Elastic Fiber Profiling in Lung Tissue Using Quantitative Imaging Analyses

Summary

Here, we present a protocol detailing tissue preparation for elastic fiber visualization and quantitative profiling in murine lung tissue. This method enables robust assessment of tissue architecture, spatial fiber organization, and extracellular matrix remodeling in physiological and pathological conditions.

Abstract

Remodeling of the pulmonary extracellular matrix (ECM) is a key feature of lung injury and disease. Elastic fibers are a central component of the lung ECM, providing resilience and recoil necessary for normal respiratory function, and their disruption contributes to impaired tissue mechanics. This protocol describes a robust, reproducible workflow for the preparation and analysis of murine lung tissue, with a focus on the visualization and quantitative profiling of elastic fibers. It outlines optimized tissue processing and histological staining methods to preserve and detect elastin structures. In addition, the protocol details an image analysis pipeline enabling quantitative assessment of elastic fiber abundance, organization, and spatial distribution within the lung. The workflow further incorporates stain vector estimation in QuPath, color deconvolution in Fiji/ImageJ, and TWOMBLI-based image analysis to enable quantitative assessment of elastic fiber abundance, branching, and network complexity. Strategies for image acquisition, region-of-interest selection, parameter optimization, and quality control are included to ensure reproducibility and minimize user-dependent bias. Using this workflow, we successfully identified alterations in elastic fiber organization and network architecture in the lungs of obese mice, demonstrating its applicability for detecting disease-associated ECM remodeling. Together, these approaches provide a comprehensive framework to investigate elastic fiber remodeling and its contribution to tissue architecture in physiological and pathological conditions. This method facilitates the study of fibroblast-driven structural changes and supports mechanistic insights into ECM alterations underlying lung disease.

Introduction

Elastic fibers are essential components of the pulmonary extracellular matrix (ECM), composed primarily of elastin and associated microfibrillar proteins that assemble into highly organized, resilient networks¹. In the lung, elastic fibers are arranged along alveolar septa and peribronchial structures, where they provide the elasticity and recoil required for normal respiratory mechanics^1,2. As one of the most abundant structural ECM components in tissues such as the lung, aorta, and uterus, its integrity is critical for maintaining tissue compliance^1,2. Disruption of elastic fiber organization has been observed in a range of pulmonary diseases, including fibrosis, emphysema, and chronic inflammatory conditions, where altered fiber architecture contributes to impaired lung function and mechanical properties^2,3.

The lung matrisome has been recently characterized, revealing core ECM proteins such as collagens and laminins, together with matrisome-associated components including cytokines, ECM regulators, and ECM-affiliated proteins (e.g., integrin ligands, galectins)^4,5. Beyond providing structural support, these components actively regulate tissue function and form specialized niches that support resident and migrating cells^6,7. Notably, recent studies have highlighted that metabolic conditions such as obesity are associated with significant alterations in elastic fiber organization, linking systemic metabolic dysregulation to changes in lung ECM architecture and fibroblast function³. These findings underscore the importance of understanding elastic fiber remodeling as a dynamic and disease-relevant process.

Conventional assessment of elastic fibers often relies on qualitative histological evaluation, semi-quantitative scoring systems, or measurements of elastin abundance, which provide limited information regarding fiber organization, connectivity, and network complexity^8,9. While these approaches are useful for detecting gross alterations in elastin content, they may not capture subtle architectural changes associated with tissue remodeling. Quantitative image-based analyses have the potential to overcome these limitations by enabling objective and reproducible characterization of elastic fiber networks and facilitating comparisons across experimental conditions and studies. To address this, the protocol described here enables integrated analysis of pulmonary ECM organization with a specific focus on elastic fiber architecture. In addition to established approaches for ECM characterization^4,5, this workflow incorporates Elastica staining and a TWOMBLI-based¹⁰ image analysis pipeline optimized for lung tissue architecture. By combining Elastica staining with quantitative network analysis, this workflow extends beyond conventional histological assessment and enables detailed characterization of fiber abundance, fragmentation, spatial organization, and structural complexity. The protocol is particularly well-suited for studies investigating fibrosis, emphysema, obesity-associated lung remodeling, inflammatory lung diseases, aging, and other conditions characterized by extracellular matrix remodeling in the lung.

