Machine Learning-Based Classification of Transcriptome Signatures of Non-Ulcerative Bladder Pain Syndrome.

Lower urinary tract dysfunction (LUTD) presents a global health challenge with symptoms impacting a substantial percentage of the population. The absence of reliable biomarkers complicates the accurate classification of LUTD subtypes with shared symptoms such as non-ulcerative Bladder Pain Syndrome (BPS) and overactive bladder caused by bladder outlet obstruction with Detrusor Overactivity (DO). This study introduces a machine learning (ML)-based approach for the identification of mRNA signatures specific to non-ulcerative BPS. Using next-generation sequencing (NGS) transcriptome data from bladder biopsies of patients with BPS, benign prostatic obstruction with DO, and controls, our statistical approach successfully identified 13 candidate genes capable of discerning BPS from control and DO patients. This set was validated using Quantitative Polymerase Chain Reaction (QPCR) in a larger patient cohort. To confirm our findings, we applied both supervised and unsupervised ML approaches to the QPCR dataset. A three-mRNA signature TPPP3, FAT1, and NCALD, emerged as a robust classifier for non-ulcerative BPS. The ML-based framework used to define BPS classifiers establishes a solid foundation for comprehending the gene expression changes in the bladder during BPS and serves as a valuable resource and methodology for advancing signature identification in other fields. The proposed ML pipeline demonstrates its efficacy in handling challenges associated with limited sample sizes, offering a promising avenue for applications in similar domains.

De-obstruction of bladder outlet in humans reverses organ remodelling by normalizing the expression of key transcription factors.

BACKGROUND: Benign prostatic hyperplasia in elderly males often causes bladder outlet obstruction termed benign prostatic obstruction (BPO). BPO induces lower urinary tract symptoms and quantifiable urodynamic alterations in bladder function. When conservative medical treatments are exhausted, surgical interventions like transurethral resection of the prostate (TURP) are employed for bladder outlet de-obstruction. Elucidating the molecular changes in the human bladder resulting from BPO and their reversal post-de-obstruction is pivotal for defining the "point of no return", when the organ deterioration becomes irreversible. In this study we carried out a comprehensive molecular and urodynamic characterization of the bladders in men with BPO before TURP and 3 months after the relief of obstruction. METHODS: We report integrated transcriptome and proteome analysis of bladder samples from male patients with BPO before and 3 months after de-obstruction surgery (TURP). mRNA and protein profiles were correlated with urodynamic findings, specifically voiding detrusor pressure (PdetQmax) before TURP. We delineated the molecular classifiers of each group, pointing at the different pre-TURP bladder status. RESULTS: Age-matched patients with BPO without DO were divided into two groups based on the PdetQmax values recorded by UDI before de-obstruction: high and medium pressure (HP and MP) groups. Three months after de-obstruction surgery, the voiding parameters PdetQmax, Qmax and RV were significantly improved in both groups, without notable inter-group differences in the values after TURP. Patients with high PdetQmax showed less advanced remodeling and inflammatory changes than those with lower values. We detected significant dysregulation of gene expression, which was at least partially reversed by de-obstruction in both patients' groups. Transcription factor SOX21 and its target thrombospondin 4 (THBS4) demonstrated normalization post-TURP. CONCLUSIONS: Our findings reveal substantial yet incomplete reversal of cell signalling pathways three months after TURP, consistent with improved urodynamic parameters. We propose a set of biomarker genes, indicative of BPO, and possibly contributing to the bladder changes. This study unveils the stages of progressive obstruction-induced bladder decompensation and offers insights into selecting an optimal intervention point to mitigate loss of contractility.

Machine Learning-based Classification of transcriptome Signatures of non-ulcerative Bladder Pain Syndrome.

Lower urinary tract dysfunction (LUTD) presents a global health challenge with symptoms impacting a substantial percentage of the population. The absence of reliable biomarkers complicates the accurate classification of LUTD subtypes with shared symptoms such as non-ulcerative Bladder Pain Syndrome (BPS) and overactive bladder caused by bladder outlet obstruction with Detrusor Overactivity (DO). This study introduces a machine learning (ML)-based approach for the identification of mRNA signatures specific to non-ulcerative BPS. Using next-generation sequencing (NGS) transcriptome data from bladder biopsies of patients with BPS, benign prostatic obstruction with DO and controls, our statistical approach successfully identified 13 candidate genes capable of discerning BPS from control and DO patients. This set was subsequently validated using Quantitative Polymerase Chain Reaction (QPCR) in a larger patient cohort. To confirm our findings, we applied both supervised and unsupervised ML approaches to the QPCR dataset. Notably, a three-mRNA signature TPPP3, FAT1, and NCALD, emerged as a robust classifier, effectively distinguishing patients with non-ulcerative BPS from controls and patients with DO. This signature was universally selected by both supervised and unsupervised approaches. The ML-based framework used to define BPS classifiers not only establishes a solid foundation for comprehending the specific gene expression changes in the bladder of the patients with BPS but also serves as a valuable resource and methodology for advancing signature identification in other fields. The proposed ML pipeline demonstrates its efficacy in handling challenges associated with limited sample sizes, offering a promising avenue for applications in similar domains.

