- Email: [email protected]
Oncosuppressor protein p53 and cyclin-dependent kinase inhibitor p21 regulate interstitial cystitis associated gene expression
Cytokine 110 (2018) 110–115
Contents lists available at ScienceDirect
Cytokine journal homepage: www.elsevier.com/locate/cytokine
Oncosuppressor protein p53 and cyclin-dependent kinase inhibitor p21 regulate interstitial cystitis associated gene expression Susan Keaya, Shreeram C. Nallarb,d, Padmaja Gadeb, Chen-Ou Zhangc, Dhan V. Kalvakolanub,c,
T ⁎
a
Department of Medicine, University of Maryland School of Medicine, Baltimore, MD 21201, United States Department of Microbiology & Immunology, University of Maryland School of Medicine, Baltimore, MD 21201, United States c Department of Pathology, University of Maryland School of Medicine, Baltimore, MD 21201, United States d Greenebaum NCI-Comprehensive Cancer Center, University of Maryland School of Medicine, Baltimore, MD 21201, United States b
A R T I C LE I N FO
A B S T R A C T
Keywords: Bladder Cystitis Gene expression Tumor suppressor Cell adhesion
Interstitial cystitis (IC) is a chronic syndrome that affects the urinary bladder. The etiology of this disease is unclear, and no effective therapies are available at this time. Although inflammation is suspected, no clear evidence for a role of conventional mediators of inflammation, such as cytokines and their downstream molecules, has been obtained to date. Our previous studies indicated that primary cell cultures derived from IC urothelium abnormally express molecules associated with cell adhesion. Here we describe a mechanism by which transcriptional changes in tight junction and adhesion molecules are mediated. Oncosuppressor proteins p53 and cyclin-dependent protein kinase inhibitor p21 directly associate with regulatory sites on the ZO-1 and Ecadherin genes, identifying important roles for p53 and p21 in driving non-oncogenic pathologies. These data also suggest that interference with these factors offers a potential therapeutic opportunity.
1. Introduction Interstitial cystitis/bladder pain syndrome (IC/BPS) is a chronic syndrome characterized by pain with increased urinary frequency and urgency [1] that affects several million people in the USA [2–4]. Since the etiology for IC/BPS is unknown, and moderately effective treatment for patients with IC/BPS has not been found, greater understanding of the pathogenesis of this debilitating syndrome may be necessary for developing effective therapies. Our lab is interested in understanding the etiology and mechanisms of bladder epithelial thinning and/or denudation in IC/BPS, which are cardinal histopathologic findings in biopsies from IC/BPS patients [5–7]. Using immunohistochemical and gene-expression microarray analyses, we and others have found [8–11] that epithelial gene expression was consistently abnormal in primary cells derived from epithelial biopsies of IC/BPS patients suggesting that an abnormal program of epithelial gene expression and/or differentiation may occur. These cells grown from biopsies of patients with IC/BPS who fulfilled modified National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) criteria (without measurement of bladder capacity) [12] display stable and heritable abnormalities similar to those found in patient biopsies in vivo, including profoundly reduced rates of cell proliferation (which could result in thinning or ulceration) [13], specifically altered cell differentiation markers including higher
⁎
E-cadherin and lower ZO-1, occludin, and vimentin [14–16], and increased paracellular permeability [16], as compared to cells grown from biopsies of age- and gender-matched controls (i.e., normal bladder or NB cells). These explanted IC/BPS cells also uniquely produce a frizzled 8 protein-related antiproliferative factor (APF) that is not made by cells from age- and gender-matched asymptomatic controls [17,18]. Antiproliferative activity is found in urine samples from ≥90% of patients with IC/BPS including patients with Hunner’s lesions [19–22]. Indeed, primary NB epithelial cells treated with HPLC-purified and/or a synthetic APF derivative in our laboratory displayed an IC/BPS cell phenotype in every manner examined including decreased proliferation [13,18], increased E-cadherin production [14,16], decreased ZO-1, occludin, and vimentin production [14,16], and increased paracellular permeability [16]. Since the regulation of these genes during IC is poorly understood, we have investigated this aspect in the current report. Here we show that tumor suppressor proteins p53 and p21, which were both previously shown to be increased in IC/BPS cells as compared to matched normal control cells [27], participate in regulating the expression of ZO-1 and E-cadherin genes. These data present an unconventional role for onco-suppressor proteins in non-tumor chronic pathologies.
Corresponding author at: Greenebaum Cancer Center, University of Maryland School of Medicine, 660 W. Redwood St, Baltimore, MD 21201, United States. E-mail address: [email protected] (D.V. Kalvakolanu).
https://doi.org/10.1016/j.cyto.2018.04.029 Received 1 March 2018; Received in revised form 20 April 2018; Accepted 23 April 2018 1043-4666/ © 2018 Published by Elsevier Ltd.
Cytokine 110 (2018) 110–115
S. Keay et al.
2.4. RNAi-mediated knockdown
Table 1 Primers used fro PCR in this study. Primer
Sequence (5′→3′)
purpose
ZO-1 Fwd: ZO-1 Rev: β-Actin Fwd: β-Actin Rev: E-Cad Fwd: E-cad Rev: p21-Fwd: p21-Rev: p53 Fwd: p53 Rev: ZO-1 P8F: ZO-1 P8R: E-cad P3F: E-cad P3R:
cggtcctctgagcctgtaag ggatctacatgcgacgacaa acattaaggagaagctgtgc ttctgcatcctgtcggcaat cgggaatgcagttgaggatc aggatggtgtaagcgatggc gagcgatggaacttcgactt ggtacaaga cagtgacaggt agtctagagccaccgtcca aggtctgaaaatgtttcctgactca gcccagtgaaggaaacaact agccgggtaacccaagtaac agtcccacaacagcataggg ttctgaactcaaggcgatcct
mRNA quantification
To deplete the indicated transcripts, commercially available (Open Biosystems, Huntsville, AL) lentiviral vectors expressing target-specific short hairpin RNAs (shRNAs) under the control of human U6 promoter were employed. These vectors also carried a puromycin resistance gene, which allows the selection of stably-transfected cells. To deplete p21 protein, lentiviral vector (pLKO1) encoding a sequence (TRCN0000040123) specific to human p21 mRNA was used. p53 was knocked down using a previously published [25] shRNA (pLKO-p53shRNA-427) obtained from Addgene, Cambridge, MA. Lentiviral particles were prepared and used as described in our earlier studies [23]. Briefly, each shRNA expression plasmid (3 μg) was mixed with pCMV-dR8.2dvpr (2.7 μg) and pCMV-VSVg (0.3 μg) vectors and transfected into HEK-293 T cells using the Fugene 6 reagent (Roche) as described earlier (49). Medium from these cultures were collected daily for 5 days, pooled and passed through a 0.45 μm filter and used as source for lentiviral shRNAs. Knockdown of the target gene expression was assessed by performing Western blot analyses or qPCR. The empty vector and scrambled shRNA as controls to ascertain the specificity of targeting in all experiments. No significant differences in the transcript levels were observed with these two controls. Hence only one of these controls presented.
Quantitative ChIP assays
2. Materials and methods 2.1. Cell culture NB and IC/BPS cells were grown in DMEM-F12 containing 10%FBS. Cells were serum starved for 2 days prior to ChIP assays, RNA, and/or Protein extraction. ChIP grade antibodies specific for p53 (Mouse mAb #48818), p21 (Rabbit mAb #2947) and H3-K9-Ac (Rabbit mAb #9649) were obtained from Cell Signaling Technologies, Danvers, MA).
3. Results 3.1. Promoter analyses of the ZO-1 and E-cadherin genes
2.2. Transcript quantification
To identify potential transcriptional regulators for ZO-1 and Ecadherin, we selected from +1 to −1000 bp of the human ZO-1 and Ecadherin genes from the NCBI genome database and analyzed this region for the binding of potential transcription factors (TF) using publicly available search tools (http://molbiol-tools.ca/Transcriptional_ factors.htm ; http://gene-regulation.com/pub/databases.html http:// alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_ 8.3/). Data from these searches were then used to generate potential binding site maps for both genes (Fig. 1). Approximate coordinates of the TFs with binding at multiple sites were notated. We noticed that both genes had high preponderance of TP53 (p53) binding sites. Sp1 and Jun B and Jun D binding sites were also found in these response elements.