Protocol

All mouse experiments described here were performed in accordance with the approval of the Animal Ethics Committee of the state of North Rhine-Westphalia, Germany (AZ: 81-02.04.2021.A221).

NOTE: Wear appropriate personal protective equipment (lab coat, gloves, eye protection) throughout the procedure.

1. Sampling and Preparation of murine lung tissue

NOTE: Mice were euthanized by isoflurane overdose according to institutional and governmental guidelines prior to tissue collection.

1. Carefully open the thoracic cavity with a midline incision, then cut the diaphragm along its attachment, and gently expose the lungs.
2. Remove the entire lung by excising the right and left lobes together, taking care to keep all lobes intact.
3. Place the lung lobes individually into a 5 mL tube containing 2 mL of 10% neutral buffered formalin and incubate for 15–20 h at 4 °C.
   NOTE: The fixation preserves tissue architecture for subsequent histological and elastic fiber analyses. Because the primary objective of this workflow was the visualization and quantitative assessment of elastic fiber organization rather than alveolar morphometric analyses, lungs were not inflation-fixed prior to tissue processing.
   CAUTION: Neutral buffered formalin contains formaldehyde, which is toxic, carcinogenic, and may cause skin, eye, and respiratory irritation. Handle formalin in a certified chemical fume hood while wearing appropriate personal protective equipment, including gloves, a lab coat, and safety glasses.
4. Embed the fixed lung tissue in paraffin using a standard histological embedding protocol¹¹.
   NOTE: Ensure complete dehydration and paraffin infiltration to maintain tissue morphology for subsequent staining. To minimize variability between samples and ensure compatibility across experimental conditions, always use the same lung lobes for each analysis. It is recommended to use the left lobe for histology.
5. Cut FFPE lung tissue into 5–7 µm-thick sections using a microtome and mount them on standard glass slides. Bake the slides at 60 °C for 1 h before proceeding with downstream applications. Slides can be stored at RT in a dust-free environment for several weeks or at 4 °C for long-term storage.
   NOTE: Avoid moisture exposure during storage, as humidity can cause paraffin oxidation and compromise staining quality. Store slides in a sealed plastic slide box with dessicant if long-term storage is required.
   Pause Point: Store slides at RT or 4 °C until staining.

2. Elastic fiber staining

1. Perform Elastica (resorcin-fuchsin) staining on 5–7 µm paraffin-embedded lung tissue sections to visualize elastic fibers within the alveolar and vascular structures.
   NOTE: Only stain as many slides as can be handled simultaneously to ensure consistent staining intensity and reproducibility across samples.
   1. Deparaffinate slides by immersing the slides in xylene (mixed isomers) two times for 5 min each.
      CAUTION: Xylene is toxic and flammable. Handle in a chemical fume hood and wear appropriate protective equipment.
   2. Rehydrate the slides in descending ethanol series until 80% ethanol.
      1. Incubate the slide for 3 min at RT in a staining chamber filled with 100% ethanol.
      2. Transfer the slide to fresh 100% ethanol and incubate 3 min at RT.
      3. Transfer the slide to 95% ethanol and incubate for 3 min at RT.
      4. Transfer the slide to 80% ethanol and incubate for 3 min at RT.
   3. Transfer the slide into a new staining chamber containing Resorcin-Fuchsin solution and incubate for 20–30 min at RT.
      NOTE: Resorcin-Fuchsin staining solution can be used several times and is stable at RT.
      CAUTION: Resorcin-fuchsin staining solution contains potentially hazardous chemicals, including concentrated acids and dyes, which may cause skin and eye irritation. Handle the solution in a well-ventilated area or fume hood and wear appropriate personal protective equipment according to institutional safety guidelines.
   4. Place the slide in a staining chamber under running distilled water and wash for 1 min.
   5. Differentiate in 95–96% ethanol: 15–60 s, monitor under the microscope.
      NOTE: Stop when elastic fibers remain deep purple-violet, and the background is adequately cleared. Stained elastic fibers appear dark purple. Collagen fibers remain unstained and typically appear pale. Overstaining can increase background; adjust incubation time depending on tissue thickness and fixation.
   6. Stop the differentiation by transferring the slide immediately into bidistilled water and rinsing for 30 s. Optional: Counterstaining can be performed if needed for additional lung evaluation, but it may reduce fiber segmentation efficiency. We recommend performing H&E staining to evaluate the lung on adjacent cuts.
   7. Dehydrate the slides sequentially in 95% ethanol for 1–2 min, followed by 100% ethanol twice for 1–2 min each to remove residual water.
   8. Clear the sections in xylene 2 times for 3 min until the tissue appears optically transparent.
   9. Mount coverslips using a xylene-compatible permanent synthetic resin and allow slides to dry flat at RT.
      NOTE: Complete dehydration and clearing are essential to prevent cloudiness and ensure long-term preservation of the stained sections.
      Pause Point: Store stained mounted slides at RT until imaging.

