2005).
CAS PubMed Google Scholar
Gjerset, R. A. & Bandyopadhyay, K. Regulation of p14ARF through subnuclear compartmentalization. Cell Cycle 5, 686–690 (2006).
CAS PubMed Google Scholar
Colombo, E., Alcalay, M. & Pelicci, P. G. Nucleophosmin and its complex network: a possible therapeutic target in hematological diseases. Oncogene 30, 2595–2609 (2011).
CAS PubMed Google Scholar
Grisendi, S., Mecucci, C., Falini, B. & Pandolfi, P. P. Nucleophosmin and cancer. Nat. Rev. Cancer 6, 493–505 (2006).
CAS PubMed Google Scholar
Wulff, J. E., Siegrist, R. & Myers, A. G. The natural product avrainvillamide binds to the oncoprotein nucleophosmin. J. Am. Chem. Soc. 129, 14444–14451 (2007).
CAS PubMed PubMed Central Google Scholar
Falini, B. et al. Acute myeloid leukemia with mutated nucleophosmin (NPM1): any hope for a targeted therapy? Blood Rev. 25, 247–254 (2011).
CAS PubMed Google Scholar
Falini, B., Brunetti, L. & Martelli, M. P. Dactinomycin in NPM1-mutated acute myeloid leukemia. N. Engl. J. Med. 373, 1180–1182 (2015).
PubMed Google Scholar
Qi, W. et al. NSC348884, a nucleophosmin inhibitor disrupts oligomer formation and induces apoptosis in human cancer cells. Oncogene 27, 4210–4220 (2008).
CAS PubMed Google Scholar
Jian, Y. et al. RNA aptamers interfering with nucleophosmin oligomerization induce apoptosis of cancer cells. Oncogene 28, 4201–4211 (2009).
CAS PubMed Google Scholar
Duckert, H. et al. Natural product-inspired cascade synthesis yields modulators of centrosome integrity. Nat. Chem. Biol. 8, 179–184 (2012).
Google Scholar
De Cola, A. et al. AML cells carrying NPM1 mutation are resistant to nucleophosmin displacement from nucleoli caused by the G-quadruplex ligand TmPyP4. Cell Death Dis. 5, e1427 (2014).
CAS PubMed PubMed Central Google Scholar
Mukherjee, H. et al. Interactions of the natural product (+)-avrainvillamide with nucleophosmin and exportin-1 mediate the cellular localization of nucleophosmin and its AML-associated mutants. ACS Chem. Biol. 10, 855–863 (2015).
CAS PubMed PubMed Central Google Scholar
Zhao, Y. et al. Genetically encoded fluorescent sensors for intracellular NADH detection. Cell Metab. 14, 555–566 (2011).
CAS PubMed PubMed Central Google Scholar
Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126–2132 (2004).
PubMed Google Scholar
Zhang, W. et al. An airborne transmissible avian influenza H5 hemagglutinin seen at the atomic level. Science 340, 1463–1467 (2013).
ADS CAS PubMed Google Scholar

Download references

Acknowledgements

We thank Drs Benjamin Y.M. Yung of The Hong Kong Polytechnic University, Chuanmao Zhang of Peking University and Jim Remington of Oregon State University for their gifts of the plasmids, the pcDNA3.1-FLAG-B23 and the mCherry-C1-H2B and roGFP1, respectively. This work was supported by grants from the National Natural Science Foundation of China (31230037 and 31471263), the Ministry of Science and Technology of China (2013CB910900) and Shanghai Municipal Science and Technology Commission 11JC1406900, 11DZ2260200, 16ZR1418400 and 15YF1402600, the Lift Engineering for Young Talent of China Association for Science and Technology (to Y.Z.).

