We describe a method for generating Precision-cut Lung Slices (PCLS) and immunostaining them to visualize the localization of various immune cell types in the lung. Our protocol can be extended to visualize the location and function of many different cell types under a variety of conditions.
Inhalation of allergens and pathogens elicits multiple changes in a variety of immune cell types in the lung. Flow cytometry is a powerful technique for quantitative analysis of cell surface proteins on immune cells, but it provides no information on the localization and migration patterns of these cells within the lung. Similarly, chemotaxis assays can be performed to study the potential of cells to respond to chemotactic factors in vitro, but these assays do not reproduce the complex environment of the intact lung. In contrast to these aforementioned techniques, the location of individual cell types within the lung can be readily visualized by generating Precision-cut Lung Slices (PCLS), staining them with commercially available, fluorescently tagged antibodies, and visualizing the sections by confocal microscopy. PCLS can be used for both live and fixed lung tissue, and the slices can encompass areas as large as a cross section of an entire lobe. We have used this protocol to successfully visualize the location of a wide variety of cell types in the lung, including distinct types of dendritic cells, macrophages, neutrophils, T cells and B cells, as well as structural cells such as lymphatic, endothelial, and epithelial cells. The ability to visualize cellular interactions, such as those between dendritic cells and T cells, in live, three-dimensional lung tissue, can reveal how cells move within the lung and interact with one another at steady state and during inflammation. Thus, when used in combination with other procedures, such as flow cytometry and quantitative PCR, PCLS can contribute to a comprehensive understanding of cellular events that underlie allergic and inflammatory diseases of the lung.
Following inhalation of pro-inflammatory stimuli such as lipopolysaccharide (LPS), there is a coordinated movement of immune cells into, within, and from the lung. For example, neutrophils are rapidly recruited to the lung parenchyma and airway. In addition, some professional antigen presenting cells known as conventional dendritic cells (cDCs) undergo a relatively complex migration pattern1,2. cDCs can be identified using flow cytometry, based in part on their display of the surface marker, CD11c. Distinct subsets of DCs can be distinguished by the differential surface expression of CD103 and CD11b3. Upon acquiring inhaled antigen, some cDCs exit the lung and migrate through the lymphatic vessels to lung-draining Lymph Nodes (LNs) where they present peptides to antigen-specific T cells4. This is a critical early event in the initiation of adaptive immune responses. For unknown reasons, however, not all cDCs that acquire inhaled antigens leave the lung, and many of these cells remain in that organ for several months5,6. This observation can be partly explained by the developmental ancestry of these cells because monocyte-derived CD11c+ cells lacking the chemokine receptor, CCR7, are unable to migrate to regional LNs7,8. It seems likely that the migration potential of cDCs is also determined, at least in part, by their anatomical position within the lung. However, the precise localization of these different populations of cDCs in the lung is not fully characterized. An improved knowledge of immune cell localization within the lung, and of the molecules that direct it, is needed for a better understanding of how the immune system of the lung becomes activated.
PCLS are being increasingly used as an ex vivo approach to visualize cellular positioning and cell-cell interactions, while maintaining the structural integrity of the lung architecture9,10. PCLS have been used to study lungs of many species, including mice, cattle, monkeys, sheep, horses, and humans11. A major advantage of this technique is that approximately 20 slices can be prepared from a single lobe of a mouse lung, thereby reducing the number of animals needed for individual experiments. Virtually all immune cell types, including DCs, macrophages, neutrophils, and T cells, are present in PCLS and maintain their normal structures.