3. Elastic fiber imaging and network analysis

1. Image acquisition
   1. Acquire high-resolution brightfield images of lung tissue sections using a slide scanner with a 40× objective. Ensure consistent exposure and white balance settings across all samples.
      NOTE: Images were acquired using identical scanner settings throughout the study, including magnification, spatial resolution (0.25 µm/pixel), illumination, and exposure.
   2. Draw a region of interest (ROI) encompassing the entire lung tissue section to ensure complete coverage during scanning.
   3. Define a background reference area without tissue to measure and subtract background intensity during image analysis.
   4. Use the scanner’s automatic focus detection to assign an optimal number of focal points across the tissue, ensuring uniform sharpness and image quality.
   5. Acquire the image using uniform illumination and exposure settings.
      NOTE: If automatic focusing fails in dense or uneven regions, use manual fine focus adjustment to ensure that all tissue structures, including alveolar and vascular areas, are sharply captured.
2. Image Deconvolution: stain vector estimation in QuPath¹². Stain vectors represent the specific RGB absorbance values of each histological dye, allowing separation of overlapping stains during color deconvolution.
   NOTE: In this protocol, stain vectors were defined in QuPath, as predefined stain vectors for resorcin-fuchsin are not available in Fiji.
   1. Open the image in QuPath and select the brightfield image type.
      NOTE: To minimize bias, calculate stain vectors using representative images from the dataset and apply the same values to all images.
   2. Draw a region of interest (ROI) that contains predominantly elastic fibers to define the first stain component.
   3. In the Image tab, navigate to the Stain vector sections.
   4. Click Stain 1, then set stain vector from ROI to define the color vector for elastic fibers (typically purple-violet).
   5. Select an ROI containing non-elastic tissue (e.g., background or collagen) and click stain 2, then set the stain vector from the ROI to define the secondary color component.
   6. QuPath automatically calculates Stain 3, corresponding to the background.
   7. An example for vectors calculated and applied in Fiji: Resorcin-Fuchsin R:0.6201225, G:0.6541292, B:0.43308553; Unspecific signal R:0.31585205; G:0.815618; B:0.48477295; Background R:-0.105005614, G:-0.4780256, B:0.87204665.
      NOTE: These values were stained for images stained only with resorcin-fuchsin. If a counterstain is present, new stain vectors have to be calculated.
   8. Save these as custom stain vectors and apply them for consistent color deconvolution across all samples.
      NOTE: Defining custom stain vectors from representative regions improves color deconvolution accuracy, particularly for resorcin-fuchsin staining, where elastin and background hues may vary across batches.
3. Image Color Deconvolution in Fiji^13,14.
   1. Open Fiji (ImageJ) and load the brightfield image of the stained tissue section.
   2. Optional: Transform the image to RGB format (Image → Type → RGB Color) only if the image is not already in RGB (e.g., 8-bit images). This step opens a new image window; save it to preserve the original raw data.
   3. Perform color deconvolution by selecting Image > Color > Color Deconvolution > User Values.
   4. Enter the custom stain vector previously obtained from QuPath.
   5. Run the plugin to generate three separate grayscale images corresponding to elastic fibers, tissue excluding elastic fibers, and background.
   6. Use images directly for TWOMBLI without additional preprocessing. TWOMBLI requires sharp images with sufficient resolution; smoothing (e.g., Gaussian blur) should be avoided to preserve structural details.
      NOTE: Ensure that the regions selected in QuPath for vector estimation are representative of typical staining, as these custom vectors will be applied uniformly across all images for consistent comparison.
   7. Save the elastic fiber channel as a TIFF file for downstream quantification and analysis.
4. ROI Selection for Fiber Analysis
   1. For each lung section, select five unbiased square ROIs in the alveolar region and five ROIs in the peribronchiolar region (100 µm x 100 µm each).
   2. Duplicate each ROI on the raw image (Image > Duplicate) to preserve the original data.
   3. Duplicate the same ROIs on the deconvoluted elastic fiber analysis.
   4. Save all duplicated ROIs as individual TIFF files for downstream analysis.
      NOTE: Regions containing tissue folds, sectioning artifacts, tears, large blood vessels, or poorly preserved alveolar architecture were excluded from analysis. Only the deconvolved elastic fiber ROIs were used as input for TWOMBLI analysis, whereas corresponding raw image duplicates were retained as reference files for quality control and visual inspection. Regions of interest (ROIs) should be selected using systematic random sampling to ensure unbiased representation of the tissue. To further minimize operator bias, an automated ROI selection script compatible with QuPath is provided in Supplementary File 1 and Supplementary File 2, and its implementation is described in Section 3.5.
5. Applying the automated ROI selection in QuPath using an automated ROI selection script.
   1. Open the image in QuPath and set the image type to Brightfield.
