Pioglitazone induces dysplastic urothelial changes in urinary bladder of experimental diabetes

Ahmed A. M. Abdel-Hamid; Alaa El-Din L. Firgany

doi:10.4103/jmau.jmau_34_21

ORIGINAL ARTICLE

Year : 2023 | Volume : 11 | Issue : 1 | Page : 34-40

Pioglitazone induces dysplastic urothelial changes in urinary bladder of experimental diabetes

Ahmed A. M. Abdel-Hamid¹, Alaa El-Din L. Firgany²
¹ Department of Anatomy, Taibah College of Medicine, Taibah University, Almadina Almonawara, Buraydah, Kingdom of Saudi Arabia; Department of Histology and Medical Cell Biology, Faculty of Medicine, Mansoura University, Mansoura, Egypt
² Department of Histology and Medical Cell Biology, Faculty of Medicine, Mansoura University, Mansoura, Egypt; Department of Basic Medical Sciences, Unaizah College of Medicine, Qassim University, Buraydah, Kingdom of Saudi Arabia

Date of Submission	02-Apr-2021
Date of Decision	20-May-2021
Date of Acceptance	22-May-2021
Date of Web Publication	06-Aug-2021

Correspondence Address:
Dr. Ahmed A. M. Abdel-Hamid
Department of Histology and Cell Biology, Faculty of Medicine, Mansoura University, Mansoura

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jmau.jmau_34_21

Abstract

Objectives: Pioglitazone (PIO) is a widely prescribed oral antidiabetic drug that has concerns regarding a potential risk of developing carcinoma of the urinary bladder. The objective of the current study was to assess this potential risk. Materials and Methods: The potential risk of PIO-induced urinary bladder carcinoma was assessed in the current study by examining urinary bladder of rats for urothelial cytokeratin (CK) expression and proliferative activity by Ki67 immunostaining. Results: Histological examination revealed dysplastic urothelial changes in PIO per se and diabetes mellitus + PIO (diabetic rats receiving PIO). In addition, a significantly (P < 0.05) decreased CK7 and CK8 expression together with a significantly increased CK20 as well as Ki67 expression was detected in the urothelial cells of groups administrated PIO, contrary to those which did not. Conclusion: The manifestations of urothelial dysplasia evidenced by histological examination as well as by the aberrant expression in CK and Ki67 after PIO administration add supporting evidence at cellular and experimental level to the previous clinical suspicions.

Keywords: Cytokeratin, diabetes, Ki67 staining, pioglitazone, proliferation, urothelial dysplasia

How to cite this article:
Abdel-Hamid AA, Firgany AE. Pioglitazone induces dysplastic urothelial changes in urinary bladder of experimental diabetes. J Microsc Ultrastruct 2023;11:34-40

How to cite this URL:
Abdel-Hamid AA, Firgany AE. Pioglitazone induces dysplastic urothelial changes in urinary bladder of experimental diabetes. J Microsc Ultrastruct [serial online] 2023 [cited 2023 Mar 31];11:34-40. Available from: https://www.jmau.org/text.asp?2023/11/1/34/323335

Highlights

Pioglitazone (PIO) significantly decreases cytokeratin (CK) 7 and CK8 expression in urothelium
PIO significantly increases CK20 expression in urothelium
PIO significantly increases proliferative activity in urothelium
PIO provokes dysplastic changes in urothelium observed by routine histology
PIO does not affect the urothelium thickness or smooth muscle/collagen ratio.

Introduction

Although pioglitazone (PIO), a thiazolidinedione, is a widely used oral antidiabetic drug in patients with type 2 diabetes mellitus (DM),^[1] a contradictory evidence has been emerged regarding its administration and the association with urinary bladder cancer.^[2] Some reports proposed that the cumulative use of PIO is not associated with the incidence of bladder cancer;^[3],[4] nonetheless, other studies have suggested an increased risk of bladder cancer with its use.^[1],[5],[6] Therefore, additional researches are required to provide more precise evidence for the potential implications of PIO administration.

