Search

Abstract Bladder cancer is one of the most common cancer types in the urinary system. Yet, current bladder cancer diagnosis and follow-up techniques are time-consuming, expensive, and invasive. In the clinical practice, the gold standard for diagnosis remains invasive biopsy followed by histopathological analysis. In recent years, costly diagnostic tests involving the use of bladder cancer biomarkers have been developed, however these tests have high false-positive and false-negative rates limiting their reliability. Hence, there is an urgent need for the development of cost-effective, and non-invasive novel diagnosis methods. To address this gap, here we propose a quick, cheap, and reliable diagnostic method. Our approach relies on an artificial intelligence (AI) model to analyze droplet patterns of blood and urine samples obtained from patients and comparing them to cancer-free control subjects. The AI-assisted model in this study uses a deep neural network, a ResNet network, pre-trained on ImageNet datasets. Recognition and classification of complex patterns formed by dried urine or blood droplets under different conditions resulted in cancer diagnosis with a high specificity and sensitivity. Our approach can be systematically applied across droplets, enabling comparisons to reveal shared spatial behaviors and underlying morphological patterns. Our results support the fact that AI-based models have a great potential for non-invasive and accurate diagnosis of malignancies, including bladder cancer.

Autophagy is a vesicular trafficking process responsible for the degradation of long-lived, misfolded or abnormal proteins, as well as damaged or surplus organelles. Abnormalities of the autophagic activity may result in the accumulation of protein aggregates, organelle dysfunction, and autophagy disorders were associated with various diseases. Hence, mechanisms of autophagy regulation are under exploration.Over-expression of hsa-miR-376a1 (shortly MIR376A) was performed to evaluate its effects on autophagy. Autophagy-related targets of the miRNA were predicted using Microcosm Targets and MIRanda bioinformatics tools and experimentally validated. Endogenous miRNA was blocked using antagomirs and the effects on target expression and autophagy were analyzed. Luciferase tests were performed to confirm that 3' UTR sequences in target genes were functional. Differential expression of MIR376A and the related MIR376B was compared using TaqMan quantitative PCR.Here, we demonstrated that, a microRNA (miRNA) from the DLK1/GTL2 gene cluster, MIR376A, played an important role in autophagy regulation. We showed that, amino acid and serum starvation-induced autophagy was blocked by MIR376A overexpression in MCF-7 and Huh7 cells. MIR376A shared the same seed sequence and had overlapping targets with MIR376B, and similarly blocked the expression of key autophagy proteins ATG4C and BECN1 (Beclin 1). Indeed, 3' UTR sequences in the mRNA of these autophagy proteins were responsive to MIR376A in luciferase assays. Antagomir tests showed that, endogenous MIR376A was participating to the control of ATG4C and BECN1 transcript and protein levels. Moreover, blockage of endogenous MIR376A accelerated starvation-induced autophagic activity. Interestingly, MIR376A and MIR376B levels were increased with different kinetics in response to starvation stress and tissue-specific level differences were also observed, pointing out to an overlapping but miRNA-specific biological role.Our findings underline the importance of miRNAs encoded by the DLK1/GTL2 gene cluster in stress-response control mechanisms, and introduce MIR376A as a new regulator of autophagy.