   2. From the Tools menu, select the desired annotation tool to create the initial annotations. Square annotations are recommended.
   3. Draw square annotations within the lung tissue. Each annotation should encompass only one region of interest (ROI) for analysis, either an alveolar region or a peribronchial area.
   4. Once the annotations have been created, navigate to Annotations and select Lock annotations to prevent accidental movement or modification.
   5. Add the classes “alveoli” and “peribronchial area” to the class list by clicking the “+” button next to Class list in the annotation panel.
   6. Assign each annotation to the appropriate class. Select an annotation, choose the corresponding class, and click Set selected. The annotation color will change, and the assigned class ("alveoli" or "peribronchial area") will be displayed in the annotation properties panel.
   7. Open the corresponding script and execute it. Upon completion, a message will indicate that the ROIs have been successfully generated. The newly created ROIs will appear in the annotation hierarchy as child annotations of the original class-defining annotation.
   8. To export the ROIs, select them and navigate to File → Export Images → Original Pixels. Save the exported ROIs as TIFF files.
      NOTE: The scripts are designed to recognize only the class names "alveoli" and "peribronchial area". These names must be entered exactly as specified. By default, the scripts randomly generate five non-overlapping square ROIs measuring 100 µm × 100 µm within each annotation. Candidate ROIs are accepted only if their corners and center are fully contained within the annotation boundaries, ensuring complete inclusion within the selected region. Both the number and dimensions of the ROIs can be modified within the script to accommodate specific experimental requirements. At least three biological replicates and a minimum of five ROIs/sample are recommended for analyses.
6. TWOMBLI analysis
   1. Save all ROI Images of elastic fibers in a dedicated folder for analysis.
   2. Create two subfolders within the working directory: a. Mask folder – for binary masks generated by TWOMBLI. b. HDM folder – for High-Density Matrix (HDM) images generated by TWOMBLI.
   3. Run the TWOMBLI plugin: Open Fiji and launch TWOMBLI (Plugins -> Macros -> Run).
   4. Select the appropriate analysis parameter file. The following TWOMBLI parameters were used for elastic fiber analysis, listed in Table 1.
      NOTE: These parameters may vary depending on staining intensity. Parameters were empirically optimized using multiple representative images from lung tissue sections stained under standardized conditions. Optimization was performed by comparing the generated masks with the original staining pattern and selecting the settings providing the best overlap between the segmented mask and the underlying elastic fiber network. Once established, the same parameters were applied to all images within the dataset.
   5. Choose the “dark fibers over light background” option.
   6. Generate and save binary masks in the Mask folder.
   7. Generate and save binary images in the HDM folder.
   8. Extract quantitative network parameters such as fiber length, branch points, curvature, density, and alignment.
      NOTE: TWOMBLI analysis requires sharp, high-resolution images (>1 pixel/µm). Poor color deconvolution, out-of-focus regions, or image compression can compromise the accuracy of fiber network quantification.
7. Quantification and Statistical Analysis
   1. Export TWOMBLI network parameters as a CSV file.
   2. Import the raw CSV data into GraphPad Prism for statistical analysis and visualization.
      NOTE: The following parameter is recommended: Mask Total Length, which reflects the total amount of elastic fibers within the ROI, allowing quantification of fiber abundance and loss. Fiber Endpoints indicates the degree of fiber fragmentation; an increased number of endpoints suggests disrupted or broken fibers. Fiber Branchpoints measure how interconnected fibers are, reflecting the structural complexity and integrity of the ECM network. A higher number of branch points indicates a more complex and interwoven network. Lacunarity quantifies the homogeneity of fiber distribution across the ROI. Low lacunarity indicates a uniform network with similarly sized gaps, whereas high lacunarity reflects a heterogeneous network with variable gap sizes describing a more complex structure. Box Counting-Fractal Dimension captures the complexity of fiber architecture. Higher fractal dimension values indicate more intricate and detailed networks, while lower values reflect simpler or disrupted structures.
   3. Generate column graphs showing mean ± SEM for each quantified network parameter (e.g., fiber length, branching, curvature, and density).
   4. Test data normality using the Shapiro-Wilk test. For parametric data, apply a two-tailed unpaired Student’s t-test. For non-parametric data, apply a Mann-Whitney U test. For comparisons involving more than two experimental groups, a one-way analysis of variance (ANOVA) followed by an appropriate post hoc multiple-comparison test, or a Kruskal–Wallis test for non-parametric data, should be applied.
      NOTE: Keep all raw CSV files unaltered to ensure reproducibility and data traceability. Report the number of ROIs and biological replicates analyzed for each condition in the figure legends.