Ki67, a nuclear marker of cellular proliferation, has been shown to correlate with the growth of many neoplasms, including transitional cell carcinoma (TCC). It may be used as a marker of urothelial cells as its expression paralleled that of p53 in TCC.^[7] In addition, cytokeratin (CK) family consists of at least 20 types of cytoplasmic intermediate filaments found in the epithelial cells.^[8],[9] Its abnormal expression has been found in various forms of neoplasia.^[10] CK20, a low-molecular-weight CK, is specifically limited to the superficial umbrella cells in the normal urothelium.^{[11],[12],[13]} Aberrant CK20 expression has been documented as a useful diagnostic marker in urothelial carcinoma.^[9]

CK7 and CK8 are also intermediate filaments found in the urothelial neoplasia of the urinary bladder.^[14] CK7 together with CK20 has a diagnostic role in differentiating origin of carcinomas, including urothelial carcinoma of the urinary bladder.^[15],[16] In addition, CK8 may play a key role in tumor cell invasion, although the exact mechanism is not yet clarified.^[17]

Morphologically, the noninvasive, intraepithelial neoplasms of the urothelium may be papillary or flat.^[18] One of these flat lesions is urothelial dysplasia, a premalignant condition that should be clearly identified from other flat urothelial lesions as flat hyperplasia, reactive atypia, atypia of unknown significance (AUS), and carcinoma in situ (CIS). In most cases, morphologic features alone suffice for diagnosis; yet, immunohistochemical (IHC) markers by CK and proliferative markers may be required for the precise diagnosis.^[19],[20]

The objective of the current study was to investigate the effect of PIO administration on urothelium of urinary bladder focusing on the histological changes including those occurring in the expression of a panel of CK as well as in the proliferative marker, Ki67.

Materials and Methods

Design of the experiment

Adult Sprague–Dawley rats weighing 190–230 g were provided by the Animal Housing Center of the Mansoura College of Pharmacy. Standard plastic cages were used to place the animals at room temperature of 23°C–25°C and under 12/12-h light/dark cycle. All the experimental maneuvers were approved by the Ethical Committee of Mansoura College of Medicine and were conducted according to their guidelines.

The animals were equally divided (20 animals per group) into four groups: I (control group), II (DM group), III (PIO group), and IV (DM + PIO group). During the whole period of experiment control and PIO per se, animals received standard laboratory animal's chow and water ad libitum. To induce DM in Group II and IV, rats were injected with streptozotocin (STZ, 15 mg/kg, intraperitoneally) after feeding on high-fat diet (containing fat which represents 30% of calories) for 2 months.^[21] Blood glucose analyses were carried out to get sure that rats become diabetic; rats having fasting blood glucose ≥ 160 mg/dl 1 week after STZ injection for 2 successive measurements were considered diabetic. In addition, PIO (20 mg/kg) was administered orally for 28 days to rats Group II and Group IV.^[22]

Histological and immunohistochemical evaluation

After 28 days of the start of PIO treatment, rats were sacrificed. Their urinary bladder was irrigated by a transurethral catheter with 10% formalin solution and kept distended overnight after ligation of both ureters. Rinsing of fixed bladder tissues with PBS and dehydration in graded concentration of alcohol, with subsequent paraffin embedding, were performed. Then, their sections (thickness = 4 μm) were processed for staining by hematoxylin and eosin (H and E), and various IHC assays were processed for anticipated histopathological changes. Meanwhile, Masson's trichrome staining was used to evaluate contents of smooth muscles and collagen in sections of the urinary bladder.

Morphometric assessment

Urothelial thickness

Urothelial thickness was evaluated in the H and E-stained sections of the urinary bladder (×1000). The distance (in μm) between the urothelial basement membrane to the apical border was calculated with ImageJ program v 1.43 (Bethesda, Maryland, USA). In each section, five fields were assessed with the estimation of 10 measurements in each field.

Smooth muscle/collagen ratio

The smooth muscle as well as collagen areas were quantitatively measured in sections of urinary bladder stained by Masson's trichrome. This was also accomplished by ImageJ program using color segmentation technique as a basis for evaluation. The ratio between the smooth muscle area% to that of collagen was estimated, considering the positive area for Masson's stain as the parameter of evaluation (10 fields were assessed/section in five sections for each animal).

