Elsevier

Journal of Pediatric Urology

Volume 15, Issue 2, April 2019, Pages 158.e1-158.e10
Journal of Pediatric Urology

Hypoxic changes to the urothelium as a bystander of end-stage bladder disease

https://doi.org/10.1016/j.jpurol.2019.01.012Get rights and content

Summary

Introduction

Urothelial cells harvested from benign diseased bladders have a compromised capacity to propagate or differentiate in vitro, potentially limiting their application in autologous tissue engineering approaches. The causative pathways behind this altered phenotype are unknown.

The hypothesis is that hypoxic damage to the urothelium occurs as a bystander to chronic or recurrent episodes of infection and inflammation.

Objective

The aim of this study was to assess immunohistochemically detected nuclear hypoxia-inducible factor 1 alpha (HIF-1α) and vascular endothelial growth factor in the urothelium when exposed to hypoxia.

Study design

Human bladder sections from a total of 29 adult and paediatric patients, representing a variety of different pathologies including neuropathy (n = 15), were analysed. Tissues from adults with bladder outlet obstruction secondary to prostatic disease (n = 1), urothelial carcinoma (n = 1) and tonsil (n = 1) were used as positive tissue controls for immunohistochemistry.

Hypoxia-inducible factor 1 alpha–labelled sections were scanned using a Zeiss AxioScan Z1 slide scanner. Analysis of urothelial nuclear HIF-1α labelling was performed using HistoQuest image analysis software (TissueGnostics). Comparison of nuclear HIF-1α labelling between neuropathic and non-neuropathic sections was performed using one-way analysis of variance with the post hoc Tukey honestly significant difference (HSD) test. Patient urodynamic studies performed before tissue sample harvest were evaluated and correlated to the HIF-1α intensity using Spearman's rank correlation.

Results

Hypoxia-inducible factor 1 alpha appeared more intense in the urothelial compartment from neuropathic bladder samples (n = 15) than in the control tissues, including non-obstructed samples (n = 9). Image analysis supported this; median nuclear HIF-1α labelling was 29.98 ± 3.10 (standard deviation [SD]) (n = 9) in controls and 74.29 ± 7.55 (SD) in neuropathic samples (n = 15). A statistically significant difference between the control and neuropathic tissue groups was shown (P < 0.05).

Of the 15 neuropathic samples, 11 had traceable urodynamic studies. Both initial and maximum detrusor pressures indicated a positive relationship when plotted against HIF-1α labelling. Spearman's rank correlation, with no missing events, confirmed significant correlations between both initial or maximum detrusor pressure and nuclear HIF-1α labelling intensity (median score); P ≤ 0.046 and P ≤ 0.05, respectively. The null hypothesis was accordingly rejected.

Conclusions

This study indicates that urothelial nuclear HIF-1α may be a biomarker of hypoxia and a common feature in end-stage bladder disease associated with high-pressure systems.

Introduction

The normal urinary bladder experiences variation in perfusion and hence oxygenation throughout the filling and voiding phases of the micturition cycle. Compression of the intramural vasculature during the filling phase resulting in intermittent hypoxia is generally accepted, but there is scant information regarding physiological oxygen tensions in the functioning human bladder. In an in-vivo guinea pig bladder model [1], oxygen saturation during the filling phase was high (92–95%), with a slight reduction in oxygen saturation seen during the voiding phase. These reported fluctuations in oxygen saturation were negatively impacted by detrusor overactivity and outlet obstruction.

Chronic interruption of the vasculature or nervous supply can progress to end-stage bladder disease, typified by the small, fibrotic bladder of limited capacity and increased risk of upper urinary tract damage [2]. Various animal models have been established to examine the compensatory and decompensatory effects of partial outflow obstruction on bladder function, and these have been used variously to report tissue oxygen tensions [1], [3], [4]. A porcine outflow obstruction model used to investigate the effects of hypoxia on the bladder [5] found similar oxygen tensions (2.5–5.5%) to those quoted generically for smooth muscles [6]. The authors have suggested that chronic bladder outlet obstruction (BOO) led to the detrusor experiencing both a reduction in blood flow and increased hypoxia, potentially contributing to the functional and structural changes characteristic of obstructed bladders. Hypoxia-dependent pathways were upregulated in a murine model of chronic intermittent bladder over-distension [3], [7], with a 2.6-fold increase in hypoxia-inducible factor 1 alpha (HIF-1α) transcript and downstream genes associated ontologically with fibrosis and inflammation [7].