Real-time PCR analyses were performed with specific primers using the JumpStart SYBR Green Master Mix (Sigma-Aldrich) in a Stratagene Mx3005P real time PCR machine. Expression differences of specific transcripts were quantified using the ACTB as an internal control by the ΔΔCT method. Each PCR reaction was had 5 replicates and each experiment was repeated with atleast 2 independent batches of RNA preparations. Primer sequences for real-time PCR are described in Table 1. 2.3. Chromatin Immunoprecipitation (ChIP) assays Standard ChIP assays were performed using a commercially available kit from the Upstate Biotech, Inc as recommended by the manufacturer and the ChIPed DNA was used for normal or real time PCR as described in our previous studies [23,24]. After cross linking the chromatin with 1% formaldehyde for 5 min, cells were collected. The cells were sonicated for 15 s for 7 times with 30 s interval under ice with Bronson sonicator. The average fragment size was 500 bp under these conditions. An equal quantity of chromatin from each sample was incubated with at least 5 μg of either protein-specific IgG or non-specific IgG (Cell Signaling Technologies) at 4 °C overnight. IP products were collected after incubating with magnetic beads coated with protein G (Active motif, Inc). The beads were washed; protein-DNA cross-links were reversed; and DNA was purified after phenol-chloroform extraction and ethanol precipitation. Purified DNA from the input and IP samples were subjected to qPCR with specific primers (Table 1).
3.2. Increased binding of p53 and p21 to response elements for the ZO-1 gene in IC cells Since ZO-1 and E-cadherin are regulated in a diametrically opposing manner during IC/BPS, and a high number of p53 binding sites were found for both genes, we next examined if p53 directly bound to these enhancers using quantitative Chromatin Immunoprecipitation (ChIP) assays. Multiple primer sets (n = 8) spanning the length (between −1000 to +1 bp) of ZO-1 and E-cadherin genes were synthesized and PCR was performed to determine a specific amplification of the target sites. In the initial ChIP assays p53-specific antibodies were tested for specificity against two control ChIP reactions with: (1) no antibody; and (2) a non-specific IgG control antibody. Parallel control reactions, where total DNA from soluble chromatin was used as a template for Fig. 1. Line diagrams showing the organizations of human ZO-1 and E-cadherin genes. Approximate distance of DNAbinding sites of various transcription factors upstream of the transcriptional start sites (arrow at the 3′ end) are indicated. Locations of specific primers used for the ChIP assays are indicated.
111
Cytokine 110 (2018) 110–115
S. Keay et al.
Fig. 2. Panels A-C: Quantitative ChIP PCR analyses of the p53, p21 and H3-K9Ac to the ZO-1 enhancer. The P3F and P3R primers were used for qPCR analyses. Chromatin was immunoprecipitated with the indicated antibodies. Equal inputs of chromatin were ensured parallel PCR reactions with DNA isolated from each sample. A number of specificity controls, such as sham-antibody and nonreactive rabbit and mouse IgG were used for these studies. None of them yielded any detectable signals (not shown). Each bar represents mean abundance of ZO-1 promoter fragments ± SD of 6 separate reactions from 2 to independent experiments. The significance of the differences between samples was determined using Student’s t-test and p values were indicated where appropriate. Panels D-F: Quantitative ChIP PCR analyses of the p53, p21 and H3-K9Ac to the E-cadherin enhancer. The P8F and P8R primers were used for qPCR analyses. Chromatin was immunoprecipitated with the indicated antibodies as in Fig. 2. Equal inputs of chromatin were ensured parallel PCR reactions with DNA isolated from each sample. Each bar represents mean abundance of E-cadherin promoter fragments ± SD of 6 separate reactions from 2 to independent experiments. The significance of the differences between samples was determined using Student’s t-test and p values were indicated where appropriate.
expression, with acetylation of the K9 residue of histone H3 being a marker for open chromatin, which can facilitate binding of regulatory proteins to response elements. Thus, a higher deposition of H3-K9-Ac at a gene enhancer can be an indicator for an open chromatin at a target gene. So, we next examined the deposition of H3-K9-Ac using ChIP assays, and found a significantly reduced deposition of H3-K9-Ac at the ZO-1 enhancer in IC cells as compared to the control NB cells. Therefore, primary epithelial cell explants from patients with interstitial cystitis have a significantly increased p53 and p21 recruitment and a decreased deposition of H3-K9-Ac at ZO-1 response elements as compared to explants from normal controls.
PCR, were also used to determine comparable inputs. Qualified primer sets that yielded the expected PCR products with p53-IP, but not in the control reactions, were further used for quantitative ChIP assays. Two primer sets, p6 and p8, yielded ZO-1 enhancer specific products in multiple tests (data not shown). Among these, the p8 primer set produced consistent p53 binding patterns in multiple primary normal bladder (NB) and IC cells. Therefore, we chose this primer set for further experiments. We then analyzed p53 binding to ZO-1enhancer in epithelial cell explants from 5 separate interstitial cystitis patients vs. explants from their 5 age- and gender-matched normal controls (Fig. 2A). In all five IC cases, a dramatically increased binding of p53 to ZO-1 response element at the p8 site was observed compared to the NB cells. One of the consequences of elevated p53 activity is the increased expression of p21. Since p21 is generally not recognized as a DNA binding protein we used it as a negative control for the ChIP study. Surprisingly, ChIP assays with p21-specific antibodies also yielded similar pattern of DNA binding as p53 (Fig. 2B). Histone modifications play a major role in regulating gene
3.3. Decreased binding of p53 and p21 to the E-cadherin enhancer in IC cells We next examined the patterns of p53 and p21 binding to E-cadherin enhancer. Our initial qualitative studies utilized multiple primer sets spanning the enhancer of E-cadherin gene. Among the 4 primer sets tested, qualitative ChIP analyses with the p3 primer set produced a 112
Cytokine 110 (2018) 110–115
S. Keay et al.
specific shRNAs that can target these transcripts. A vector that expressed a “non-targeting” scrambled (Sc) shRNA was used as a control for the knockdown and matched NB cell explants were used as controls for the study. The resultant transfectants were selected with puromycin for 3 days to eliminate any non-transfected cells. In the initial analyses, we measured the levels of p53 and p21 transcripts in the transfected cells using qPCR (Fig. 4A). Transfection of Sc-shRNA alone did not alter p53 and p21 transcript expression. However, the p53 and p21 specific shRNAs significantly depleted the corresponding transcripts in IC9 cells, when compared to Sc-shRNA control. We next determined the impact of p53 and p21 depletion on the expression of ZO-1 (Fig. 4B) and E-cadherin (Fig. 4C) transcripts in the same IC cell explants. The ScshRNA-transfected IC9 cells had basal levels of ZO-1 transcript. Upon depletion of either p53 or p21 in the IC cells robustly enhanced ZO-1 expression in IC9 cells, indicating a repressive effect of p53 and p21 on ZO-1 gene expression. A reverse scenario was observed with E-cadherin expression. Depletion of either p53 or p21 caused a significant loss of Ecadherin expression, indicating that these factors server as positive regulators of E-cadherin expression. Similar results were obtained with IC11 cells upon RNAi of p53 or p21 (data not shown).
specific PCR product, which was absent in the no-antibody and nonspecific IgG control ratios. This primer set was chosen for all quantitative ChIP assays. In contrast to the ZO-1 enhancer, binding of p53 to the E-cadherin enhancer was high in NB cells as compared to cells explanted from IC patients (Fig. 2D). In a similar manner, binding of p21 to the E-cadherin enhancer found in NB cells was significantly decreased in IC cells (Fig. 2E). In contrast, H3-K9-Ac deposition at this enhancer was significantly increased in IC cells compared to the corresponding NB cell controls (Fig. 2F). Thus, loss of p53- and p21binding and enhanced H3-K9-Ac were found in IC cells, a pattern opposite to that of the ZO-1 in the same cells. 3.4. Correlation of p53 and p21 binding with ZO-1 and E-cadherin expression Because the above experiments show only the binding of p53 and p21 to the specific gene enhancers, we next determined the biological relevance of these findings by correlating them with mRNA levels. qPCR analyses for the ZO-1 and E-cadherin transcripts in two representative pairs of NB and IC/BPS cells showed that ZO-1 expression was significantly downregulated in IC cells, when compared to the corresponding NB cell control (Fig. 3A). In contrast E-cadherin expression was upregulated in IC cells, relative to the NB cell controls (Fig. 3B). Thus, increased p53 and p21 binding to ZO-1 and E-cadherin enhancers distinctly impacts their expression correlates with decreased expression of both genes. We have also measured p53 and p21 transcript levels in these cells (Fig. 3C and D). We noticed a significant increase in the expression of both these transcripts in IC/BPS cells.
