B79-21 Spatial Proteomic Measurements in Fibrotic Human Lung Slices Reveal Drug-induced Tissue Softening Mechanisms

Rationale

To uncover new therapeutic mechanisms in fibrotic human lungs, we have combined 1) spatial proteomics with 2) tissue stiffness measurements within the same drug-treated fibrotic human precision cut lung slices (PCLS). We hypothesize that this novel approach will reveal spatially regulated molecular mechanisms underlying drug-induced tissue softening.

Methods

We utilized the MechanoWell (Kim et al Sci Adv, 2023) to measure PCLS stiffness and then fixed, paraffin-embedded, and sliced the PCLS to yield approximately thirty 5 µm thick section. We selected three structurally diverse sections (corresponding to top, middle, and bottom of the PCLS) and using cyclic immunofluorescence microscopy, analyzed each section for spatial variations of Type I Collagen, α-Smooth Muscle Actin, TIMP1, MMP9, Fibronectin, LOX, vWF, and KRT17.

Results

1) Compared to n = 3 PCLS derived from a non-fibrotic donor lung treated with PCLS media alone (96 hr), n = 3 PCLS from a fibrotic donor lung treated with PCLS media alone (96 hr) were ∼3 fold stiffer, and had significantly enhanced tissue mass, protein fluorescence intensity, and protein expression (Figures 1 a-c). 2) Compared to n = 3 PCLS derived from a fibrotic donor lung treated with PCLS media alone (96 hr), n = 3 PCLS derived from the same fibrotic donor lung treated with Nintedanib (0.3uM, 96 hr) were ∼31% softer, with substantially reduced TIMP-1 expression (Figure 1d) which may have allowed MMP-1 to soften the collagen network and hence the PCLS. A similar comparative analysis between vehicle and Pirfenidone (100mM, 96 hr) treated PCLS from the same fibrotic donor revealed an ∼19% drug-induced PCLS softening but no significant changes in TIMP-1 expression. Instead, Pirfenidone significantly reduced Type I Collagen’s ability to form large stiff clusters leading to a reduced PCLS stiffness (Figure 1e).

Conclusions

We have established foundational methods to elucidate structure-function relationships in drug-treated fibrotic human PCLS. Using these methods, we have revealed distinct mechanisms by which clinically used drugs can induce tissue softening. Figure 1: Spatial proteomic measurements of drug-induced PCLS changes including (a) collagen mass (sum of collagen intensities) within the tissue, (b) fluorescent antibody intensity, and (c) example images of spatial expression. While (d) TIMP1 expression (fluorescent intensity) was significantly reduced by Nintedanib treatment (Nint), (e) compared to vehicle (IPF), the probability of finding large clusters of connected Col1α1 in the parenchyma was significantly lower (p < 0.00001) after Pirfenidone (Pirf) treatment. Scatter points within the interquartile range (IQR) shown in black and points outside the IQR shown in red.

This abstract is funded by: Pfizer Inc./Mechanobiologix