Background
This laboratory has proposed the third isoform of the metallothionein gene family as a potential biomarker for the development of human bladder cancer [
1,
2]. This was first suggested by a retrospective immunohistochemical analysis of MT-3 expression on a modest sample set of archival diagnostic specimens composed of benign and cancerous lesions of the bladder [
1]. The cells of the normal bladder were shown to have no immunoreactivity for the MT-3 protein, and no expression of MT-3 mRNA or protein were noted in extracts prepared from samples from surgically removed normal bladder tissue. In contrast, all specimens of urothelial cancer were immunoreactive for the MT-3 protein, and the intensity of staining correlated to tumor grade. This was later expanded to a more robust retrospective study using archival diagnostic tissue [
2]. This study showed that only 2 of 63 (3.17%) benign bladder specimens had even weak immunostaining for the MT-3 protein. In contrast, 103 of 107 (96.26%) high grade urothelial cancers and 17 of 17 (100%) specimens of carcinoma
in situ stained positive for the MT-3 protein. For low grade urothelial cancer, 30 of 48 specimens (62.5%) expressed the MT-3 protein.
The laboratory has used the UROtsa cell line as a model system to elucidate the differences in the expression of the MT-3 gene between normal and malignant urothelium. The UROtsa cell line is derived from a primary culture of human urothelial cells that was immortalized using the SV40 large T-antigen [
3,
4]. The UROtsa cells retain a normal cytogenetic profile, grow as a contact inhibited monolayer, and are not tumorigenic as judged by the inability to form colonies in soft agar and tumors in nude mice. This laboratory showed that UROtsa cells grown in a serum-free growth medium displayed features consistent with the intermediate layer of the urothelium [
5]. Identical to that of normal
in situ urothelium, the UROtsa cell line was shown to have no basal expression of MT-3 mRNA or protein. The laboratory has also directly malignantly transformed the UROtsa cell line by exposure to Cd
+2 or As
+3 and shown that the tumor transplants produced by the transformed cells had histologic features consistent with human urothelial cancer [
6]. An interesting finding in subsequent studies was that MT-3 mRNA and protein was not expressed in the Cd
+2 and As
+3 transformed cell lines, but was expressed in the tumor transplants generated by these cell lines in immunocompromised mice [
2]. That this was not an anomaly of the UROtsa cell line was suggested by identical findings between cell lines and tumor transplants for the MCF-7, T-47 D, Hs 578T, MDA-MB-231 breast cancer cell lines and the PC-3 prostate cancer cell lines [
2]. The first goal of the present study was to determine if epigenetic modifications were responsible for gene silencing of MT-3 in the parental UROtsa cell line. The second goal of the study was to determine if the accessibility of the MRE of the MT-3 promoter to the MTF-1 transcription factor was different between the parental UROtsa cell line and the UROtsa cell lines malignantly transformed by either Cd
+2 or As
+3. The third goal was to determine if histone modifications were different between the parental UROtsa cell line and the transformed cell lines. The last goal was to perform a preliminary analysis to determine if MT-3 expression might translate clinically as a possible biomarker for malignant urothelial cells released into the urine by patients with urothelial cancer.
Discussion
The initial goal of this study was to determine if epigenetic modification was responsible for the silencing of the MT-3 gene in the parental UROtsa cell line. Treatment of the parental UROtsa cells with 5-AZC, a commonly used agent to determine DNA methylation status, was shown to have no effect on MT-3 mRNA expression. This provides evidence that the MT-3 gene was not silenced by a mechanism involving DNA methylation in the parental UROtsa cells. The treatment of the cells with MS-275, a histone deacetylase inhibitor, was shown to result in the expression of MT-3 mRNA by the parental UROtsa cell line. MS-275 has been shown to preferentially inhibit HDAC 1 compared to HDAC 3 and has little or no effect on HDAC 6 and 8 [
7‐
9]. This finding provides strong evidence that MT-3 expression is silenced in the parental UROtsa cell line through a mechanism involving histone modification. The MT-3 gene is also silent in cell lines derived from the UROtsa parent that have been malignantly transformed by either Cd
+2 or As
+3 [
2,
6]. A pattern of MT-3 mRNA expression similar to that for the parental UROtsa cells was found following treatment of the Cd
+2 and As
+3 transformed cell lines with 5-AZC and MS-275. The only exception being that the expression of MT-3 mRNA was several fold higher following MS-275 treatment in the Cd
+2 and As
+3 transformed cell lines compared to the parental UROtsa cells. These findings suggest that MT-3 gene expression is silenced in both the parental UROtsa cells and the Cd
+2 and As
+3 transformed counterparts through a mechanism involving histone modification.
