Human liver afferent and efferent nerves revealed by 3-D/Airyscan super-resolution imaging.

Neural regulation of hepatic metabolism has long been recognized. However, the detailed afferent and efferent innervation of the human liver has not been systematically characterized. This is largely due to the liver's high lipid and pigment contents, causing false-negative (light scattering and absorption) and false-positive (autofluorescence) results in in-depth fluorescence imaging. Here, to avoid the artifacts in three-dimensional (3-D) liver neurohistology, we embed the bleached human liver in the high-refractive-index polymer for tissue clearing and antifade 3-D/Airyscan super-resolution imaging. Importantly, using the paired substance P (SP, sensory marker) and PGP9.5 (pan-neuronal marker) labeling, we detect the sensory nerves in the portal space, featuring the SP+ varicosities in the PGP9.5+ nerve bundles/fibers, confirming the afferent liver innervation. Also, using the tyrosine hydroxylase (TH, sympathetic marker) labeling, we identify 1) condensed TH+ sympathetic nerves in the portal space, 2) extension of sympathetic nerves from the portal to the intralobular space, in which the TH+ nerve density is 2.6 ± 0.7-fold higher than that of the intralobular space in the human pancreas, and 3) the TH+ nerve fibers and varicosities contacting the ballooning cells, implicating potential sympathetic influence on hepatocytes with macrovesicular fatty change. Finally, using the vesicular acetylcholine transporter (VAChT, parasympathetic marker), PGP9.5, and CK19 (epithelial marker) labeling with panoramic-to-Airyscan super-resolution imaging, we detect and confirm the parasympathetic innervation of the septal bile duct. Overall, our labeling and 3-D/Airyscan imaging approach reveal the hepatic sensory (afferent) and sympathetic and parasympathetic (efferent) innervation, establishing a clinically related setting for high-resolution 3-D liver neurohistology.NEW & NOTEWORTHY We embed the human liver (vs. pancreas, positive control) in the high-refractive-index polymer for tissue clearing and antifade 3-D/Airyscan super-resolution neurohistology. The pancreas-liver comparison reveals: 1) sensory nerves in the hepatoportal space; 2) intralobular sympathetic innervation, including the nerve fibers and varicosities contacting the ballooning hepatocytes; and 3) parasympathetic innervation of the septal bile duct. Our results highlight the sensitivity and resolving power of 3-D/Airyscan super-resolution imaging in human liver neurohistology.

Oncogenic KRAS, Mucin 4, and Activin A-Mediated Fibroblast Activation Cooperate for PanIN Initiation.

Over 90% of patients with pancreatic ductal adenocarcinoma (PDAC) have oncogenic KRAS mutations. Nevertheless, mutated KRAS alone is insufficient to initiate pancreatic intraepithelial neoplasia (PanIN), the precursor of PDAC. The identities of the other factors/events required to drive PanIN formation remain elusive. Here, optic-clear 3D histology is used to analyze entire pancreases of 2-week-old Pdx1-Cre; LSL-KrasG12D/+ (KC) mice to detect the earliest emergence of PanIN and observed that the occurrence is independent of physical location. Instead, it is found that the earliest PanINs overexpress Muc4 and associate with αSMA+ fibroblasts in both transgenic mice and human specimens. Mechanistically, KrasG12D/+ pancreatic cells upregulate Muc4 through genetic alterations to increase proliferation and fibroblast recruitments via Activin A secretion and consequently enhance cell transformation for PanIN formation. Inhibition of Activin A signaling using Follistatin (FST) diminishes early PanIN-associated fibroblast recruitment, effectively curtailing PanIN initiation and growth in KC mice. These findings emphasize the vital role of interactions between oncogenic KrasG12D/+ -driven genetic alterations and induced microenvironmental changes in PanIN initiation, suggesting potential avenues for early PDAC diagnostic and management approaches.

Transparent tissue in solid state for solvent-free and antifade 3D imaging.

