Summary
The present protocol optimizes the liver in situ perfusion/decellularization and two-photon microscopy methods to establish a reliable platform to visualize the dynamics of extracellular matrix (ECM) remodeling during non-alcoholic steatohepatitis (NASH).
Abstract
Non-alcoholic steatohepatitis (NASH) is the most common chronic liver disease in the United States, affecting more than 70 million Americans. NASH can progress to fibrosis and eventually to cirrhosis, a significant risk factor for hepatocellular carcinoma. The extracellular matrix (ECM) provides structural support and maintains liver homeostasis via matricellular signals. Liver fibrosis results from an imbalance in the dynamic ECM remodeling process and is characterized by excessive accumulation of structural elements and associated changes in glycosaminoglycans. The typical fibrosis pattern of NASH is called "chicken wire," which usually consists of zone 3 perisinusoidal/pericellular fibrosis, based on features observed by Masson's trichrome stain and Picrosirius Red stains. However, these traditional thin two-dimensional (2D) tissue slide-based imaging techniques cannot demonstrate the detailed three-dimensional (3D) ECM structural changes, limiting the understanding of the dynamic ECM remodeling in liver fibrosis.
The current work optimized a fast and efficient protocol to image the native ECM structure in the liver via decellularization to address the above challenges. Mice were fed either with chow or fast-food diet for 14 weeks. Decellularization was performed after in situ portal vein perfusion, and the two-photon microscopy techniques were applied to image and analyze changes in the native ECM. The 3D images of the normal and NASH livers were reconstituted and analyzed. Performing in situ perfusion decellularization and analyzing the scaffold by two-photon microscopy provided a practical and reliable platform to visualize the dynamic ECM remodeling in the liver.
Introduction
Non-alcoholic fatty liver disease (NAFLD) is the most common liver disease, affecting 20%-25% of the adult population. 25% of NAFLD patients progress to non-alcoholic steatohepatitis (NASH), where the risk of cirrhosis, liver failure, and hepatocellular carcinoma increases1. In the next 20 years, it is estimated that NASH will account for 2 million liver-related deaths in U.S2. As there are no approved treatments, there is an urgent need to decipher the mechanisms that cause liver fibrosis in NASH patients and develop targeted treatment3.
The extracellular matrix (ECM) is a dynamic, complex microenvironment that exerts bi-directional communication with cells to regulate tissue homeostasis4. The liver ECM is composed of structural elements such as proteoglycans, collagens, fibronectin, elastin, and other non-structural proteins (e.g., olfactomedin and thrombospondin) to provide physical and structural support4.
Liver fibrosis is a chronic wound-healing response to liver damage of various etiologies, including NASH3. It results from an imbalance in the dynamic ECM matrix remodeling process and is characterized by excessive structural proteins in the injured liver4. Fibrogenesis depends on the dynamic cell-cell communication among different hepatic cell types. Hepatic stellate cells (HSCs), when activated, differentiate into Smooth Muscle Alpha 2 Actin-expressing, migrating, and proliferating myofibroblast-like cells and synthesize ECM proteins as a wound-closing action. Activated HSCs are the central collagen-producing cells in the liver1.
The molecular mechanism of ECM remodeling, patterns of fibrosis, and their relationship with cellular events are not clear. A better understanding of the three-dimensional (3D) ECM structure is still needed, even though mass spectrometry techniques have helped analyze ECM protein composition4. Traditionally, Masson's trichrome stain, Picro Sirius Red stains, and second harmonic generation (SHG) imaging have been performed on two-dimensional (2D) thin liver sections. The typical fibrosis pattern of NASH is called "chicken wire," which extends to zone 3 and is perisinusoidal/pericellular fibrosis5,6. However, there has been a lack of studies focusing on the 3D structure of the native liver, particularly those that do not involve tissue sectioning. Robust imaging approaches to identify patterns and characteristics of fibrosis throughout dynamic ECM remodeling in liver fibrosis would significantly strengthen the understanding of NASH mechanisms and identify new therapeutic targets.
To address these challenges, a fast and efficient protocol was optimized to image the native liver ECM via decellularization7. Whole-liver decellularization is an approach to remove the hepatic cellular content while maintaining the native 3D ECM network through detergent perfusion. Mice were fed either chow or fast-food diet (FFD) for 14 weeks. Decellularization was performed after in situ portal vein perfusion with mild detergent and low flow rates to preserve triple-helical and native fibrillar collagen structures. Two-photon microscopy was applied to analyze changes in collagen structures in ECM. The 3D images of the native ECM structure in normal and NASH livers were reconstituted and analyzed. Performing in situ perfusion decellularization and analyzing the scaffold by two-photon microscopy provides a practical and affordable platform to visualize the dynamic ECM remodeling in the liver.
