Importance of PNO1 for growth and survival of urinary bladder carcinoma: Role in core‐regulatory circuitry

Abstract PNO1 (partner of Nob1) was known as a RNA‐binding protein in humans, and its ortholog PNO1 was reported to participate ribosome and proteasome biogenesis in yeasts. Yet there have been few studies about its functions in mammalian cells, and so far its role in human cells has never been reported, especially in urinary bladder cancer (UBC).We interrogated the cellular functions and clinical significance of PNO1 in, and its molecular mechanism through microarrays and bioinformatics analysis. Our findings support that PNO1 participates in promoting proliferation and colonogenesis, while reducing apoptosis of UBC cells, and is also predicted to be associated with the migration and metastasis of UBC PNO1 knockdown (KD) attenuated the tumorigenesis ability of UBC in mouse. PNO1 KD led to the altered expression of 1543 genes that are involved in a number of signalling pathways, biological functions and regulation networks. CD44, PTGS2, cyclin D1, CDK1, IL‐8, FRA1, as well as mTOR, p70 S6 kinase, p38 and Caspase‐3 proteins were all down‐regulated in PNO1 KD cells, suggesting the involvement of PNO1 in inflammatory responses, cell cycle regulation, chemotaxis, cell growth and proliferation, apoptosis, cell migration and invasiveness. This study will enhance our understanding of the molecular mechanism of UBC and may eventually provide novel targets for individualized cancer therapy.


| INTRODUC TI ON
Urinary bladder cancer (UBC) is the ninth most common malignancy and the thirteenth most common cause of cancer-related death worldwide. 1 In 2018, there have been an estimated 81 190 new cases of UBC in the United States, causing a total of 17 240 deaths, with the incidence ratio of male vs female being 3.3 vs 1. 2 In China, the total incidence of UBC was 80 500, causing 32 900 deaths annually, with the male vs female ratio being 3.4 vs 1. 3 UBC is a highly immunogenic cancer with a higher rate of mutation than other types of cancers. 4 The current trend of 'personalized and precision care' in cancer therapy requires multiple layers of molecular profiling of biomarkers for accurate diagnosis and prediction of treatment responses. 4 It is therefore important to identify novel biomarkers and oncogenes for UBC and thus can improve personalized and precision care.
PNO1 was first characterized as a novel RNA-binding gene partner of NOB1 isolated from human kidney. 5 PNO1 is highly conserved, all the way from yeasts up to mammals. 6 In yeasts, PNO1 is reportedly involved in both ribosome and proteasome biogenesis. 7,8 In mammalian cells, it is localized to the nucleus, especially within the nucleoli. 5 In humans, PNO1 is most abundantly expressed in the thyroid, adrenals, appendix, placenta, bone marrow, urinary bladder and testes (NCBI Gene Database, ID: 56902). Currently, there have been few studies about its functions in mammalian cells, and so far its role in humans has not been reported.
To this end, the aim of the study was to identify the potential involvement of PNO1 in human UBC. The association of PNO1 with UBC was studied both in vitro and in vivo, and its molecular mechanism was predicted through microarray and bioinformatics analysis.

| MATERIAL S AND ME THODS
The human and animal subjects and materials of the paper were approved by the Yantai Yu Huang Ding Hospital's ethical committee.

| RNA isolation and quantitative real-time PCR (qRT-PCR)
Total RNA was extracted from cells using SuperfecTRI total RNA isolation reagent (Pufei), according to the manufacturer's instructions.

| Western blot
Cellular protein extraction and Western blot were performed as previously reported. 9 Proteins were identified with antibodies

| Automated cell counting
Lentivirus-infected cells were seeded with GFP fluorescence in plates at an appropriate concentration and cultured under routine conditions. Plates were read on a Celigo® Image Cytometer

| MTT assay
Cell viability was analysed using an MTT assay. Two thousand cells were seeded in each well of the 96-well plates and cultured under routine conditions. At each time-point, 20 μL of 5 mg/mL MTT (Genview) was added to each well, followed by a 4-hour incubation after which the culture medium was replaced by 100 μL DMSO.
Following 2-5 minutes agitation, the optical densities at 490 nm wavelength (OD490) were measured on an Infinite® M2009PR microplate reader (Tecan). Growth curves based on OD490 or fold changes against the OD490 on day 1 were plotted.

| Colony formation assay
Cells infected with lentivirus for 3 days were seeded in 6-well plates at a concentration of 800 per well and cultured for around 10 days. Cells in each well were fixed with 1 mL 4% paraformaldehyde for 30-60 minutes and stained with 500 μL GIEMSA Stain (Sigma-Aldrich) for 10-20 minutes. Following several washes by ddH 2 O, images were taken and colony numbers were counted.