4. Discussion Previous studies from our group and others identified a number of genes associated with this disease [8–20], however the mechanisms altering epithelial gene expression in IC are unclear. Some of these genes are involved in cell adhesion (E-cadherin, upregulated in IC/BPS) and tight junction formation (ZO-1, downregulated in IC/BPS). In addition, we have reported that anti-proliferative factor (APF), a short glycopeptide made by IC/BPS cells and found in the urine of IC/PBS patients, suppresses cell proliferation. Therefore, it was of interest to investigate how the expression of these IC associated genes is regulated. As APF can increase the expression of p53 and p21 in T24 bladder carcinoma cells as well as TRT-HU1 immortalized human bladder
3.5. RNAi-mediated depletion of p53 and p21 alters ZO-1 and E-cadherin gene expression To study the impact of p53 and p21 on ZO-1 and E-cadherin gene expression, we depleted them in IC9 cell explants using RNAi with
Fig. 3. Quantitative PCR analyses of the indicated transcripts in NB and IC cell pairs. Each bar represents mean abundance specific ± SD of 7 separate reactions from 2 to independent experiments. The significance of the differences between samples was determined using Student’s t-test and p values were indicated where appropriate. 113
Cytokine 110 (2018) 110–115
S. Keay et al.
Fig. 4. Quantitative PCR analyses of the indicated transcripts in IC9 cells. Each bar represents mean abundance specific ± SD of 6 separate reactions from 2 to independent experiments. The level of expression of each transcript in Empty vector (EV) transfected cells was considered as 1. Levels of each transcript observed in the shRNA-transfected samples were determined by expressing them as a fraction of the EV control. The significance of the differences between samples was determined using Student’s t-test and the corresponding p values were indicated.
bladder epithelial cells by APF, a negative growth regulator that appears to be made by bladder epithelial cells explanted uniquely from patients with IC and that also induces abnormalities in bladder epithelial gene expression to mimic abnormalities seen in IC cell explants [13,14,16,17–20]]. In this paper we have shown that like APF-treated T24 bladder carcinoma cells or TRT-HU1 immortalized human bladder epithelial cells, cells explanted from IC patients also display increased expression of both p53 and p21 as compared to explants from age- and gender-matched normal donors, and that both of these proteins participate in transcriptional control of IC-associated gene expression. Decreased binding of p21 and p53 to the E-cadherin enhancer correlated with an upregulation of its expression. This is highlighted by the deposition of acetylated H3K9 at the same region in multiple different IC cells compared to the corresponding NB cells. Indeed, knock down of p53 and p21 promoted E-cadherin expression. There was no direct interaction between p53 and p21 (not shown) indicating that this might occur via interaction through other proteins in the transcriptional complexes. In contrast increased binding of p53 and p21 to the ZO-1 gene-enhancer inhibited its expression in IC cells, compared to the normal controls. Consistent with this, RNAi-mediated depletion of either p53 or p21 significantly suppressed ZO-1 expression. How p21 regulates transcription is unclear at this stage. One possibility is that its post-translational modifications may allow its specific association with various cellular factors. Indeed, it is phosphorylated by multiple different protein kinases such as the members of the MAPkinase superfamily-ASK1, ERK1, ERK2, JNK1, p38α, and cyclindependent kinase family Cdk2 and Cdk6, the death-associated protein kinase-3, Protein kinases PKC and PKA, Pim1 and Pim2 and Glycogen Synthase Kinase-3b. Future studies are required for critically evaluating each of these kinases and the phosphorylation sites in the context of IC disease. In addition, some evidence indicates that p21 can bind to DNA and/or other DNA-binding proteins such as cyclin-dependent kinase 2 (CKD2), altering specific gene expression in lung carcinoma cells [30], suggesting the possibility that it may also function directly in the regulation of abnormal gene expression in IC/BPS”. In summary, in this paper we show evidence for a novel function of p53 and p21 proteins in regulating ZO-1 and E-cadherin gene expression in bladder epithelial cell explants from IC patients. These findings suggest additional potential targets for therapeutic intervention of this poorly controlled disorder.
epithelial cells [26,27], and response elements for p53 are present in the E-cadherin and ZO-1 genes, we investigated whether p53 may play a role in regulating these genes. Transcription factor p53 is known generally to participate in cell growth regulatory pathways including a blockade of cell division cycle progression and cell migration, angiogenesis, promotion of senescence, apoptosis and autophagy, and ROSmediated DNA-damage repair [28]. Many of these activities, one way or another, are connected to its tumor suppressive function. However, several questions remain concerning its physiological non-cancer functions, including its role in mediating changes in the expression of specific genes (including genes for tight junction and adherens proteins), as well as cell proliferation, in bladder epithelial cells from patients with interstitial cystitis. In this paper we show evidence that p53 binds to ZO-1 and E-cadherin gene enhancers, and regulates the expression of these genes in bladder epithelial cells explants. Importantly, this p53 binding to these gene enhancer elements differs significantly between IC and NB cells in a direction compatible with negative regulation of expression for both genes. Perhaps somewhat surprisingly, CDKN1A/p21, also bound to the ZO-1 and E-cadherin gene enhancers and altered expression of the same genes in the same direction as p53. While this protein (also known as Waf-1/Cip-1) is well known to be a p53-induced gene product that functions as an inhibitor of cyclin-dependent kinases during G1phase of the cell cycle [29], it has also been found to bind to DNA and regulate gene expression via coupling with other proteins [30]. In this paper we have shown that both P53 and p21 may be involved in the abnormal regulation of genes in IC/BPS. Diametrically, p53 both positively and negative regulated gene expression in this model. p21/ CDKN1A (also known as Waf-1/Cip-1) inhibits Cyclin-dependent kinases during G1phase [29,31,32]. It was discovered as a p53 induced gene product, although it can also be induced in a p53-independent manner. p21 is a versatile protein that regulates cell cycle, apoptosis, differentiation, and migration, DNA repair, senescence and aging, and therefore there is a significant overlap between the anti-oncogenic functions of p53 and p21 [29,31,32]. Other studies have shown that p21 regulates reprogramming of induced pluripotent stem cells. In addition, p21 has a pathophysiological role in different diseases, atherosclerosis and HIV infection, beyond its prominent role in cancer [31]. Like p53, p21 was previously shown to be stimulated by treatment of T24 bladder carcinoma cells or TRT-HU1 immortalized human 114
