Engineered ACE2 decoy mitigates lung injury and death induced by SARS-CoV-2 variants

The COVID-19 pandemic, caused by SARS-CoV-2, resulted in millions of deaths. While vaccines provide significant protection, vaccine hesitancy, limited vaccine access, breakthrough infections, and the emergence of SARS-CoV-2 variants of concern (VOCs) necessitate the development of effective therapeutics. These VOCs, such as Alpha, Beta, Gamma, and Delta, carry S protein mutations that enhance ACE2 receptor binding and potentially evade vaccine-induced immunity. The SARS-CoV-2 S protein, a class I fusion protein, is cleaved into S1 and S2 subunits. The S1 subunit's receptor-binding domain (RBD) interacts with ACE2 for cell entry. Monoclonal antibodies (mAbs) targeting S protein epitopes have been developed, but their neutralization capacity against VOCs is often reduced. Soluble ACE2 (sACE2) offers an alternative strategy, acting as a decoy for the viral S protein. However, wild-type sACE2 has shown limited efficacy in clinical trials. This study focuses on a next-generation engineered sACE2 derivative, sACE2_2.v2.4, with three mutations (T27Y, L79T, and N330Y) shown to enhance S protein binding affinity. The efficacy of this engineered sACE2 derivative is evaluated *in vivo* using a K18-hACE2 transgenic mouse model, which develops SARS-like lung injury after SARS-CoV-2 infection. This model allows for the study of the pathophysiology of COVID-19, including acute lung injury (ALI) and acute respiratory distress syndrome (ARDS), characterized by lung vascular endothelial barrier breakdown, endothelial cell denudation, VE-cadherin disruption, lung edema, and respiratory failure. The study aims to determine the pharmacokinetics (PK) of sACE2_2.v2.4-IgG1 and assess its prophylactic and therapeutic efficacy against SARS-CoV-2 infection, particularly against VOCs.

The literature extensively documents the challenges posed by SARS-CoV-2 variants and vaccine hesitancy. Studies highlight the escape of VOCs from neutralizing antibodies induced by vaccines (Garcia-Beltran et al., 2021; Hoffmann et al., 2021). The structure and function of the SARS-CoV-2 spike glycoprotein, its role in viral entry via ACE2, and the development of mAbs with high affinity have been described (Walls et al., 2020; Shang et al., 2020; Wec et al., 2020; Wu et al., 2020). However, limitations of mAbs against VOCs prompted the exploration of sACE2 as a decoy (Monteil et al., 2020). Previous studies investigated the use of wild-type sACE2, but its efficacy was limited (Chan et al., 2020; Iwanaga et al., 2020). Deep mutagenesis identified amino acid substitutions in ACE2 leading to enhanced S protein affinity (Chan et al., 2020), paving the way for the development of the engineered sACE2_2.v2.4 used in this study.

The study employed various methods, including molecular dynamics (MD) simulations to analyze the molecular basis of enhanced binding affinity of sACE2_2.v2.4, pharmacokinetic (PK) studies to determine the protein's behavior in vivo, and functional assays, such as pseudovirus and live virus infection studies in K18-hACE2 transgenic mice, to evaluate the therapeutic efficacy of sACE2_2.v2.4-IgG1. MD simulations used approximately 400 µs of all-atom data for WT and v2.4 ACE2 in the unbound (apo) state and approximately 530 µs for RBD-bound complexes. Time-lagged independent component analysis (TICA) was applied to identify critical residue movements. Markov state models (MSMs) were used to analyze the stability of newly formed interactions. For PK studies, sACE2_2.v2.4-IgG1 was administered intravenously, intratracheally, and via inhalation in mice, with serum and lung concentrations measured over time. The efficacy of sACE2_2.v2.4-IgG1 was assessed in K18-hACE2 mice infected with SARS-CoV-2 WA-1/2020 and P.1 variants. Prophylactic and therapeutic administration strategies were evaluated. Lung vascular permeability (using Evans blue-albumin tracer), lung edema (wet/dry ratio), and viral load were measured. Histology and flow cytometry were used to assess lung injury and immune cell infiltration. Binding of sACE2_2.v2.4-IgG1 to S proteins from various VOCs was assessed using flow cytometry and biolayer interferometry (BLI). Comparisons were made with clinically used mAbs. Statistical analyses included Student's t-test, one-way ANOVA, and two-way ANOVA.

MD simulations revealed that the three mutations in sACE2_2.v2.4 enhance binding affinity through tighter steric packing and strengthened hydrogen bonds with the RBD. PK studies showed that intravenous administration of sACE2_2.v2.4-IgG1 resulted in a serum half-life exceeding 7 days. Intratracheal and inhalation delivery also resulted in sustained lung concentrations. Prophylactic intravenous administration of sACE2_2.v2.4-IgG1 (10 mg/kg) 12h before infection with SARS-CoV-2 WA-1/2020 completely protected K18-hACE2 mice from death and lung injury. Therapeutic administration of sACE2_2.v2.4-IgG1 (10 mg/kg and 15 mg/kg started at 12h and 24h post-infection, respectively) significantly improved survival rates (50-60%) compared to the control group in mice infected with SARS-CoV-2 WA-1/2020. Similarly, therapeutic treatment with sACE2_2.v2.4-IgG1 significantly mitigated lung injury and improved survival in mice infected with the P.1 variant. Flow cytometry and BLI studies showed that sACE2_2.v2.4-IgG1 binds to the S proteins of various SARS-CoV-2 VOCs (Alpha, Beta, Gamma, Delta), with affinity comparable to or exceeding clinically used mAbs. The engineered decoy also bound to the S protein of SARS-CoV-1.

The findings demonstrate that the engineered sACE2_2.v2.4-IgG1 decoy effectively mitigates SARS-CoV-2-induced lung injury and improves survival in a mouse model. The high affinity of sACE2_2.v2.4-IgG1 for the S protein, even in VOCs, explains its broad efficacy. The ability to improve survival even with therapeutic administration starting 24h after infection highlights its potential as a treatment for COVID-19. The study's focus on lung endothelial injury, a key factor in ARDS development, provides valuable insights into the pathogenesis of COVID-19. The results also suggest that the multi-mechanistic action of sACE2_2.v2.4-IgG1, including both decoy function and potential modulation of the renin-angiotensin system, contributes to its efficacy. The comparison with clinically used mAbs underscores the potential advantages of sACE2_2.v2.4-IgG1 as a broad-spectrum therapeutic.

This study demonstrates the *in vivo* efficacy of an engineered ACE2 decoy protein, sACE2_2.v2.4-IgG1, against multiple SARS-CoV-2 variants. Its high affinity for the S protein and efficacy in preventing lung injury and improving survival highlight its significant therapeutic potential for COVID-19. Future studies in non-human primates could optimize dosing and delivery strategies before human clinical trials. The broad-spectrum activity and multi-mechanistic action of this decoy protein offer advantages over mAb-based therapies.

The study utilizes a mouse model, which may not fully recapitulate the complexities of human COVID-19. The efficacy of sACE2_2.v2.4-IgG1 in humans needs to be confirmed through clinical trials. The long-term effects of sACE2_2.v2.4-IgG1 treatment require further investigation. Further research is needed to evaluate the optimal route and timing of administration.