Icon Legend

This session is not in your schedule.

This session is in your schedule. Click again to remove it.

Bioinformatics, Computational and Systems Biology

Bioinformatics, Computational and Systems Biology - Poster Session C & D

S19 - Machine Learning-Based Quantification of LRP1-Regulated Cardiac Fibroblast Proliferation After Myocardial Infarction

Friday, October 10, 2025

10:00 AM - 11:00 AM PDT

Location: Exhibit Hall F, G & H

Presenting Author(s)

Gordina Hodibert

Biomedical Engineering Student
University of Virginia, United States

Co-Author(s)

AA

Antonio Abbate

Ruth C. Heede Professor of Cardiology
University of Virginia, United States
ST

Stefano Toldo

Associate Professor of Medicine, Cardiovascular Medicine
University of Virginia, United States
AL

Alice Luanpaisanon

Biomedical Engineering, Graduate Student
University of Virginia, United States

Last Author(s)

JS

Jeffrey J. Saucerman, Ph.D.

Associate Professor
University of Virginia
Charllotsville, Virginia, United States

Primary Investigator(s)

JS

Jeffrey J. Saucerman, Ph.D.

Associate Professor
University of Virginia
Charllotsville, Virginia, United States

Introduction: : After myocardial infarction (MI), fibroblasts play a significant role in healing the heart by proliferating and producing scar tissue. While scar formation is necessary to prevent rupture, excessive fibroblast proliferation can drive maladaptive remodeling and lead to heart failure. LRP1 (Low-Density Lipoprotein Receptor Related Protein 1) is a multifunctional receptor expressed in fibroblasts and has been implicated in wound healing and fibrosis. However, its role in driving fibroblast proliferation in the heart following MI is still poorly defined. In this study, we developed a machine learning-based image analysis pipeline to quantify the effect of LRP1 in regulating fibroblast proliferation in both genetic knockout and pharmacologic inhibition from mouse cardiac tissue sections.

Materials and

Methods: : Ten-week-old mice underwent transient coronary artery ligation for 30 minutes, followed by 7 days of reperfusion. Experimental groups included fibroblast-specific LRP1 knockout (LRP1-/-) and littermate controls (LRP1+/+), or wild-type mice treated every 72 hours with an LRP1-blocking antibody (LRP1 Ab) or control IgG. Hearts were fixed in formalin, paraffin-embedded, and sectioned at 5 µm thickness. Tissue slices were stained to assess fibroblast proliferation using immunostaining for Proliferating Cell Nuclear Antigen (PCNA, proliferation) and Platelet-Derived Growth Factor Receptor Alpha (PDGFRα, fibroblasts), and counterstained with 4′,6-diamidino-2-phenylindole (DAPI, nuclei). Images were processed using Ilastik’s Pixel Classification tool to generate probability maps for four labels: DAPI, PDGFR−/PCNA+ (non-fibroblast proliferating cells), PDGFR+/PCNA+ (proliferating fibroblasts), and background. CellProfiler was used to segment nuclei, apply thresholds for fibroblast-associated proliferation, and extract object-level data. Representative immunofluorescent images and their corresponding probability maps are shown in Figure 1a–d, demonstrating the input and output used for classification. Only nuclei located in dense cell regions and exceeding a minimum intensity threshold were classified as proliferating fibroblasts. Student t-tests were performed to evaluate statistically significant differences between groups.

Results, Conclusions, and Discussions::

Results: Compared to LRP1+/+ mice, LRP1-/- animals showed a substantial reduction in PDGFRα+/PCNA+ fibroblast counts (Figure 2b), accompanied by a smaller scar area (Figure 2c). Notably, the total population of PDGFRα−/PCNA+ cells did not differ significantly between LRP1-/- and LRP1+/+ groups (Figure 2a), suggesting a fibroblast-specific effect. In the antibody-treated cohort, LRP1 Ab mice showed similar reductions in fibroblast proliferation (Figure 2e) and smaller scar areas (Figure 2f) compared to Ctrl Ab controls. PDGFRα−/PCNA+ counts again showed no significant difference (Figure 2d), reinforcing the specificity of LRP1’s effect on fibroblast regulation.

Conclusions: Both genetic deletion and antibody-mediated inhibition of LRP1 led to decreased fibroblast proliferation and reduced scar formation post-MI, without significantly affecting non-fibroblast proliferation. These results suggest that LRP1 selectively promotes the expansion of fibroblasts during infarct healing.

Discussion: Our findings support a model where LRP1 drives fibroblast proliferation post-MI in a cell-type-specific manner. This was consistently observed across both experimental approaches (LRP1-/- and antibody treatment) strengthening the conclusion that LRP1 is functionally necessary for fibroblast expansion. The lack of effect on PDGFR−/PCNA+ populations indicates that LRP1 signaling does not broadly influence proliferation across all interstitial cells. By using a high-throughput, image-based quantification pipeline, we were able to robustly analyze proliferation at the single-cell level. Future work may focus on identifying downstream effectors of LRP1 and testing whether its inhibition improves long-term cardiac function. These insights may guide therapeutic strategies aimed at limiting pathological fibrosis after infarction.

Acknowledgements and/or References (Optional): :