| Subcutaneous xenotransplantation of human bladder cancer cells in mice
Twenty-four-week-old BALB/c female nude mice were divided into a normal control (NC) group and a knockdown (KD) group, with 10 mice in each group. 1 × 10 7 T24 lentivirus-infected cells carrying shCtrl (NC group) or shPNO1 (KD group) for 7 days were subcutaneously injected into the right arm pit of each mouse. The length (L) and width (W) of tumours were measured from day 24 post-inoculation and every 3-5 days until day 36 (tumour volume = 3.14/6 × L × W × W).
All mice were then killed by injection of an overdose of 2% pentobarbital sodium followed by cervical vertebra dislocation, and tumours were excised and measured for volume and weight.
Before the mice were killed, tumour sizes were measured by

| DNA microarray and bioinformatics analysis
Total RNA extracted from T24 cells expressing shCtrl or shPNO1 was analysed with the Agilent 2100 Bioanalyzer system (Agilent) and Canonical pathway analysis, upstream regulation analysis, disease and function analysis, regulator effect analysis and gene interaction network analysis were all carried out using the Ingenuity® Pathway Analysis (IPA) software (Qiagen).

| Antibody microarray
Antibody microarray was performed using the PathScan® Intracellular Signaling Antibody Array Kit (Chemiluminescent Readout) according to the manufacturer's instructions. T24NC and T24KD cells were washed twice with ice-cold PBS and lysed with 1× PathScan® Sandwich ELISA Lysis Buffer (Cell Signaling), containing 1 mmol/L PMSF for 2 minutes on ice. Cell lysate was diluted to 0.2-1.0 mg/mL with the Array Dilution Buffer, which was used in PathScan® Array tests. Arrays were exposed with 5% ChemiScope 5300 (Clinx Science Instruments Co. Ltd).

| Statistical analysis
Statistical analysis was performed using the GraphPad Prism 7.0 software. The results were expressed as mean ± standard deviation.
The data were analysed initially by F test to check the equality of variances. Data with F-value < 0.05 were subjected to two-tailed Welch's t test, and those with F-value > 0.05 were subjected to twotailed Student's t test. P < .05 in the t test suggested statistically significant difference.

| Clinicopathological factors associated with PNO1 expression in bladder cancer tissues
We first evaluated PNO1 expression in 56 bladder urothelial carcinomas by immunohistochemistry (IHC). The staining of PNO1 was low or moderate in low-grade tumours ( Figure 1A,B), but strong in highgrade tumours ( Figure 1C,D). Based on the percentage for PNO1 immune-positive tumour cells, a score of 1 was given when ≤5% of cells were positive, 2 when 6%-25%, 3 when 26%-50% and 4 when ≥ 50% of cells were positive. Staining intensity was scored as 0 (negative), 1 (weak), 2 (moderate) and 3 (strong). Both scores were multiplied, and the resulting score was used to trichotomize PNO1 expression as Low (≤4), Moderate (>5, ≤8) and High (>8). We then analysed the correlation between clinicopathological parameters and PNO1 expression in the 56 tumour tissue samples ( Figure 1E). The expression level of PNO1, particularly localized in the nucleus of tumour cells, correlates with high grade (P = .0091).

| Role of PNO1 in growth and survival of bladder cancer cells
We was also confirmed that for both cell lines in the KD group, the colonogenesis ability was significantly attenuated (P < .001), while cell apoptosis was significantly activated (P < .001; Figure 2G,H). These results indicated that PNO1 was important for the proliferation and survival of bladder cancer cells.

| Attenuation tumour growth in vivo after PNO1 knockdown
To study the function of PNO1 in bladder cancer in vivo used the in the KD group. The size of tumours was also measured by bioluminance live imaging before sacrificing the animal, which showed lower total radiant efficiency in KD group of mice (P < .001; Figure 3D). Our results confirmed that in a mouse model, the PNO1 KD cells formed significantly smaller and lighter tumours than the control cells, further confirming the involvement of PNO1 in tumour growth in vivo.

| RNA microarray identified DEGs in PNO1 KD cells
Gene expression in T24KD and T24NC cells was analysed by RNA microarray. A total of 1543 DEGs were identified, including 675 upregulated genes and 868 down-regulated genes in T24KD cells compared with T24NC cells, with PNO1 down-regulated by 3.42-fold (P = 2.97 × 10 −5 ). Distribution of DEGs between groups is shown in Figure 4A,B. RNA microarray data were subjected to IPA analysis.

| Canonical pathway analysis
Sixteen significantly altered canonical signalling and metabolic pathways correlated with DEGs (−log P > 1.3 and |Z-score| > 2) and were summarized by IPA ( Figure 4C, Table 1). The Acute-phase response signalling pathway was the most significantly enriched and down-regulated pathway involving the identified DEGs (−log P = 4.54), whose  Figure S1 and Figure S2. Colorectal cancer metastasis signalling was the most down-regulated pathway (Z-score = 3.53). Peroxisome proliferator-activated receptors (PPAR) signalling was the only up-regulated pathway.

