PCNA and JNK1‐Stat3 pathways respectively promotes and inhibits diabetes‐associated centrosome amplification by targeting at the ROCK1/14‐3‐3σ complex in human colon cancer HCT116 cells

Abstract We have recently reported that type 2 diabetes promotes centrosome amplification via enhancing the expression, biding, and centrosome translocation of rho‐associated coiled‐coil containing protein kinase 1 (ROCK1)/14‐3‐3σ complex in HCT116 cells. In the functional proteomic study, we further investigated the molecular pathways underlying the centrosome amplification using HCT116 cells. We found that treatment of HCT116 cells with high glucose, insulin, and palmitic acid triggered the centrosome amplification and increased the expressions of proliferating cell nuclear antigen (PCNA), nucleophosmin (NPM), and 14‐3‐3σ. Individual knockdown of PCNA, NPM, or 14‐3‐3σ inhibited the centrosome amplification. Knockdown of PCNA inhibited the treatment‐increased expression of ROCK1, whereas knockdown of ROCK1 did not affect the PCNA expression. High glucose, insulin, and palmitic acid also increased the expressions of c‐Jun N‐terminal kinase‐1 (JNK1) and signal transducer and activator of transcription 3 (Stat3), individual knockdown of which upregulated the treatment‐increased expression of 14‐3‐3σ and promoted the centrosome amplification. In contrast, overexpression of JNK1 inhibited the centrosome amplification. Knockdown of Stat3 enhanced the centrosome translocation of 14‐3‐3σ. Moreover, we showed that knockdown of JNK1 inhibited the treatment‐increased expression of Stat3. Knockdown of PCNA, JNK1, or Stat3 did not have an effect on NPM and vice versa. In conclusion, our results suggest that PCNA and JNK1‐Stat3 pathways respectively promotes and feedback inhibits the centrosome amplification by targeting at the ROCK1/14‐3‐3σ complex, and NPM serves as an independent signal for the centrosome amplification.