Cytokine 110 (2018) 110–115
S. Keay et al.
Acknowledgments
antiproliferative factor treatment, Physiol. Genom. 14 (2) (2003) 107–115. [15] J. Southgate, C.L. Varley, M.A. Garthwaite, J. Hinley, F. Marsh, J. Stahlschmidt, L.K. Trejdosiewicz, I. Eardley, Differentiation potential of urothelium from patients with benign bladder dysfunction, BJU Int. 99 (6) (2007) 1506–1516. [16] C.O. Zhang, J.Y. Wang, K.R. Koch, S. Keay, Regulation of tight junction proteins and bladder epithelial paracellular permeability by an antiproliferative factor from patients with interstitial cystitis, J. Urol. 174 (6) (2005) 2382–2387. [17] S. Keay, M. Kleinberg, C.O. Zhang, M.K. Hise, J.W. Warren, Bladder epithelial cells from patients with interstitial cystitis produce an inhibitor of heparin-binding epidermal growth factor-like growth factor production, J. Urol. 164 (6) (2000) 2112–2118. [18] S.K. Keay, Z. Szekely, T.P. Conrads, T.D. Veenstra, J.J. Barchi Jr., C.O. Zhang, K.R. Koch, C.J. Michejda, An antiproliferative factor from interstitial cystitis patients is a frizzled 8 protein-related sialoglycopeptide, Proc. Natl. Acad. Sci. USA 101 (32) (2004) 11803–11808. [19] S. Keay, C.O. Zhang, A.L. Trifillis, M.K. Hise, J.R. Hebel, S.C. Jacobs, J.W. Warren, Decreased 3H-thymidine incorporation by human bladder epithelial cells following exposure to urine from interstitial cystitis patients, J Urol 156 (6) (1996) 2073–2078. [20] S.K. Keay, C.O. Zhang, J. Shoenfelt, D.R. Erickson, K. Whitmore, J.W. Warren, R. Marvel, T. Chai, Sensitivity and specificity of antiproliferative factor, heparinbinding epidermal growth factor-like growth factor, and epidermal growth factor as urine markers for interstitial cystitis, Urology 57 (6 Suppl 1) (2001) 9–14. [21] C.O. Zhang, Z.L. Li, C.Z. Kong, APF, HB-EGF, and EGF biomarkers in patients with ulcerative vs. non-ulcerative interstitial cystitis, BMC Urol. 5 (2005) 7. [22] D.R. Erickson, S.X. Xie, V.P. Bhavanandan, M.A. Wheeler, R.E. Hurst, L.M. Demers, L. Kushner, S.K. Keay, A comparison of multiple urine markers for interstitial cystitis, J. Urol. 167 (6) (2002) 2461–2469. [23] P. Gade, S.K. Roy, H. Li, S.C. Nallar, D.V. Kalvakolanu, Critical role for transcription factor C/EBP-beta in regulating the expression of death-associated protein kinase 1, Mol. Cell. Biol. 28 (8) (2008) 2528–2548. [24] P. Gade, D.V. Kalvakolanu, Chromatin immunoprecipitation assay as a tool for analyzing transcription factor activity, Methods Mol. Biol. 809 (2012) 85–104. [25] J.S. Kim, C. Lee, C.L. Bonifant, H. Ressom, T. Waldman, Activation of p53-dependent growth suppression in human cells by mutations in PTEN or PIK3CA, Mol. Cell. Biol. 27 (2) (2007) 662–677. [26] J. Kim, M. Ji, J.A. DiDonato, R.R. Rackley, M. Kuang, P.C. Sadhukhan, J.R. Mauney, S.K. Keay, M.R. Freeman, L.S. Liou, R.M. Adam, An hTERT-immortalized human urothelial cell line that responds to anti-proliferative factor, Vitro Cell Dev. Biol. Anim. 47 (1) (2011) 2–9. [27] J. Kim, S.K. Keay, J.D. Dimitrakov, M.R. Freeman, p53 mediates interstitial cystitis antiproliferative factor (APF)-induced growth inhibition of human urothelial cells, FEBS Lett. 581 (20) (2007) 3795–3799. [28] K.T. Bieging, L.D. Attardi, Deconstructing p53 transcriptional networks in tumor suppression, Trends Cell. Biol. 22 (2) (2012) 97–106. [29] A.L. Gartel, p21(WAF1/CIP1) and cancer: a shifting paradigm? Biofactors 35 (2) (2009) 161–164. [30] Y.C. Lin, Y.N. Chen, K.F. Lin, F.F. Wang, T.Y. Chou, M.Y. Chen, Association of p21 with NF-YA suppresses the expression of Polo-like kinase 1 and prevents mitotic death in response to DNA damage, Cell Death Dis. 5 (2014) e987. [31] T. Abbas, A. Dutta, p21 in cancer: intricate networks and multiple activities, Nat. Rev. Cancer 9 (6) (2009) 400–414. [32] A. Karimian, Y. Ahmadi, B. Yousefi, Multiple functions of p21 in cell cycle, apoptosis and transcriptional regulation after DNA damage, DNA Repair (Amst) 42 (2016) 63–71.
This work was supported by MERIT review funding from the Veterans Administration to SK; and University of Maryland tobacco restitution funds to DVK. References [1] P.M. Hanno, D.A. Burks, J.Q. Clemens, R.R. Dmochowski, D. Erickson, M.P. Fitzgerald, J.B. Forrest, B. Gordon, M. Gray, R.D. Mayer, D. Newman, L. Nyberg Jr., C.K. Payne, U. Wesselmann, M.M. Faraday, E. Interstitial Cystitis Guidelines Panel of the American Urological Association, I. Research, AUA guideline for the diagnosis and treatment of interstitial cystitis/bladder pain syndrome, J. Urol. 185 (6) (2011) 2162–2170. [2] S.H. Berry, M.N. Elliott, M. Suttorp, L.M. Bogart, M.A. Stoto, P. Eggers, L. Nyberg, J.Q. Clemens, Prevalence of symptoms of bladder pain syndrome/interstitial cystitis among adult females in the United States, J. Urol. 186 (2) (2011) 540–544. [3] G.C. Curhan, F.E. Speizer, D.J. Hunter, S.G. Curhan, M.J. Stampfer, Epidemiology of interstitial cystitis: a population based study, J. Urol. 161 (2) (1999) 549–552. [4] A.M. Suskind, S.H. Berry, B.A. Ewing, M.N. Elliott, M.J. Suttorp, J.Q. Clemens, The prevalence and overlap of interstitial cystitis/bladder pain syndrome and chronic prostatitis/chronic pelvic pain syndrome in men: results of the RAND Interstitial Cystitis Epidemiology male study, J. Urol. 189 (1) (2013) 141–145. [5] S.L. Johansson, M. Fall, Clinical features and spectrum of light microscopic changes in interstitial cystitis, J. Urol. 143 (6) (1990) 1118–1124. [6] T.L. Ratliff, C.G. Klutke, E.M. McDougall, The etiology of interstitial cystitis, Urol. Clin. North Am. 21 (1) (1994) 21–30. [7] J.E. Tomaszewski, J.R. Landis, V. Russack, T.M. Williams, L.P. Wang, C. Hardy, C. Brensinger, Y.L. Matthews, S.T. Abele, J.W. Kusek, L.M. Nyberg, G. Interstitial Cystitis Database Study, Biopsy features are associated with primary symptoms in interstitial cystitis: results from the interstitial cystitis database study, Urology 57 (6 Suppl 1) (2001) 67–81. [8] P.J. Hauser, M.G. Dozmorov, B.L. Bane, G. Slobodov, D.J. Culkin, R.E. Hurst, Abnormal expression of differentiation related proteins and proteoglycan core proteins in the urothelium of patients with interstitial cystitis, J. Urol. 179 (2) (2008) 764–769. [9] P. Laguna, F. Smedts, J. Nordling, T. Horn, K. Bouchelouche, A. Hopman, J. de la Rosette, Keratin expression profiling of transitional epithelium in the painful bladder syndrome/interstitial cystitis, Am. J. Clin. Pathol. 125 (1) (2006) 105–110. [10] V. Sanchez Freire, F.C. Burkhard, T.M. Kessler, A. Kuhn, A. Draeger, K. Monastyrskaya, MicroRNAs may mediate the down-regulation of neurokinin-1 receptor in chronic bladder pain syndrome, Am. J. Pathol. 176 (1) (2010) 288–303. [11] G. Slobodov, M. Feloney, C. Gran, K.D. Kyker, R.E. Hurst, D.J. Culkin, Abnormal expression of molecular markers for bladder impermeability and differentiation in the urothelium of patients with interstitial cystitis, J. Urol. 171 (4) (2004) 1554–1558. [12] J.Y. Gillenwater, A.J. Wei, Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Workshop on Interstitial Cystitis. NIH, Bethesda, MD, Aug. 28–29, 1987, J. Urol. 140 (1988) 203–206. [13] S. Keay, C.O. Zhang, J.L. Shoenfelt, T.C. Chai, Decreased in vitro proliferation of bladder epithelial cells from patients with interstitial cystitis, Urology 61 (6) (2003) 1278–1284. [14] S. Keay, F. Seillier-Moiseiwitsch, C.O. Zhang, T.C. Chai, J. Zhang, Changes in human bladder epithelial cell gene expression associated with interstitial cystitis or
115
Contents lists available at ScienceDirect
Cytokine journal homepage: www.elsevier.com/locate/cytokine
Oncosuppressor protein p53 and cyclin-dependent kinase inhibitor p21 regulate interstitial cystitis associated gene expression Susan Keaya, Shreeram C. Nallarb,d, Padmaja Gadeb, Chen-Ou Zhangc, Dhan V. Kalvakolanub,c,
T ⁎
a
Department of Medicine, University of Maryland School of Medicine, Baltimore, MD 21201, United States Department of Microbiology & Immunology, University of Maryland School of Medicine, Baltimore, MD 21201, United States c Department of Pathology, University of Maryland School of Medicine, Baltimore, MD 21201, United States d Greenebaum NCI-Comprehensive Cancer Center, University of Maryland School of Medicine, Baltimore, MD 21201, United States b
A R T I C LE I N FO
A B S T R A C T
Keywords: Bladder Cystitis Gene expression Tumor suppressor Cell adhesion
Interstitial cystitis (IC) is a chronic syndrome that affects the urinary bladder. The etiology of this disease is unclear, and no effective therapies are available at this time. Although inflammation is suspected, no clear evidence for a role of conventional mediators of inflammation, such as cytokines and their downstream molecules, has been obtained to date. Our previous studies indicated that primary cell cultures derived from IC urothelium abnormally express molecules associated with cell adhesion. Here we describe a mechanism by which transcriptional changes in tight junction and adhesion molecules are mediated. Oncosuppressor proteins p53 and cyclin-dependent protein kinase inhibitor p21 directly associate with regulatory sites on the ZO-1 and Ecadherin genes, identifying important roles for p53 and p21 in driving non-oncogenic pathologies. These data also suggest that interference with these factors offers a potential therapeutic opportunity.
