ENDO-LYSOSOME-TARGETED NANOPARTICLE DELIVERY OF ANTIVIRAL THERAPY FOR CORONAVIRUS INFECTIONS
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Review Commons)
Abstract
SARS-CoV-2 can infect cells through endocytic uptake, a process which can be targeted by inhibition of lysosomal proteases. However, clinically this approach fared poorly with an oral regimen of hydroxychloroquine that was accompanied by significant toxicity due to off-target effects. We rationalized that an organelle-targeted approach will avoid toxicity while increasing the concentration of the drug at the target. Here we describe a lysosome-targeted, mefloquine-loaded poly(glycerol monostearate-co-ε-caprolactone) nanoparticle (MFQ-NP) for pulmonary delivery via inhalation. Mefloquine is a more effective inhibitor of viral endocytosis than hydroxychloroquine in cellular models of COVID-19. MFQ-NPs are less toxic than molecular mefloquine, 100-150 nm in diameter, and possess a negative surface charge which facilitates uptake via endocytosis allowing inhibition of lysosomal proteases. MFQ-NPs inhibit coronavirus infection in mouse MHV-A59 and human OC43 coronavirus model systems and inhibit SARS-CoV-2-WA1 and its Omicron variant in a human lung epithelium model. This study demonstrates that organelle-targeted delivery is an effective means to inhibit viral infection.
Article activity feed
-
-
-
-
Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
First and foremost, we would like to extend our thanks to the two referees and managing editor of Review Commons for their constructive feedback and careful consideration of this manuscript. We have taken into consideration all of the suggestions from the reviewers and address all points below and in the following sections.
Minor comments from Reviewer #1:
“It is not clear why the authors used cell number as a measure of viability compared to the MTS assay used in figure 2.”
Response:
In Fig. 2, both MTS assay and CellTiter-Blue assay are used to assess cell viability at 24h and 72h, respectively, whereas counterstaining with DAPI is used to quantify cell …
Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
First and foremost, we would like to extend our thanks to the two referees and managing editor of Review Commons for their constructive feedback and careful consideration of this manuscript. We have taken into consideration all of the suggestions from the reviewers and address all points below and in the following sections.
Minor comments from Reviewer #1:
“It is not clear why the authors used cell number as a measure of viability compared to the MTS assay used in figure 2.”
Response:
In Fig. 2, both MTS assay and CellTiter-Blue assay are used to assess cell viability at 24h and 72h, respectively, whereas counterstaining with DAPI is used to quantify cell number in viral infection assays. Quantification of viral infection using immunofluorescence requires fixation and permeabilization of the cells which is not compatible with assessment of metabolic activity through either the MTS assay or CellTiter-Blue assay post-imaging. One can perform these assays (e.g., MTS and CellTiter-Blue) prior to imaging, however there were concerns regarding viral assay image quantification after running these assays due to the metabolic demand of dye processing which may influence susceptibility to viral infection/propagation, and fluorescence of the metabolic sensor interfering with subsequent IF staining.
DAPI staining is not meant to show viability per se, however, one can gate for fragmented cell nuclei (indicative of lysed cell) to remove dead cells or nuclear debris from the measurement. However, pre-apoptotic or dying cells cannot be accounted for in this measurement.
In our case, due to the rapid doubling time of the immortalized cell lines used (e.g., Vero E6 ~24h, Calu-3 ~35-48h, L929 ~24h), we are able to see large differences in cell number as infected or lysed cells fail to replicate over the 48h experiment. Thus, DAPI staining can be used as a proxy to determine the overall health of the culture but is not directly a measure of cell viability.
Minor comments from Reviewer #1:
“Why did the authors adopt a different pre-treatment infection protocol for SARS-CoV-2 compared to MHV and OC43?”
Response:
The pre-treatment protocol was developed by collaborators in the BSL3 facility as standard practice for rapid treatment/screening of multiple compounds in a BSL3 lab where hands-on time was limited due to immense research focus on infectious disease (e.g., SARS-CoV-2) at the time. We screened many different nanoparticle formulations with this protocol before assessing nanoparticle effect on MHV and OC43 where we also used standardized protocols.
In reference to comments from Reviewer #2:
“The substance of the work is very interesting, but experience shows that the idea that by testing one isolate you can generalize about all other viruses is not accurate. Viruses mutate. SARS-CoV-2 has shown an exceptional capacity for mutation, and the mechanism by which viruses enter cells by endocytosis or membrane fusion plays a role in their exposure to products concentrated in the lysosome. Viruses that enter by membrane fusion (including certain isolates of SARS-CoV-2) should not a priori be as sensitive because they are not subject to phagolysosome fusion.”
“In this sense, it is important to evaluate efficacy on several viral strains and not on a single strain, as has long been the case for acute viral infections. As with HIV, viruses can present natural or acquired resistances linked to evolution under selection pressure or not.
In practice, we cannot rely on testing two viruses, one that has disappeared (Wuhan) and the first virus of the Omicron generation (B1), to assess the therapeutic capacity of a new strategy.”Response:
We would like to highlight that MFQ-NP efficacy was evaluated in several different coronavirus models (MHV, HCoV-OC43 and SARS-CoV-2) in addition to two distinct viral strains (SARS-CoV-2 WT-WA1 and Omicron BA.1) in this study. Not only this, but we have selected viruses from distinct Betacoronavirus lineages which infect different species (homo sapiens and mus musculus). At the time, we had chosen Omicron BA.1 as our model strain as it was the dominant strain in circulation and was temporally separated and genetically distinct from the original WT-WA1 strain.