| INTRODUCTION
Empty two characters Centrosomes are small nonmembrane organelles composed of two centrioles surrounded by pericentriolar material in mammalian and lower plant cells. They are the microtubule organizing centers, which are important for cell division and the maintenance of genomic stability (Nigg & Raff, 2009). A normal mitotic cell contains two centrosomes oriented on opposite poles of the cell. These centrosomes are responsible for generating bipolar mitotic spindles which promote proper segregation of chromosomes into two equal daughter cells. Each of the daughter cells contains one centrosome (D'Assoro, Lingle, & Salisbury, 2002).
Centrosome amplification, acquisition of more than two centrosomes in a cell, has been described in various types of cancers, including both solid tumors and hematological malignancies (Nigg, 2006). It severely disturbs mitotic process and cytokinesis due to formation of more than two spindle poles, resulting in chromosome instability due to chromosome missegregation. Notably, Basto et al. (2008) reported that centrosomes amplification is sufficient to initiate tumorigenesis in flies. In a transgenic mice model, centrosome amplification is sufficient to initiate spontaneous tumorigenesis (Levine et al., 2017). These observations strongly suggest that there is a cause and consequence relationship between centrosome amplification and tumorigenesis. Moreover, Godinho et al. (2014) have shown that centrosome amplification can promote cancer cell invasion, which may explain the association between centrosome amplification and poor prognosis in cancer patients (Chan, 2011).
Type 2 diabetes is a common disease, which confers an increased risk for all-site cancer (Giovannucci et al., 2010) and worsens the cancer prognosis (Mills, Bellows, Hoffman, Kelly, & Gagliardi, 2013). In the development of type 2 diabetes (DeFronzo et al., 2015), it is currently recognized that insulin resistance occurs in prediabetic state that shows hyperinsulinemia due to the compensation for reduced insulin sensitivity. Clinical diabetes appears when fasting blood glucose is abnormally elevated. Hyperinsulinemia persists at this state; however, elevated insulin level is already unable to maintain normal fasting blood glucose level in light of insulin resistance. At the later stage of type 2 diabetes, insulin deficiency develops due to the loss of islet beta cells, with persisting hyperglycemia. Continuous hypersecretion of insulin for compensating insulin insensitivity exhausts beta cells, which leads to beta cell death. Elevated level of free fatty acids is noted since the prediabetic state. Hyperinsulinemia, hyperglycemia, and increase in free fatty acids are present over a long period of time in the course of type 2 diabetes. Palmitic acid, the most common saturated free fatty acid, is often used to investigate the effects of free fatty acids, in particular the adverse effects (Mancini et al., 2015).
Recently, we have reported that type 2 diabetes is associated with an increased level of centrosome amplification in peripheral blood mononuclear cells (PBMC). In HCT116 cell line model in vitro, high glucose, insulin, and palmitic acid increase the expression, binding, and centrosomal translocation of rho-associated coiled-coil containing protein kinase 1 (ROCK1) and 14-3-3σ. These molecular events are also seen in PBMC from patients with type 2 diabetes (Wang et al., 2018). Our results suggest that centrosome amplification is a candidate biological link between type 2 diabetes and cancer development.
At the cellular level, it is believed that there are four major potential mechanisms for centrosome amplification: (1) Multiple centrosome duplication in one cell cycle can occur when a cell division and centrosome duplication cycle become uncoupled and the centrosome duplication cycle continues whereas the cell cycle is stalled (Fukasawa, 2005); (2) the failure of cytokinesis when a cell fails to complete cytokinesis at the end of mitosis and centrosomes remain in a cell rather then enter into daughter cells (Fukasawa, 2005); (3) abnormal centriole splitting, when two G2 phase centrosomes split to form four centrosomes with one centriole (Fukasawa, 2005); (4) the acentriolar centrosome formation, which is the generation of centrosomes that do not contain centrioles (Fukasawa, 2005).
At the molecular level, many genes and proteins have been shown to promote centrosome amplification. In flies and humans, the overexpression of SAK/PLK4 can drive centriole over-duplication (Holland et al., 2012). Aurora-A does not deregulate centrosome duplication but gives rise to extra centrosomes through defects in cell division and consequent tetraploidization (Meraldi, Honda, & Nigg, 2002). Overexpression of Aurora-A, which is frequently amplified in tumors (Lens, Voest, & Medema, 2010), has been shown to override the spindle assembly checkpoint and thereby results in tetraploid cells which are due to cell division defects (Anand, Penrhyn-Lowe, & Venkitaraman, 2003;Meraldi et al., 2002). The loss of p53 and the deregulated expression of its regulators Mdm2 as well as downstream targets p21 Cip1 and Gadd45 have been linked to centrosome amplification (Meraldi et al., 2002). BRCA1/2 and ATR also affect centrosome numbers in response to DNA damage (Tutt et al., 1999).
Despite that many genes and proteins are shown to promote centrosome amplification, limited amount of data can be found on signal transduction pathway. Arquint and Nigg (2016) reported that STIL and SAS-6 are downstream signals of Plk4, forming a Plk4-STIL-SAS-6 pathway. The interaction of the pathway initiated centriole duplication, and overexpression of any one of the components of the pathway resulted centrosome amplification (Arquint & Nigg, 2016).
Moreover, in breast cancer cells, E2F regulated centrosome amplification in part through Nek2. Another finding is that Cdk4/Nek2 signaling modulated centrosome amplification (Harrison Pitner & Saavedra, 2013;Lee, Moreno, & Saavedra, 2014). The present study further investigated the signals and signal transduction pathways underlying the centrosome amplification by high glucose, insulin, and palmitic acid using functional proteomic analysis in combination with candidate protein characterization.