1. Introduction Interstitial cystitis/bladder pain syndrome (IC/BPS) is a chronic syndrome characterized by pain with increased urinary frequency and urgency [1] that affects several million people in the USA [2–4]. Since the etiology for IC/BPS is unknown, and moderately effective treatment for patients with IC/BPS has not been found, greater understanding of the pathogenesis of this debilitating syndrome may be necessary for developing effective therapies. Our lab is interested in understanding the etiology and mechanisms of bladder epithelial thinning and/or denudation in IC/BPS, which are cardinal histopathologic findings in biopsies from IC/BPS patients [5–7]. Using immunohistochemical and gene-expression microarray analyses, we and others have found [8–11] that epithelial gene expression was consistently abnormal in primary cells derived from epithelial biopsies of IC/BPS patients suggesting that an abnormal program of epithelial gene expression and/or differentiation may occur. These cells grown from biopsies of patients with IC/BPS who fulfilled modified National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) criteria (without measurement of bladder capacity) [12] display stable and heritable abnormalities similar to those found in patient biopsies in vivo, including profoundly reduced rates of cell proliferation (which could result in thinning or ulceration) [13], specifically altered cell differentiation markers including higher
⁎
E-cadherin and lower ZO-1, occludin, and vimentin [14–16], and increased paracellular permeability [16], as compared to cells grown from biopsies of age- and gender-matched controls (i.e., normal bladder or NB cells). These explanted IC/BPS cells also uniquely produce a frizzled 8 protein-related antiproliferative factor (APF) that is not made by cells from age- and gender-matched asymptomatic controls [17,18]. Antiproliferative activity is found in urine samples from ≥90% of patients with IC/BPS including patients with Hunner’s lesions [19–22]. Indeed, primary NB epithelial cells treated with HPLC-purified and/or a synthetic APF derivative in our laboratory displayed an IC/BPS cell phenotype in every manner examined including decreased proliferation [13,18], increased E-cadherin production [14,16], decreased ZO-1, occludin, and vimentin production [14,16], and increased paracellular permeability [16]. Since the regulation of these genes during IC is poorly understood, we have investigated this aspect in the current report. Here we show that tumor suppressor proteins p53 and p21, which were both previously shown to be increased in IC/BPS cells as compared to matched normal control cells [27], participate in regulating the expression of ZO-1 and E-cadherin genes. These data present an unconventional role for onco-suppressor proteins in non-tumor chronic pathologies.
Corresponding author at: Greenebaum Cancer Center, University of Maryland School of Medicine, 660 W. Redwood St, Baltimore, MD 21201, United States. E-mail address: [email protected] (D.V. Kalvakolanu).
https://doi.org/10.1016/j.cyto.2018.04.029 Received 1 March 2018; Received in revised form 20 April 2018; Accepted 23 April 2018 1043-4666/ © 2018 Published by Elsevier Ltd.
Cytokine 110 (2018) 110–115
S. Keay et al.
2.4. RNAi-mediated knockdown
Table 1 Primers used fro PCR in this study. Primer
Sequence (5′→3′)
purpose
ZO-1 Fwd: ZO-1 Rev: β-Actin Fwd: β-Actin Rev: E-Cad Fwd: E-cad Rev: p21-Fwd: p21-Rev: p53 Fwd: p53 Rev: ZO-1 P8F: ZO-1 P8R: E-cad P3F: E-cad P3R:
cggtcctctgagcctgtaag ggatctacatgcgacgacaa acattaaggagaagctgtgc ttctgcatcctgtcggcaat cgggaatgcagttgaggatc aggatggtgtaagcgatggc gagcgatggaacttcgactt ggtacaaga cagtgacaggt agtctagagccaccgtcca aggtctgaaaatgtttcctgactca gcccagtgaaggaaacaact agccgggtaacccaagtaac agtcccacaacagcataggg ttctgaactcaaggcgatcct
mRNA quantification
To deplete the indicated transcripts, commercially available (Open Biosystems, Huntsville, AL) lentiviral vectors expressing target-specific short hairpin RNAs (shRNAs) under the control of human U6 promoter were employed. These vectors also carried a puromycin resistance gene, which allows the selection of stably-transfected cells. To deplete p21 protein, lentiviral vector (pLKO1) encoding a sequence (TRCN0000040123) specific to human p21 mRNA was used. p53 was knocked down using a previously published [25] shRNA (pLKO-p53shRNA-427) obtained from Addgene, Cambridge, MA. Lentiviral particles were prepared and used as described in our earlier studies [23]. Briefly, each shRNA expression plasmid (3 μg) was mixed with pCMV-dR8.2dvpr (2.7 μg) and pCMV-VSVg (0.3 μg) vectors and transfected into HEK-293 T cells using the Fugene 6 reagent (Roche) as described earlier (49). Medium from these cultures were collected daily for 5 days, pooled and passed through a 0.45 μm filter and used as source for lentiviral shRNAs. Knockdown of the target gene expression was assessed by performing Western blot analyses or qPCR. The empty vector and scrambled shRNA as controls to ascertain the specificity of targeting in all experiments. No significant differences in the transcript levels were observed with these two controls. Hence only one of these controls presented.
Quantitative ChIP assays
2. Materials and methods 2.1. Cell culture NB and IC/BPS cells were grown in DMEM-F12 containing 10%FBS. Cells were serum starved for 2 days prior to ChIP assays, RNA, and/or Protein extraction. ChIP grade antibodies specific for p53 (Mouse mAb #48818), p21 (Rabbit mAb #2947) and H3-K9-Ac (Rabbit mAb #9649) were obtained from Cell Signaling Technologies, Danvers, MA).
3. Results 3.1. Promoter analyses of the ZO-1 and E-cadherin genes
2.2. Transcript quantification
To identify potential transcriptional regulators for ZO-1 and Ecadherin, we selected from +1 to −1000 bp of the human ZO-1 and Ecadherin genes from the NCBI genome database and analyzed this region for the binding of potential transcription factors (TF) using publicly available search tools (http://molbiol-tools.ca/Transcriptional_ factors.htm ; http://gene-regulation.com/pub/databases.html http:// alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_ 8.3/). Data from these searches were then used to generate potential binding site maps for both genes (Fig. 1). Approximate coordinates of the TFs with binding at multiple sites were notated. We noticed that both genes had high preponderance of TP53 (p53) binding sites. Sp1 and Jun B and Jun D binding sites were also found in these response elements.