Optical clearing with high-refractive-index (high-n) reagents is essential for 3D tissue imaging. However, the current liquid-based clearing condition and dye environment suffer from solvent evaporation and photobleaching, causing difficulties in maintaining the tissue optical and fluorescent features. Here, using the Gladstone-Dale equation [(n-1)/density=constant] as a design concept, we develop a solid (solvent-free) high-n acrylamide-based copolymer to embed mouse and human tissues for clearing and imaging. In the solid state, the fluorescent dye-labeled tissue matrices are filled and packed with the high-n copolymer, minimizing scattering in in-depth imaging and dye fading. This transparent, liquid-free condition provides a friendly tissue and cellular environment to facilitate high/super-resolution 3D imaging, preservation, transfer, and sharing among laboratories to investigate the morphologies of interest in experimental and clinical conditions.

Multimodal 3-D/2-D human islet and duct imaging in exocrine and endocrine lesion environment: associated pancreas tissue remodeling.

Pancreatic intraepithelial neoplasia (PanIN) and islet cell microadenoma are exocrine and endocrine neoplasms of human pancreas that have been linked to pancreatic ductal adenocarcinoma (PDAC) and neuroendocrine tumor, respectively. However, in health and at the surgical margin of pancreatic cancer, it remains unresolved how to simultaneously characterize duct and islet remodeling to investigate the exocrine-endocrine association in the lesion microenvironment. Here, we develop a new vibratome-based approach to detect, confirm, and analyze the two types of pancreas remodeling via stereo/three-dimensional (3-D) and classic/two-dimensional (2-D) histology. Surgical margins of PDAC (n = 10, distal) and cadaveric donor pancreases (n = 10, consecutive cases) were fixed, sectioned by vibratome (350 µm), and surveyed for PanIN and microadenoma via stereomicroscopy. After lesion detection, PanIN and microadenoma were analyzed with 3-D fluorescence imaging and clinical microtome-based histology for confirmation and assessment of microenvironment. Multimodal imaging of PDAC surgical margins and cadaveric donor pancreases detected the peri-PanIN islet aggregation with duct-islet cell clusters. Organ-wide survey of cadaveric donor pancreases shows a marked 2.3-fold increase in the lesion size with the PanIN-islet association vs. without the association. In the survey, we unexpectedly detected the islet cell microadenoma adjacent to (<2 mm) PanIN. Overall, among the 53 early lesions in the cadaveric donor pancreases (PanINs and microadenomas), 81% are featured with the associated exocrine-endocrine tissue remodeling. Multimodal 3-D/2-D tissue imaging reveals local and simultaneous duct and islet remodeling in the cancer surgical margin and cadaveric donor pancreas. In the cadaveric donor pancreas, the peri-PanIN islet aggregation and PanIN-microadenoma association are two major features of pancreas remodeling in the early lesion microenvironment.NEW & NOTEWORTHY We develop a new multimodal 3-D/2-D imaging approach (matched stereomicroscopic, fluorescence, and H&E signals) to examine human duct-islet association in the PDAC surgical margin and cadaveric donor pancreas. In both conditions, peri-PanIN islet aggregation with duct-islet cell clusters was identified. The PanIN-islet cell microadenoma association was unexpectedly detected in the donor pancreas. Our work provides the technical and morphological foundations to simultaneously characterize human islets and ducts to study their association in health and disease.

Local islet remodelling associated with duct lesion-islet complex in adult human pancreas.