Ki67 immunohistochemistry

Paraffin sections of the urinary bladder were IHC stained by anti-Ki67 (1/25, monoclonal, MIB-5, Dako, Glostrup, Denmark) antibodies. Antibodies were applied on paraffin-embedded sections with 20-min heat-induced epitope retrieval in Dako Target Retrieval Solution, citrate pH 6, and 30-min incubation at room temperature with the primary antibody. Combination of biotinylated rabbit anti-mouse immunoglobulins (Dako, code No. E0464) diluted 1:50, and streptavidin/HRP, diluted 1:300, was used for visualization using DAB⁺ (DakoCytomation) as chromogen.

Cytokeratin 7

To assess Ck7 IHC expression, we used anti-CK-7 antibody (Monoclonal, EPR17078, 1/8000 dilution, Abcam) in the current study. Antigen retrieval was performed heat-mediated procedure with employing Tris/EDTA buffer, at pH 9.0; then, further steps of the protocol IHC technique were commenced. Then, paraffin-embedded urinary bladder was labeled with CK7 ab181598, followed by goat anti-rabbit immunoglobulin G H&L (HRP), at 1/500 dilution. Counterstaining was performed with hematoxylin.

Cytokeratin 8

In the current study, anti-CK 8 antibody (Abcam, M20, Monoclonal, at concentration of 4 μg/ml) was employed to detect CK8 IHC expression with an automated system DAKO Autostainer Plus. Sections of the urinary bladder were rehydrated, and antigen was retrieved by Dako 3-in-1 AR buffer EDTA pH 9.0 in a DAKO PT Link. Slides were blocked in 3% H₂O₂ in methanol and then in Dako Protein block (containing casein 0.25% in PBS, each for 10 min). Afterward, they were incubated with the primary antibody (for 20 min) and the colorimetric reaction was detected by Diaminobenzidine Dako using Envision Flex amplification kit (30 min). Finally, counterstaining of slides with haematoxylin and coverslipping were commenced.

Cytokeratin 20

Anti-CK 20 antibody (Abcam, SP33 ab64090, Rabbit monoclonal, 1/100 dilution) was purchased to evaluate CK20 IHC expression. Urinary bladder sections were incubated with the primary antibody for 30 min at room temperature. Heat mediated antigen retrieval with Tris-EDTA buffer (pH 6.0, epitope retrieval solution 1) for 10 min. We used an HRP/DAB detection kit as a secondary antibody. Hematoxylin was used as a counterstain.

Statistical analysis

Statistical analysis was performed by SPSS 20 software (Chicago, IL, USA). Mean and standard deviations were used for data presentation and ANOVA test to compare between various groups regarding Ki67 and various histological parameters, whereas Mann–Whitney test was selected for CK quantification. A P < 0.05 was considered statistically significant.

Results

Effect of pioglitazone on routine histology

H and E-stained sections of the control [Figure 1]a and DM [Figure 1]b groups displayed that urothelium is formed of the typical 4–7 layers, with the top one having the umbrella-shaped cells and then the intermediate layer on the underling basal one. The cells exhibited vesicular nuclei with visible nucleolus and frequent cytoplasmic spaces. Although we did not detect a frank sinister pathology in the form of invasive carcinoma as TCC, yet atypical changes of urothelium have been observed in the PIO [Figure 1]c and DM + PIO [Figure 1]d groups. These included loss of the frequent empty spaces seen in the cytoplasm together with relative nuclear augmentation compared to the cytoplasmic compartment (N/C ratio), loss of nuclear polarity, as well as nuclear irregularity, and hyperchromasia. Unexpectedly, mitotic figures were scarcely seen in the urothelium of various groups. On the other hand, the loose connective tissue of the corium remained unchanged in all groups without evidence of inflammatory changes [Figure 1].

Figure 1: Representative bladder micrographs showing atypical characters in the pioglitazone and diabetes mellitus + pioglitazone (c and d respectively) urothelia contrary to those of nondiabetic and diabetes mellitus (a and b respectively) animals. These include partial loss of the frequent empty cytoplasmic spaces, together with relatively augmented N/C ratio and loss of nuclear polarity. Interestingly, occasional binucleation of superficial cells was observed in the diabetes mellitus group. Mitotic figures were not frequently encountered in urothelium of various groups. The corium appears relatively unchanged in various study groups. (H and E, ×400, scale bar = 20 μm)

Click here to view

Concerning the urothelium thickness and the smooth muscle/collagen ratio, no significant changes were detected in either PIO or DM + PIO groups compared to others [Figure 2]a and [Figure 2]b.