Real-time PCR analyses were performed with specific primers using the JumpStart SYBR Green Master Mix (Sigma-Aldrich) in a Stratagene Mx3005P real time PCR machine. Expression differences of specific transcripts were quantified using the ACTB as an internal control by the ΔΔCT method. Each PCR reaction was had 5 replicates and each experiment was repeated with atleast 2 independent batches of RNA preparations. Primer sequences for real-time PCR are described in Table 1. 2.3. Chromatin Immunoprecipitation (ChIP) assays Standard ChIP assays were performed using a commercially available kit from the Upstate Biotech, Inc as recommended by the manufacturer and the ChIPed DNA was used for normal or real time PCR as described in our previous studies [23,24]. After cross linking the chromatin with 1% formaldehyde for 5 min, cells were collected. The cells were sonicated for 15 s for 7 times with 30 s interval under ice with Bronson sonicator. The average fragment size was 500 bp under these conditions. An equal quantity of chromatin from each sample was incubated with at least 5 μg of either protein-specific IgG or non-specific IgG (Cell Signaling Technologies) at 4 °C overnight. IP products were collected after incubating with magnetic beads coated with protein G (Active motif, Inc). The beads were washed; protein-DNA cross-links were reversed; and DNA was purified after phenol-chloroform extraction and ethanol precipitation. Purified DNA from the input and IP samples were subjected to qPCR with specific primers (Table 1).
3.2. Increased binding of p53 and p21 to response elements for the ZO-1 gene in IC cells Since ZO-1 and E-cadherin are regulated in a diametrically opposing manner during IC/BPS, and a high number of p53 binding sites were found for both genes, we next examined if p53 directly bound to these enhancers using quantitative Chromatin Immunoprecipitation (ChIP) assays. Multiple primer sets (n = 8) spanning the length (between −1000 to +1 bp) of ZO-1 and E-cadherin genes were synthesized and PCR was performed to determine a specific amplification of the target sites. In the initial ChIP assays p53-specific antibodies were tested for specificity against two control ChIP reactions with: (1) no antibody; and (2) a non-specific IgG control antibody. Parallel control reactions, where total DNA from soluble chromatin was used as a template for Fig. 1. Line diagrams showing the organizations of human ZO-1 and E-cadherin genes. Approximate distance of DNAbinding sites of various transcription factors upstream of the transcriptional start sites (arrow at the 3′ end) are indicated. Locations of specific primers used for the ChIP assays are indicated.
111
Cytokine 110 (2018) 110–115
S. Keay et al.
Fig. 2. Panels A-C: Quantitative ChIP PCR analyses of the p53, p21 and H3-K9Ac to the ZO-1 enhancer. The P3F and P3R primers were used for qPCR analyses. Chromatin was immunoprecipitated with the indicated antibodies. Equal inputs of chromatin were ensured parallel PCR reactions with DNA isolated from each sample. A number of specificity controls, such as sham-antibody and nonreactive rabbit and mouse IgG were used for these studies. None of them yielded any detectable signals (not shown). Each bar represents mean abundance of ZO-1 promoter fragments ± SD of 6 separate reactions from 2 to independent experiments. The significance of the differences between samples was determined using Student’s t-test and p values were indicated where appropriate. Panels D-F: Quantitative ChIP PCR analyses of the p53, p21 and H3-K9Ac to the E-cadherin enhancer. The P8F and P8R primers were used for qPCR analyses. Chromatin was immunoprecipitated with the indicated antibodies as in Fig. 2. Equal inputs of chromatin were ensured parallel PCR reactions with DNA isolated from each sample. Each bar represents mean abundance of E-cadherin promoter fragments ± SD of 6 separate reactions from 2 to independent experiments. The significance of the differences between samples was determined using Student’s t-test and p values were indicated where appropriate.
expression, with acetylation of the K9 residue of histone H3 being a marker for open chromatin, which can facilitate binding of regulatory proteins to response elements. Thus, a higher deposition of H3-K9-Ac at a gene enhancer can be an indicator for an open chromatin at a target gene. So, we next examined the deposition of H3-K9-Ac using ChIP assays, and found a significantly reduced deposition of H3-K9-Ac at the ZO-1 enhancer in IC cells as compared to the control NB cells. Therefore, primary epithelial cell explants from patients with interstitial cystitis have a significantly increased p53 and p21 recruitment and a decreased deposition of H3-K9-Ac at ZO-1 response elements as compared to explants from normal controls.
PCR, were also used to determine comparable inputs. Qualified primer sets that yielded the expected PCR products with p53-IP, but not in the control reactions, were further used for quantitative ChIP assays. Two primer sets, p6 and p8, yielded ZO-1 enhancer specific products in multiple tests (data not shown). Among these, the p8 primer set produced consistent p53 binding patterns in multiple primary normal bladder (NB) and IC cells. Therefore, we chose this primer set for further experiments. We then analyzed p53 binding to ZO-1enhancer in epithelial cell explants from 5 separate interstitial cystitis patients vs. explants from their 5 age- and gender-matched normal controls (Fig. 2A). In all five IC cases, a dramatically increased binding of p53 to ZO-1 response element at the p8 site was observed compared to the NB cells. One of the consequences of elevated p53 activity is the increased expression of p21. Since p21 is generally not recognized as a DNA binding protein we used it as a negative control for the ChIP study. Surprisingly, ChIP assays with p21-specific antibodies also yielded similar pattern of DNA binding as p53 (Fig. 2B). Histone modifications play a major role in regulating gene
3.3. Decreased binding of p53 and p21 to the E-cadherin enhancer in IC cells We next examined the patterns of p53 and p21 binding to E-cadherin enhancer. Our initial qualitative studies utilized multiple primer sets spanning the enhancer of E-cadherin gene. Among the 4 primer sets tested, qualitative ChIP analyses with the p3 primer set produced a 112
Cytokine 110 (2018) 110–115
S. Keay et al.
specific shRNAs that can target these transcripts. A vector that expressed a “non-targeting” scrambled (Sc) shRNA was used as a control for the knockdown and matched NB cell explants were used as controls for the study. The resultant transfectants were selected with puromycin for 3 days to eliminate any non-transfected cells. In the initial analyses, we measured the levels of p53 and p21 transcripts in the transfected cells using qPCR (Fig. 4A). Transfection of Sc-shRNA alone did not alter p53 and p21 transcript expression. However, the p53 and p21 specific shRNAs significantly depleted the corresponding transcripts in IC9 cells, when compared to Sc-shRNA control. We next determined the impact of p53 and p21 depletion on the expression of ZO-1 (Fig. 4B) and E-cadherin (Fig. 4C) transcripts in the same IC cell explants. The ScshRNA-transfected IC9 cells had basal levels of ZO-1 transcript. Upon depletion of either p53 or p21 in the IC cells robustly enhanced ZO-1 expression in IC9 cells, indicating a repressive effect of p53 and p21 on ZO-1 gene expression. A reverse scenario was observed with E-cadherin expression. Depletion of either p53 or p21 caused a significant loss of Ecadherin expression, indicating that these factors server as positive regulators of E-cadherin expression. Similar results were obtained with IC11 cells upon RNAi of p53 or p21 (data not shown).
specific PCR product, which was absent in the no-antibody and nonspecific IgG control ratios. This primer set was chosen for all quantitative ChIP assays. In contrast to the ZO-1 enhancer, binding of p53 to the E-cadherin enhancer was high in NB cells as compared to cells explanted from IC patients (Fig. 2D). In a similar manner, binding of p21 to the E-cadherin enhancer found in NB cells was significantly decreased in IC cells (Fig. 2E). In contrast, H3-K9-Ac deposition at this enhancer was significantly increased in IC cells compared to the corresponding NB cell controls (Fig. 2F). Thus, loss of p53- and p21binding and enhanced H3-K9-Ac were found in IC cells, a pattern opposite to that of the ZO-1 in the same cells. 3.4. Correlation of p53 and p21 binding with ZO-1 and E-cadherin expression Because the above experiments show only the binding of p53 and p21 to the specific gene enhancers, we next determined the biological relevance of these findings by correlating them with mRNA levels. qPCR analyses for the ZO-1 and E-cadherin transcripts in two representative pairs of NB and IC/BPS cells showed that ZO-1 expression was significantly downregulated in IC cells, when compared to the corresponding NB cell control (Fig. 3A). In contrast E-cadherin expression was upregulated in IC cells, relative to the NB cell controls (Fig. 3B). Thus, increased p53 and p21 binding to ZO-1 and E-cadherin enhancers distinctly impacts their expression correlates with decreased expression of both genes. We have also measured p53 and p21 transcript levels in these cells (Fig. 3C and D). We noticed a significant increase in the expression of both these transcripts in IC/BPS cells.