The second goal of the study was to determine if the accessibility of the MREs of the MT-3 promoter to a transcription factor were different between the parental UROtsa cell line and the UROtsa cell lines malignantly transformed by either Cd
+2 or As
+3. The initial indication that the integrity of the MT-3 promoter may be different between the parent and transformed UROtsa cells, was that MT-3 mRNA expression could be further induced by Zn
+2 in the transformed cell lines following treatment with MS-275, but was not induced by an identical treatment in the parental UROtsa cell line. This observation was extended by an analysis of the accessibility of the MREs within the MT-3 promoter to binding of MTF-1. MTF-1 is a constitutively expressed transcription factor that is activated by diverse stress stimuli, the most notable being metal load [
10]. Upon stimulation MTF-1 translocates to the nucleus where it binds to the enhancers/promoters of target genes that harbor one or multiple copies of the specific recognition sequence, called MREs. The best characterized of these target genes are the metallothioneins [
11]. The analysis was performed in the presence of 100 μM Zn
+2 because Zn
+2 is necessary for the activation of MTF-1 and 100 μM is the concentration commonly utilized to determine MTF-1 activation. ChIP analysis showed that there was no binding of MTF-1 to MREa and MREb of the MT-3 promoter in the parental UROtsa cell line before or after treatment with MS-275. In contrast, there was MTF-1 binding to MREa and MREb of the MT-3 promoter in the Cd
+2 and As
+3 transformed cell lines under basal conditions, with a further increase in binding following treatment with MS-275. A similar analysis of MTF-1 binding to MREc in the MT-3 promoter showed the parental cells to have limited binding under basal conditions and an increased interaction following treatment with MS-275. In contrast, the Cd
+2 and As
+3 transformed cell lines were shown to have increased binding of MTF-1 to MREc of the MT-3 promoter under both basal conditions with no increase in interaction following treatment with MS-275. An identical analysis of MREe, f and g of the MT-3 promoter with MTF-1 showed no interaction in the parental UROtsa cell under basal conditions and an increase in binding following treatment with MS-275. In contrast, MREe, f, g of the MT-3 promoter were able to bind MTF-1 under basal conditions, which was increased following treatment with MS-275. These studies show that there is a fundamental difference in the accessibility of MREs to MTF-1 binding within the MT-3 promoter between the parental UROtsa cells and the Cd
+2 and As
+3 transformed cell lines. Under basal conditions, the MREs of the MT-3 promoter are not accessible to MTF-1 binding in the parental UROtsa cells. In contrast, the MREs of the MT-3 promoter are accessible for MTF-1 binding under basal conditions in the Cd
+2 and As
+3 transformed cell lines.
Several common histone modifications, acetyl H4, trimethyl H3K4, trimethyl H3K27, and trimethyl H3K9, associated with gene activation were analyzed in two regions of the MT-3 promoter for the parental UROtsa cells and the Cd
+2 and As
+3 transformed cell lines. The level of histone H4 acetylation was always increased in both the parental and transformed cell lines in the presence of MT-275. In addition, it was also found to be increased in the more proximal region of the Cd
+2 and As
+3 transformed cell lines not treated with MS-275 in comparison to the parent cell line. The increase in H4 acetylation correlated with the increase in MT-3 expression and it is known that H4 acetylation is associated with transcriptional activation [
12‐
14]. The antibody used for H4 acetylation does not distinguish among the four potentially acetylated lysines 5, 8, 12, and 16, but all are thought to be involved in transcriptional activation. Similarly, the above noted increases in MT-3 expression in the parental and transformed cell lines also was associated with methylation of H3K4, which is a modification also known to occur in promoters of actively transcribing genes [
12‐
14]. Together, these findings give an indication that the MT-3 promoter in the transformed cells has histone modifications that are positive for transcription of the MT-3 gene. In contrast to the above the findings which support a "transcription ready" state, are the findings of increased histone H3K9 and H3K27 methylation, which are both associated with a transcriptionally repressed state [
12‐
14]. Taken together, these findings can be interpreted to suggest that the MT-3 promoter in the Cd
+2 and As
+3 transformed cells has gained bivalent chromatin structure, that is having elements of being "transcriptionally repressed" and "transcription ready", when compared to parental UROtsa cells [
15].