Figure 2: Graphs representing urothelial thickness (a, b μm) and the ratio between smooth muscle and collagen, which remained insignificantly changed in various groups. The lower panel of the figure shows the % of Ki67-stained cells (c) and their intensity score (arbitrary unit, d) which are significantly increased in animals administrated pioglitazone, contrary to those which did not. Values are expressed as mean ± standard deviations (n = 20 in each group). *: Significant versus control group, $: Significant versus diabetes mellitus group, P < 0.05 is significant

Click here to view

Effect of pioglitazone on urothelial proliferation

To assess the proliferative activity of urothelium, we stained Urinary bladder (UB) sections with anti-Ki67 immunostaining. A minimal to nearly negative Ki67 immunoreactivity was observed in both the nondiabetic [Figure 3]a and DM [Figure 3]b urothelia. Nonetheless, a positive Ki67 immunostaining was detected in urothelial cells of the PIO [Figure 3]c and DM + PIO [Figure 3]d groups which involved its full thickness from top layer to basal one [Figure 3]. This was statistically confirmed by the significant increase in both Ki67% and intensity in urothelial cells of PIO as well as DM + PIO groups versus the other ones [Figure 2]c and [Figure 2]d.

Figure 3: Representative bladder micrographs in various groups immunostained with anti-Ki67. Urothelia of both the nondiabetic and diabetes mellitus (a and b, respectively) animals display a nearly negative Ki67 immunoreactivity. An intense positive Ki67 immunoreactivity is seen in urothelial cells of pioglitazone and diabetes mellitus + pioglitazone (c and d, respectively) groups including the whole epithelial thickness. (Anti-Ki67 immunostaining, ×400, scale bar = 20 μm)

Click here to view

Effect of pioglitazone on cytokeratin expression

To confirm the dysplastic changes observed by routine histology as well as Ki67 immunostaining, we stained bladder sections of various groups with CK7, CK8, and CK20 immunostaining. The urothelium of both control [Figure 4]a and DM [Figure 4]b groups displayed a positive CK7 reactivity, particularly in the most superficial cells involving the umbrella-shaped ones; nonetheless, PIO [Figure 4]c and DM + PIO [Figure 4]d urothelia showed a weak-to-negative immunoreactivity. In addition, the urothelial cells of both nondiabetic [Figure 5]a and DM [Figure 5]b groups showed a diffuse CK8 reaction including all its layers, whereas that of the PIO [Figure 5]c and DM + PIO [Figure 5]d groups displayed minimal positive-to-negative reaction. On the contrary, the urothelium of both control [Figure 6]a and DM [Figure 6]b groups exhibited CK20 reaction limited to top layer including the umbrella cells. Yet, extension of the CK20 immunoreactivity toward the deeper layer was observed in urothelium of the PIO [Figure 6]c and DM + PIO [Figure 6]d groups.

Figure 4: Representative bladder micrographs showing an intense CK7 reactivity in urothelia of both the control (a) and diabetes mellitus (b) groups, particularly in the most superficial cells involving the umbrella-shaped ones. On the other hand, those of pioglitazone and diabetes mellitus + pioglitazone (c and d, respectively) groups show a negative reactivity towards anti-CK7 antibodies. (Anti-CK7 immunostaining, ×400, scale bar = 20 μm)

Click here to view

Figure 5: Representative bladder micrographs displaying a diffuse and intense CK8 reactivity in urothelium of both the control and diabetes mellitus (a and b, respectively) groups including all its layers. On the other hand, that of the pioglitazone and diabetes mellitus + pioglitazone (c and d respectively) groups displays minimal to negative reaction towards anti-CK8 antibodies. (Anti-CK8 immunostaining, ×400, scale bar = 20 μm)

Click here to view

Figure 6: Representative bladder micrographs showing CK20 immunoreactivity limited to top layer including the umbrella cells in urothelium of both the control (a) and diabetes mellitus (b) groups. On the other hand, extension of the CK20 immunoreactivity toward the deeper layers is seen in urothelium of the pioglitazone and diabetes mellitus + pioglitazone (c and d, respectively) groups. (Anti-CK20 immunostaining, ×400, scale bar = 20 μm)

Click here to view

Statistical validation of expression of CK7, CK8, and CK20 was determined by comparing the corresponding IHC scoring in various groups. A significant decrease in the expression of CK7 [Figure 7]a and CK8 [Figure 7]b was detected in PIO and DM + PIO groups compared to control and DM groups. On the other hand, the score of CK20 expression was significantly increased in the PIO as well as DM + PIO groups versus those which did not administrate PIO whether diabetic or not [Figure 7]c.