4. Discussion Previous studies from our group and others identified a number of genes associated with this disease [8–20], however the mechanisms altering epithelial gene expression in IC are unclear. Some of these genes are involved in cell adhesion (E-cadherin, upregulated in IC/BPS) and tight junction formation (ZO-1, downregulated in IC/BPS). In addition, we have reported that anti-proliferative factor (APF), a short glycopeptide made by IC/BPS cells and found in the urine of IC/PBS patients, suppresses cell proliferation. Therefore, it was of interest to investigate how the expression of these IC associated genes is regulated. As APF can increase the expression of p53 and p21 in T24 bladder carcinoma cells as well as TRT-HU1 immortalized human bladder
3.5. RNAi-mediated depletion of p53 and p21 alters ZO-1 and E-cadherin gene expression To study the impact of p53 and p21 on ZO-1 and E-cadherin gene expression, we depleted them in IC9 cell explants using RNAi with
Fig. 3. Quantitative PCR analyses of the indicated transcripts in NB and IC cell pairs. Each bar represents mean abundance specific ± SD of 7 separate reactions from 2 to independent experiments. The significance of the differences between samples was determined using Student’s t-test and p values were indicated where appropriate. 113
Cytokine 110 (2018) 110–115
S. Keay et al.
Fig. 4. Quantitative PCR analyses of the indicated transcripts in IC9 cells. Each bar represents mean abundance specific ± SD of 6 separate reactions from 2 to independent experiments. The level of expression of each transcript in Empty vector (EV) transfected cells was considered as 1. Levels of each transcript observed in the shRNA-transfected samples were determined by expressing them as a fraction of the EV control. The significance of the differences between samples was determined using Student’s t-test and the corresponding p values were indicated.
bladder epithelial cells by APF, a negative growth regulator that appears to be made by bladder epithelial cells explanted uniquely from patients with IC and that also induces abnormalities in bladder epithelial gene expression to mimic abnormalities seen in IC cell explants [13,14,16,17–20]]. In this paper we have shown that like APF-treated T24 bladder carcinoma cells or TRT-HU1 immortalized human bladder epithelial cells, cells explanted from IC patients also display increased expression of both p53 and p21 as compared to explants from age- and gender-matched normal donors, and that both of these proteins participate in transcriptional control of IC-associated gene expression. Decreased binding of p21 and p53 to the E-cadherin enhancer correlated with an upregulation of its expression. This is highlighted by the deposition of acetylated H3K9 at the same region in multiple different IC cells compared to the corresponding NB cells. Indeed, knock down of p53 and p21 promoted E-cadherin expression. There was no direct interaction between p53 and p21 (not shown) indicating that this might occur via interaction through other proteins in the transcriptional complexes. In contrast increased binding of p53 and p21 to the ZO-1 gene-enhancer inhibited its expression in IC cells, compared to the normal controls. Consistent with this, RNAi-mediated depletion of either p53 or p21 significantly suppressed ZO-1 expression. How p21 regulates transcription is unclear at this stage. One possibility is that its post-translational modifications may allow its specific association with various cellular factors. Indeed, it is phosphorylated by multiple different protein kinases such as the members of the MAPkinase superfamily-ASK1, ERK1, ERK2, JNK1, p38α, and cyclindependent kinase family Cdk2 and Cdk6, the death-associated protein kinase-3, Protein kinases PKC and PKA, Pim1 and Pim2 and Glycogen Synthase Kinase-3b. Future studies are required for critically evaluating each of these kinases and the phosphorylation sites in the context of IC disease. In addition, some evidence indicates that p21 can bind to DNA and/or other DNA-binding proteins such as cyclin-dependent kinase 2 (CKD2), altering specific gene expression in lung carcinoma cells [30], suggesting the possibility that it may also function directly in the regulation of abnormal gene expression in IC/BPS”. In summary, in this paper we show evidence for a novel function of p53 and p21 proteins in regulating ZO-1 and E-cadherin gene expression in bladder epithelial cell explants from IC patients. These findings suggest additional potential targets for therapeutic intervention of this poorly controlled disorder.
epithelial cells [26,27], and response elements for p53 are present in the E-cadherin and ZO-1 genes, we investigated whether p53 may play a role in regulating these genes. Transcription factor p53 is known generally to participate in cell growth regulatory pathways including a blockade of cell division cycle progression and cell migration, angiogenesis, promotion of senescence, apoptosis and autophagy, and ROSmediated DNA-damage repair [28]. Many of these activities, one way or another, are connected to its tumor suppressive function. However, several questions remain concerning its physiological non-cancer functions, including its role in mediating changes in the expression of specific genes (including genes for tight junction and adherens proteins), as well as cell proliferation, in bladder epithelial cells from patients with interstitial cystitis. In this paper we show evidence that p53 binds to ZO-1 and E-cadherin gene enhancers, and regulates the expression of these genes in bladder epithelial cells explants. Importantly, this p53 binding to these gene enhancer elements differs significantly between IC and NB cells in a direction compatible with negative regulation of expression for both genes. Perhaps somewhat surprisingly, CDKN1A/p21, also bound to the ZO-1 and E-cadherin gene enhancers and altered expression of the same genes in the same direction as p53. While this protein (also known as Waf-1/Cip-1) is well known to be a p53-induced gene product that functions as an inhibitor of cyclin-dependent kinases during G1phase of the cell cycle [29], it has also been found to bind to DNA and regulate gene expression via coupling with other proteins [30]. In this paper we have shown that both P53 and p21 may be involved in the abnormal regulation of genes in IC/BPS. Diametrically, p53 both positively and negative regulated gene expression in this model. p21/ CDKN1A (also known as Waf-1/Cip-1) inhibits Cyclin-dependent kinases during G1phase [29,31,32]. It was discovered as a p53 induced gene product, although it can also be induced in a p53-independent manner. p21 is a versatile protein that regulates cell cycle, apoptosis, differentiation, and migration, DNA repair, senescence and aging, and therefore there is a significant overlap between the anti-oncogenic functions of p53 and p21 [29,31,32]. Other studies have shown that p21 regulates reprogramming of induced pluripotent stem cells. In addition, p21 has a pathophysiological role in different diseases, atherosclerosis and HIV infection, beyond its prominent role in cancer [31]. Like p53, p21 was previously shown to be stimulated by treatment of T24 bladder carcinoma cells or TRT-HU1 immortalized human 114