| Two dimensional gel electrophoresis (2DE)
Two dimensional gel electrophoresis (2-DE) analyses were performed using immobilized pH gradient (IPG) with a linear pH gradient from 4 to 7 as the first dimensional gels. IPG strips were rehydrated in re-swelling buffer (8 M urea, 2% CHAPS, 0.5% pharmalyte, 0.2% DTT) for first dimensional electrophoresis in an IPGphor machine (500 v 1 hr, 1,000 v 1 hr, 3,500 v 1 hr, gradient 8,000 v 1 hr, and 24,000 vh). Total cellular protein samples were applied using sample loading cup method. Focused IPG strips were frozen at −20°C till use.
Before the second-dimensional run, IPG strips were equilibrated

| Confocal microscopy
A cover slip was placed in a well of a 6-well plate. HCT116 cells were plated at a density of 50,000 cells per well. Cells grown on the cover slips were fixed in cold methanol and acetone (1:1; v:v) for 6 min at −20°C, followed by three washes with phosphate-buffered saline (PBS; 10 min each time). Then, the cells were incubated with 0.1% Triton X-100 for 15 min and 3% bovine serum albumin (BSA) for 1 hr.
The cells were incubated with a primary antibody in 3% BSA in PBS overnight at 4°C, washed two times with PBS, and incubated with a FITC-conjugated secondary antibody in 3% BSA in 1 × PBS for 1 hr at room temperature in the dark. Finally, the cells were mounted with mounting medium. Confocal microscopy was performed using the Zeiss LSM880 microscope (Oberkochen, Germany) with a 1.4 NA oilimmersion lens, and image processing was performed with Zen software (Oberkochen, Germany).

| Western blot analysis
HCT116 cells were lysed in radioimmunoprecipitation assay (RIPA) buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.2, 1% Triton X-100%, and 0.1% SDS). Proteins were separated by polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride (PVDF) membrane. After blocking for 1 hr at room temperature with TBST containing 0.05% (v/v) Tween-20% and 5% (w/v) nonfat milk, the membranes were incubated with primary antibodies overnight at 4°C, followed by washes with TBST containing 0.05% Tween-20. The membranes were then incubated with a horseradish peroxidase-conjugated secondary antibody for 1 hr at room temperature. ECL reagents (Thermo Biosciences, MA) were used to visualize the protein bands which were captured on X-ray film.

| Statistical analysis
All the experiments were performed in triplicate. The data are expressed as the mean ± SD. Multigroup comparisons were performed using one-way analysis of variance (ANOVA) analysis. The statistical analysis software package SPSS21 was used for the statistical comparisons. A p value < 0.05 was considered to be statistically significant. ) samples were separated on two dimensional gels, and the images of control and treated samples were compared. We found that nine peptide spots were differentially expressed, with two peptide spots downregulated and seven peptide spots upregulated (Figures 2a-c). The identities of these proteins were obtained using the PMF technology (Table 1). The calibrated molecular weight and isoelectric point (pI) values were comparable to the theoretical molecular and pI values of the proteins (Table 1), which supported the identification by PMF.
Amongst the nine proteins, we were interested in PCNA, NPM, and 14-3-3σ, which were chosen for further functional analyses.
Western blot analyses confirmed that the expression levels of PCNA,

| PCNA targets at ROCK1 to promote the centrosome amplification
We have recently reported that the ROCK1/14-3-3σ complex mediated the centrosome amplification (Wang et al., 2018). In the present study, we were interested to investigate the relationships between ROCK1/14-3-3σ complex and PCNA as well as NPM. We found that NPM was not upstream or downstream of others.
However, the knockdown of PCNA inhibited the treatment-  Figures 4b,c). In contrast, the knockdown of ROCK1 did not change the expression level of PCNA (Supporting Information Figure S1). F I G U R E 2 Proteomic analysis identifies nine peptides associated with the centrosome amplification. (a) and (b) Global view of the peptide spots on the two dimensional gels of the control and treated samples, respectively. The differentially expressed spots were labeled. (c) Localized view of the differentially expressed peptide spots on the two dimensional gels. (d-f) Western blot analyses confirmed that the protein levels of PCNA, NPM, and 14-3-3σ were upregulated after the treatment, respectively, which was inhibited by their siRNA. Glu: glucose, 15 mM; Ins: insulin, 5 nM; Pal: palmitic acid, 150 μM. NPM: nucleophosmin; PCNA: proliferating cell nuclear antigen; siRNA: small interfering RNA LU ET AL.