Hypoxia inducible factors (HIFs) are a family of transcription factors that mediate the adaptive response to hypoxia in cells and tissues. Active HIF is heterodimeric, consisting of the constitutively expressed HIF-1β subunit (also known as aryl hydrocarbon receptor nuclear translocator) and one of the three oxygen-sensitive alpha subunits 1α, 2α or 3α [8]. During normoxia, HIF-1α subunits are degraded in the cytoplasm by the hydroxylation-ubiquitination-proteasomal system [9]. Under hypoxic conditions, nuclear translocation of HIF-1α and heterodimerisation with HIF-1β result in the induction of target genes responsible for angiogenesis, glucose metabolism, cell survival and microenvironment remodelling [10].

Hypoxia has been associated with adult human bladder disease, with the detection of nuclear HIF in the bladder wall being used as a marker of hypoxia pathway activation in both benign and malignant conditions, including BOO [11]. In the translation of novel bladder reconstructive techniques [12], it was unexpectedly found that urothelial cells from children with end-stage bladder diseases showed a compromised phenotype in vitro even though the urothelium itself was not implicated in the disease process [13].

Surgery or therapy, such as intravesical Botox or augmentation, to alleviate high pressures may reduce active hypoxia signalling and temporarily relieve symptoms. However, the possible phenotypic changes already undergone by the urothelium due to the prior chronic hypoxia exposure also remain poorly understood.

The aim of this work was to test the hypothesis that the urothelium itself encounters hypoxia, associated with high pressures, during end-stage bladder disease. Immunohistochemistry of archived tissues was used to quantitatively examine nuclear expression of HIF-1α in the urothelium and correlate this to the results of urodynamic investigations.

Additional immunohistochemistry (non-quantified) was performed on samples to examine the expression of one of the downstream targets of HIF-1α, vascular endothelial growth factor (VEGF). VEGF is an autocrine and paracrine signalling molecule whose primary role is the activation of angiogenesis. However, it has also been implicated in tissue inflammatory responses and, specifically, the recruitment of monocytes and macrophages [14], [15].

The link between hypoxia, VEGF and inflammation, coupled with the knowledge that inflammation is a feature of the neuropathic bladder, suggests that VEGF may be upregulated in neuropathic bladders.

Section snippets

Tissue samples

All human urothelial tissue samples used were collected from three different hospitals between 2007 and 2015. All specimens were covered by national health service (NHS) Research Ethics Committee approvals, with stipulated patient consent, and are indicated as follows: REC 12/YH/0507, REC 99/095, REC99/04/003 and REC04/Q1206/143.

The resection specimens of bladder were collected from paediatric patients with neuropathic bladders (n = 15): 14 due to myelomeningocele and one due to non-neuropathic

Immunolabelling of human urothelial tissues for evidence of hypoxia

Microscopically nuclear labelling of HIF-1α appeared more intense in the urothelial compartment from the neuropathic bladder samples (n = 15) than in the various control tissues, including non-obstructed, samples (n = 9) (Fig. 1a). Evaluation of VEGF expression in neuropathic bladder versus bladder tissue from patients with VUR and VUJO is shown in Fig. 1b. Subjectively, there may be a difference in expression, which is predominantly in the cytoplasm, but no image analysis was performed.

Image analysis

In

Discussion

Bladders from neuropathic patients frequently acquire the features, such as compromised capacity, fibrosis and raised pressures, of end-stage bladder disease, resulting increased risk to the upper renal tracts. The effect on the urothelium is unknown. Here, evidence in the form of significantly increased nuclear HIF-1α expression indicates that in paediatric neuropathic bladder disease, the urothelium is exposed to hypoxia-related pathway activation, which is further supported by the subjective

Conclusion

The study indicates that the presence of urothelial nuclear HIF-1α may be a biomarker of end-stage bladder disease. In particular, nuclear expression of this transcription factor was significantly associated with high-pressure systems. Hypoxia-inducible factor 1 alpha as evidence of the pathological endpoint does not help clarify the multifactorial prequel involving infection, recurrent inflammation, fibrosis and so on. The understanding of these feeding pathways would help identify modifiable

Acknowledgements

The authors would like to thank the following people for help with the supply of tissues: Mr. Alex Turner of Leeds Children's Hospital for his advice and support with collecting tissues; Mr Niaz Ahmad of the Urology department of Leeds Teaching Hospitals NHS Trust, St. James's University Hospital, Beckett Street, Leeds, LS9 7 TF; Mr Derek Rosario of the Department of Oncology & Metabolism, The Medical School, Beech Hill Road, Sheffield S10 2RX, Mr Imran Mushtaq of the Department of Urology,

Cited by (5)

View full text