Terabyte-scale supervised 3D training and benchmarking dataset of the mouse kidney.

The performance of machine learning algorithms, when used for segmenting 3D biomedical images, does not reach the level expected based on results achieved with 2D photos. This may be explained by the comparative lack of high-volume, high-quality training datasets, which require state-of-the-art imaging facilities, domain experts for annotation and large computational and personal resources. The HR-Kidney dataset presented in this work bridges this gap by providing 1.7 TB of artefact-corrected synchrotron radiation-based X-ray phase-contrast microtomography images of whole mouse kidneys and validated segmentations of 33 729 glomeruli, which corresponds to a one to two orders of magnitude increase over currently available biomedical datasets. The image sets also contain the underlying raw data, threshold- and morphology-based semi-automatic segmentations of renal vasculature and uriniferous tubules, as well as true 3D manual annotations. We therewith provide a broad basis for the scientific community to build upon and expand in the fields of image processing, data augmentation and machine learning, in particular unsupervised and semi-supervised learning investigations, as well as transfer learning and generative adversarial networks.

Evolution of hypoxia and hypoxia-inducible factor asparaginyl hydroxylase regulation in chronic kidney disease.

BACKGROUND: The roles of hypoxia and hypoxia inducible factor (HIF) during chronic kidney disease (CKD) are much debated. Interventional studies with HIF-α activation in rodents have yielded contradictory results. The HIF pathway is regulated by prolyl and asparaginyl hydroxylases. While prolyl hydroxylase inhibition is a well-known method to stabilize HIF-α, little is known about the effect asparaginyl hydroxylase factor inhibiting HIF (FIH). METHODS: We used a model of progressive proteinuric CKD and a model of obstructive nephropathy with unilateral fibrosis. In these models we assessed hypoxia with pimonidazole and vascularization with three-dimensional micro-computed tomography imaging. We analysed a database of 217 CKD biopsies from stage 1 to 5 and we randomly collected 15 CKD biopsies of various severity degrees to assess FIH expression. Finally, we modulated FIH activity in vitro and in vivo using a pharmacologic approach to assess its relevance in CKD. RESULTS: In our model of proteinuric CKD, we show that early CKD stages are not characterized by hypoxia or HIF activation. At late CKD stages, some areas of hypoxia are observed, but these are not colocalizing with fibrosis. In mice and in humans, we observed a downregulation of the HIF pathway, together with an increased FIH expression in CKD, according to its severity. Modulating FIH in vitro affects cellular metabolism, as described previously. In vivo, pharmacologic FIH inhibition increases the glomerular filtration rate of control and CKD animals and is associated with decreased development of fibrosis. CONCLUSIONS: The causative role of hypoxia and HIF activation in CKD progression is questioned. A pharmacological approach of FIH downregulation seems promising in proteinuric kidney disease.

Virtual histology of an entire mouse brain from formalin fixation to paraffin embedding. Part 2: Volumetric strain fields and local contrast changes.

BACKGROUND: Fixation and embedding of post mortem brain tissue is a pre-requisite for both gold-standard conventional histology and X-ray virtual histology. This process alters the morphology and density of the brain microanatomy. NEW METHOD: To quantify these changes, we employed synchrotron radiation-based hard X-ray tomography with 3 µm voxel length to visualize the same mouse brain after fixation in 4% formalin, immersion in ethanol solutions (50%, 70%, 80%, 90%, and 100%), xylene, and finally after embedding in a paraffin block. The volumetric data were non-rigidly registered to the initial formalin-fixed state to align the microanatomy within the entire mouse brain. RESULTS: Volumetric strain fields were used to characterize local shrinkage, which was found to depend on the anatomical region and distance to external surface. X-ray contrast was altered and enhanced by preparation-induced inter-tissue density changes. The preparation step can be selected to highlight specific anatomical features. For example, fiber tract contrast is amplified in 100% ethanol. COMPARISON WITH EXISTING METHODS: Our method provides volumetric strain fields, unlike approaches based on feature-to-feature or volume measurements. Volumetric strain fields are produced by non-rigid registration, which is less labor-intensive and observer-dependent than volume change measurements based on manual segmentations. X-ray microtomography provides spatial resolution at least an order of magnitude higher than magnetic resonance microscopy, allowing for analysis of morphology and density changes within the brain's microanatomy. CONCLUSION: Our approach belongs to three-dimensional virtual histology with isotropic micrometer spatial resolution and therefore complements atlases based on a combination of magnetic resonance microscopy and optical micrographs of serial histological sections.

Virtual histology of an entire mouse brain from formalin fixation to paraffin embedding. Part 1: Data acquisition, anatomical feature segmentation, tracking global volume and density changes.

