AI models for automated segmentation of engineered polycystic kidney tubules.

Autosomal dominant polycystic kidney disease (ADPKD) is a monogenic, rare disease, characterized by the formation of multiple cysts that grow out of the renal tubules. Despite intensive attempts to develop new drugs or repurpose existing ones, there is currently no definitive cure for ADPKD. This is primarily due to the complex and variable pathogenesis of the disease and the lack of models that can faithfully reproduce the human phenotype. Therefore, the development of models that allow automated detection of cysts' growth directly on human kidney tissue is a crucial step in the search for efficient therapeutic solutions. Artificial Intelligence methods, and deep learning algorithms in particular, can provide powerful and effective solutions to such tasks, and indeed various architectures have been proposed in the literature in recent years. Here, we comparatively review state-of-the-art deep learning segmentation models, using as a testbed a set of sequential RGB immunofluorescence images from 4 in vitro experiments with 32 engineered polycystic kidney tubules. To gain a deeper understanding of the detection process, we implemented both pixel-wise and cyst-wise performance metrics to evaluate the algorithms. Overall, two models stand out as the best performing, namely UNet++ and UACANet: the latter uses a self-attention mechanism introducing some explainability aspects that can be further exploited in future developments, thus making it the most promising algorithm to build upon towards a more refined cyst-detection platform. UACANet model achieves a cyst-wise Intersection over Union of 0.83, 0.91 for Recall, and 0.92 for Precision when applied to detect large-size cysts. On all-size cysts, UACANet averages at 0.624 pixel-wise Intersection over Union. The code to reproduce all results is freely available in a public GitHub repository.

A protocol for the generation of EPO - Producing neural crest cells from human induced pluripotent stem cells.

Insufficient production of erythropoietin (EPO) leads to anaemia. Developing methods for the generation and transplantation of EPO-producing cells would allow scientists to design personalised therapeutic solutions. Here we present a simple and highly reproducible protocol for the generation of neural crest cells (NCCs) that can produce and secrete erythropoiesis-competent EPO in response to hypoxia.

Post-translational modifications by SIRT3 de-2-hydroxyisobutyrylase activity regulate glycolysis and enable nephrogenesis.

Abnormal kidney development leads to lower nephron number, predisposing to renal diseases in adulthood. In embryonic kidneys, nephron endowment is dictated by the availability of nephron progenitors, whose self-renewal and differentiation require a relatively repressed chromatin state. More recently, NAD+-dependent deacetylase sirtuins (SIRTs) have emerged as possible regulators that link epigenetic processes to the metabolism. Here, we discovered a novel role for the NAD+-dependent deacylase SIRT3 in kidney development. In the embryonic kidney, SIRT3 was highly expressed only as a short isoform, with nuclear and extra-nuclear localisation. The nuclear SIRT3 did not act as deacetylase but exerted de-2-hydroxyisobutyrylase activity on lysine residues of histone proteins. Extra-nuclear SIRT3 regulated lysine 2-hydroxyisobutyrylation (Khib) levels of phosphofructokinase (PFK) and Sirt3 deficiency increased PFK Khib levels, inducing a glycolysis boost. This altered Khib landscape in Sirt3-/- metanephroi was associated with decreased nephron progenitors, impaired nephrogenesis and a reduced number of nephrons. These data describe an unprecedented role of SIRT3 in controlling early renal development through the regulation of epigenetics and metabolic processes.

Human iPSC-derived neural crest stem cells can produce EPO and induce erythropoiesis in anemic mice.

Inadequate production of erythropoietin (EPO) leads to anemia. Although erythropoiesis-stimulating agents can be used to treat anemia, these approaches are limited by high costs, adverse effects, and the need for frequent injections. Developing methods for the generation and transplantation of EPO-producing cells would allow for the design of personalized and complication-free therapeutic solutions. In mice, the first EPO source are neural crest cells (NCCs), which ultimately migrate to the fetal kidney to differentiate into EPO-producing fibroblasts. In humans however, it remains unknown whether NCCs can produce EPO in response to hypoxia. Here, we developed a new protocol to differentiate human induced pluripotent stem cells (hiPSCs) into NCCs and showed that cthese cells can produce functional EPO that can induce human CD34+ hematopoietic progenitor differentiation into erythroblasts in vitro. Moreover, we showed that hiPSC-derived NCCs can be embedded in clinical-grade atelocollagen scaffolds and subcutaneously transplanted into anemic mice to produce human EPO, accelerate hematocrit recovery, and induce erythropoiesis in the spleen. Our findings provide unprecedented evidence of the ability of human NCCs to produce functional EPO in response to hypoxia, and proof-of-concept for the potential clinical use of NCC-containing scaffolds as cell therapy for renal and non-renal anemia.