Robust and generalizable segmentation of human functional tissue units

The Human BioMolecular Atlas Program aims to compile a reference atlas for the healthy human adult body at the cellular level. Functional tissue units (FTU, e.g., renal glomeruli and colonic crypts) are of pathobiological significance and relevant for modeling and understanding disease progression. Yet, annotation of FTUs is time consuming and expensive when done manually and existing algorithms achieve low accuracy and do not generalize well. This paper compares the five winning algorithms from the "Hacking the Kidney" Kaggle competition to which more than a thousand teams from sixty countries contributed. We compare the accuracy and performance of the algorithms on a large-scale renal glomerulus Periodic acid-Schiff stain dataset and their generalizability to a colonic crypts hematoxylin and eosin stain dataset. Results help to characterize how the number of FTUs per unit area differs in relationship to their position in kidney and colon with respect to age, sex, body mass index (BMI), and other clinical data and are relevant for advancing pathology, anatomy, and surgery.

Calmodulin Antagonist W-7 Enhances Intermediate Conductance Ca2+- Sensitive Basolateral Potassium Channel (IKCa) Activity in Human Colonic Crypts

Intermediate conductance potassium (IKCa) channels are exquisitively Ca2+ sensitive, intracellular Ca2+ regulating channel activity by complexing with calmodulin (CaM), which is bound to the cytosolic carboxyl tail. Although CaM antagonists might be expected to decrease IKCa channel activity, the effect of W-7 in human T lymphocytes are conflicting. We therefore evaluated the effect of W-7 on basolateral IKCa channels in human colonic crypt cells. Intact crypts obtained from normal human colonic biopsies by Ca2+ chelation were used for patch clamp studies of basolateral IKCa channels in the cell-attached configuration. IKCa channel activity was studied when the bath Ca2+ concentration was changed from 1.2 mmol/L to 100 μmol/L and back to 1.2 mmol/L, as well as from 100 μmol/L to 1.2 mmol/L and back to 100 μmol/L, both in the absence and presence of 25 μmol/L W-7. Decreasing bath Ca2+ from 1.2 mmol/L to 100 μmol/L decreased IKCa channel activity reversibly in the absence of W-7, whereas there was a uniformly high level of channel activity at both bath Ca2+ concentrations in the presence of W-7. In separate experiments, increasing bath Ca2+ from 100 μmol/L to 1.2 mmol/L increased IKCa channel activity reversibly in the absence of W-7, whereas there was again a uniformly high level of channel activity at both bath Ca2+ concentrations in the presence of W-7. We, therefore, propose that W-7 has a specific stimulatory effect on basolateral IKCa channel activity, despite its ability to inhibit Ca2+/CaM-mediated, IKCa channel-dependent Cl− secretion in human colonic epithelial cells. Graphic Abstract

3D cyclorama for digital unrolling and visualisation of deformed tubes

Colonic crypts are tubular glands that multiply through a symmetric branching process called crypt fission. During the early stages of colorectal cancer, the normal fission process is disturbed, leading to asymmetrical branching or budding. The challenging shapes of the budding crypts make it difficult to prepare paraffin sections for conventional histology, resulting in colonic cross sections with crypts that are only partially visible. To study crypt budding in situ and in three dimensions (3D), we employ X-ray micro-computed tomography to image intact colons, and a new method we developed (3D cyclorama) to digitally unroll them. Here, we present, verify and validate our ‘3D cyclorama’ method that digitally unrolls deformed tubes of non-uniform thickness. It employs principles from electrostatics to reform the tube into a series of onion-like surfaces, which are mapped onto planar panoramic views. This enables the study of features extending over several layers of the tube’s depth, demonstrated here by two case studies: (i) microvilli in the human placenta and (ii) 3D-printed adhesive films for drug delivery. Our 3D cyclorama method can provide novel insights into a wide spectrum of applications where digital unrolling or flattening is necessary, including long bones, teeth roots and ancient scrolls.

Growth Performance, Biochemical Blood Indices, and Large Intestine Physiology of Rats Fed Diets with Alfalfa Protein-Xanthophyll Concentrate

The effect of dietary levels of alfalfa protein-xanthophyll concentrate (PXC) was determined in growing rats. Three groups of eight four-week-old male Wistar rats, with an average initial body weight of 61 g, were fed for 28 days either natural-ingredient diets without PXC or supplemented with 1.5% or 3% PXC. Growth performance, blood biochemistry, caecal fermentation, morphology of the large intestine, and mucin gene expression were evaluated. PXC did not affect growth performance but tended to decrease relative liver weight. Among biochemical blood parameters, only bilirubin decreased and uric acid increased in response to 1.5% and 3% PXC, respectively. Caecal fermentation was not affected, with the exception of isovaleric acid concentration, which tended to be higher in rats fed the diet containing 3% PXC. Colonic crypts tended to be deeper in rats fed the 3% PXC diet and the thickness of the colonic mucus layer was reduced by both PXC levels. In conclusion, PXC did not affect growth performance or caecal fermentation but decreased thickness of the protective mucus layer in the colon.