It has been shown previously that the Cd
+2 and As
+3 transformed cell lines have no expression of MT-3 mRNA under cell culture conditions, but gain MT-3 expression when transplanted as tumors in immune compromised mice [
2]. Based on the above histone modifications in the cell lines, this finding would suggest that transplantation of the Cd
+2 and As
+3 transformed cell lines into an
in vivo environment further alters the chromatin structure of the MT-3 promoter to a state capable of active transcription of the MT-3 gene. This would suggest that the
in vivo environment is providing a factor/s that is capable of advancing bivalent chromatin to a fully active state. There is no literature base that allows one to speculate what this factor might be or if it would be expected to be soluble or an insoluble component of the cell matrix.
The last goal of this study was to perform a preliminary analysis to determine if MT-3 expression might translate clinically as a possible biomarker for malignant urothelial cells released into the urine by patients with urothelial cancer. This was tested by the collection of urothelial cells from the urine of patients attending their regularly scheduled appointment in the urology clinic. There was no clinical information available regarding the possible exposure of the patients to metals. Urinary cytologies were prepared using standard clinical laboratory methods and the cells subsequently immunostained for MT-3 positive cells using an MT-3 antibody. The hypothesis was that patients with urothelial cancer would shed MT-3 positive cells into their urine and that the shedding of MT-3 positive cells might identify patients with urothelial cancer and also those whose disease had relapsed to an active state. The present diagnosis of urothelial cancer relies on the visual examination of the bladder using a cystoscope. The results of the present study did not support this initial hypothesis for either newly diagnosed patients or for those being assessed for recurrence of urothelial cancer. Urinary cytology documented MT-3 positive cells in only a subset of patients confirmed to have bladder cancer by cystoscopy and also found many instances of MT-3 positive cells in patients having been diagnosed with urothelial cancer and having no evidence of recurrence upon cytoscopic examination. Despite not advancing the initial hypothesis, there were some potentially important findings in the study. First, it was shown that patients without a diagnosis of urothelial cancer rarely had MT-3 positive cells in their urine. The low rate in the control population is significant since these samples were collected in the urology clinic and there are no or few disease-free patients in such a specialized clinic. This indicates a very low rate of MT-3 expression in individuals without urothelial cancer. Second, the results also showed that a subset of urothelial cancer patients did shed MT-3 positive cells into their urine and those with more progressive urothelial cancer were more prone to shed MT-3 positive cells. This may indicate that MT-3 staining in cytologies from newly diagnosed and recurrent urothelial cancer patients may have promise as a prognostic marker for disease progression. There are two rationales in support of this concept. The first is that urinary cytology depends on the loss of strong cell-to-cell contact between adjacent cells, allowing cells to shed into the urine. As such, MT-3 positive cells in the urine may define urothelial cancers where there has been an extensive loss in cell-to-cell contact and interaction with the surrounding tissue environment. These would be expected to define more aggressive cancers prone to invasion of the bladder wall. A second related rationale involves a field effect of "normal" tissue adjacent to the urothelial cancer that may have expression of MT-3. This would explain the presence of MT-3 positive cells in the urine from individuals negative for a recurrence of bladder cancer when examined by cytoscopy. The field effect would contain pre-malignant cells that are positive for MT-3. A long term clinical follow-up of current patients and further analysis of archival tissue will be necessary to advance these possibilities.
Methods
Cell culture
Stock cultures of the parent UROtsa cell line and the transformed Cd
+2 and As
+3cell lines were maintained in 75 cm
2 tissue culture flasks using Dulbecco's modified Eagles' medium (DMEM) containing 5% v/v fetal calf serum in a 37°C, 5% CO
2: 95% air atmosphere [
2]. Confluent flasks were sub-cultured at a 1:4 ratio using trypsin-EDTA (0.05%, 0.02%) and the cells were fed fresh growth medium every 3 days.
