1. Introduction
Since its first appearance in Wuhan, China in December 2019, severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) spread worldwide posing a threat to public health [
1,
2]. Four years later and despite massive improvements largely because of the vaccination efforts, new cases are frequently reported [
3,
4]. Omicron variants currently are the most commonly variants seen worldwide [
5,
6,
7]. SARS-CoV-2, like other coronaviruses including its predecessor SARS-CoV-1, uses membrane bound Angiotensin Converting Enzyme 2 (ACE2) as its primary cell entry receptor [
8,
9,
10,
11]. The complex multistep process of SARS-CoV-2 cell entry starts with interaction of the receptor binding domain (RBD) of the viral S1 spike protein with ACE2 on the cell surface [
12,
13,
14,
15]. Conformational changes then lead to priming of the viral spike protein by proteases like transmembrane protease serine 2 (TMPRSS2) which results in fusion pore formation and ultimately cell infection [
9,
16,
17,
18].
At the beginning of the COVID-19 pandemic in early 2020 our group proposed the use of soluble ACE2 proteins to intercept the binding of SARS-CoV-2 to its receptor, membrane bound ACE2, via the so called “decoy effect” [
19,
20]. We had bioengineered a human soluble ACE2 protein that was shortened from 740 to 618 amino acids and fused with an albumin binding domain (ABD) to increase its duration of action [
21]. Later a dodecapeptide (DDC) motif that contains four cysteine residues was introduced to enhance binding affinity for SARS-CoV-2 via formation of a disulfide bonded dimeric protein [
21,
22]. In human kidney organoids and the k18-hACE2 mouse model of lethal SARS-CoV-2 infection we demonstrated the potential of our bioengineered soluble ACE2 proteins to neutralize SARS-CoV-2, markedly improve survival and provide
in-vivo lung, brain and kidney protection [
21,
22]. We established preventative, intranasal delivery, starting one hour before viral inoculation followed by two additional doses at 24 and 48 hours post viral inoculation, as an effective mode of ACE2-618-DDC-ABD delivery achieving optimal protection in terms of survival and organ injury in the k18-hACE2 mouse model infected with ancestral SARS-CoV-2 [
23].
As new variants of SARS-CoV-2 emerge, mutational escape and resistance to treatment are critical issues and both antibodies and antiviral drugs have been shown to be associated with the development of mutational escape of SARS-CoV-2 [
24,
25]. Mutations of SARS-CoV-2 that would decrease ACE2 decoy affinity while simultaneously maintaining efficacious binding of SARS-CoV-2 to the membrane bound ACE2 receptor have not been described and are unlikely to occur [
26,
27]. This highlights an advantage of soluble ACE2 proteins over monoclonal antibodies [
28]. Studies in permissive cells have shown that soluble ACE2 proteins moreover are effective against different SARS-CoV-2 variants [
10,
21,
22,
23,
29,
30,
31]. Here we investigated intranasal administration of ACE2-618-DDC-ABD to k18-hACE2 mice infected with the SARS-CoV-2 delta variant which causes severe disease in humans to further demonstrate the universal efficacy of our unique soluble ACE2 protein. Moreover, ACE2-618-DDC-ABD was applied as early as six hours prior viral inoculation to further determine its lasting protective effect and extended duration of action.
4. Discussion
This study demonstrates a dramatic protective effect of the soluble ACE2 decoy protein termed ACE2-618-DDC-ABD in the lethal k18-hACE2 mouse model of SARS-CoV-2 infection caused by the SARS-CoV-2 delta variant. Animals that received ACE2-618-DDC-ABD six hours prior to as well as 24 and 48 hours after viral inoculation had 90% survival as compared to uniform lethality in the PBS-control group. Disease severity as evaluated by a clinical score and weight loss were markedly improved in the ACE2 618-DDC-ABD treated group. Lung and brain viral titers moreover were markedly reduced in animals receiving ACE2-618-DDC-ABD as compared to the PBS-control group. At the end of the study period on day 14 post viral inoculation all remaining animals in the ACE2-618-DDC-ABD treated group had no detectable lung and brain viral titers Taken together with our previous studies in the same mouse model infected with the ancestral SARS-CoV-2 variant (Washington isolate) [
22,
23] our data demonstrates remarkable efficacy of ACE2-618-DDC-ABD in preventing mortality and organ infectivity in the aggressive delta variant. Furthermore, we also show that intranasal administration of ACE2-618-DDC-ABD prior to viral inoculation results in a protective effect when administered at least six hours prior to viral inoculation which expands our previous work when ACE2-618-DDC-ABD was administered close to viral inoculation with the ancestral SARS-CoV-2 variant (one hour before).