| Upstream regulation analysis
The upstream regulating genes of DEGs were analysed. One hundred and fifty-eight genes were predicted to significantly up-regulate the DEGs, among which the kinase inhibitor chemical U0126 was the upstream regulator (Activation Z-score = 5.905, P = 1.45 × 10 −11 ) to be activated most strongly, along with 48 unanimously activated genes. Three hundred and fifty-nine genes were predicted to significantly down-regulate the DEGs, among which phorbol myristate acetate was the upstream regulator (Activation Z-score = −6.063, P = 1.90 × 10 −9 ) to be inhibited most strongly, compared with 80 unanimously inhibited genes. The regulation network of phorbol myristate acetate and its downstream genes are shown in Figure S3.

| Disease and function analysis
The inhibition or activation of diseases and physiological func-

| Regulator effect analysis
The possible routes of DEGs participating in the upstream regula-  Figure S4).

| Interaction network among DEGs
The interaction networks among all DEGs were also analysed, and results showed that the top diseases and functions significantly correlated with interacting molecules within the network were as follows: haematological disease, hereditary disorder, organismal injury and abnormalities, which involved 34 DEGs ( Table 2). The interaction of molecules within the top-scored network was presented in Figure 4E.

| PNO1-regulated gene expression
Based on the above informatics analysis, we chose PNO1 and some of its downstream genes (CD44, PTGS2, CCND1, CDK1, F3, CXCL8 and FOSL1) for interaction analysis, and the gene interaction network is shown in Figure 5A. The protein expression of these genes was further confirmed by Western blot in T24NC and T24KD cells ( Figure 5B). Apart from F3, the expression of all genes was significantly reduced as a result of PNO1 knockdown, which was consistent with the microarray result.
In addition, the effect of PNO1 on 18 key node proteins of intracellular signalling pathways that are closely associated with diseases was studied by antibody array. Results showed that, among these proteins, mTOR, p70 S6 kinase, p38 and Caspase-3 were significantly down-regulated in PNO1 KD cells ( Figure 5C).

| D ISCUSS I ON
Previously PNO1 was well known as a RNA-binding protein in humans, and its ortholog PNO1 was reported to participate in ribo-  Note: −log(P-value) > 1.3, which meant P < .05, represented significant correlation between DEGs and the pathway. Z-score > 2 suggested significant activation of the pathway while Z-score < −2 suggested significant repression of the pathway.

TA B L E 1
List of significantly up-or down-regulated regulated canonical pathways that were significantly related to differentially expressed genes (DEGs) between NC and PNO1 KD T24 cells tumour cells were also down-regulated in PNO1 KD cells, indicating that PNO1 might also participate in the metastasis of cancer cells. The acute-phase proteins (APPs), C-reactive protein (CRP) and orosomucoid (ORM) were reported to be important prognostic or diagnostic biomarkers for UBC. 17,18 Comparing the identified activation/inhibition status of proteins in the acute-phase response signalling based on our test result ( Figure S1) and that reported by  CD44 antigen is a cell-surface glycoprotein involved in cell-cell interactions, cell adhesion and cell migration. It is stimulated by IL-6 and was implicated with higher clinical stage, radio-resistance, cancer stem cell-like property and resistance to apoptosis in UBC. [19][20][21] Prostaglandin-endoperoxide synthase 2 (PTGS2) is also known as cyclooxygenase-2 (COX-2), 22  This study also revealed a down-regulation of mTOR, p70 S6 kinase, p38 and caspase-3 proteins in PNO1 KD cells, suggesting the association of PNO1 with several intracellular signalling pathways.
mTOR and p70 S6 kinase are both key components of the mTOR signalling pathway, which promotes cell growth and proliferation in eukaryotic cells. 28 Altered mTOR pathway activity has been noted in a variety of human tumours, including urothelial carcinoma. 29 mTOR pathway activation was reported to be involved in UBC tumorigenesis and was a predictor of cancer progression and poor survival. 30 p38 mitogen-activated protein kinases belong to the MAPK family and are activated by stress stimuli, such as cytokines, ultraviolet irradiation, heat shock and osmotic shock. The activation of p38 MAPKs has been reported to contribute to the epithelial-mesenchymal transition of cells in the primary tumour, to the acquisition of invasion and migrating capabilities and to the extravasation of migrating tumour cells, while p38 MAPK inhibition has been correlated with the resistance to anoikis. 31 Caspase-3 belongs to the caspase family and plays an indispensable role in the execution-phase of cell apoptosis. 32 However, the significantly increased apoptotic activity in T24KD cells that was observed contradicted the down-regulation of caspase-3. We suspected that this might be caused by experimental errors.
This is the first time the cellular functions and clinical significance of PNO1 in UBC was investigated, and its molecular mechanism explored through microarrays and bioinformatics analysis. Our

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflict of interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.