The role of migration in mutant evolution in fragmented populations

Mutant evolution in fragmented populations has been studied extensively in evolutionary biology. With an increased focus on evolutionary dynamics in medical research, quantification of mutant load in fragmented populations with varying levels of migration has become especially important. Examples of fragmented populations are hematopoietic stem cell niches in the bone marrow where cells can re-circulate between niches through the blood, or colonic crypts where movement of cells across different crypts is not thought to be common. Here we use a combination of experiments and theory to investigate the role of migration in mutant distribution. In the case of neutral mutants, the experiments confirmed that while the mean number of mutants is not influenced by migration, the probability distribution is, which manifested itself in a change in the skewedness of the distribution of the mutant numbers in the demes. In the case of disadvantageous mutants, we investigated the phenomenon of the increase in the expected number of mutants compared to that of the selection-mutation balance. In a single deme, this increase is observed when the deme size is lower than the critical size, $N_c$. In a fragmented system that consists of connected demes with a probability of migration, the increase in mutant numbers above the selection-mutation balance can be maintained in small ($N<N_c$) demes as long as the migration rate is sufficiently small. The migration rate above which the mutants approach the selection-mutation balance decays exponentially with $N/N_c$. These findings are relevant in the context of the complex and poorly understood processes that may lead to changes in the clonal composition in tissues and tumors.

A Non-redundant Role for T cell-derived IL-22 in Antibacterial Defense of Colonic Crypts

IL-22 is a key cytokine in immune defense against pathogens at barrier sites. In response to enteric attaching and effacing bacteria, IL-22 produced by type 3 innate lymphoid cells (ILC3s) is thought to be important early for induction of antimicrobial peptides (AMPs) that protect intestinal epithelial cells (IECs) in advance of T cell-derived IL-22 that arises later. Yet, the basis for a requirement for both innate and adaptive IL-22-producing immune cells in protecting the intestinal mucosa is unknown. Here, using novel mice that both report IL-22 expression and can be targeted for its lineage-specific deletion, we show that mice with deficiency of IL-22 targeted to innate immune cells, including ILC3s, have impaired STAT3 activation of surface colonic IECs colonized by bacteria early in infection. In contrast, mice with IL-22 deficiency limited to T cells have complete loss of STAT3 activation in IECs lining colonic crypts and fail to protect the crypts from bacterial invasion late despite ongoing production of IL-22 from ILC3s. T cell-derived IL-22 is required for upregulation of many host-protective genes by crypt IECs, including those encoding AMPs, neutrophil-recruiting chemokines, and mucins and mucin-related molecules, while also restricting pro-inflammatory genes downstream of IFNγ and TNF signals. Thus, T cell-derived IL-22 is indispensable for antibacterial defense and damage control of intestinal crypts.

The Collagen-Modifying Enzyme PLOD2 Is Induced and Required during L1-Mediated Colon Cancer Progression

The overactivation of Wnt/b-catenin signaling is a hallmark of colorectal cancer (CRC) development. We identified the cell adhesion molecule L1CAM (L1) as a target of b-catenin-TCF transactivation in CRC cells. The overexpression of L1 in CRC cells confers enhanced proliferation, motility, tumorigenesis and liver metastasis, and L1 is exclusively localized in the invasive areas of human CRC tissue. A number of genes are induced after L1 transfection into CRC cells by a mechanism involving the cytoskeletal protein ezrin and the NF-kB pathway. When studying the changes in gene expression in CRC cells overexpressing L1 in which ezrin levels were suppressed by shRNA to ezrin, we discovered the collagen-modifying enzyme lysyl hydroxylase 2 (PLOD2) among these genes. We found that increased PLOD2 expression was required for the cellular processes conferred by L1, including enhanced proliferation, motility, tumorigenesis and liver metastasis, since the suppression of endogenous PLOD2 expression, or its enzymatic activity, blocked the enhanced tumorigenic properties conferred by L1. The mechanism involved in increased PLOD2 expression by L1 involves ezrin signaling and PLOD2 that affect the SMAD2/3 pathway. We found that PLOD2 is localized in the colonic crypts in the stem cell compartment of the normal mucosa and is found at increased levels in invasive areas of the tumor and, in some cases, throughout the tumor tissue. The therapeutic strategies to target PLOD2 expression might provide a useful approach for CRC treatment.