Author information

yijing@shsmu.edu.cn

Authors and Affiliations

Department of Biochemistry and Molecular Cell Biology, Shanghai Key Laboratory for Tumor Microenvironment and Inflammation, Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, Shanghai Jiao Tong University School of Medicine, 280 South Chongqing Road, Shanghai, 200025, China
Kai Yang, Ming Wang, Xuxu Sun, Huilin Chu, Jie Yang & Jing Yi
Synthetic Biology and Biotechnology Laboratory, State Key Laboratory of Bioreactor Engineering, School of Pharmacy, East China University of Science and Technology, 130 Mei Long Road, Shanghai, 200237, China
Yuzheng Zhao, Yi Yang & Xie Li
Department of Pathophysiology, Shanghai Jiao Tong University School of Medicine, 280 South Chongqing Road, Shanghai, 200025, China
Aiwu Zhou
Shanghai Institute of Materia Medica, 555 Zu Chong Zhi Road, Zhang Jiang Hi-Tech Park, Shanghai, 201203, China
Hu Zhou
Department of Pharmacology, Shanghai Jiao Tong University School of Medicine, 280 South Chongqing Road, Shanghai, 200025, China
Jianrong Xu
School of Life Sciences, University of Science and Technology of China, 96 Jinzhai Road, Hefei, 230022, Anhui, China
Mian Wu

Authors

Kai Yang
View author publications
You can also search for this author in PubMed Google Scholar
Ming Wang
View author publications
You can also search for this author in PubMed Google Scholar
Yuzheng Zhao
View author publications
You can also search for this author in PubMed Google Scholar
Xuxu Sun
View author publications
You can also search for this author in PubMed Google Scholar
Yi Yang
View author publications
You can also search for this author in PubMed Google Scholar
Xie Li
View author publications
You can also search for this author in PubMed Google Scholar
Aiwu Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Huilin Chu
View author publications
You can also search for this author in PubMed Google Scholar
Hu Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Jianrong Xu
View author publications
You can also search for this author in PubMed Google Scholar
Mian Wu
View author publications
You can also search for this author in PubMed Google Scholar
Jie Yang
View author publications
You can also search for this author in PubMed Google Scholar
Jing Yi
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

K.Y., J. Yi, J. Yang and Y.Y. designed experiments. K.Y., M.W., Y.Z., X.S., X.L. and H.C. conducted experiments. A.Z. analysed structure and mutation. H.Z. analysed mass spectrum data. J.X. conducted SPR. Y.Y. and M.W. provided plasmids. K.Y. and J. Yi wrote the manuscript.

Corresponding authors

Correspondence to Jie Yang or Jing Yi.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Information

Supplementary Figure 1-7, Supplementary Table 1 (PDF 27917 kb)

Peer Review File (PDF 4645 kb)

Rights and permissions

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Reprints and permissions

About this article

Cite this article

Yang, K., Wang, M., Zhao, Y. et al. A redox mechanism underlying nucleolar stress sensing by nucleophosmin. Nat Commun 7, 13599 (2016). https://doi.org/10.1038/ncomms13599

Download citation

Received: 12 January 2016
Accepted: 18 October 2016
Published: 25 November 2016
DOI: https://doi.org/10.1038/ncomms13599

This article is cited by

Structural Changes in Rat Hepatocyte Nucleolus under Nucleolar Stress Caused by Hypothermia
- I. P. Bobrov
- A. Yu. Dolgatov
- E. L. Lushnikova
Bulletin of Experimental Biology and Medicine (2024)
The thioesterase APT1 is a bidirectional-adjustment redox sensor
- Tuo Ji
- Lihua Zheng
- Tao Wang
Nature Communications (2023)
Targeting selenoprotein H in the nucleolus suppresses tumors and metastases by Isovalerylspiramycin I
- Jing Cui
- Jingcheng Zhou
- Zhengping Zhuang
Journal of Experimental & Clinical Cancer Research (2022)
A role for the NPM1/PTPN14/YAP axis in mediating hypoxia-induced chemoresistance to sorafenib in hepatocellular carcinoma
- Dengke Zhang
- Fazong Wu
- Jiansong Ji
Cancer Cell International (2022)
Nucleolar protein NOC4L inhibits tumorigenesis and progression by attenuating SIRT1-mediated p53 deacetylation
- Xiaomeng Jia
- Huijiao Liu
- Xiangdong Li
Oncogene (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.