Treatment of UROtsa cells with 5-Aza-2'-deoxycytidine and histone deacetylase inhibitor MS-275
Parent and transformed UROtsa cells (106)were seeded at a 1:10 ratio and the next day they were treated with 1 or 3 μM 5-AZC (Sigma-Aldrich, St. Louis, MO) or 1, 3 or 10 μM MS-275 (ALEXIS Biochemicals, Lausen, Switzerland). The cells were allowed to grow to confluency (48 h) and then harvested for RNA isolation. For the exposure and recovery experiment, the cells were exposed to 3 or 10 μM MS-275 until they reached confluency (2-3 days), fed fresh media without drug for 24 h, and then dosed with 100 μM ZnSO4(Sigma-Aldrich) for 24 h and harvested for RNA isolation.
RNA isolation and RT-PCR analysis
Total RNA was isolated from the cells according to the protocol supplied with TRI REAGENT (Molecular Research Center, Inc., Cincinnati, OH) as described previously by this laboratory [
16]. Real time RT-PCR was used to measure the expression level of MT-3 mRNA levels utilizing a previously described MT-3 isoform-specific primer [
16]. For analysis, 1 μg was subjected to complementary DNAsynthesis using the iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA) in a total volume of 20 μl. Real-time PCR was performed utilizing the SYBR Green kit (Bio-Rad Laboratories) with 2 μl of cDNA, 0.2 μM primers in a total volume of 20 μl in an iCycler iQ real-time detection system (Bio-Rad Laboratories). Amplification was monitored by SYBR Green fluorescence and compared to that of a standard curve of the MT-3 isoform gene cloned into pcDNA3.1/hygro (+) and linearized with
Fsp I. Cycling parameters consisted of denaturation at 95°C for 30 s and annealing at 65°C for 45 s which gave optimal amplification efficiency of each standard. The level of MT-3 expression was normalized to that of β-actin assessed by the same assay with the primer sequences being sense, CGACAACGGCTCCGGCATGT, and antisense, TGCCGTGCTCGATGGGGTACT, with the cycling parameters of annealing/extension at 62°C for 45 s and denaturation at 95°C for 15 s. Semiquantitative RT-PCR was also performed for MT- 3 expression using the GeneAmp RNA PCR Kit (Applied Biosystem, Foster city, CA) as described previously [
17].
ChIP (Chromatin Immunoprecipitation) assay
ChIP assays were carried out using the ChIP-IT™ Express kit (Active Motif, Carlsbad, CA). The protocols and reagents were supplied by the manufacturer. UROtsa parent and the transformed cell lines were seeded at 10
6 cells/75 cm
2 flask and 24 hrs later treated with 10 μM MS-275. Following incubation for 48 hrs, the cells were fixed with 1% formaldehyde for 10 min. Cross linking was stopped by the addition of glycine stop solution (0.125 M). The cells were scraped in 2 ml phosphate buffered saline containing 0.5 mM PMSF. The cells were pelleted and resuspended in ice cold lysis buffer and homogenized in an ice-cold dounce homogenizer. The released nuclei were pelleted and resuspended in a digestion buffer supplemented with PMSF and protease inhibitor cocktail. The chromatin was sheared using the enzymatic shearing cocktail at 37°C for 5 min to an average length of 200-1500 bp. Approximately 7 μg of sheared chromatin was used to coat the protein G-coated magnetic beads along with 3 μg of the antibody. The following antibodies were used in the immunoprecipitations: MTF-1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), Histone H3 trimethyl Lys9, Histone H3 trimethyl Lys4, Histone H3 trimethyl Lys27, (Active Motif) and Anti-acetyl-Histone H4 (Millipore, Billerica, MA). The negative control IgG was purchased from Active Motif. The coating was performed overnight at 4°C following which the beads were washed and the immune complexes were eluted using the elution buffer and the cross linking was reversed using the reverse cross linking buffer. The immunoprecipitated DNA was analyzed by real time PCR using the iQ™ SYBR Green Supermix kit from Bio-Rad and semi quantitative PCR using the Gene Amp PCR core kit from Applied Biosystems. The primers for the MT-3 promoter were designed to span certain segments of the MT-3 promoter as depicted in Figure
4, and the sequences and annealing temperatures are indicated in Table
2. For quantitative PCR analysis, the quantity of the PCR template found in each specific precipitate was normalized to the amount of the corresponding DNA sequence found in the fragmented chromatin solution present before antibody-based precipitation (normalized to percentage of DNA input).