Figure 7: Graphs representing CK7, CK8, and CK20 IHC score in various groups. A significantly decreased expression of CK7 (a) and CK8 (b) is detected in groups administered pioglitazone, contrary to those which did not. On the other hand, a significantly increased CK20 expression (c) is detected in urothelia of pioglitazone as well as diabetes mellitus + Pioglitazone compared to controls and diabetes mellitus groups. Expression of CK7, CK8, and CK20 in each group, based on the percentage of cells with strong staining. *: Significant versus control group, $: Significant versus diabetes mellitus group, P < 0.05 by Kruskal–Wallis and Wilcoxon rank-sum test

Click here to view

Discussion

Routine bladder histology of the control and DM groups displayed that urothelium is formed of 4–7 layers; the most superficial displayed the characteristic cells of umbrella-shaped cells and then an intermediate layer lying on basal cells. Urothelial nuclei appeared vesicular with frequent spaces in the cytoplasm. These findings coincide with what was previously reported.^[19],[20] The observed clear cytoplasmic areas occur as a result of the dissolving of its glycogen.^[13]

In the current study, atypical changes of urothelium were observed in the PIO and DM + PIO groups including loss of the frequent empty cytoplasmic spaces together with relative nuclear augmentation, compared to the cytoplasmic fragment, loss of nuclear polarity, as well as nuclear irregularity and hyperchromasia without significant changes in urothelial thickness. These findings are typically the morphological manifestations reported in urothelial dysplasia.^{[12],[23],[24]} Cells of the intermediate layer classically demonstrate apical/basal nuclear polarity with the longitudinal axis lying perpendicularly on the basement membrane. Crowding of cells together with a longitudinal axis of their nuclei parallel to the basement membrane is stigmata of loss of this polarity.^[13]

Urothelial dysplasia denotes an early morphologic sign of progressive changes between normal urothelium and CIS. It is of clinical significance as up to one-fifth of its cases can progress to further lesions.^[25] In contrast, reactive atypia does not progress to dysplasia, CIS, or urothelial carcinoma.^[19],[20] It usually follows cystitis, stones, or instrumentation and is manifested by mild nuclear changes occurring in inflamed urothelia with frequent basal mitoses but without loss of the normal nuclear polarity.^[25],[26]

Our data revealed occasional binucleation of the superficial cells in the DM group; nonetheless, mitotic figures were not frequently encountered in urothelium of various groups. The occasional binucleation of the superficial cells should not be misinterpreted as dysplastic change.^[13] We have previously demonstrated this binucleation in normal nondysplastic urothelium.^[27] Similar to our findings, Hodges et al. mentioned that mitotic figures are not usually seen in normal urothelium apart from its basal layer.^[13]

Interestingly, the corium remained unchanged in all groups of the current study without evidence of inflammatory reaction or changes in smooth muscle/collagen ratio. This was consistent with former researches which proposed that dysplasia is unlike AUS where the corium beneath urothelium is typically inflamed and acquires neovascularity.^[19],[20]

A statistically significant increase in Ki67 immunoreactivity involving full urothelial thickness was detected in the PIO and DM + PIO groups, contrary to the minimal or virtually negative immunoreactivity in the control and DM groups. Ki67 is normally absent or <10% positive in the basal urothelium, indicating a low proliferative rate.^{[24],[28],[29]} The higher percentage and intensity of Ki67 expression in dysplastic urothelial cells comparable to that detected in TCC is supporting evidence that it could progress to invasive TCC.^[7]

We detected a significantly decreased IHC score of CK7 and CK8 expression in the groups administrated PIO compared to those which did not. The urothelium of the latter displayed a positive CK7 reactivity, particularly in the top layer and a rather diffused CK8 expression compared to the weak-to-negative reaction of the PIO and DM + PIO groups. CK7 and CK8 expression has been demonstrated in most bladder tumors,^[14] and CK7 is detected in a major fraction of carcinomas of the bladder.^[9] Yet, Wilkerson mentioned that CK7 and CK8 are detectable in normal, reactive, as well as neoplastic urothelium but nearly undetectable in urothelial dysplasia.^[30] Loss of CK8 expression, a characteristic of simple epithelia, may be also seen in interstitial cystitis.^[31] Moreover, a decrease in CK7 and CK8 expression can occur in some higher-grade TCCs.^[17]