PCLS can also be used to study calcium signaling and contractility of airway and smooth muscle cells after treatment with acetylcholine12 or methacholine13. In this approach, only a small portion of the lung is analyzed microscopically, but one study reported that measurements of airway contraction in PCLS vary only about 10% from slice to slice, and this variance is comparable to that seen using lung function tests in intact animals14. Other investigators have used PCLS as an ex vivo approach to study changes in cytokine expression and cell surface markers after incubation with LPS15. PCLS have also been used in an ex vivo model of hypoxic pulmonary vasoconstriction in small intra-acinar arteries. These vessels are located in the part of the lung that cannot be reached using other procedures, including recordings from dissected arterial segments or analysis of subpleural vessels16. Our lab has primarily used PCLS to visualize immune cell localization in live lung tissue at steady state and following an in vivo inflammatory stimulus. The procedures we have developed for this are as follows.
Animal experimental procedures described in this paper were approved by the NIEHS Animal Care and Use Committee (IACUC).
1. Lung Preparation
2. Lung Slicing
NOTE: Make slices using an automated slicer, metal cooling block, plunger and metal syringe, according to the manufacturer's instructions.
3. Antibody Staining
4. Static Imaging of PCLS
5. Live Cell Imaging of PCLS
To identify the location of two DC subsets, CD11bhi cDCs and CD103+ cDCs, PCLS from C57BL/6 mice were cut and stained with monoclonal antibodies (mAbs) specific to CD11c, CD88, CD103, and CD324 (E-cadherin). Antibodies to CD324 stain airway epithelial cells, and CD88 is displayed on macrophages and neutrophils, but not cDCs8. This allowed us to distinguish cDCs from CD11c+ macrophages, and to observe the spatial relationship of each cell type to the airways (Figure 1). We found that CD11bhi cDCs localize in the parenchyma, whereas CD103+ cDCs reside primarily around the airways and the subpleural area. Although CD103+ cDCs were readily detected by direct staining of cell surface CD103 in this experiment, the detection of CD11bhi cDCs relied on the absence of CD88 and CD103 staining. To directly identify CD11bhi cDCs, we first attempted to stain PCLS using an anti-CD11b mAb (clone M1/70), which is widely used in immunohistochemistry and flow cytometry. However, this mAb had a high background and low specificity in this application, regardless of which fluorochromes were conjugated to the mAb (Figure 2A, and data not shown). By contrast, antibodies to CD172a (SIRP1α) stained CD11bhi cDCs18, but not structural cells or CD103+ cDCs (Figure 2B, and data not shown). On their own, antibodies to CD172a could not discriminate between cDCs and monocytes or macrophages. However, by co-staining with antibodies against CD88 and CD172a, we were able to distinguish alveolar macrophages (CD172a–CD88+) from interstitial macrophages (CD172a+CD88+), and CD11bhi cDCs (CD172a+CD88–), and show that the latter cells preferentially localize in the parenchyma, not the sub-epithelial area (Figure 2C – F), in agreement with our results obtained by indirect staining of these cells.
Study of intra-tissue migration of leukocytes requires visualization of structural cells, including those of the lymphatics, through which DCs traffic from peripheral tissues to regional LNs. In the skin and LNs, lymphatic vessels are often identified using mAbs to LYVE-1 or Podoplanin. However, we found that in PCLS, mAbs to LYVE-1 and Podoplanin stain a variety of other cell types, including vascular endothelium and alveolar epithelium (data not shown). These mAbs are therefore of limited practical use for specifically identifying lymphatic vessels in PCLS. However, mAbs to CD90.2 (Thy1.2), a well-known T cell marker, labeled lymphatic structures, as shown previously in other tissues19,20 (Figures 3A, B), as did TdTomato fluorescent protein encoded by a transgene under transcriptional control of the Prox1 promoter21 (Figure 3A, C). PROX1 is a master transcription factor that is necessary for lymphangiogenesis22. Prox1TdTomato transgenic mice exhibit brightly fluorescent lymphatic vessels in the lungs and other tissues such as the liver, lens, dentate gyrus and neuroendocrine cells of the adrenal medulla21. Indeed, these two labeling methods gave overlapping staining patterns (Figure 3D). Using these labeling approaches, we were able to determine that CD103+ cDCs were present in several anatomical areas, including CD324-expressing airway epithelium, the lymphatics, and the subpleural area (Figure 3A, E).