Subscription Required. Please recommend JoVE to your librarian.
Protocol
Animal experiments are performed according to the experimental procedures approved by the institutional animal care and use committees (IACUCs) of Stanford University and the Veterans Affairs Hospital in Palo Alto. 6-8 week-old male C57BL/6J mice were fed either chow or a fast-food diet supplemented with 4.2% high-fructose corn syrup (see Table of Materials) in drinking water for 14 weeks5. The mice were kept in standard cages at a 12 h dark/light cycle.
1. Surgical preparation and procedures of liver perfusion
- Weigh the mouse before the procedure.
- Anesthetize the mouse by administering Ketamine (90 mg/kg) and Xylazine (10 mg/kg) (see Table of Materials) via an intraperitoneal injection. Ensure that the mouse is fully anesthetized by toe pinching before proceeding to the next step.
NOTE: Isoflurane is not needed here since this is a non-survival surgery. - Shave the ventral area using a hair clipper and disinfect the skin with 70% ethanol. Place the mouse on its back, and immobilize the legs on the surgical board with tape.
- After reconfirming the depth of anesthesia by a toe-pinch, use Mayo scissors and Adson forceps (see Table of Materials) to make a 3 cm incision laterally in the lower abdomen and then a 5 cm incision vertically from the low abdomen to the xiphoid process. Be careful not to open the chest cavity.
- After opening the peritoneum, gently place the intestines to the animal's left, and elevate the right lobe of the liver with a Rayon tipped applicator (see Table of Materials) to expose the portal vein.
- Catheterize the portal vein using a 24 G IV catheter. Introduce the catheter into the distal end of the portal vein. Place the catheter tip before the portal vein branches out into the hepatic lobes.
- Withdraw the needle immediately when the catheter enters the vein. Check that the catheter is correctly positioned inside the vein by perfusing the liver with PBS using a 1mL syringe via the catheter.
NOTE: Upon PBS perfusion, all lobes should immediately blanch. If the catheter is correctly placed, venous blood will quickly flow through the catheter. - Use Dumont micro forceps, a Castroviejo microneedle holder, and 4-0 suture (see Table of Materials) to place a stitch below the branching of the portal vein and a second stitch 1 cm below to secure the catheter.
2. Tissue decellularization
- Connect the catheter to a silicone tubing (~1 m length, 3 mm inner diameter, and 4.1 mm outer diameter) with a Luer connector. Connect the silicone tubing to a peristaltic pump and a reservoir containing deionized water (see Table of Materials).
- Carefully remove the air bubbles inside the tube and avoid generating new bubbles during the perfusion. Set the peristaltic pump at a flow output of 0.2 mL/min (i.e., 288 mL/24 h).
- Perfuse first with deionized water (~50 mL/mouse) for 2 h. The color of the liver will change from red to yellow during perfusion.
NOTE: The animal will expire after 10 minutes of perfusion. - Switch the perfusion solution to 0.5% (wt/vol) sodium deoxycholate (DOC, see Table of Materials) and continue overnight (18 h, ~250mL/mouse).The liver will become white at the end of perfusion (Figure 1).
- Switch the perfusion solution to deionized water (~50 mL/mouse) and perfuse for 2 h.
- Use Dumont micro forceps and micro-spring scissors (see Table of Materials) to collect the decellularized liver and wash it carefully in a Petri-dish with PBS.
- Cut the decellularized tissue into small pieces for immediate imaging or freeze at -80 °C for further biochemical analysis such as tandem mass spectrometry, ELISA, or western blot.
3. Slide preparation for imaging (with a multiphoton/confocal fluorescence microscope)
- Transfer a small piece of decellularized liver (<3 x 3 x 3 mm) to a microscope slide and place the tissue in the middle.
- Add 10 µL of antifade mounting medium (see Table of Materials) to the tissue with a pipette to avoid tissue drying.
- Cover the tissue with a cover glass and apply a small force to flatten the sample. Seal the edges of the cover glass with colorless and transparent nail polish (Figure 2).
NOTE: Due to the setting of the upright two-photon through the microscope, a cover glass was used on the sample. However, samples could be placed differently if an inverted microscope is used.
4. Image acquisition
- Place a drop of immersion oil on the top of the cover glass and place the slide on the microscope slide holder. Lower the 20x oil objective lens until it contacts the immersion oil.