AIMS/HYPOTHESIS: Islets are thought to be stably present in the adult human pancreas to maintain glucose homeostasis. However, identification of the pancreatic intraepithelial neoplasia (PanIN)-islet complex in mice and the presence of PanIN lesions in adult humans suggest that similar remodelling of islet structure and environment may occur in the human pancreas. To identify islet remodelling in a clinically related setting, we examine human donor pancreases with 3D histology to detect and characterise the human PanIN-islet complex. METHODS: Cadaveric donor pancreases (26-65 years old, n = 10) were fixed and sectioned (350 µm) for tissue labelling, clearing and microscopy to detect local islet remodelling for 3D analysis of the microenvironment. The remodelled microenvironment was subsequently examined via microtome-based histology for clinical assessment. RESULTS: In nine pancreases, we identified the unique peri-lobular islet aggregation associated with the PanIN lesion (16 lesion-islet complexes detected; size: 3.18 ± 1.34 mm). Important features of the lesion-islet microenvironment include: (1) formation of intra-islet ducts, (2) acinar atrophy, (3) adipocyte association, (4) inflammation (CD45+), (5) stromal accumulation (α-SMA+), (6) increase in Ki-67 proliferation index but absence of Ki-67+ alpha/beta cells and (7) in-depth and continuous duct-islet cell contacts, forming a cluster. The duct-islet cell cluster and intra-islet ducts suggest likely islet cell neogenesis but not replication. CONCLUSIONS/INTERPRETATION: We identify local islet remodelling associated with PanIN-islet complex in the adult human pancreas. The tissue remodelling and the evidence of inflammation and stromal accumulation suggest that the PanIN-islet complex is derived from tissue repair after a local injury.

Heterogeneity and neurovascular integration of intraportally transplanted islets revealed by 3-D mouse liver histology.

Intraportal islet transplantation has been clinically applied for treatment of unstable type 1 diabetes. However, in the liver, systematic assessment of the dispersed islet grafts and the graft-hepatic integration remains difficult, even in animal models. This is due to the lack of global and in-depth analyses of the transplanted islets and their microenvironment. Here, we apply three-dimensional (3-D) mouse liver histology to investigate the islet graft microstructure, vasculature, and innervation. Streptozotocin-induced diabetic mice were used in syngeneic intraportal islet transplantation to achieve euglycemia. Optically cleared livers were prepared to enable 3-D morphological and quantitative analyses of the engrafted islets. 3-D image data reveal the clot- and plaque-like islet grafts in the liver: the former are derived from islet emboli and associated with ischemia, whereas the latter (minority) resemble the plaques on the walls of portal vessels (e.g., at the bifurcation) with mild, if any, perigraft tissue damage. Three weeks after transplantation, both types of grafts are revascularized, yet significantly more lymphatics are associated with the plaque- than clot-like grafts. Regarding the islet reinnervation, both types of grafts connect to the periportal nerve plexus and develop peri- and intragraft innervation. Specifically, the sympathetic axons and varicosities contact the α-cells, highlighting the graft-host neural integration. We present the heterogeneity of the intraportally transplanted islets and the graft-host neurovascular integration in mice. Our work provides the technical and morphological foundation for future high-definitional 3-D tissue and cellular analyses of human islet grafts in the liver.NEW & NOTEWORTHY Modern 3-D histology identifies the clot- and plaque-like islet grafts in the mouse liver, which otherwise cannot be distinguished with the standard microtome-based histology. The two types of grafts are similar in blood microvessel density and sympathetic reinnervation. Their differences, however, are their locations, severity of associated liver injury, and access to lymphatic vessels. Our work provides the technical and morphological foundation for future high-definitional 3-D tissue/cellular analyses of human islet grafts in the liver.

Lymphatic vessel remodeling and invasion in pancreatic cancer progression.

BACKGROUND: The lymphatic system is involved in metastasis in pancreatic cancer progression. In cancer staging, lymphatic spread has been used to assess the invasiveness of tumor cells. However, from the endothelium's perspective, the analysis downplays the peri-lesional activities of lymphatic vessels. This unintended bias is largely due to the lack of 3-dimensional (3-D) tissue information to depict the lesion microstructure and vasculature in a global and integrated fashion. METHODS: We targeted the pancreas as the model organ to investigate lymphatic vessel remodeling in cancer lesion progression. Transparent pancreases were prepared by tissue clearing to facilitate deep-tissue, tile-scanning microscopy for 3-D lymphatic network imaging. FINDINGS: In human pancreatic ductal adenocarcinoma, we identify the close association between the pancreatic intraepithelial neoplasia (PanIN) lesions and the lymphatic network. In mouse models of PanIN (elastase-CreER;LSL-KrasG12D and elastase-CreER;LSL-KrasG12D;p53+/-), the 3-D image data reveal the peri-lesional lymphangiogenesis, endothelial invagination, formation of the bridge/valve-like luminal tubules, vasodilation, and luminal invasion. In the orthotopic mouse model of pancreatic cancer, we identify the localized, graft-induced lymphangiogenesis and the peri- and intra-tumoral lymphatic vessel invasion. INTERPRETATION: The integrated view of duct lesions and vascular remodeling suggests an active role, rather than a passive target, of lymphatic vessels in the metastasis of pancreatic cancer. Our 3-D image data provide insights into the pancreatic cancer microenvironment and establish the technical and morphological foundation for systematic detection and 3-D analysis of lymphatic vessel invasion. FUND: Taiwan Academia Sinica (AS-107-TP-L15 and AS-105-TP-B15), Ministry of Science and Technology (MOST 106-2321-B-001-048, 106-0210-01-15-02, 106-2321-B-002-034, and 106-2314-B-007-004-MY2), and Taiwan National Health Research Institutes (NHRI EX107-10524EI).