SpheroScan: A User-Friendly Deep Learning Tool for Spheroid Image Analysis.

Background: In recent years, three-dimensional (3D) spheroid models have become increasingly popular in scientific research as they provide a more physiologically relevant microenvironment that mimics in vivo conditions. The use of 3D spheroid assays has proven to be advantageous as it offers a better understanding of the cellular behavior, drug efficacy, and toxicity as compared to traditional two-dimensional cell culture methods. However, the use of 3D spheroid assays is impeded by the absence of automated and user-friendly tools for spheroid image analysis, which adversely affects the reproducibility and throughput of these assays. Results: To address these issues, we have developed a fully automated, web-based tool called SpheroScan, which uses the deep learning framework called Mask Regions with Convolutional Neural Networks (R-CNN) for image detection and segmentation. To develop a deep learning model that could be applied to spheroid images from a range of experimental conditions, we trained the model using spheroid images captured using IncuCyte Live-Cell Analysis System and a conventional microscope. Performance evaluation of the trained model using validation and test datasets shows promising results. Conclusion: SpheroScan allows for easy analysis of large numbers of images and provides interactive visualization features for a more in-depth understanding of the data. Our tool represents a significant advancement in the analysis of spheroid images and will facilitate the widespread adoption of 3D spheroid models in scientific research. The source code and a detailed tutorial for SpheroScan are available at https://github.com/FunctionalUrology/SpheroScan.

Molecular Characterization of Non-Neurogenic and Neurogenic Lower Urinary Tract Dysfunction (LUTD) in SCI-Induced and Partial Bladder Outlet Obstruction Mouse Models.

We examined bladder function following spinal cord injury (SCI) by repeated urodynamic investigation (UDI), including external urethral sphincter (EUS) electromyography (EMG) in awake restrained mice and correlated micturition parameters to gene expression and morphological changes in the bladder. A partial bladder outlet obstruction (pBOO) model was used for comparison to elucidate both the common and specific features of obstructive and neurogenic lower urinary tract dysfunction (LUTD). Thirty female C57Bl/6J mice in each group received an implanted bladder catheter with additional electrodes placed next to the EUS in the SCI group. UDI assessments were performed weekly for 7 weeks (pBOO group) or 8 weeks (SCI group), after which bladders were harvested for histological and transcriptome analysis. SCI mice developed detrusor sphincter dyssynergia (DSD) one week after injury with high-pressure oscillations and a significantly increased maximal bladder pressure Pmax and were unable to void spontaneously during the whole observation period. They showed an increased bladder-to-bodyweight ratio, bladder fibrosis, and transcriptome changes indicative of extracellular matrix remodeling and alterations of neuronal signaling and muscle contraction. In contrast, pBOO led to a significantly increased Pmax after one week, which normalized at later time points. Increased bladder-to-bodyweight ratio and pronounced gene expression changes involving immune and inflammatory pathways were observed 7 weeks after pBOO. Comparative transcriptome analysis of SCI and pBOO bladders revealed the activation of Wnt and TGF-beta signaling in both the neurogenic and obstructive LUTD and highlighted FGF2 as a major upregulated transcription factor during organ remodeling. We conclude that SCI-induced DSD in mice leads to profound changes in neuronal signaling and muscle contractility, leading to bladder fibrosis. In a similar time frame, significant bladder remodeling following pBOO allowed for functional compensation, preserving normal micturition parameters.

SpheroScan: a user-friendly deep learning tool for spheroid image analysis.