- Switch to the violet (405 nm) channel, turn on the shutter, and focus on the sample. Navigate and position the area of interest for imaging. Turn off the shutter before moving to image acquisition using laser light.
- Switch to the computer control and start the two-photon laser. Turn on the two-photon laser first (Figure 3A), then turn on the two-photon laser controller (see Table of Materials) and shutter (Figure 3B,C). Ensure that the output power is higher than 2.5 W.
Reduce the laser power to prevent fluorescence quenching. Select and adjust detectors (both photomultipliers and the hybrid) to accommodate the chosen fluorophores.
NOTE: Two-photon imaging was performed at an excitation wavelength of 860 nm. Two channels were selected for collagen SHG and two-photon excited fluorescence (TPEF) images, respectively. One channel corresponding to the wavelength range of 415-445 nm detected the detailed structure of collagen from SHG signals. Another channel covered the range of 465-669 nm to collect TPEF signals. - Choose simultaneous or sequential image acquisitions. Adjust the pinhole to the highest value. Adjust laser wavelengths, gain, power, and offset. Adjust pixel dwell time, the pixel size, averaging, and the zoom (Figure 4A).
- Scroll the computer z-controller and set the z-dimensions, define the start and endpoints, and choose the numbers of images given within a volume (z-stack). Acquire images.
NOTE: Here, a 20 µm Z volume and 1 µm Z-step size was chosen (Figure 4B).
Subscription Required. Please recommend JoVE to your librarian.
Representative Results
The collagen fibers were detected with second harmonic generation and two-photon microscopy. The signal is from the frangible triple-helical and native fibrillar collagen structures. Specific antibodies were not used to analyze collagen subtypes; however, this could be added to the imaging technique.
When the liver tissue is studied without decellularization, it is challenging to get high-resolution images of the collagen network (Figure 5A). In decellularized ECM, collagen fibers (green) appear parallel or interweaving (Figure 5B). There were apparent differences in the morphology and spatial distribution of the ECM components between Chow and FFD livers. Collagen fibers in mice on a chow diet exhibited an interweaving and well-organized network. In mice on FFD, however, collagen bundles were observed with less connectivity (Figure 5B).
To further study the 3D structure of liver ECM, images of the slices were captured with a 20 µm of Z-volume (thickness) and 1 µm of Z-step size. Collagen fibers in mice on chow diet showed a well-organized network but not in the mice on FFD (Figure 6). To confirm the quality of the fibrillar collagen in the decellularized ECM, the slides were fixed and embedded in paraffin and stained with Hematoxylin and eosin (H&E) and Picrosirius red (PSR) (see Table of Materials). H&E shows that all the cells have been removed. The PSR is a commonly used histological technique to highlight collagen in paraffin-embedded tissue sections (Figure 7).
Figure 1: Mouse liver during perfusion. The liver becomes white and semi-transparent at the end of the perfusion. Please click here to view a larger version of this figure.
Figure 2: Slide preparation for imaging. The tissue is covered with a cover glass, and a small force is applied to flatten the sample. The edges of the cover glass are sealed with colorless and transparent nail polish. Please click here to view a larger version of this figure.
Figure 3: Screenshot of the two-photon laser software, starting the two-photon laser. (A) First, the two-photon laser is turned on. (B)(C) The two-photon laser controller and the shutter are turned on. Please click here to view a larger version of this figure.
Figure 4: Screenshot of the two-photon laser software during image acquisition. (A) Ensure to set the pinhole to the highest value (red arrows). Adjust laser wavelengths, gain, power, and offset. Adjust pixel dwell time, the pixel size, averaging, and the zoom. (B) Set the Z volume to 20 µm and Z-step size to 1 µm. Please click here to view a larger version of this figure.
Figure 5: Decellularized ECM imaged with two-photon microscopy. (A) SHG images of collagen fibers from a 6 µm liver slice (fresh frozen) without decellularization. (B) SHG images depict a well-organized collagen network in livers of mice on normal control, chow diet. Remodeled collagen network in mice on fast-food diet (FFD). Arrow, fibrosis area. Star, remodeled collagen network. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 6: 3D structure of liver collagen network. The collagen fibers in mice on the chow diet showed a well-organized network but not in fast-food diet (FFD) mice. The images were captured with a 20 µm Z-volume (thickness) and 1 µm Z-step size. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 7: Hematoxylin and eosin staining (H&E) and Picrosirius red (PSR) staining of the decellularized ECM. H&E shows that all the cells have been removed. In the PSR staining images, the ECM from the chow-fed mice revealed a well-organized network, and in contrast, mice on fast-food diet (FFD) showed denser collagen fibers. Z-volume (thickness) = 6 µm. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Subscription Required. Please recommend JoVE to your librarian.