Human pancreatic afferent and efferent nerves: mapping and 3-D illustration of exocrine, endocrine, and adipose innervation.

The pancreas consists of both the exocrine (acini and ducts) and endocrine (islets) compartments to participate in and regulate the body's digestive and metabolic activities. These activities are subjected to neural modulation, but characterization of the human pancreatic afferent and efferent nerves remains difficult because of the lack of three-dimensional (3-D) image data. Here we prepare transparent human donor pancreases for 3-D histology to reveal the pancreatic microstructure, vasculature, and innervation in a global and integrated fashion. The pancreatic neural network consists of the substance P (SP)-positive sensory (afferent) nerves, the vesicular acetylcholine transporter (VAChT)-positive parasympathetic (efferent) nerves, and the tyrosine hydroxylase (TH)-positive sympathetic (efferent) nerves. The SP+ afferent nerves were found residing along the basal domain of the interlobular ducts. The VAChT+ and TH+ efferent nerves were identified at the peri-acinar and perivascular spaces, which follow the blood vessels to the islets. In the intrapancreatic ganglia, the SP+ (scattered minority, ~7%) and VAChT+ neurons co-localize, suggesting a local afferent-efferent interaction. Compared with the mouse pancreas, the human pancreas differs in 1) the lack of SP+ afferent nerves in the islet, 2) the lower ganglionic density, and 3) the obvious presence of VAChT+ and TH+ nerves around the intralobular adipocytes. The latter implicates the neural influence on the pancreatic steatosis. Overall, our 3-D image data reveal the human pancreatic afferent and efferent innervation patterns and provide the anatomical foundation for future high-definition analyses of neural remodeling in human pancreatic diseases.NEW & NOTEWORTHY Modern three-dimensional (3-D) histology with multiplex optical signals identifies the afferent and efferent innervation patterns of human pancreas, which otherwise cannot be defined with standard histology. Our 3-D image data reveal the unexpected association of sensory and parasympathetic nerves/neurons in the intrapancreatic ganglia and identify the sympathetic and parasympathetic nerve contacts with the infiltrated adipocytes. The multiplex approach offers a new way to characterize the human pancreas in remodeling (e.g., fatty infiltration and duct lesion progression).

Renal metastasis from papillary thyroid microcarcinoma.

Papillary or follicular microcarcinoma of the thyroid comprises 10-20% of all thyroid malignancies. Most of these diseases exhibit slow progression and have a favorable prognosis. Distant metastasis caused by thyroid microcarcinoma is uncommon, and is usually found in the lung or bone. Thyroid microcarcinoma with metastasis to the kidney has not previously been reported. Clinically detectable well-differentiated metastatic thyroid carcinoma to the kidney is rare, and only 16 cases have been reported. Herein we describe a case of metastatic papillary thyroid microcarcinoma to the kidney in a patient with a pelvic fracture and pulmonary metastasis. The patient underwent a total thyroidectomy, radiotherapy, ablation treatment with 131I and thyroxine suppressive therapy. In conclusion, microcarcinoma cannot be ignored. Although uncommon, metastases may arise.