The SARS-CoV-2 delta variant first emerged in October 2020 and was shown to cause severe disease in humans [
40,
41,
42,
43]. Thirteen amino acid mutations have been found in the viral spike protein of the SARS-CoV-2 delta variant as compared to the ancestral virus genome [
38]. Of particular interest are the mutations L452R, T478K, P681R, and D614G, which have been reported to promote infectivity and immune escape by increasing the stability of the spike protein and RBD-ACE2 interaction, as well as facilitating cleavage of the spike precursor protein and promoting viral replication and transmission [
44,
45,
46,
47,
48,
49,
50,
51,
52]. Infection of k18-hACE2 mice with the SARS-CoV-2 delta variant induces distinct pathogenic patterns of respiratory disease [
38,
39]. As compared to infection with the SARS-CoV-2 alpha variant significantly more inflammation, increased production of antiviral cytokines and a greater number of activated immunological pathways have been demonstrated which underlines the uniqueness of the host response against different SARS-CoV-2 variants [
39]. Of note, we found that in k18-hACE2 mice infected with the SARS-CoV-2 delta variant brain viral titers were markedly higher than lung viral titers. This is in agreement with our findings in the same model infected with the ancestral virus variant and strongly suggests that the high lethality is due to brain viral invasion [
23].
We have previously reported that ACE2-618-DDC-ABD, administered to k18-hACE2 mice both intranasally and systemically one hour before as well as 24 and 48 hours after viral inoculation with the wildtype SARS-CoV-2 variant markedly improved survival and mitigated disease severity as assessed by clinical score and weight [
22]. ACE2-618-DDC-ABD also provided
in-vivo lung and kidney protection and reduced lung and brain viral titers [
22]. As recently demonstrated by Hassler et al. the best protective effect can be achieved when administering ACE2-618-DDC-ABD intranasally to k18-hACE2 mice both before and after viral inoculation which is superior to systemic administration at the same time-points [
23]. Administration of ACE2 618-DDC-ABD only post viral inoculation provided only partial protection as compared to the untreated control group and was markedly less efficacious than administration of ACE2 618-DDC-ABD both before and after viral inoculation [
23]. Specifically, intranasal administration of ACE2-618-DDC-ABD before and after viral inoculation resulted in 90% survival on day 6 post viral inoculation, brain viral titers were undetectable in all mice, and lung and brain histopathology were essentially normal [
23]. In the present study that followed the same protocol of intranasal ACE2-618-DDC-ABD administration brain viral titers were similarly undetectable in the ACE2-618-DDC-ABD treated group except for the one animal that had to be humanly euthanized and considered a fatality on day six post viral inoculation. This further supports the idea that brain invasion is likely the main cause of lethality in this model.
The present study was aimed at evaluating the protective effect of ACE2-618-DDC-ABD in k18-hACE2 mice infected with the SARS-CoV-2 delta variant to establish the universality of the
in-vivo protective effect of this ACE2 decoy protein. In previous
in-vitro studies ACE2-618-DDC-ABD exerted a neutralizing effect on infection of Vero E6 cells by the SARS-CoV-2 wildtype, delta and gamma variants at high concentrations [
10,
22,
23]. Infection of A549 cells with the SARS-CoV-2 omicron BA.1 variant was neutralized by 20-fold lower concentrations of ACE2 618-DDC-ABD than those required to neutralize the wildtype variant [
23]. Some SARS-CoV-2 variants, like omicron BA.1, may therefore be neutralized by markedly lower concentrations of ACE2 618-ABD-DDC.
The previous
in-vitro cell studies together with the
in-vivo data in this study show a universal protective effect of ACE2-618-DDC-ABD against SARS-CoV-2 variants. Mutational changes during viral evolution may lead to the development of resistance to current treatment approaches [
24]. Strategies that have been employed to combat SARS-CoV-2 include monoclonal antibodies and specific antiviral drugs [
53,
54,
55]. For those approaches applied in clinical practice the development of mutations that compromise treatment efficacy has been described.
In-vitro studies have demonstrated the development of resistance towards the antiviral drugs Remdesivir and Paxlovid [
56,
57,
58]. The emergence of resistant mutations has moreover been shown when
in-vitro passing the virus in the presence of SARS-CoV-2 specific monoclonal antibodies highlighting the potential of rapidly occurring mutational changes [
28,
54,
59]. Further studies have shown the development of resistant mutations in patients infected with the SARS-CoV-2 delta and omicron variants that received the monoclonal antibody Sotrovimab [
60,
61,
62]. Mutations causing resistance against the antibody cocktails Casirivimab/Imdevimab (REGN-CoV-2) and Bamlanivimab/Etesevimab (LY-CoV555/LY-CoV016), moreover, have also been shown [
63,
64]. These drugs consists of two monoclonal antibodies targeting distinct epitopes [
65,
66,
67], which is particularly interesting as combination therapies are typically less prone to mutational escape and decreased treatment efficiency [
68,
69,
70,
71]. The possibility of mutational escape has also been described for polyclonal antibodies, although unlike monoclonal antibodies a variety of different epitopes is being targeted [
72,
73].