Together with Ki67, CK20 may be helpful in differential consideration for urothelial dysplasia from reactive atypia.^[32],[33] Ki67 is well known marker for epithelial proliferation in vascular, [34] metabolic[35-40] as well as pre-neopalstic lesions.[41-43] CK20 expression is restricted to the umbrella cells in normal urothelium.^[13],[29] In dysplasia, aberrant expression of CK20 is seen in the deeper layers of the urothelium.^[19],[20] This was in agreement with our findings of a significantly increased score of CK20 expression with deeper extension of its immunoreactivity in urothelium of the PIO and DM + PIO compared to other groups.

Although some studies suggested that no evidence of association between the risk of bladder cancer and PIO administration,^[3],[4] others have reported a possible increased risk.^[1],[5],[44] This was particularly proposed in people after long-term and high-dose cumulative administration.^[2],[45] Therefore, cessation of PIO and substitution with a suitable alternative might be recommended in patients with having risk factors, such as smoking, family history, or previous chemotherapy, besides those with newly diagnosed bladder cancer.^[1]

Conclusion

Although we did not find a frank sinister pathology as carcinoma, PIO had clearly induced dysplastic changes in urothelium of the urinary bladder manifested by histological alterations besides the aberrant expression of the CK as well as ki67. These findings provide supporting evidence at experimental level for potentially premalignant changes occurred in the urothelium after PIO administration. This warrants both further investigation and a selective and judicious prescription of PIO, particularly in those with high risk of developing bladder carcinoma.

Recommendations

Based on the potential evidence of dysplastic urothelial changes induced by PIO in urinary bladder of experimental diabetes, we recommend further experimental research as well as clinical trials to confirm this potential risk.

Ethical approval

Approval of procedures of this work (Proposal Code: MS.18.05.160-2018/05/30) was obtained from Mansoura College of Medicine Ethical Committee.

