Abstract. This paper introduces a network for volumetric segmentation that learns from sparsely annotated volumetric images. We outline two attractive use cases of this method: (1) In a semi-automated setup, the user annotates some slices in the volume to be segmented. The network learns from these sparse annotations and provides a dense 3D segmentation. (2) In a fully-automated setup, we assume that a representative, sparsely annotated training set exists. Trained on this data set, the network densely segments new volumetric images. The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts. The implementation performs on-the-fly elastic deformations for efficient data augmentation during training. It is trained end-to-end from scratch, i.e., no pre-trained network is required. We test the performance of the proposed method on a complex, highly variable 3D structure, the Xenopus kidney, and achieve good results for both use cases.

Many genetic diseases have been linked to the dysfunction of primary cilia, which occur nearly ubiquitously in the body and act as solitary cellular mechanosensory organelles. The list of clinical manifestations and affected tissues in cilia-related disorders (ciliopathies) such as nephronophthisis is broad and has been attributed to the wide expression pattern of ciliary proteins. However, little is known about the molecular mechanisms leading to this dramatic diversity of phenotypes. We recently reported hypomorphic NPHP3 mutations in children and young adults with isolated nephronophthisis and associated hepatic fibrosis or tapetoretinal degeneration. Here, we chose a combinatorial approach in mice and humans to define the phenotypic spectrum of NPHP3/Nphp3 mutations and the role of the nephrocystin-3 protein. We demonstrate that the pcy mutation generates a hypomorphic Nphp3 allele that is responsible for the cystic kidney disease phenotype, whereas complete loss of Nphp3 function results in situs inversus, congenital heart defects, and embryonic lethality in mice. In humans, we show that NPHP3 mutations can cause a broad clinical spectrum of early embryonic patterning defects comprising situs inversus, polydactyly, central nervous system malformations, structural heart defects, preauricular fistulas, and a wide range of congenital anomalies of the kidney and urinary tract (CAKUT). On the functional level, we show that nephrocystin-3 directly interacts with inversin and can inhibit like inversin canonical Wnt signaling, whereas nephrocystin-3 deficiency leads in Xenopus laevis to typical planar cell polarity defects, suggesting a role in the control of canonical and noncanonical (planar cell polarity) Wnt signaling.

Cystic kidney diseases are a global public health burden, affecting over 12 million people1. Although much is known about the genetics of kidney development and disease, the cellular mechanisms driving normal kidney tubule elongation remain unclear 2,3. Here, we used in vivo imaging to demonstrate for the first time that mediolaterally-oriented cell intercalation is fundamental to vertebrate kidney morphogenesis. Surprisingly, kidney tubule elongation is driven in large part by a myosin-dependent, multi-cellular rosette-based mechanism, previously only described in Drosophila. In contrast to Drosophila, however, non-canonical Wnt/PCP signaling is required to control rosette topology and orientation during vertebrate kidney tubule elongation. These data resolve longstanding questions concerning the role of PCP signaling in the developing kidney and moreover establish rosette-based intercalation as a deeply conserved cellular engine for epithelial morphogenesis.

Nephronophthisis (NPH) is an autosomal recessive cystic kidney disease that leads to renal failure in childhood or adolescence. Most NPHP gene products form molecular networks. We have identified ANKS6 as a new NPHP family member that connects NEK8 (NPHP9) to INVERSIN (INVS, NPHP2) and NPHP3 to form a distinct NPHP module. ANKS6 localizes to the proximal cilium and knockdown experiments in zebrafish and Xenopus confirmed a role in renal development. Genetic screening identified six families with ANKS6 mutations and NPH, including severe cardiovascular abnormalities, liver fibrosis and situs inversus. The oxygen sensor HIF1AN (FIH) hydroxylates ANKS6 and INVS, while knockdown of Hif1an in Xenopus resembled the loss of other NPHP proteins. HIF1AN altered the composition of the ANKS6/INVS/NPHP3 module. Network analyses, uncovering additional putative NPHP-associated genes, placed ANKS6 at the center of the NPHP module, explaining the overlapping disease manifestation caused by mutations of either ANKS6, NEK8, INVS or NPHP3.