BACKGROUND: Micrometer-resolution neuroimaging with gold-standard conventional histology requires tissue fixation and embedding. The exchange of solvents for the creation of sectionable paraffin blocks modifies tissue density and generates non-uniform brain shrinkage. NEW METHOD: We employed synchrotron radiation-based X-ray microtomography for slicing- and label-free virtual histology of the mouse brain at different stages of the standard preparation protocol from formalin fixation via ascending ethanol solutions and xylene to paraffin embedding. Segmentation of anatomical regions allowed us to quantify non-uniform tissue shrinkage. Global and local changes in X-ray absorption gave insight into contrast enhancement for virtual histology. RESULTS: The volume of the entire mouse brain was 60%, 56%, and 40% of that in formalin for, respectively, 100% ethanol, xylene, and paraffin. The volume changes of anatomical regions such as the hippocampus, anterior commissure, and ventricles differ from the global volume change. X-ray absorption of the full brain decreased, while local absorption differences increased, resulting in enhanced contrast for virtual histology. These trends were also observed with laboratory microtomography measurements. COMPARISON WITH EXISTING METHODS: Microtomography provided sub-10 µm spatial resolution with sufficient density resolution to resolve anatomical structures at each step of the embedding protocol. The spatial resolution of conventional computed tomography and magnetic resonance microscopy is an order of magnitude lower and both do not match the contrast of microtomography over the entire embedding protocol. Unlike feature-to-feature or total volume measurements, our approach allows for calculation of volume change based on segmentation. CONCLUSION: We present isotropic micrometer-resolution imaging to quantify morphology and composition changes in a mouse brain during the standard histological preparation. The proposed method can be employed to identify the most appropriate embedding medium for anatomical feature visualization, to reveal the basis for the dramatic X-ray contrast enhancement observed in numerous embedded tissues, and to quantify morphological changes during tissue fixation and embedding.

Simultaneous Three-Dimensional Vascular and Tubular Imaging of Whole Mouse Kidneys With X-ray µCT.

Concurrent three-dimensional imaging of the renal vascular and tubular systems on the whole-kidney scale with capillary level resolution is labor-intensive and technically difficult. Approaches based on vascular corrosion casting and X-ray micro computed tomography (µCT), for example, suffer from vascular filling artifacts and necessitate imaging with an additional modality to acquire tubules. In this work, we report on a new sample preparation, image acquisition, and quantification protocol for simultaneous vascular and tubular µCT imaging of whole, uncorroded mouse kidneys. The protocol consists of vascular perfusion with the water-soluble, aldehyde-fixable, polymeric X-ray contrast agent XlinCA, followed by laboratory-source µCT imaging and structural analysis using the freely available Fiji/ImageJ software. We achieved consistent filling of the entire capillary bed and staining of the tubules in the cortex and outer medulla. After imaging at isotropic voxel sizes of 3.3 and 4.4 µm, we segmented vascular and tubular systems and quantified luminal volumes, surface areas, diffusion distances, and vessel path lengths. This protocol permits the analysis of vascular and tubular parameters with higher reliability than vascular corrosion casting, less labor than serial sectioning and leaves tissue intact for subsequent histological examination with light and electron microscopy.

Crosslinkable polymeric contrast agent for high-resolution X-ray imaging of the vascular system.

A contrast agent for X-ray micro computed tomography (µCT), called XlinCA, that combines reliable perfusion and permanent retention and contrast properties, was developed for ex vivo imaging. The new imaging agent XlinCA features high molecular weight, low viscosity and a high iodine content.

Generation of renal Epo-producing cell lines by conditional gene tagging reveals rapid HIF-2 driven Epo kinetics, cell autonomous feedback regulation, and a telocyte phenotype.

Erythropoietin (Epo) is essential for erythropoiesis and is mainly produced by the fetal liver and the adult kidney following hypoxic stimulation. Epo regulation is commonly studied in hepatoma cell lines, but differences in Epo regulation between kidney and liver limit the understanding of Epo dysregulation in polycythaemia and anaemia. To overcome this limitation, we have generated a novel transgenic mouse model expressing Cre recombinase specifically in the active fraction of renal Epo-producing (REP) cells. Crossing with reporter mice confirmed the inducible and highly specific tagging of REP cells, located in the corticomedullary border region where there is a steep drop in oxygen bioavailability. A novel method was developed to selectively grow primary REP cells in culture and to generate immortalized clonal cell lines, called fibroblastoid atypical interstitial kidney (FAIK) cells. FAIK cells show very early hypoxia-inducible factor (HIF)-2α induction, which precedes Epo transcription. Epo induction in FAIK cells reverses rapidly despite ongoing hypoxia, suggesting a cell autonomous feedback mechanism. In contrast, HIF stabilizing drugs resulted in chronic Epo induction in FAIK cells. RNA sequencing of three FAIK cell lines derived from independent kidneys revealed a high degree of overlap and suggests that REP cells represent a unique cell type with properties of pericytes, fibroblasts, and neurons, known as telocytes. These novel cell lines may be helpful to investigate myofibroblast differentiation in chronic kidney disease and to elucidate the molecular mechanisms of HIF stabilizing drugs currently in phase III studies to treat anemia in end-stage kidney disease.

