Published: February 12th, 2019
Here, we present a protocol for the preparation of agarose-filled human precision-cut lung slices from resected patient tissue that are suitable for generating 3D lung tissue cultures to model human lung diseases in biological and biomedical studies.
Translation of novel discoveries to human disease is limited by the availability of human tissue-based models of disease. Precision-cut lung slices (PCLS) used as 3D lung tissue cultures (3D-LTCs) represent an elegant and biologically highly relevant 3D cell culture model, which highly resemble in situ tissue due to their complexity, biomechanics and molecular composition. Tissue slicing is widely applied in various animal models. 3D-LTCs derived from human PCLS can be used to analyze responses to novel drugs, which might further help to better understand the mechanisms and functional effects of drugs in human tissue. The preparation of PCLS from surgically resected lung tissue samples of patients, who experienced lung lobectomy, increases the accessibility of diseased and peritumoral tissue. Here, we describe a detailed protocol for the generation of human PCLS from surgically resected soft-elastic patient lung tissue. Agarose was introduced into the bronchoalveolar space of the resectates, thus preserving lung structure and increasing the tissue's stiffness, which is crucial for subsequent slicing. 500 µm thick slices were prepared from the tissue block with a vibratome. Biopsy punches taken from PCLS ensure comparable tissue sample sizes and further increase the amount of tissue samples. The generated lung tissue cultures can be applied in a variety of studies in human lung biology, including the pathophysiology and mechanisms of different diseases, such as fibrotic processes at its best at (sub-)cellular levels. The highest benefit of the 3D-LTC ex vivo model is its close representation of the in situ human lung in respect of 3D tissue architecture, cell type diversity and lung anatomy as well as the potential for assessment of tissue from individual patients, which is relevant to further develop novel strategies for precision medicine.
Chronic and acute lung diseases are a major cause of morbidity and mortality worldwide1. For patients with chronic lung diseases like obstructive pulmonary disease (COPD)2, severe asthma3, lung cancer4 and diffuse parenchymal lung diseases5, curative therapies are currently not available. Although studies in animal models for lung diseases have deepened the understanding of disease pathomechanisms6 and have led to the identification of potential novel therapeutic targets7,8,9, these models exhibit relevant biological and physiological differences when compared to humans10. To overcome these discrepancies between murine and human biology as well as anatomy, human ex vivo 3D lung tissue culture (3D-LTC) systems are used in various areas of biomedical research. These 3D-LTC culture systems are based on precision-cut lung slices (PCLS). The generation of PCLS ex vivo allows analysis of a third spatial dimensionality, which allows for the investigation of the spatial and functional relationships of cells in entire alveoli and airways11, as well as the interstitium, vasculature and mesothelium. Notably, PCLS ex vivo models are multicellular, meaning that they contain most functional cells of in situ lungs, thus closely representing the cells' native biological environment and thus overcoming the limited cell-cell and cell-matrix interaction in most 2D cell culture approaches. Up to now, ex vivo murine PCLS were used to model pulmonary diseases, like COPD12, lung fibrosis13, lung cancer14, viral infection15,16, bronchopulmonary dysplasia17, and asthma18. However, a considerable proportion of novel drug therapies in human lung diseases that were investigated in clinical trials do not translate to the clinic due to their lack of efficacy or safety, assumingly due to yet considerable differences between human and murine biology and disease19,20,21.
Over several years, human PCLS have been largely used to assess lung toxicity of chemicals and drugs. Only recently, human lung tissue has been used from patients with COPD22,23, asthma24, and lung fibrosis25, to pursue pathophysiological and pharmacological studies. By using resected patient organ material and generating PCLS thereof, one can recapitulate major disease hallmarks in a complex 3D tissue environment22 representing and maintaining most of the native cellular diversity of the organ. Moreover, diseased tissue applied in a variety of experimental setups was shown to mimic disease-like changes in liver, intestine and kidney26,27,28,29.
However, processing of lung tissue remains challenging for several reasons. Unlike solid tissue, native lung parenchyma tends to collapse without ventilation and exhibits lower tissue stiffness. These properties impede the slicing of the tissue. Thus, filling of airways and the alveolar space with low-melting point agarose preserves the native lung structure and provides the stiffness needed for precision-cut slicing of murine and human lungs30. Human lung resectates donated for research purposes are by their nature anatomically, genetically and physiologically highly diverse, thus often presenting a high inter-patient variability when performing experiments25. In contrast to the whole lobe or whole lung explants, lung samples resected by means of thoracic surgery do not necessarily follow the anatomical segments and, therefore, require special preparation. In this article, we provide a detailed and optimized protocol for the generation of human PCLS from resected lung tissue and their subsequent cultivation and experimental use to model lung disease.