Table 2
Sequences and PCR conditions for primers
MRE A & B
| AAAGAGCGGGCGCGGTGC | GACGCGCGGCTTGGCTAGTGG | 111 | 75 |
MRE C
| GGCCCCGGCAGTGCACA | GCGCACGCACTGCATCTGTCG | 55 | 75 |
MRE E, F, G
| ATGGTACGTGCGCGCTTCC | CATCCGCGTGCACGACCCACT | 124 | 70 |
Region 1
| GAACAGATCTGGCGTCCTG | GCGCACGTACCATCTCCGA | 111 | 63 |
Region 2
| GTCGGGCTCATCGTGA | ATTCTCCAGGACGCCAGAT | 77 | 61 |
Urinary cytology and immunostaining for MT-3
The collection of urine and access to clinical data was reviewed and approved by both the IRB at the University of North Dakota and the IRB of Sanford Health. All participants signed an informed consent document. The procedures for the collection of urine and preparation for urinary cytology were identical to those procedures used for clinical diagnosis of urinary samples in the Sanford Health Urology Clinic and the Sanford Health Cytology Laboratory in Fargo, ND. The Sanford Health Laboratory is fully accredited by the College of American Pathologists (CAP) and meets all standards of the Clinical Laboratory Improvement Act (CLIA). Briefly, urine samples were accessioned with time and date stamp upon arrival in the laboratory. Color, clarity and amount were recorded for each sample. The sample was centrifuged for 5 min at 2,000 rpm (Eppendorf 5810R) and the specimen decanted, leaving cellular material and 2 - 5 ml of supernatant. An equal volume of PreservCyt was added and 2 to 5 ThinPrep® slides prepared from each sample. The slides were spray fixed immediately after preparation and allowed to dry completely. Prior to immunostaining, sections were immersed in preheated Target Retrieval Solution (Dako, Carpinteria, CA) and heated in a steamer for 20 minutes. The sections were allowed to cool to room temperature and immersed into Tris-buffered saline containing Tween 20 for 5 minutes. The immunostaining was performed on a Dako autostainer universal staining system. A primary anti-rabbit MT-3 antibody generated and characterized by this laboratory was used to localize MT-3 protein expression [
2,
18]. The primary antibody was localized using the Dakocytomation EnVision+ System-HRP for rabbit primary antibodies. Liquid diaminobenzidine was used for visualization (DakoCytomation liquid DAB substrate chromogen system). Slides were rinsed in distilled water, dehydrated in graded ethanol, cleared in xylene, and coverslipped. The presence and degree of MT-3 immunoreactivity was judged by two pathologists. Sections of human kidney served as a positive control for MT-3 staining.
Statistics
Statistical analysis for the promoter studies consisted of ANOVA with Tukey post-hoc testing performed by GraphPad PRISM 4. All statistical significance is denoted at p < 0.05.
For the urine cytology experiments, statistical analysis was performed with the aid of PASW Statistics 18 (SPSS, Inc., Chicago). Pearson Chi-square was used to calculate the distribution of MT-3 positive or negative counts in each group, as well as to evaluate the correlations of frequency of MT-3 positive or negative between each group. Kaplan-Meier method was applied for survival analysis, Log-rank and Tarone-Ware tests were used to analyze for statistical significance. A value of p < 0.05 was considered statistically significant.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
SS: Cell Culture and MRE ChIP Analysis. SG: Histone Modifications. CT: Informed Consent, Patient Urine Samples, Clinic Data. XDZ: Interpretation of Urinary Cytologies. YZ: Statistics and Patient Database. AA: Measurement of MT-3 Expression. MAS: Interpretation of Urinary Cytologies. DAS: Wrote the paper, Designed Study, Data Interpretation, IRB monitoring. All the authors read and approved the final manuscript.