In contrast to monoclonal antibodies,
in-vitro passaging of SARS-CoV-2 in the presence of a soluble ACE2 decoy protein did not lead to the development of resistant mutations [
28]. Importantly, saturation mutagenesis of the SARS-CoV-2 RBD followed by
in vitro selection with full-length ACE2 and an engineered ACE2 decoy protein did not show any RBD mutations that discriminated in favor of full-length ACE2 as compared to the decoy protein [
26]. Although the development of mutations that compromise the affinity for ACE2 decoy binding is theoretically possible, ACE2 decoy proteins clearly have less susceptibility to resistance caused by SARS-CoV-2 mutations [
26,
27]. Mutations that decrease decoy affinity would simultaneously decrease affinity for full-length membrane bound ACE2 which is essential for cell entry and infection of all currently known SARS-CoV-2 variants [
8,
9,
10]. In fact, mutational escape from the neutralizing effect of ACE2 decoy proteins would therefore come at the expense of diminished viral infectivity and virulence [
26,
27].
During the viral evolution different mutations have been described that increase affinity of the SARS-CoV-2 RBD for ACE2 receptor binding as for example the amino acid exchanges N501Y in the SARS-CoV-2 alpha and beta variants as well as L452R in the SARS-CoV-2 delta variant [
44,
74,
75]. The currently dominant circulating SARS-CoV-2 omicron variants also harbor various spike protein mutations leading to increased ACE2 receptor binding while efficiently evading neutralizing antibodies [
76,
77,
78]. Consistent with this, as we have previously shown and noted above, concentrations of ACE2-618-DDC-ABD required to neutralize the SARS-CoV-2 omicron BA.1 variant
in-vitro are significantly lower than those needed to neutralize the wild-type variant [
23]. Of note, while this might be explained by higher binding affinity of the SARS-CoV-2 RBD to ACE2,
in-vitro studies with SARS-CoV-2 omicron variants have also shown decreased fusion and syncytium formation which may in turn result in lower concentrations being effective for neutralization of the SARS-CoV-2 omicron variant than those needed for other SARS-CoV-2 variants [
79,
80,
81].
The present study also shows the absence of kidney SARS-CoV-2 titers in k18-hACE2 mice infected with the SARS-CoV-2 delta variant. While renal involvement has been recognized as a frequent complication of COVID-19 with increased occurrence of acute kidney injury and associated higher mortality in hospitalized COVID-19 patients, the contributing molecular mechanisms remain unclear [
82,
83,
84,
85]. Direct SARS-CoV-2 infection and viral replication in the kidney parenchyma has been discussed as ACE2 is highly expressed in many cell types present in the kidney like proximal tubular cells, podocytes, mesangial and endothelial cells [
86,
87,
88]. SARS-CoV-2 has been shown to directly infect human kidney organoids and the presence of SARS-CoV-2 in kidney parenchyma of patients with COVID-19 has been demonstrated by methods like immunohistochemistry (IHC), immunofluorescence (IF), real-time PCR, and single cell RNA sequencing [
89]. This data, however, originates mainly from autopsies where the postmortem interval from death to autopsy ranges from several hours to days during which SARS-CoV-2 could spread to and within the kidney post-mortem [
89]. Many studies have shown negative findings regarding the presence of SARS-CoV-2 in the kidney including a large series of kidney biopsy samples (n=284) of patients with COVID-19 where no direct kidney SARS-CoV-2 infection could be demonstrated [
89,
90]. Moreover, results from our group in the k18-hACE2 mouse model infected with the ancestral SARS-CoV-2 variant also showed lack of evidence for kidney invasion [
22,
91]. Kidney tissue of k18hACE2 mice infected with the ancestral SARS-CoV-2 variant showed no detectable viral titers by plaque assay as well as no evidence for viral spike and nucleoprotein by IHC and IF staining [
22,
91]. Further analysis using single molecule fluorescence in situ hybridization was also negative for SARS-CoV-2 RNA [
91]. The absence of kidney viral titers in k18-hACE2 mice infected with the delta SARS-CoV-2 variant supports the findings from our previous study. Considering conflicting data throughout the literature, the question whether direct kidney invasion by SARS-CoV-2 occurs remains not fully elucidated. Multiple factors, such as timing between confirmation of SARS-CoV-2 infection and obtaining of the kidney sample, the type of sample obtained (biopsy or autopsy) as well as sensitivity of the methods to detect potential presence of SARS-CoV-2 may influence the detection of SARS-CoV-2 in the kidney [
89,
92].
In conclusion, the present study demonstrates the efficacy of ACE2-618-DDC-ABD to protect k18-hACE2 mice from lethal infection with the aggressive SARS CoV-2 delta variant. Animals that received ACE2-618-DDC-ABD before and after viral inoculation had significantly improved survival, mitigated weight loss and improved clinical scores. Lung and brain SARS-CoV-2 titers, moreover, were significantly reduced in the ACE2 618-DDC-ABD treated group. Importantly, brain viral titers were much higher than lung viral titers and appear much more likely to be the cause of lethality in the k18-hACE2 model. Taken together with previous
in-vivo experiments with wildtype SARS-CoV-2 and
in-vitro data on the neutralizing effect on various variants including wildtype, gamma, delta, and omicron BA.1 [
22,
23], this study demonstrates the universality of the protective effect of ACE2-618-DDC-ABD against infection with SARS-CoV-2 variants. Its therapeutic and preventive potential in humans should be investigated for current and future emerging coronaviruses that use ACE2 as their main receptor for cell entry.