BACKGROUND: In recent years, 3-dimensional (3D) spheroid models have become increasingly popular in scientific research as they provide a more physiologically relevant microenvironment that mimics in vivo conditions. The use of 3D spheroid assays has proven to be advantageous as it offers a better understanding of the cellular behavior, drug efficacy, and toxicity as compared to traditional 2-dimensional cell culture methods. However, the use of 3D spheroid assays is impeded by the absence of automated and user-friendly tools for spheroid image analysis, which adversely affects the reproducibility and throughput of these assays. RESULTS: To address these issues, we have developed a fully automated, web-based tool called SpheroScan, which uses the deep learning framework called Mask Regions with Convolutional Neural Networks (R-CNN) for image detection and segmentation. To develop a deep learning model that could be applied to spheroid images from a range of experimental conditions, we trained the model using spheroid images captured using IncuCyte Live-Cell Analysis System and a conventional microscope. Performance evaluation of the trained model using validation and test datasets shows promising results. CONCLUSION: SpheroScan allows for easy analysis of large numbers of images and provides interactive visualization features for a more in-depth understanding of the data. Our tool represents a significant advancement in the analysis of spheroid images and will facilitate the widespread adoption of 3D spheroid models in scientific research. The source code and a detailed tutorial for SpheroScan are available at https://github.com/FunctionalUrology/SpheroScan.

Integrated mRNA-miRNA transcriptome analysis of bladder biopsies from patients with bladder pain syndrome identifies signaling alterations contributing to the disease pathogenesis.

BACKGROUND: Interstitial cystitis, or bladder pain syndrome (IC/BPS), is a chronic bladder disorder characterized by lower abdominal pain associated with the urinary bladder and accompanied by urinary frequency and urgency in the absence of identifiable causes. IC/PBS can be separated into the classic Hunner's ulcerative type and the more prevalent non-ulcerative disease. Our aim was to unravel the biological processes and dysregulated cell signaling pathways leading to the bladder remodeling in non-ulcerative bladder pain syndrome (BPS) by studying the gene expression changes in the patients' biopsies. METHODS: We performed paired microRNA (miRNA) and mRNA expression profiling in the bladder biopsies of BPS patients with non-Hunner interstitial cystitis phenotype, using comprehensive Next-generation sequencing (NGS) and studied the activated pathways and altered biological processes based on the global gene expression changes. Paired mRNA-miRNA transcriptome analysis delineated the regulatory role of the dysregulated miRNAs by identifying their targets in the disease-induced pathways. RESULTS: EIF2 Signaling and Regulation of eIF4 and p70S6K Signaling, activated in response to cellular stress, were among the most significantly regulated processes during BPS. Leukotriene Biosynthesis nociceptive pathway, important in inflammatory diseases and neuropathic pain, was also significantly activated. The biological processes identified using Gene Ontology over-representation analysis were clustered into six main functional groups: cell cycle regulation, chemotaxis of immune cells, muscle development, muscle contraction, remodeling of extracellular matrix and peripheral nervous system organization and development. Compared to the Hunner's ulcerative type IC, activation of the immune pathways was modest in non-ulcerative BPS, limited to neutrophil chemotaxis and IFN-γ-mediated signaling. We identified 62 miRNAs, regulated and abundant in BPS and show that they target the mRNAs implicated in eIF2 signalling pathway. CONCLUSIONS: The bladders of non-ulcerative BPS patients recruited in this study had alterations consistent with a strong cell proliferative response and an up-regulation of smooth muscle contractility, while the contribution of inflammatory processes was modest. Pathway analysis of the integrated mRNA-miRNA NGS dataset pinpointed important regulatory miRNAs whose dysregulation might contribute to the pathogenesis. Observed molecular changes in the peripheral nervous system organization and development indicate the potential role of local bladder innervation in the pain perceived in this type of BPS.

Urinary miRNA profiles discriminate between obstruction-induced bladder dysfunction and healthy controls.

Urgency, frequency and incomplete emptying are the troublesome symptoms often shared between benign prostatic obstruction-induced (BLUTD) and neurogenic (NLUTD) lower urinary tract dysfunction. Previously, using bladder biopsies, we suggested a panel of miRNA biomarkers for different functional phenotypes of the bladder. Urine is a good source of circulating miRNAs, but sex- and age-matched controls are important for urinary metabolite comparison. In two groups of healthy subjects (average age 32 and 57 years old, respectively) the total protein and RNA content was very similar between age groups, but the number of secreted extracellular vesicles (uEVs) and expression of several miRNAs were higher in the young healthy male volunteers. Timing of urine collection was not important for these parameters. We also evaluated the suitability of urinary miRNAs for non-invasive diagnosis of bladder outlet obstruction (BOO). A three urinary miRNA signature (miR-10a-5p, miR-301b-3p and miR-363-3p) could discriminate between controls and patients with LUTD (BLUTD and NLUTD). This panel of representative miRNAs can be further explored to develop a non-invasive diagnostic test for BOO. The age-related discrepancy in the urinary miRNA content observed in this study points to the importance of selecting appropriate, age-matched controls.