Cytokine 110 (2018) 110–115
S. Keay et al.
Acknowledgments
antiproliferative factor treatment, Physiol. Genom. 14 (2) (2003) 107–115. [15] J. Southgate, C.L. Varley, M.A. Garthwaite, J. Hinley, F. Marsh, J. Stahlschmidt, L.K. Trejdosiewicz, I. Eardley, Differentiation potential of urothelium from patients with benign bladder dysfunction, BJU Int. 99 (6) (2007) 1506–1516. [16] C.O. Zhang, J.Y. Wang, K.R. Koch, S. Keay, Regulation of tight junction proteins and bladder epithelial paracellular permeability by an antiproliferative factor from patients with interstitial cystitis, J. Urol. 174 (6) (2005) 2382–2387. [17] S. Keay, M. Kleinberg, C.O. Zhang, M.K. Hise, J.W. Warren, Bladder epithelial cells from patients with interstitial cystitis produce an inhibitor of heparin-binding epidermal growth factor-like growth factor production, J. Urol. 164 (6) (2000) 2112–2118. [18] S.K. Keay, Z. Szekely, T.P. Conrads, T.D. Veenstra, J.J. Barchi Jr., C.O. Zhang, K.R. Koch, C.J. Michejda, An antiproliferative factor from interstitial cystitis patients is a frizzled 8 protein-related sialoglycopeptide, Proc. Natl. Acad. Sci. USA 101 (32) (2004) 11803–11808. [19] S. Keay, C.O. Zhang, A.L. Trifillis, M.K. Hise, J.R. Hebel, S.C. Jacobs, J.W. Warren, Decreased 3H-thymidine incorporation by human bladder epithelial cells following exposure to urine from interstitial cystitis patients, J Urol 156 (6) (1996) 2073–2078. [20] S.K. Keay, C.O. Zhang, J. Shoenfelt, D.R. Erickson, K. Whitmore, J.W. Warren, R. Marvel, T. Chai, Sensitivity and specificity of antiproliferative factor, heparinbinding epidermal growth factor-like growth factor, and epidermal growth factor as urine markers for interstitial cystitis, Urology 57 (6 Suppl 1) (2001) 9–14. [21] C.O. Zhang, Z.L. Li, C.Z. Kong, APF, HB-EGF, and EGF biomarkers in patients with ulcerative vs. non-ulcerative interstitial cystitis, BMC Urol. 5 (2005) 7. [22] D.R. Erickson, S.X. Xie, V.P. Bhavanandan, M.A. Wheeler, R.E. Hurst, L.M. Demers, L. Kushner, S.K. Keay, A comparison of multiple urine markers for interstitial cystitis, J. Urol. 167 (6) (2002) 2461–2469. [23] P. Gade, S.K. Roy, H. Li, S.C. Nallar, D.V. Kalvakolanu, Critical role for transcription factor C/EBP-beta in regulating the expression of death-associated protein kinase 1, Mol. Cell. Biol. 28 (8) (2008) 2528–2548. [24] P. Gade, D.V. Kalvakolanu, Chromatin immunoprecipitation assay as a tool for analyzing transcription factor activity, Methods Mol. Biol. 809 (2012) 85–104. [25] J.S. Kim, C. Lee, C.L. Bonifant, H. Ressom, T. Waldman, Activation of p53-dependent growth suppression in human cells by mutations in PTEN or PIK3CA, Mol. Cell. Biol. 27 (2) (2007) 662–677. [26] J. Kim, M. Ji, J.A. DiDonato, R.R. Rackley, M. Kuang, P.C. Sadhukhan, J.R. Mauney, S.K. Keay, M.R. Freeman, L.S. Liou, R.M. Adam, An hTERT-immortalized human urothelial cell line that responds to anti-proliferative factor, Vitro Cell Dev. Biol. Anim. 47 (1) (2011) 2–9. [27] J. Kim, S.K. Keay, J.D. Dimitrakov, M.R. Freeman, p53 mediates interstitial cystitis antiproliferative factor (APF)-induced growth inhibition of human urothelial cells, FEBS Lett. 581 (20) (2007) 3795–3799. [28] K.T. Bieging, L.D. Attardi, Deconstructing p53 transcriptional networks in tumor suppression, Trends Cell. Biol. 22 (2) (2012) 97–106. [29] A.L. Gartel, p21(WAF1/CIP1) and cancer: a shifting paradigm? Biofactors 35 (2) (2009) 161–164. [30] Y.C. Lin, Y.N. Chen, K.F. Lin, F.F. Wang, T.Y. Chou, M.Y. Chen, Association of p21 with NF-YA suppresses the expression of Polo-like kinase 1 and prevents mitotic death in response to DNA damage, Cell Death Dis. 5 (2014) e987. [31] T. Abbas, A. Dutta, p21 in cancer: intricate networks and multiple activities, Nat. Rev. Cancer 9 (6) (2009) 400–414. [32] A. Karimian, Y. Ahmadi, B. Yousefi, Multiple functions of p21 in cell cycle, apoptosis and transcriptional regulation after DNA damage, DNA Repair (Amst) 42 (2016) 63–71.
This work was supported by MERIT review funding from the Veterans Administration to SK; and University of Maryland tobacco restitution funds to DVK. References [1] P.M. Hanno, D.A. Burks, J.Q. Clemens, R.R. Dmochowski, D. Erickson, M.P. Fitzgerald, J.B. Forrest, B. Gordon, M. Gray, R.D. Mayer, D. Newman, L. Nyberg Jr., C.K. Payne, U. Wesselmann, M.M. Faraday, E. Interstitial Cystitis Guidelines Panel of the American Urological Association, I. Research, AUA guideline for the diagnosis and treatment of interstitial cystitis/bladder pain syndrome, J. Urol. 185 (6) (2011) 2162–2170. [2] S.H. Berry, M.N. Elliott, M. Suttorp, L.M. Bogart, M.A. Stoto, P. Eggers, L. Nyberg, J.Q. Clemens, Prevalence of symptoms of bladder pain syndrome/interstitial cystitis among adult females in the United States, J. Urol. 186 (2) (2011) 540–544. [3] G.C. Curhan, F.E. Speizer, D.J. Hunter, S.G. Curhan, M.J. Stampfer, Epidemiology of interstitial cystitis: a population based study, J. Urol. 161 (2) (1999) 549–552. [4] A.M. Suskind, S.H. Berry, B.A. Ewing, M.N. Elliott, M.J. Suttorp, J.Q. Clemens, The prevalence and overlap of interstitial cystitis/bladder pain syndrome and chronic prostatitis/chronic pelvic pain syndrome in men: results of the RAND Interstitial Cystitis Epidemiology male study, J. Urol. 189 (1) (2013) 141–145. [5] S.L. Johansson, M. Fall, Clinical features and spectrum of light microscopic changes in interstitial cystitis, J. Urol. 143 (6) (1990) 1118–1124. [6] T.L. Ratliff, C.G. Klutke, E.M. McDougall, The etiology of interstitial cystitis, Urol. Clin. North Am. 21 (1) (1994) 21–30. [7] J.E. Tomaszewski, J.R. Landis, V. Russack, T.M. Williams, L.P. Wang, C. Hardy, C. Brensinger, Y.L. Matthews, S.T. Abele, J.W. Kusek, L.M. Nyberg, G. Interstitial Cystitis Database Study, Biopsy features are associated with primary symptoms in interstitial cystitis: results from the interstitial cystitis database study, Urology 57 (6 Suppl 1) (2001) 67–81. [8] P.J. Hauser, M.G. Dozmorov, B.L. Bane, G. Slobodov, D.J. Culkin, R.E. Hurst, Abnormal expression of differentiation related proteins and proteoglycan core proteins in the urothelium of patients with interstitial cystitis, J. Urol. 179 (2) (2008) 764–769. [9] P. Laguna, F. Smedts, J. Nordling, T. Horn, K. Bouchelouche, A. Hopman, J. de la Rosette, Keratin expression profiling of transitional epithelium in the painful bladder syndrome/interstitial cystitis, Am. J. Clin. Pathol. 125 (1) (2006) 105–110. [10] V. Sanchez Freire, F.C. Burkhard, T.M. Kessler, A. Kuhn, A. Draeger, K. Monastyrskaya, MicroRNAs may mediate the down-regulation of neurokinin-1 receptor in chronic bladder pain syndrome, Am. J. Pathol. 176 (1) (2010) 288–303. [11] G. Slobodov, M. Feloney, C. Gran, K.D. Kyker, R.E. Hurst, D.J. Culkin, Abnormal expression of molecular markers for bladder impermeability and differentiation in the urothelium of patients with interstitial cystitis, J. Urol. 171 (4) (2004) 1554–1558. [12] J.Y. Gillenwater, A.J. Wei, Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Workshop on Interstitial Cystitis. NIH, Bethesda, MD, Aug. 28–29, 1987, J. Urol. 140 (1988) 203–206. [13] S. Keay, C.O. Zhang, J.L. Shoenfelt, T.C. Chai, Decreased in vitro proliferation of bladder epithelial cells from patients with interstitial cystitis, Urology 61 (6) (2003) 1278–1284. [14] S. Keay, F. Seillier-Moiseiwitsch, C.O. Zhang, T.C. Chai, J. Zhang, Changes in human bladder epithelial cell gene expression associated with interstitial cystitis or
115