Discussion
The present protocol shows that decellularization through a low flow rate DOC in situ perfusion preserves the frangible triple-helical and native fibrillar collagen structures, providing a reliable and cost-effective platform to capture dynamic ECM remodeling in NASH liver fibrosis. Although decellularization was performed in normal and fibrotic livers before identifying ECM components or generating biological scaffolds for cell culture, the dynamics of ECM remodeling in liver fibrosis have not been well studied8,9,10. This method allows investigators to draw comparisons across various fibrosis models.
Currently, there are varieties of perfusion protocols using DNase, sodium dodecyl sulfate (SDS), or Triton X-10011. The cost of DNase is high because large volumes of solutions are required. SDS is a harsh denaturing solution, an anionic detergent that disrupts the cellular membranes and denatures proteins. Therefore, preserving the entire collagen network after SDS perfusion is challenging. In contrast, Triton X-100, as a mild non-denaturing, non-ionic detergent, disrupts lipid-lipid and lipid-protein interactions and preserves protein-protein interactions. However, it has been reported that Triton X-100 perturbs the ECM scaffolds, making the decellularized liver ECM unsuitable for further mass spectrometry analysis11.
The perfusion protocol with DOC was modified from the previous work7, demonstrating better preservation of the triple-helical and native fibrillar collagen structures in most tissues. The collagen network was studied with second harmonic generation (SHG) microscopy without antibodies, reflecting the triple-helical and native fibrillar collagen structures. The decellularized liver ECM from FFD mice was also tested, and structural changes in NASH were confirmed.
The most critical steps in the protocol are catheter placement and setting up the perfusion system. The portal vein catheterization is vital, but the inferior vena cava could also be used if the portal vein is damaged. Bubbles in the buffer and tubes must be avoided because they affect perfusion. Because an upright two-photon microscope was used, the sample size was limited. This can be modified if an inverted microscope is to be used.
There is an increasing interest in the remodeled ECM in chronic liver diseases and hepatocarcinogenesis12,13,14. Decellularized ECM scaffolds have been employed in surgical repair and regenerative medicine as an appropriate biological scaffold material15,16,17. However, this mild decellularization process also has limitations, as some cellular components such as DNA or lipid remnants may not be entirely removed. Longer DOC perfusion time and an increase in flow rate may help get ultra-purified decellularized scaffolds. Adding DNase into the washing solution after the DOC perfusion may also help remove DNA. The protocol can be adjusted for human liver biopsies, thus enhancing the understanding of ECM alterations during NASH progression.
Subscription Required. Please recommend JoVE to your librarian.
Disclosures
The authors declare that there are no conflicts of interest regarding this article.
Acknowledgments
We thank Hyesuk Park for the technical help. This research was supported by funding from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), NIH (R01 2DK083283, to NJT), the National Institute on Aging (NIA), NIH (1R01AG060726, to NJT). We gratefully acknowledge Jon Mulholland and Kitty Lee of the Cell Sciences Imaging Facility in the Beckman Center for technical assistance with the two-photon microscopy imaging.