Renal arteriovenous oxygen shunting.

PURPOSE OF REVIEW: Renal arteriovenous oxygen shunting has been proposed as a mechanism by which oxygen supplied to the kidney can bypass the renal parenchyma. Shunting could, therefore, play a crucial role in renal hypoxia and hyperoxia. In the absence of suitable quantitative experimental methods, computational modeling has been employed in recent years to estimate the extent and potential impact of oxygen shunting. RECENT FINDINGS: Overestimation of the separation distance between arteries and veins was suggested to be responsible for previous findings that only negligible amounts of oxygen are shunted in the preglomerular vasculature. However, models considering the correct separation distance and wrapping of artery-vein pairs still showed shunting at negligible levels of less than 1% of total renal oxygen delivery. The effect of reverse CO2 shunting on the oxygen-hemoglobin dissociation curve was found to impair, rather than promote, preglomerular oxygen shunting. SUMMARY: Oxygen is unlikely to be shunted along the preglomerular vasculature in sufficient quantities to affect renal oxygenation. There may be substantial shunting at the level of the postglomerular vasculature, but more extensive efforts in structural imaging and computational modeling are needed to quantify it reliably.

Novel bilayer bacterial nanocellulose scaffold supports neocartilage formation in vitro and in vivo.

Tissue engineering provides a promising alternative therapy to the complex surgical reconstruction of auricular cartilage by using ear-shaped autologous costal cartilage. Bacterial nanocellulose (BNC) is proposed as a promising scaffold material for auricular cartilage reconstruction, as it exhibits excellent biocompatibility and secures tissue integration. Thus, this study evaluates a novel bilayer BNC scaffold for auricular cartilage tissue engineering. Bilayer BNC scaffolds, composed of a dense nanocellulose layer joined with a macroporous composite layer of nanocellulose and alginate, were seeded with human nasoseptal chondrocytes (NC) and cultured in vitro for up to 6 weeks. To scale up for clinical translation, bilayer BNC scaffolds were seeded with a low number of freshly isolated (uncultured) human NCs combined with freshly isolated human mononuclear cells (MNC) from bone marrow in alginate and subcutaneously implanted in nude mice for 8 weeks. 3D morphometric analysis showed that bilayer BNC scaffolds have a porosity of 75% and mean pore size of 50 ± 25 µm. Furthermore, endotoxin analysis and in vitro cytotoxicity testing revealed that the produced bilayer BNC scaffolds were non-pyrogenic (0.15 ± 0.09 EU/ml) and non-cytotoxic (cell viability: 97.8 ± 4.7%). This study demonstrates that bilayer BNC scaffolds offer a good mechanical stability and maintain a structural integrity while providing a porous architecture that supports cell ingrowth. Moreover, bilayer BNC scaffolds provide a suitable environment for culture-expanded NCs as well as a combination of freshly isolated NCs and MNCs to form cartilage in vitro and in vivo as demonstrated by immunohistochemistry, biochemical and biomechanical analyses.

Magnetic resonance imaging of the ear for patient-specific reconstructive surgery.

INTRODUCTION: Like a fingerprint, ear shape is a unique personal feature that should be reconstructed with a high fidelity during reconstructive surgery. Ear cartilage tissue engineering (TE) advantageously offers the possibility to use novel 3D manufacturing techniques to reconstruct the ear, thus allowing for a detailed auricular shape. However it also requires detailed patient-specific images of the 3D cartilage structures of the patient's intact contralateral ear (if available). Therefore the aim of this study was to develop and evaluate an imaging strategy for acquiring patient-specific ear cartilage shape, with sufficient precision and accuracy for use in a clinical setting. METHODS AND MATERIALS: Magnetic resonance imaging (MRI) was performed on 14 volunteer and six cadaveric auricles and manually segmented. Reproducibility of cartilage volume (Cg.V), surface (Cg.S) and thickness (Cg.Th) was assessed, to determine whether raters could repeatedly define the same volume of interest. Additionally, six cadaveric auricles were harvested, scanned and segmented using the same procedure, then dissected and scanned using high resolution micro-CT. Correlation between MR and micro-CT measurements was assessed to determine accuracy. RESULTS: Good inter- and intra-rater reproducibility was observed (precision errors <4% for Cg.S and <9% for Cg.V and Cg.Th). Intraclass correlations were good for Cg.V and Cg.S (>0.82), but low for Cg.Th (<0.23) due to similar average Cg.Th between patients. However Pearson's coefficients showed that the ability to detect local cartilage shape variations is unaffected. Good correlation between clinical MRI and micro-CT (r>0.95) demonstrated high accuracy. DISCUSSION AND CONCLUSION: This study demonstrated that precision and accuracy of the proposed method was high enough to detect patient-specific variation in ear cartilage geometry. The present study provides a clinical strategy to access the necessary information required for the production of 3D ear scaffolds for TE purposes, including detailed patient-specific shape. Furthermore, the protocol is applicable in daily clinical practice with existing infrastructure.