| JNK1 and Stat3 inhibit the centrosome amplification
In an experiment which examined candidate proteins for centrosome amplification. We found that the expression and activation levels of JNK1 (Figure 5a), Stat3 (Figure 5b) were both upregulated by high glucose, insulin, and palmitic acid, which were inhibited by their specific siRNA (Figures 5a,b). Interestingly, the knockdown of JNK1

| DISCUSSION
We have recently reported that increased expression and centrosomal translocation of ROCK1/14-3-3σ complex mediates centrosome amplification by high glucose, insulin, and palmitic acid (Wang et al., 2018). In the present study, we confirmed that high glucose, insulin, and palmitic acid could induce centrosome amplification (Figures 1a,   b) 14-3-3σ is a signal mediator (Figures 2f and 3c), agreeing with our previous report (Wang et al., 2018). Through proteomic analysis, we identified nine proteins associated with the centrosome amplification, which included PCNA, NPM, and 14-3-3σ (Figures 2a-c; Table   1). Results from functional assays showed that PCNA targeted at ROCK1 to promote centrosome amplification (Figures 3a and 4a-c).
NPM remained to be an independent signal, since knockdown of others did not change its expression or phosphorylation, and knockdown of NPM did not affect other signals (Supporting Information Figure S2).
Knockdown of Stat3 enhanced the treatment-increased centrosome translocation of 14-3-3σ (Figures 6d,e). In contrast, the overexpression of JNK1 inhibited the centrosome amplification. Thus, the experimental treatment activated the JNK1-Stat3 pathway (Figures   5a,b) as a feedback loop that targeted at 14-3-3σ in the ROCK1/14-3-3σ complex to inhibit the centrosome amplification. Therefore, the experimental treatment activated PCNA-ROCK1 and NPM pathways to promote centrosome amplification. It simultaneously activated a feedback JNK1-Stat3 pathway to inhibit the centrosome amplification ( Figure 7). Therefore, it is assumed that the feedback inhibition loop is not activated powerfully enough to counteract the actions of the centrosome amplification-promoting pathways, thus, centrosome amplification occurs upon the treatment.
PCNA is a protein that is found to be in the nucleus, which functions as a cofactor of DNA polymerase for DNA synthesis T A B L E 1 Identification of proteins of interests using peptide mass fingerprinting  (Zhao et al., 2012). However, in colorectal cancer, the expression of Aurora members does not correlate with the PCNA expression (Takahashi et al., 2000). Our results showed that PCNA controlled the expression of 14-3-3σ and its translocation to the centrosomes (Figures 4a,b) for centrosome amplification. This, for the NPM, also named as B23, localized in granular regions of the nucleolus and has been shown to be associated with preribosomal particles (Okuda et al., 2000). It is identified as one of the targets of CDK2/cyclin E, which can enhance centrosome duplication. Nonphosphorylated NPM binds to centrosomes (Tokuyama, Horn, Kawamura, Tarapore, & Fukasawa, 2001). After Cdk2 and Cyclin E phosphorylate Thr199 in NPM, NPM dissociates from centrosomes, which in turn triggers centrosome duplication (Grisendi et al., 2005).
NPM loss leads to unrestricted centrosome duplication and genomic instability in mouse embryonic fibroblasts (Xia et al., 2013). In the present experiment, we found that the expression and phosphorylation levels of NPM were upregulated by the treatment of high glucose, insulin, and palmitic acid, which concomitantly occurred with enhanced centrosome amplification (Figures 2e and 3b). Knockdown of NPM using siRNA inhibited the treatment-induced upregulation of NPM expression, NPM phosphorylation, and centrosome amplification ( Figure 2e). Our results suggest that NPM mediates centrosome amplification, which is in agreement with the observations by others (Grisendi et al., 2005).
The c-Jun N-terminal kinases, a member of mitogen-activated protein kinase family of serine/threonine kinases, are ubiquitously expressed intracellular signaling molecules that are activated by cytokines and exposure to environmental stresses such as osmotic stress, redox stress, and radiation (Kyriakis & Avruch, 2001). More recent experimental data have shown that JNK signaling is involved in the regulation of cellular processes, including apoptosis and survival signaling, T-cell maturation, brain development, cardiac hypertrophy, and cell cycle (Davis, 2000). Little is known about the However, there are correlates with JNK and centrosome.
Inhibition of centrosomal JNK by HSP70 contributes to the repair of heat shock-induced damage to centrosomes (Brown, Hong-Brown, Doxsey, & Welch, 1996). Apoptosis of acentrosomal cells is mediated by JNK signaling, which also drives compensatory proliferation to maintain tissue integrity and viability (Poulton, Cuningham, & Peifer, 2014). In addition, the guts irradiated with 2Gy shows increased JNK and AKT activities as well as centrosome amplification in intestinal stem cells of adult midgets (Pyo et al., 2014). In the present study, we found that JNK1 is activated during treatment-induced centrosome amplification (Figures 5a,c). Inhibition and overexpression of JNK1 Obesity, a worldwide epidemic, is a type of health impairment resulting from aberrant or excessive adipose accumulation (McClean, Kee, Young, & Elborn, 2008). There is an association between type 2 diabetes and obesity; a significant proportion of patients with type 2 diabetes have increased body mass index . Adipose tissue is composed of adipocytes and a vascular-stromal fraction, which contains macrophages, fibroblasts, endothelial cells, and necrosis factor-α, interleukin-6, monocyte chemotactic protein-1, and adipocyte fatty acid-binding protein (Hsu, Wu, Chang, & Lin, 2015).