Materials
| Name | Company | Catalog Number | Comments |
| 4-0 MONOCRYL UNDYED 1 x 18" P-3 | MONOCRYL | Y494G | |
| 4-0 suture | fisher scientific | 10-000-649 | https://www.fishersci.com/shop/products/monomid-nylon-non-absorbable-sutures-7/10000649?keyword=true |
| AnaSed Injection (xylazine) | AnaSed | NDC 59399-110-20 | this drug to use by or on the order of a licensed veterinarian. |
| BD INSYTE AUTOGUARD I.V. CATHETER WITH BC TECHNOLOGY | BD | 382612 | |
| Chow diet | Envigo | # 2918 | Control diet. A fixed formula, non-autoclavable diet manufactured with high quality ingredients and designed to support gestation, lactation, and growth of rodents. |
| Fast-food diet (AIN76A Western Diet) | Test Diet | 1810060 | https://www.testdiet.com/cs/groups/lolweb/@testdiet/documents/web_content/mdrf/mdux/~edisp/ducm04_051601.pdf |
| Hematoxylin and Eosin Stain Kit | vectorlabs | H-3502 | https://vectorlabs.com/hematoxylin-and-eosin-stain-kit.html |
| Kent Scientific Rat Surgical Kit | fisher scientific | 13-005-205 | https://www.fishersci.com/shop/products/rat-surgical-kit/13005205#?keyword=mouse%20surgery%20kit |
| KETAMINE HYDROCHLORIDE INJECTION | Vedco | NDC 50989-996-06 - 10 mL - vial. | KetaVed has been clinically studied in subhuman primates in addition to those species listed under Administration and Dosage. |
| Leica SP5 upright Confocal, multi-photon | Leica | SP5 | |
| Luer connector (Three-way stopcock with SPIN-LOCK®) | bbraun | D300 | https://www.bbraunusa.com/en/products/b0/three-way-stopcockwithspin-lock.html |
| Picrosirius Red Stain Kit | Polysciences, Inc. | 24901 | https://www.polysciences.com/default/picrosirius-red-stain-kit-40771 |
| Rayon tipped applicator | puritan | 25-806 1PR | |
| Sodium deoxycholate | sigmaaldrich | D6750-100G | |
| Syrup | www.target.com | 24 fl oz | https://www.target.com/p/pancake-syrup-24-fl-oz-market-pantry-8482/-/A-13007801 |
| Variable Speed Peristaltic Pump | INTLLAB | BT100 | https://www.amazon.com/gp/product/B082K97W5W/ref=ox_sc_saved_title_2?smid=A12NUUP87ZRRAR&psc=1 |
| VECTASHIELD Antifade Mounting Medium | vectorlabs | H-1000-10 | https://vectorlabs.com/vectashield-mounting-medium.html |
References
- Friedman, S. L., Pinzani, M. Hepatic fibrosis: 2022 unmet needs and a blueprint for the future. Hepatology. 75 (2), 473-488 (2021).
- Ye, Q., et al. Global prevalence, incidence, and outcomes of non-obese or lean non-alcoholic fatty liver disease: A systematic review and meta-analysis. The Lancet Gastroenterology and Hepatology. 5 (8), 739-752 (2020).
- Schwabe, R. F., Tabas, I., Pajvani, U. B. Mechanisms of fibrosis development in non-alcoholic steatohepatitis. Gastroenterology. 158 (7), 1913-1928 (2020).
- Arteel, G. E., Naba, A. The liver matrisome - looking beyond collagens. JHEP Reports. 2 (4), 100115 (2020).
- Jiang, J. X., et al. Nonphagocytic activation of NOX2 is implicated in progressive non-alcoholic steatohepatitis during aging. Hepatology. 72 (4), 1204-1218 (2020).
- Dehnad, A., et al. AGER1 downregulation associates with fibrosis in non-alcoholic steatohepatitis and type 2 diabetes. Journal of Clinical Investigation. 130 (8), 4320-4330 (2020).
- Mayorca-Guiliani, A. E., et al. Decellularization and antibody staining of mouse tissues to map native extracellular matrix structures in 3D. Nature Protocols. 14 (12), 3395-3425 (2019).
- Mazza, G., et al. Cirrhotic human liver extracellular matrix 3D scaffolds promote smad-dependent tgf-beta1 epithelial mesenchymal transition. Cells. 9 (1), 83 (2019).
- Klaas, M., et al. The alterations in the extracellular matrix composition guide the repair of damaged liver tissue. Scientific Reports. 6, 27398 (2016).
- Mattei, G., et al. Mechanostructure and composition of highly reproducible decellularized liver matrices. Acta Biomaterialia. 10 (2), 875-882 (2014).
- Ren, H., et al. Evaluation of two decellularization methods in the development of a whole-organ decellularized rat liver scaffold. Liver International. 33 (3), 448-458 (2013).
- Piersma, B., Hayward, M. K., Weaver, V. M. Fibrosis and cancer: A strained relationship. Biochimica et Biophysica Acta - Reviews on Cancer. 1873 (2), 188356 (2020).
- Cox, T. R.
- Mirdamadi, E. S., Kalhori, D., Zakeri, N., Azarpira, N., Solati-Hashjin, M. Liver tissue engineering as an emerging alternative for liver disease treatment. Tissue Engineering Part B: Reviews. 26 (2), 145-163 (2020).
- Mazza, G., et al. Decellularized human liver as a natural 3D-scaffold for liver bioengineering and transplantation. Scientific Reports. 5, 13079 (2015).
- Jia, Z., et al. 3D culture system for liver tissue mimicking hepatic plates for improvement of human hepatocyte (C3A) function and polarity. BioMed Research International. 2020, 6354183 (2020).
- Shimoda, H., et al. Decellularized liver scaffolds promote liver regeneration after partial hepatectomy. Scientific Reports. 9 (1), 12543 (2019).
