Alexa fluor 647 conjugated anti rabbit antibody

Thin sections of lung tissue were subjected to deparaffinization, rehydration and antigen recovery. Endogenous peroxidases were then blocked, followed by tissue permeabilization using 0.4% Triton X-100 in PBS. Nonspecific binding was blocked, and the thin sections were washed twice with PBS containing 0.05% Tween 20 and incubated with anti-SARS-CoV-2 spike glycoprotein rabbit antibody (Abcam, ab272504, 1:100) for one hour at room temperature. The thin sections were then washed with PBS containing 0.05% Tween 20 twice and incubated with the Alexa Fluor 647-conjugated anti-rabbit antibody (Thermo Fisher Scientific; A21244, 1:100) for 30 min in a dark chamber at room temperature. After washing, anti-horse IgG FITC (Sigma-Aldrich; F-7759, 1:100) was added and incubated for one hour at room temperature. The thin sections were washed, and the slides were mounted using Hoechst (Thermo Fisher Scientific; 62249, 1:1000) and ProLong™ Glass Antifade Mountant (Thermo Fisher Scientific; P36980). The sections were analysed under a confocal microscope (Leica TSC SP8 DSL Hyvolution). The images were assembled and analysed in 3D using Imaris Viewer software.

The peptide microarrays were generated and assayed as described previously [37 (link), 38 (link)], except that the arrays contained four triplicate spots of each peptide. Briefly, GST-tagged proteins were diluted to 0.5–2 μM in phosphate-buffered saline (PBS) supplemented with 0.1% (v/v) Tween-20 (PBST) and 5% (w/v) bovine serum albumin (BSA, EMD Millipore Omnipure Fraction V) and incubated with peptide microarrays overnight at 4 °C. Arrays were washed three times with PBS and then probed with an anti-GST antibody (EpiCypher Inc.; Cat. No. 13-0022) diluted to 1:1000 in PBST + 5% BSA. Arrays were washed again 3× with PBS and then probed with an Alexa Fluor 647-conjugated anti-rabbit antibody at 1:10,000 (ThermoFisher). Arrays were imaged using a Typhoon Scanner and protein binding was determined as previously described [37 (link), 38 (link)]. The average signal intensity for each peptide was normalized to the most intense binding within an array, and normalized binding was averaged for at least two independent replicates for each protein. Heat maps of relative binding were generated using JavaTree View (version 1.16r4) after normalizing the relative binding within the subset of peptides selected for the heat map.