Leptin induces matrix metalloproteinase 7 expression to promote ovarian cancer cell invasion by activating ERK and JNK pathways (Ghasemi, Hashemy, Aghaei, & Panjehpour, 2018). Genistein inhibited JNK activation, which inhibited the TNF-α-mediated downregulation of adiponectin (Yanagisawa et al., 2012). Visfatin may represent a pro-inflammatory cytokine that is influenced by insulin/insulin sensitivity via the NF-κB and JNK pathways (McGee et al., 2011). Although these adipokines are associated with JNK activity or signaling pathways, whether these adipokines can alter the diabetes-associated centrosome amplification remains to be further studied, in the light the JNK was found to inhibit the centrosome amplification in the study (Figure 5f).
Signal transducer and activator of transcription 3, a member of the STAT protein family, is a transcription factor which in humans is encoded by the STAT gene (Akira et al., 1994). Stat3 mediates the expression of a variety of genes in response to cell stimuli, and thus plays a key role in many cellular processes such as cell growth and apoptosis (Yuan et al., 2004). A recent study suggests that Stat3 is also involved in centrosome duplication. For example, Metge et al. reported that Stat3 apparently regulates γ-tubulin levels (Metge, Ofori-Acquah, Stevens, & Balczon, 2004). Furthermore, Morris et al. reported that Stat3/Stathmin/PLK1 can regulate centrosomal γtubulin levels which allowed centrosome to position themselves F I G U R E 6 JNK1-Stat3 is a pathway targeting at 14-3-3σ. (a) Treatment-increased Stat3 phosphorylation was inhibited by JNK1 siRNA. (b) 14-3-3σ expression was enhanced by JNK1 siRNA. (c) Stat3 siRNA increases protein level of 14-3-3σ. (d) and (e) Stat3 siRNA increased the centrosome translocation of 14-3-3σ. Glu: glucose, 15 mM; Ins: insulin, 5 nM; Pal: palmitic acid, 150 μM. **p < 0.01, compared with that in the control group; ## p < 0.01, compared with that in the samples treated with Glu, Ins, and Pal. White scale bar represents 5 μm. JNK1: c-Jun N-terminal kinase; siRNA: small interfering RNA; Stat3: signal transducer and activator of transcription 3 [Color figure can be viewed at wileyonlinelibrary.com] together, which result centrosome clustering and bipolar spindle in cancer cells (Morris et al., 2017). In the present study, we found that activation of Stat3 inhibited the centrosome amplification by high glucose, insulin and palmitic acid (Figures 5b,f) which clearly shows that Stat3 can negatively regulate centrosome amplification.
In our recent report (Wang et al., 2018), we showed that the protein levels of ROCK1 and 14-3-3σ as well as centrosome amplification were all increased in PBMC from the patients with type 2 diabetes. In HCT116 cells treated with high glucose, insulin, and palmitic acid, the protein levels of ROCK1 and 14-3-3σ were as well as centrosome amplification were also all upregulated. Knockdown of ROCK1 or 14-3-3σ attenuated the treatment-elicited centrosome amplification. The results suggest that type 2 diabetes promotes centrosome amplification via ROCK1 and 14-3-3σ, with high glucose, insulin, and palmitic acid as the triggers.
However, the in vivo relevance of the pathways discovered in vitro in the present study requires verification under in vivo setting.
We are investigating the protein levels of PCNA and p-NPM in the PBMC samples from healthy subjects and those with type 2 diabetes, to obtain evidence that can link PCNA and NPM to centrosome amplification in type 2 diabetes (Wang et al., 2018). Similarly, we are building a diabetic mice model using streptozotocin, in an attempt for correlating the levels centrosome amplification with PCNA and p-NPM. It is difficult to assess the role of JNK1 and Stat3 in the diabetes-associated centrosome amplification in vivo. Perhaps, genetically modified animal models (i.e., knockout or overexpression) are useful, which is beyond our ability. Importantly, we found that when we treated IEC-6 rat colon epithelial cells, which are immortalized but noncancerous cells, with high glucose, insulin, and palmitic acid, the levels PCNA, p-NPM, ROCK1, and 14-3-3σ as well as centrosome amplification were all upregulated significantly (Lu et al. unpublished data). We are designing protocols to inoculate the IEC-6 cell with or without treatment-increased centrosome amplification into nude mice, for more direct evidence for a cause and consequence relationship amongst the activation of the signaling pathways, centrosome amplification and tumorigenesis.
Discovery of the molecular pathways for the type 2 diabetesassociated centrosome will be beneficial for development of methods for inhibiting increased centrosome amplification in type 2 diabetes (Wang et al., 2018), that is, by inhibiting the centrosome amplification-promoting pathways and/or by activating the feedback inhibition loop. It remains to be explored whether the increased centrosome amplification in type 2 diabetes (Wang et al., 2018) contributes to cancer development. If so, inhibition of centrosome amplification in type 2 diabetes would be meaningful for cancer prevention for patients with type 2 diabetes.