The use of human tissue was approved by the ethics committee of the Ludwig-Maximillian University [Munich, Germany (project number 455-12)]. Tumor-free human lung resections were provided by the Asklepios Biobank for Lung Diseases (Gauting, Germany, project number 333-10).
NOTE: All procedures of human PCLS production (Figure 1) are done under a sterile laminar flow hood.
1. Preparation of Instruments and Materials
2. Resected Lung Tissue
3. Inspection and Selection of Resected Tissue Prior to Agarose Filling
4. Lung Tissue Inflation by Agarose Filling
5. Precision-cut Lung Slicing
6. Generation of PCLS Punches
7. Tissue Culture and Sample Harvesting
The generation of PCLS can be separated into four essential steps: surgical lung tissue resection, agarose filling, vibratome-based PCLS generation, and culture of PCLS. The resected lung tissue is filled with low-melting point agarose, which adds the required stiffness to the lung tissue for slicing and preserves the native lung structure and architecture. Of note, PCLS generation is highly time consuming, thus often overnight storage of the filled lung tissue in DMEM F-12 medium can be included as an additional step and PCLS generation is started on the next day. Depending on the following experimental setup, generated PCLS can be incubated overnight in standard cell culture medium containing 0.1% (w/v) fetal bovine serum, before experimental conditions are applied. 3D-LTCs were viable and exhibited cellular functionality (such as surfactant protein secretion) for up to 120 h22 in the culture conditions outlined in this protocol (Figure 1) and might be optimized upon further improvement thereof.
For agarose filling of the tissue, a cannula of a peripheral venous catheter with a 1.3 mm diameter attached to the agarose-filled syringe was inserted into a bronchus at the surface of the cut tissue (Figure 2A). Bronchi are often localized near to a pulmonary artery. While the arteries have thinner walls and tend to collapse, bronchi exhibited a good visible lumen. Depending on the tissue's integrity, the catheter can be advanced through several generations of the respiratory tree into the periphery of the lung. The penetrated bronchus was sealed around the cannula by using tweezers (Figure 2B). The pulmonary artery can be clamped with the tweezers at the same time. Afterwards, the tissue is lifted up and liquid agarose is gently instilled into the airways.
Depending on the position of the catheter, a majority of the tissue can be filled with liquid agarose (Figure 2D). Optionally, cone like parts of the lung tissue, which reflect the lung's parenchyma ventilated by the penetrated bronchus, might get filled with the agarose (Figure 2C). In both scenarios, a characteristic pattern of solidly filled tissue regions can be observed: first, a major part of the tissue is filled in wedges (Figure 2D), or secondly, smaller protruding round areas of thoroughly filled tissue regions appear (Figure 2C). If parts of the airways obstruct due to agarose clots or other causes, parts of the tissue might not be properly filled with agarose. Thus, only parts of the tissue might be applicable for slicing. In case of leakages during the agarose filling procedure, parts of the filled respiratory tree might get perforated and filling of the lung tissue gets nearly impossible However, possible workarounds include the filling via a more peripheral bronchus, a deeper penetration of the cannula into the distal airways (Figure 2G), or potential clamping of the leakage area (Figure 2H).
Precision-cut lung slicing
Tissue blocks at a length and width of 1-1.5 cm were excised from tissue regions, which were completely filled with solidified agarose (Figure 3A-3B). Next, the individual tissue blocks were glued on the tissue holder of the vibratome (Figure 3C). 500 µm thick PCLS were generated, whereas the tissue block at the vibratome was advancing forward with speeds between 3-12 µm/s. (Figure 3D-3F). Finally, the PCLS were submerged in cell culture medium containing 0.1% (w/v) fetal bovine serum and cultured at standard cell culture conditions, as outlined step 7.
Experimental readouts of human 3D-LTC after 48h of culturing
A representative immunofluorescence staining, as previously described by Alsafadi et al.25, is shown in Figure 4A-4C. Immunolabeling of fibronectin (red) and cell nuclei (DAPI, blue), allowed for imaging of the preserved alveolar structure in the human 3D-LTC ex vivo. Treatment of the human PCLS punches with a profibrotic cytokine cocktail (including transforming growth factor beta 1, platelet derived growth factor AB, lipophosphatidyl acid, and tumor necrosis factor alpha) for 48 h resulted in fibrosis-like changes in human 3D-LTCs. By qPCR, a significant induction of the fibrosis-relevant extracellular matrix components collagen type 1 and fibronectin genes in 3D-LTC punches was observed upon treatment with the profibrotic cocktail (Figure 4D). Additionally, protein levels of the mesenchymal marker vimentin were found upregulated in 3 out of 4 patients after treatment of 3D-LTC punches (Figure 4E).