Acknowledgments

The authors have not made any acknowledgment.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Ferwana M, Firwana B, Hasan R, Al-Mallah MH, Kim S, Montori VM, et al. Pioglitazone and risk of bladder cancer: A meta-analysis of controlled studies. Diabet Med 2013;30:1026-32.
2.	He S, Tang YH, Zhao G, Yang X, Wang D, Zhang Y. Pioglitazone prescription increases risk of bladder cancer in patients with type 2 diabetes: An updated meta-analysis. Tumour Biol 2014;35:2095-102.
3.	Levin D, Bell S, Sund R, Hartikainen SA, Tuomilehto J, Pukkala E, et al. Pioglitazone and bladder cancer risk: A multipopulation pooled, cumulative exposure analysis. Diabetologia 2015;58:493-504.
4.	Korhonen P, Heintjes EM, Williams R, Hoti F, Christopher S, Majak M, et al. Pioglitazone use and risk of bladder cancer in patients with type 2 diabetes: Retrospective cohort study using datasets from four European countries. BMJ 2016;354:i3903.
5.	Neumann A, Weill A, Ricordeau P, Fagot JP, Alla F, Allemand H. Pioglitazone and risk of bladder cancer among diabetic patients in France: A population-based cohort study. Diabetologia 2012;55:1953-62.
6.	Bosetti C, Rosato V, Buniato D, Zambon A, La Vecchia C, Corrao G. Cancer risk for patients using thiazolidinediones for type 2 diabetes: A meta-analysis. Oncologist 2013;18:148-56.
7.	Sun W, Zhang PL, Herrera GA. p53 protein and Ki-67 overexpression in urothelial dysplasia of bladder. Appl Immunohistochem Mol Morphol 2002;10:327-31.
8.	Southgate J, Harnden P, Trejdosiewicz L. Invited reviews-cytokeratin expression patterns in normal and malignant urothelium: A review of the biological and diagnostic implications. Histol Histopathol 1999;14:657-64.
9.	Jiang J, Ulbright TM, Younger C, Sanchez K, Bostwick DG, Koch MO, et al. Cytokeratin 7 and cytokeratin 20 in primary urinary bladder carcinoma and matched lymph node metastasis. Arch Pathol Lab Med 2001;125:921-3.
10.	Lane EB, Alexander CM. Use of keratin antibodies in tumor diagnosis. Semin Cancer Biol 1990;1:165-79.
11.	Moll R, Schiller DL, Franke WW. Identification of protein IT of the intestinal cytoskeleton as a novel type I cytokeratin with unusual properties and expression patterns. J Cell Biol 1990;111:567-80.
12.	Chandramouly G, Abad PC, Knowles DW, Lelièvre SA. The control of tissue architecture over nuclear organization is crucial for epithelial cell fate. J Cell Sci 2007;120:1596-606.
13.	Hodges KB, Lopez-Beltran A, Davidson DD, Montironi R, Cheng L. Urothelial dysplasia and other flat lesions of the urinary bladder: clinicopathologic and molecular features. Hum Pathol 2010;41:155-62.
14.	Riedel I, Czernobilsky B, Lifschitz-Mercer B, Roth LM, Wu XR, Sun TT, et al. Brenner tumors but not transitional cell carcinomas of the ovary show urothelial differentiation: immunohistochemical staining of urothelial markers, including cytokeratins and uroplakins. Virchows Arch 2001;438:181-91.
15.	Wang N, Zee S, Zarbo R, Bacchi C, Gown A. Coordinate expression of cytokeratins 7 and 20 defines unique subsets of carcinomas. Appl Immunohistochem 1995;3:99-107.
16.	Chu P, Wu E, Weiss LM. Cytokeratin 7 and cytokeratin 20 expression in epithelial neoplasms: A survey of 435 cases. Mod Pathol 2000;13:962-72.
17.	Schaafsma HE, Ramaekers FC, van Muijen GN, Lane EB, Leigh IM, Robben H, et al. Distribution of cytokeratin polypeptides in human transitional cell carcinomas, with special emphasis on changing expression patterns during tumor progression. Am J Pathol 1990;136:329-43.
18.	Montironi R, Lopez-Beltran A, Mazzucchelli R, Bostwick DG. Classification and grading of the non-invasive urothelial neoplasms: Recent advances and controversies. J Clin Pathol 2003;56:91-5.
19.	Lopez-Beltran A, Montironi R, Vidal A, Scarpelli M, Cheng L. Urothelial dysplasia of the bladder: Diagnostic features and clinical significance. Anal Quant Cytopathol Histpathol 2013;35:121-9.
20.	Lopez-Beltran A, Marques RC, Montironi R, Reymundo C, Fonseca J, Cheng L. Dysplasia and carcinoma in situ of the urinary bladder. Anal Quant Cytopathol Histpathol 2015;37:29-38.
21.	Zhang F, Ye C, Li G, Ding W, Zhou W, Zhu H, et al. The rat model of type 2 diabetic mellitus and its glycometabolism characters. Exp Anim 2003;52:401-7.
22.	Ding SY, Shen ZF, Chen YT, Sun SJ, Liu Q, Xie MZ. Pioglitazone can ameliorate insulin resistance in low-dose streptozotocin and high sucrose-fat diet induced obese rats. Acta Pharmacol Sin 2005;26:575-80.
