The SARS-CoV-2 receptor-binding domain facilitates neutrophil transepithelial migration and nanoparticle uptake in the mice airways

SARS-CoV-2-induced infection is still dangerous. Mouse models are convenient to the investigation of virus-activated immune response mechanisms. However, mice are not proper model organisms to study COVID-19 due to decreased interaction affinity between the SARS-CoV-2 receptor-binding domain (RBD) and mouse angiotensin-converting enzyme 2 (ACE2) compared with human ACE2. In the present study, we propose a mouse model that allows estimating the influence of SARS-CoV-2 on the immune system. To mimic the effects of RBD– ACE2 high-affinity interaction, mice received the ACE2 inhibitor MLN-4760. To simulate virus loading, we applied 100 nm particles suspended in the solution of RBD via the oropharyngeal route to mice. In this model, MLN-4760 application enhanced neutrophil egress from the bone marrow to the bloodstream and RBD attracted neutrophils to the luminal side of the conducting airway epithelium. By contrast, inert 100 nm particles were not potent to stimulate neutrophil recruitment to the conducting airway mucosa. Using this model, and by altering the dosage of the ACE2 inhibitor, nanoparticles, and RBD, one can adapt it to investigate different COVID-19 states characterized with mild or severe airway inflammation. Statement This study presents a mouse model that allows estimating the influence of SARS-CoV-2 on the immune system and investigates immune cell-model virus particle interactions in the conducting airway mucosa.

In this study, using immunohistochemistry and three-dimensional imaging of the whole-mount 61 conducting airways, we investigated the ability of 100 nm fluorescent particles to attract 62 neutrophils after oropharyngeal application to mice. We checked the effects of RBD and the ACE2 63 inhibition on the neutrophil recruitment and nanoparticle internalization. This study presents a 64 model that mimics some aspects of COVID-19 and affords to investigate the effects of SARS- 65 CoV-2 on the immune system without using humanized mice and virulent pathogens.  69 response to inhaled nanoparticles 70 To investigate the conducting airway immune response to SARS-CoV2, mice received an 71 oropharyngeal application of 100 nm particles -FluoSpheres, comparable in size with  CoV2 virus particles ( Bar-On et al., 2020). Before application to mice, FluoSpheres were 73 resuspended either in PBS or RBD solution. In both cases, 24 h after the application, 74 FluoSpheres were detected on the luminal side of the conducting airway epithelium (Fig. 1A, B). 75 In both cases, FluoSpheres were mostly internalized by CD169 + actin-rich cells; however, free 76 FluoSpheres were also detected. The application of FluoSpheres suspended in PBS did not 77 induce measurable neutrophil migration to the conducting airway mucosa. Both anti-Ly6G and 78 anti-CD11b antibodies revealed that Ly6G + and CD11b + cells were mostly located beneath the 79 smooth muscle layer in the submucosal compartment ( Fig. S1A-C). Thus, 24 h after 80 FluoSpheres application, only a few CD11b + were observed in the conducting airway wall, 81 between the epithelium and the smooth muscle layer (Fig. 1A). No CD11b + cells were detected 82 in the luminal side of the epithelium close to FluoSpheres (Fig. 1A). Application of FluoSpheres 83 suspended in RBD solution induced CD11b + cells migration to the conducting airway mucosa 84 6 and specifically to the luminal side of the epithelium (Fig. 1B). Moreover, migrating through the 85 epithelial barrier CD11b + cells were detected in close proximity to FluoSpheres (Fig. 1B,C, 86 arrows). The precise analysis showed that CD11b + cells together with CD169 + cells participated 87 in the internalization of FluoSpheres in the luminal side of the epithelium (Fig. 1D). When

88
FluoSpheres were applied to mice in BSA solution, they did not induce CD11b + cell infiltration 89 to the conducting airway mucosa (Fig. S2B). The RBD solution alone (without FluoSpheres) did 90 not induce neutrophil recruitment to the conducting airway mucosa (Fig. S2C Fig. 2A,B). We showed that a single MLN-4760 intraperitoneal (i.p.) injection significantly 106 increased the blood myeloid cell and neutrophil proportions 2 h after the application (Fig. 2C,D).

107
As the injection itself (as a trauma) can alter the blood neutrophil number, the control group of 108 7 mice received a PBS i.p. injection. Two hours after, the blood neutrophil proportion in mice that 109 received the PBS injection was significantly higher than that of the intact mice but significantly 110 lower than that of mice injected with D).

111
As expected, the i.p. injection of MLN-4760 did not influence the CD11b + cell migration to the     Thus, here we report the approach that allows mimicking the RBD-ACE2 interaction in mice and   The peripheral blood was collected from the tail vein. The blood was collected to the 1.5 mL 305 test-tube with 50 µL heparin (VelPharm, Russia) and transferred to the 10 mL of the hemolysis 306 buffer (155 mM NH4Cl (Reachem, Russia); 0.1 mM Na2EDTA (Sigma-Aldrich); 10 mM 307 NaHCO3 (PanReac Applichem); pH 7.3, stored at +4°C, and warmed to RT before use) in 50 mL 308 tube. After 5 min at RT, 20 mL of PBS was added, and the tube was centrifuged at 380 g 5 min.

309
The supernatant was replaced with 5 mL of the hemolysis buffer. After 5 min at RT, cells were were used for isotype-control staining (Fig. S4A, B). The airway specimens were then washed 332 with PBS, permeabilized with 0.3% Triton X-100, and blocked with 1% BSA and 4% normal  The image stacks were analyzed using Imaris. Based on maximum intensity projections, CD11b + 367 cells were processed via three-dimensional surface rendering of the APC channel, using several 368 thresholds, as previously described (Veres et al., 2009). CD11b + cell surfaces were quantified 369 automatically; the inspection of the results was made each time manually. The images of at least 370 two proximate to trachea regions from each specimen were acquired (Shevchenko et al., 2013). The data are presented as the scattered dot plots with the median and interquartile range (i.q.r.) 374 for at least 4 mice per group. The differences between two groups were analyzed with the Mann-375 Whitney test using GraphPad Prism software (GraphPad Software, San Diego, CA). A p-value 376 less than 0.05 was considered statistically significant.   The data are pooled from two independent experiments, the total number of mice n=4, and 541 28 shown as median and i.q.r. Significant difference between the indicated group and the intact 542 mice or between the indicated groups was detected using Mann-Whitney test *: p ≤ 0.05. E, F.  FluoSpheres and MLN-4760 alone. The data are shown as median and i.q.r., n=5 mice.

588
Significant difference between the indicated group and the intact mice was detected using Mann-589 Whitney test *: p ≤ 0.05.