Figure 1: Workflow of PCLS generation. Tumor-free areas of lung resections are thoroughly inspected due to their tissue integrity. If the tissue is scored suitable for further use (scoring is explained in detail in the material and methods section), it is next filled with liquid agarose. Tissue blocks filled with solidified agarose are subsequently sliced with a vibratome. Submerged in cell culture medium, 3D-LTC are cultured up to 120 h after their generation. Downstream analyses of the 3D-LTCs involve protein- or RNA-expression, live-tissue fluorescence imaging, as well as immunofluorescence staining after fixation of the tissue. Please click here to view a larger version of this figure.
Figure 2: Filling the lung tissue with low-melting point agarose. The lung tissue is cannulated with a peripheral venous catheter which is inserted into a bronchus adjacent to the pulmonary artery (Figure 2A). Tweezers are used to fix the cannula in the bronchus and to clamp the pulmonary artery to avoid leaking of the liquid agarose. Liquid agarose at 42 °C is poured into the lung tissue with a 30 mL syringe (Figure 2B). A distal positioning of the cannula during filling will result in small areas of filled tissue (Figure 2C), while proximal positioning will ensure the filling of a larger tissue volume (Figure 2D). Any obstructions of the airways will reduce the amount of tissue volume that can be filled (Figure 2E). In case of agarose leaking, a distal cannula positioning and/or clamping of the leakage side enables proper agarose filling of the lung tissue (Figure 2F-2H). Please click here to view a larger version of this figure.
Figure 3: Precision-cut lung slicing. A successfully agarose-filled lung tissue is used to excise a piece of a tissue block (1 cm x 1.5 cm x 1 cm) with a scalpel (Figure 3A). Next the excised tissue block is glued to the tissue holder, scale bar indicates 1 cm (Figure 3B). Preferably, the tissue is glued with its pleural surface to the surface of the tissue holder as shown in Figure 3C. 500 µm thick slices are cut by the vibratome with a sapphire knife in a 10°-15° angle relative to the tissue (Figure 3D and 3E). The cutting procedure results in 2-3 cm3 large intact lung slices, Scale bar = 5 mm (Figure 3F). Additionally, by using a biopsy puncher, small reproducible punches with a diameter of 4 mm can be generated. Please click here to view a larger version of this figure.
Figure 4: Experimental readouts of human 3D-LTCs after 48h of culture. A human 3D-LTC punch of 4 mm diameter was immunostained for fibronectin (in red) and DAPI (in blue) (Figure 4A-4C). Scale bars = 1,000 µm. Figure 4C shows the merged image. RNA analysis of PCLS by quantitative RT-PCR shows significant increases of COL1A1 and FN1 gene expression by the profibrotic cocktail25. Figure 4E displays an immunoblot of whole protein lysates of PCLS, which were treated with a fibrotic cocktail25. Probing for Vimentin and β-Actin demonstrated an increased protein expression of the mesenchymal marker (vimentin) after treatment with profibrotic factors in patient samples 1, 3, and 4. Please click here to view a larger version of this figure.
|The tissue sample has intact pleural surface.
|The tissue sample seems macroscopically intact, lacking incisions, squashing, ruptures and distortions.
|The tissue sample contains at least one bronchus with a diameter >1mm.
|The tissue sample contains no or only little amounts of blood.
|The tissue sample was stored fully in medium and shows no obvious sign of atelectasis.
|The tissue sample was resected within the last four hours.
|The tissue sample is larger than 5cm in its largest diameter.
|Score in sum:
Table 1: Lung Agarose Filling Score. The Lung Agarose Filling Score (LAFS) correlates with the success-rate to fill a tissue resection with agarose for its subsequent vibratome-based PCLS production. The score sums up all points of criteria met by the tissue. An LAFS equal or above 72 predicts good agarose filling properties, a score below 60 predicts a highly probable failure of agarose filling of the tissue.