In addition to static imaging, cell movement and cell to cell interactions can be recorded using live cell imaging of PCLS (Movie 1). PCLS were freshly prepared from mice that had received injections of GFP-expressing OT-II T cells and instillation of OVA and LPS, stained with mAbs against CD103 (red) and the airway epithelium (CD324, blue), and were imaged to track the movement of CD103+ cDCs and T cells 16 h after OVA/LPS instillation. Over the course of the video (4 h), the dynamic movement of CD103+ DCs (red) and the adoptively transferred OVA-specific T cells (green) and their interaction are clearly visible (Movie 1).
Figure 1. Distinct Anatomic Locations of cDC Subsets as Revealed by PCLS Staining and Confocal Microscopy. A) Whole PCLS from an untreated C57BL/6 mouse lung stained with various antibodies. The color of each molecule is indicated by its font color (top right). Individual and combinatorial staining gives the following colors for cell types of interest: CD103+ cDCs (CD103+CD11c+CD88–; white), CD11bhi cDCs (CD11c+CD103–CD88–; red), macrophages (CD11c+CD88+; yellow), neutrophils (CD11c–CD88+; green) and airway epithelial cells (CD324+; blue). B–D) Higher magnification of the insets in A. B) CD11bhi cDCs localize in the parenchyma. C) CD103+ cDCs around the airways. D) CD103+ cDCs underneath the visceral pleura. White bars denote 50 µm. Please click here to view a larger version of this figure.
Figure 2. Staining of DCs and Macrophages. A) Staining of PCLS with antibodies to CD11b (blue) gives a high background in PCLS. B) Antibodies to CD172a (green) have a relatively low background in PCLS and greater specificity for CD11b+ cells. C) Whole PCLS of C57BL/6 mice stained with mAbs whose color is indicated by font color (top right). CD11c (red), CD88 (cyan), CD172a (green) and CD324 (blue). CD11bhi cDCs (CD11c+CD88–CD172a+; yellow), alveolar macrophages (CD11c+CD88+CD172a–; pink), interstitial macrophages (CD11c+CD88+CD172a+; white), neutrophils (CD11c–CD88+; cyan) and airways (CD324+; blue). D – F) Higher magnification of the insets depicted in C. D) Sub-pleural area. E) Sub-epithelial area. F) Interstitium. Interstitial macrophages localize in the parenchyma, and alveolar macrophages are in alveoli. Unlabeled white bars denote 50 µm. Please click here to view a larger version of this figure.
Figure 3. Labeling of Structural Cells in PCLS. A) Whole PCLS prepared from Prox1tdTomato mouse, and stained with mAbs to CD90.2 (green), CD103 (white) and CD324 (blue). Prox1-expressing cells (red) are genetically labeled with TdTomato under the regulation of Prox1 promoter (Prox1-TdTomato). B–E) Higher magnification of the inset in A. B) CD90.2 (green), C) Prox1-TdTomato (red), and D) combination of CD90.2 and Prox1-TdTomato. E) Combination of CD90.2 (green), CD103 (red) and CD324 (blue). Unlabeled white bars denote 50 µm. Please click here to view a larger version of this figure.
Movie 1. T Cells Interact with CD103+ cDCs in the Lung. Live cell imaging of CD103+ cDCs (red) interacting with T cells (green). CD4+ T cells isolated from OVA-specific OT-II Nur77gfp mice were stimulated with splenic DCs and OVA in vitro, and adoptively transferred into Rag -/- mouse. 2 h later, the recipient mouse was treated with OVA/LPS instillation to the airway. 16 h after OVA/LPS treatment, PCLS were made, stained and imaged for 4 h at 37 °C. Time recording image (min) is shown. Please click here to view this video. (Right-click to download.)