| CONCLUSIONS
In the present study, we have identified two novel signal transduction pathways that regulate the centrosome amplification by high glucose, insulin, and palmitic acid, which target at ROCK1/14-3-3σ complex in the centrosome. PCNA-ROCK1 pathway promotes the centrosome amplification, whereas JNK1-Stat3-14-3-3σ feedback inhibits the centrosome amplification. NPM serves as an independent promoting pathway (Figure 7).

ACKNOWLEDGMENTS
The authors thank Dr. ZY Li for her help in establishing our cell culture facility. The centrosome detection experiments were supported by the staffs in the Scientific Instrument Center of the Shanxi University. This study was supported by grants from Shanxi University (No. 113533901005 and 113545017) and the Jiangsu F I G U R E 7 Summary scheme: PCNA-ROCK1, JNK1-Stat3-14-3-3σ, and NPM pathways regulate the centrosome amplification triggered by high glucose, insulin, and palmitic acid. The experimental treatment promotes centrosome amplification via PCNA-ROCK1, and NPM pathways and activates a feedback inhibition pathway of JNK1-Stat3-14-3-3σ. JNK1: c-Jun N-terminal kinase; NPM: nucleophosmin; PCNA: proliferating cell nuclear antigen; ROCK1: rho-associated coiled-coil containing protein kinase 1; Stat3: signal transducer and activator of transcription 3 [Color figure can be viewed at wileyonlinelibrary.com]