23.	Cheng L, Cheville JC, Neumann RM, Bostwick DG. Natural history of urothelial dysplasia of the bladder. Am J Surg Pathol 1999;23:443-7.
24.	Montironi R, Lopez-Beltran A, Scarpelli M, Mazzucchelli R, Cheng L. Morphological classification and definition of benign, preneoplastic and non-invasive neoplastic lesions of the urinary bladder. Histopathology 2008;53:621-33.
25.	Cheng L, Cheville JC, Neumann RM, Bostwick DG. Flat intraepithelial lesions of the urinary bladder. Cancer 2000;88:625-31.
26.	Lopez-Beltran A, Cheng L, Andersson L, Brausi M, de Matteis A, Montironi R, et al. Preneoplastic non-papillary lesions and conditions of the urinary bladder: An update based on the Ancona International Consultation. Virchows Arch 2002;440:3-11.
27.	Abdel-Hamid AA, Ali EM. Effect of testosterone therapy on the urinary bladder in experimental hypogonadism of rats. J Mol Histol 2015;46:263-72.
28.	McKenney JK, Desai S, Cohen C, Amin MB. Discriminatory immunohistochemical staining of urothelial carcinoma in situ and non-neoplastic urothelium: An analysis of cytokeratin 20, p53, and CD44 antigens. Am J Surg Pathol 2001;25:1074-8.
29.	Mallofré C, Castillo M, Morente V, Solé M. Immunohistochemical expression of CK20, p53, and Ki-67 as objective markers of urothelial dysplasia. Mod Pathol 2003;16:187-91.
30.	Wilkerson M. Urinary bladder. In: Lin F, Prichard J, editors. Handbook of Practical Immunohistochemistry: Frequently Asked Questions. New York: Springer; 2011. p. 321-33.
31.	Laguna P, Smedts F, Nordling J, Horn T, Bouchelouche K, Hopman A, et al. Keratin expression profiling of transitional epithelium in the painful bladder syndrome/interstitial cystitis. Am J Clin Pathol 2006;125:105-10.
32.	Harnden P, Eardley I, Joyce AD, Southgate J. Cytokeratin 20 as an objective marker of urothelial dysplasia. Br J Urol 1996;78:870-5.
33.	Abdel-Hamid AAM, Firgany AEL. Dysplastic urothelial changes accompany empagliflozin administration in urinary bladder of experimental diabetes, International journal of experimental pathology 2019;100:369-77.
34.	Abdel-Hamid AAM, Firgany AEL, Mesbah Y, Soliman MF. Vascular and cellular changes accompany altered expression of angiopoietins in placenta of non-complicated ART pregnancies, Experimental and molecular pathology 2017;102:284-9.
35.	Abdel-Hamid AA, El-Firgany Ael D. Hydroxychloroquine hindering of diabetic isletopathy carries its signature on the inflammatory cytokines, Journal of molecular histology 2016;47:183-93.
36.	Abdel-Hamid AAM, Firgany AEL. Favorable outcomes of hydroxychloroquine in insulin resistance may be accomplished by adjustment of the endothelial dysfunction as well as the skewed balance of adipokines, Acta histochemica 2016;118:560-73.
37.	Abdel-Hamid AAM, Firgany AEL. Modulatory effect of empagliflozin on cellular parameters of endocrine pancreas in experimental pre-diabetes, Annals of Anatomy - Anatomischer Anzeiger 2019.
38.	Abdel-Hamid AAM, Atef H, Zalata KR, Abdel-Latif A. Correlation between testicular mast cell count and spermatogenic epithelium in non-obstructive azoospermia, International journal of experimental pathology 2018;99:22-8.
39.	Abdel-Hamid AAM, A.E.-D.L. Firgany, Correlation between pancreatic mast cells and the low grade inflammation in adipose tissue of experimental prediabetes, Acta histochemica 121(1) (2019) 35-42.
40.	Abdel-Hamid AAM, Firgany AEL, Increased mast cell number is associated with a decrease in beta-cell mass and regeneration in type 2 diabetic rats, Acta histochemica 2019;121:508-515.
41.	Abdel-Hamid AA, Mesbah Y, Farouk MF. Tubal cytokine changes accompany the epithelial atypia of letrozole-stimulated ovaries, Acta histochemica 2016;118:236-43.
42.	Abdel-Hamid AA, Mesbah Y, Soliman MF. Reversal of tubo-ovarian atypical epithelial patterns after cessation of ovarian stimulation by letrozole, International journal of experimental pathology 2016;97:329-36.
43.	Atef E, Zalata KR, Atef H, Abdel-Hamid AA. Increased Proliferative Activity Accompanies the Local Inflammatory Response of Gastric Mucosa After Intragastric Balloon Insertion, Digestive diseases and sciences 2016;61:3498-505.
44.	Zhu Z, Shen Z, Lu Y, Zhong S, Xu C. Increased risk of bladder cancer with pioglitazone therapy in patients with diabetes: A meta-analysis. Diabetes Res Clin Pract 2012;98:159-63.
45.	Lewis JD, Ferrara A, Peng T, Hedderson M, Bilker WB, Quesenberry CP Jr., et al. Risk of bladder cancer among diabetic patients treated with pioglitazone: Interim report of a longitudinal cohort study. Diabetes Care 2011;34:916-22.