The protocol described here was originally developed to visualize the locations of two subsets of cDCs within the lung. However, this protocol can be readily adapted to study many different cell types, while maintaining cell viability and the three-dimensional architecture of the lung. The latter feature is an important advantage over cell culture systems and facilitates identification of rare cell types. The method relies on the generation of PCLS from the lung, and an appropriate combination of antibodies to identify specific cell types while minimizing background staining. To a large extent, this is an empirical exercise because antibodies that work well in other applications, including flow cytometry, do not necessarily work well for staining PCLS. We have already made considerable progress in this regard, but investigators that wish to study cell types not addressed here may have to test different antibodies and optimize fluorochromes and concentrations.
Using the protocol described here, we found that CD103+ cDCs and CD11bhi cDCs reside at distinct localizations in the lung. At steady state, CD103+ cDCs were detected around the airways and in the subpleural region, whereas CD11bhi cDCs were primarily found in the parenchyma. Although the protocol we describe here is not suitable for staining intracellular molecules, it is useful for imaging cell surface markers, as well as secretory molecules. Thus, we were able to detect the chemokines, CCL21 and CCL19, in PCLS (data not shown). If only weak fluorescent signals are seen using primary antibodies, it is possible to use secondary antibodies to amplify those signals.
A major challenge when conducting live cell video-imaging of PCLS was to sufficiently anchor the tissue in culture medium to prevent significant x-y- and z-drift during the 4-h microscopy period. To address this, we experimented with different ratios of Leibovitz's medium to extracellular matrix, which is solid at room temperature, and determined a 1:1 ratio maintained a stable, 3D structure with cell viability. During post-image processing, we also utilized a series of ImageJ plug-ins, called TurboReg and StackReg that eliminate any additional incidental x-y or z-shift (http://www.einstein.yu.edu/segallLab/page.aspx?id=18226). Using the auto-focus tool of the confocal microscope, the region of interest was kept in focus, especially in the z-range, during the entire scanning period. Another challenge during the live cell imaging of PCLS was utilizing the tile scan, z-stack, time series, and autofocus functions of the microscope. To minimize the time between each frame, a balance must be struck between the time spent scanning each tile. To optimize the scan time in each research project, z-stack number, tiles, and scanning speeds need to be adjusted.
The authors have nothing to disclose.
We thank Jeff Tucker, Erica Scappini, and Agnes Janoshazi for their help with microscopy, Ligon Perrow for her management of the mouse colony, and Jun Chen and Michael Sanderson for help with the tissue slicer, and Michael Fessler and Derek Cain for critical reading of the manuscript. This work was funded by the intramural branch of the NIEHS, NIH (ZIA ES102025-09), which is in turn sponsored by the Department of Health and Human Services.
C57BL/6J mice | Jackson Laboratory | 000664 | |
Prox1-TdTomato transgenic mice | Jackson Laboratory | 018128 | B6;129S-Tg(Prox1-tdTomato)12Nrud/J |