Results

This protocol provides quantitative parameters of the elastic fiber network across different lung regions and experimental conditions. TWOMBLI analysis, as previously described, yields parameters such as fiber length, branching density, lacunarity, and fractal dimension for extracellular matrix assessment¹⁰. Building upon this framework, our approach enables a comprehensive characterization of elastic fiber organization, allowing assessment of changes in fiber abundance, connectivity, uniformity, ...

Discussion

This protocol provides a robust, reproducible workflow for investigating murine pulmonary extracellular matrix (ECM) architecture, with a particular focus on fibroblast-driven elastic fiber remodeling. Elastic fibers are essential for maintaining lung compliance and recoil, and their disruption is a hallmark of multiple pulmonary diseases, including fibrosis, lung injury, aging, and metabolic disorders². Despite their central functional role, quantitative assessment of elastic fiber organization a...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Axio CS2 Slide Scanner	Carl Zeiss
Coplin Glass Staining jar	Sigma-Aldrich	S6016
DPX Mountant	Sigma-Aldrich	6522
Ethanol (100%) for histology	Carl Roth	9065.4
Fiji		https://imagej.net/software/fiji/	version 1.54t
Formalin solution, neutral buffered 10%	Sigma	HT501128
HISTOSETTE I Tissue Processing/Embedding Cassettes	Sigma-Aldrich	H0542
microtome	Thermo Fischer Scietific	HM355S
Micrsocope Slides	VWR	631-0108
Mouse: C57BL/6J	The Jackson Laboratory	RRID: IMSR_JAX:000664	Strain #:000664
paraffin embedding system	Epredia HistoStar embedding module
Prism		https://www.graphpad.com	version 11
Qupath		https://qupath.github.io	version 0.7.0
Resorcin-Fuchsin	Waldeck	2.00E-30
TWOMBLI plugin		https://github.com/wershofe/TWOMBLI
Xylene	Applichem GmBH	14020172