The protocol described in this manuscript covers the generation of PCLS from human lung tissue resectates by filling it with liquid agarose and subsequent vibratome slicing. Generation of tissue slices was demonstrated before for a couple of organs, like liver and brain, whereas the inherent stiffness of these organs allowed direct slicing without any modification of the tissue. Of note, the initial proper preparation of the lung tissue is the most crucial step in generating PCLS. Agarose filling of the lung is the method of choice to stabilize its soft and elastic nature, and to ensure a homogenous and reproducible PCLS generation. Large airways of the resected lung tissue are cannulated to provide access to the small airways, as well as to the intact lung parenchyma. The lack of an intact pleura, which makes agarose filling nearly impossible, is a major reason why lung tissue is mostly not usable for lung slicing. Prospectively, a synthetic pleura originally designed to perform functional experiments on decellularized scaffolds could potentially be applied to achieve successful agarose filling of explants that lack an intact pleura31. Resections resulting in a human lung tissue piece with intact pleura are essential for generating tissue blocks for slicing. Resected tissue is more available due to tumor-free tissue from cancer resections than fully intact lobes or whole-lung explants of patients who underwent lung transplantation.
Commonly, two systems are used to produce PCLS: the Krumdieck tissue slicer15 and vibratory microtomes (vibratomes). Tissue slicers generate slices by passing a tissue block through a metal vessel, which cuts the PCLS at 90° at the end of this vessel. Vibratomes generate PCLS by moving a vibrating knife horizontally over an anchored block of tissue that is submerged in a cooled medium bath, which compared to the Krumdieck slicer exerts less shear force on the tissue. This results in less harsh treatment of the tissue before cultivation. On the other hand, the vibratome cutting is more time and work consuming. In our hands, vibratome slicing enabled the production of a maximum of 100 PCLS or 500 PCLS punches in one day, sufficient for most experimental studies. PCLS can be cultured in various ways: (a) attached to Trans-wells, thus generating an air liquid interface (ALI) system, (b) as dynamic organ culture (DOC), or (c) submerged in cell culture medium at standard cell culture conditions. The in-detail cultivation of PCLS was previously described22,23,25; however, a common standard of cultivation conditions between their usage in various labs around the world is still missing. In particular, the culture time might be critical: as in murine PCLS, a loss of SFTPC positive alveolar type 2 cells is observed after 144 h, but not after 120 h22. In addition, metabolic activity seems to remain stable in murine22 and human PCLS25 for 120 h.
There are a couple of technical limitations for the generation of PCLS: the number and size of the resectates fluctuates over time; the efficiency of the agarose filling, which depends on the presence of intact pleura within the obtained tissue, determines the final success of the PCLS generation; and tissue destruction caused by pathological changes within the obtained (diseased) lung tissue might interfere with the PCLS preparation. Airway obstructions and fibrotic tissue lacking intact alveolar space impede with agarose filling and thus make slicing of the fibrotic tissue a demanding task. Emphysematous tissues as found in diseases such as COPD or alpha-1-anti-trypsin deficiency might not withstand the pressure of agarose filling, and will result in rupture of the alveoli and architectural artefacts. In these cases, the usage of low agarose concentration, e.g., 1% (w/v), might be useful to decrease pressure and speed during agarose filling. Overall, the disease state of the tissue can dramatically limit the use of the tissue for PCLS generation. All these parameters determine the amount of PCLS that can be generated from lung tissue, and also the amount of time it takes to produce the PCLS. Further limitations of PCLS are inconsistencies in between different lung slices with respect to size or tissue content, which requires further normalization steps for experiments. To overcome this, biopsy punches of similar regions of the same slice can be generated. This procedure is apt to reduce the tissue variability and, as an additional benefit, increase the number of PCLS samples that can be used for experiments.
In conclusion, human 3D lung tissue cultures from agarose filled PCLS provide a complex human model for studying lung physiology and diseases. The protocol provides a detailed description of the preparation of PCLS from resected lung tissue and their cultivation, and moreover addresses challenges in agarose filling of human lung resections and how to overcome them.
All authors declare no competing financial interests.
The authors are grateful to Marisa Neumann for expert technical assistance. All lung tissues were kindly provided by the CPC-M Bio-Archive. This work was supported by the German Center of Lung Research (DZL), the Helmholtz Association and CPC Research School grants.
|Vibratome Hyrax V50
|Hyrax CU 65
|Vasofix Braunüle 18G
|B. Braun Melsungen AG
|Henke Sass Wolf
|Guarded disposable scalpels, sterile
|Synthetic Single Crystal Sapphire
|Delaware Diamond Knives
|Dulbecco's Modified Eagle Medium F-12 Nutreient Mixture (Ham) + L-Glutamine + 15mM HEPES
|Gibco by Life Technologies
|Special process fetal bovine serum (Sera Plus)
|Disposable Biopsy Punch
|96 Well, Black/Clear, Tissue Culture Treated Plate, Flat Bottom with Lid, sterile
|Falcon / Corning
|Agarose, low geling temperature
Copyright © 2024 MyJoVE Corporation. All rights reserved