OT-II OVA-specific TCR x Nur77-GFP transgenic mice | Jackson Laboratory | 004194, 006617 | B6.Cg-Tg(TcraTcrb)425Cbn/J x C57BL/6-Tg(Nr4a1-EGFP/cre)820Khog/J |
Rag1 knock-out mice | Jackson Laboratory | 002216 | B6.129S7-Rag1tm1Mom/J |
Ovalbumin, Low Endo, Purified | Worthington Biochemical Corporation | LS003059 | |
Lipopolysaccharides from Escherichia coli | Sigma-Aldrich Co. | L2630-25MG | |
Polyethylene tubing (Non-Sterile) 100 ft | BD Diagnostic Systems | 427421 | 0.86 mm inside diameter, 1.27 mm outside diameter |
GeneMate Sieve GQA Low Melt Agarose | BioExpress | E-3112-125 | 2% solution dissolved in PBS at 70 °C and held at 40 °C. |
Compresstome VF-300 | Precisionary Instruments, Inc. | VF-300 | |
Double Edge Stainless Razor Blade | Electron Microscopy Sciences | 72000 | Disposable; 250/box. Blade should be changed for every lung. |
Krazy Glue All Purpose Instant Gel | VWR | 500033-484 | Commonly available for $3/tube in local drugstores |
Leibovitz's L-15 Medium, no phenol red | ThermoFisher Scientific | 21083027 | |
Normal Rat Serum | Jackson ImmunoResearch Inc. | 012-000-120 | |
Normal Mouse Serum | Jackson ImmunoResearch Inc. | 015-000-120 | |
Fetal bovine serum | Hyclone | SH30071.03HI | |
Staining Buffer | Made in House | N/A | PBS w/ 0.5% bovine serum albumin, 0.1% NaN3, pH 7.4 |
Fc Blocker (anti-CD16/32 antibodies) | Made in House | N/A | Supernatant of cultured hybridoma cell line 2.4G2 |
Anti-mouse CD11b eFluor 450 | eBioscience | 48-0112-80 | Anti-mouse CD11b eFluor 450 (clone: M1/70) |
Anti-mouse CD11c Brilliant Violet 605 | BioLegend | 101237 | Brilliant Violet 605 anti-mouse CD11c (clone: M1/70) |
Anti-mouse CD11c Phycoerythrin | eBioscience | 12-0114-82 | PE conjugated anti-mouse CD11c (clone: N418) |
Anti-mouse CD11c Allophycocyanin | BD Phamingen | 550261 | APC-labeled anti-mouse CD11c 9clone: HL3) |
Anti-mouse CD88 Phycoerythrin | BioLegend | 135806 | PE anti-mouse CD88 (clone: 20/70) |
Anti-mouse CD103 Allophycocyanin | eBioscience | 17-1031-82 | Anti-mouse CD103 APC (clone: 2E7) |
Anti-mouse CD90.2/Thy1.2 eF450 | eBioscience | 48-0902-82 | Anti-mouse CD90.2 eFluor 450 (clone: 53-2.1) |
Anti-mouse CD172a/Sirp1a Allophycocyanin | eBioscience | 17-1721-82 | Anti-mouse CD172a APC (clone: P84) |
Anti-mouse CD324 Brilliant Violet 421 | BD Horizon | 564188 | BV421 mouse anti-E Cadherin (clone: 5E8 also known as 5E8-G9-B4) |
Anti-mouse CD324 Alexa Fluor 488 | eBioscience | 53-3249-82 | Anti-CD324(E-Cadherin) Alexa Flour 488 (clone: DECMA-1) |
Anti-mouse CD324 Alexa Fluor 647 | eBioscience | 51-3249-82 | Anti-CD324(E-Cadherin) Alexa Flour 647 (clone: DECMA-1) |
Glass Bottom Microwell Dishes 35mm petri dish, 14mm Microwell, No. 1.5 coverglass | MatTek Corperation | P35G-1.5-14-C | |
Nunc Lab-Tek Chambered Coverglass | ThermoFisher Scientific | 155411PK | Pack of 16 |
15 mm Coverslip, No. 1.5 Glass Thickness | MatTek Corperation | PCS-1.5-15 | |
Bare Platinum Wire | World Precision Instruments | PTP201 | 0.020" (0.5mm) diameter cut into ~1 cm long pieces and bent into an "L" shape |
ProLong Gold Antifade Mountant | ThermoFisher Scientific | P36934 | Keep at 4 °C, warm to room tempterature before use. |
Matrigel Growth Factor Reduced (GFR) Basement Membrane Matrix, Phenol Red-Free, *LDEV-Free | Corning | 356231 | |
Zeiss 880 multi-photon laser-scanning microscope | Carl Zeiss | Zen Black software version 8.1, 2012 (Zeiss) | |
Plan-Apochromat 20x/0.8 M27 objective lends | Carl Zeiss | 420650-9901-000 |