We can appreciate that viruses mutate, and SARS-CoV-2 has exemplified this through its life cycle. Due to this rapid mutation rate, it is challenging to assess the current dominant variant (e.g., XBB.1.16 as of writing this) and complete manuscript preparation in addition to full peer-review prior to the emergence of a new, potentially more relevant, dominant variant. Additionally, it remains challenging to accurately predict the emergence and genotypic/phenotypic changes of new strains before they arise, necessitating investigation on ancestral strains or, at the least, the current dominant strain.
Lastly, we do not wish to speculate/generalize that MFQ-NPs or a similar approach works for “all other viruses”. As exemplified by efficacy studies in three Betacoronavirus lineages, and one of which using two distinct strains, we do argue that this approach is an effective means to inhibit Betacoronavirus infection. We speculate in the discussion that MFQ-NPs may be used as either a prophylactic or treatment for an array of other respiratory coronaviruses, however we neither show data or speculate efficacy against viruses outside of the coronavirus family.
In reference to comments from Reviewer #1:
“The claim that MFQ may impact cell entry is not supported by the data in the paper. At minimum, the impact of treatment on expression of viral entry receptors for all 3 viruses should be performed, viral attachment assays (see PMID 35176124) and viral pseudoparticle assays. Further the conclusion that MFQ inhibits replication as well as entry is not fully supported by the data presented. This could be improved using a single cycle infection experiment using a synchronised infection protocol. The gold standard to determine impacts on replication would be the use of a viral replicon however I appreciate the technical difficulties in performing these experiments.”
Response:
We agree that the mechanism by which MFQ-NPs inhibit coronavirus infection has not been fully interrogated through this work. We do have preliminary evidence addressed in Fig. 4 which suggests that mechanistically MFQ-NPs may work through targeting pH-dependent protease activity and lysosomal function downstream of viral uptake. However, we have not yet investigated the effect that MFQ-NPs may have on viral entry receptors or viral attachment.
To that end, we are proposing to perform RT-qPCR to measure changes in expression level of key membrane bound proteins responsible for viral uptake. For these assays, we plan to treat Calu-3 cells with MFQ-NPs, unloaded PGC-NPs, equivalent concentration of molecular MFQ, or DMSO as control to gauge expression level of ACE2 (Hs01085333_m1), TMPRSS2 (Hs01122322_m1), sialate O-acetyltransferase gene (CasD1, Hs01082700_m1), and sialic acid acetylesterase gene (SIAE, Hs00405149_m1) relative to expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Hs02786624_g1) using TaqMan probes (ThermoFisher, USA). Additionally, we will also measure the expression level of Cathepsin L (Hs00964650_m1). Cathepsin L is not a transmembrane protein, however strong evidence suggests that this lysosomal protease is essential for S protein processing and viral membrane – endolysosomal membrane fusion in the SARS-CoV-2 endocytic infection route.
These proteins are chosen as ACE2 and TMPRSS2 are known mediators of SARS-CoV-2 uptake and fusion, and HCoV-OC43 relies on uptake via sialoglycan-based receptors with 9-O-acetylated sialic acid (9-O-Ac-Sia) as a key component. Although CasD1 and SIAE themselves are not transmembrane receptors for HCoV-OC43, they regulate the addition or removal of O-acetyl ester groups from sialic acids, respectively.
Similarly, we plan to treat L929 cells with MFQ-NPs, unloaded PGC-NPs, equivalent concentration of molecular MFQ, or DMSO as control to gauge expression level of CEACAM1 (Mm04204476_m1) relative to expression of GAPDH (Mm99999915_g1). MHV spike protein binds to murine carcinoembryonic antigen-related cell adhesion molecule 1a (mCEACAM1a) facilitating infection. NP treatment duration will be consistent with viral infection assays (e.g., 48 h) prior to RNA isolation and qPCR.
Results from PCR measurements may warrant further investigation into protein level expression measured by Western Blotting. However, we plan to begin with PCR as blotting for multiple membrane bound proteins is considerably challenging and higher cost than qPCR.
In addition to expression of viral entry receptors, we will also perform a viral attachment assay and internalization assay to further interrogate MFQ’s mechanism of action. To determine whether mefloquine inhibits SARS-CoV-2 binding/attachment, we will perform viral binding assays in Calu-3/ Vero E6-TMPRSS2-T2A-ACE2 cells that express ACE2 at high levels, and HEK293T cells that express ACE2 at lower levels. Assays will be performed using SARS-CoV-2 Spike pseudo-typed lentivirus which expresses a fluorescent reporter upon mammalian cell infection. SARS-CoV-2 Spike pseudo-virus is advantageous as it mimics SARS-CoV-2 entry mechanisms; however, it can be handled at BSL2, and it can be used to accurately quantify viral uptake since this virus is not replication competent.
SARS-CoV-2 can be internalized primarily via receptor-mediated endocytosis in cells which do not express TMPRSS2 (e.g., Vero E6) or via direct plasma membrane fusion in cells which do express TMPRSS2 (e.g., Calu-3 and Vero E6-TMPRSS2-T2A-ACE2). To test whether mefloquine inhibits the endocytosis of SARS-CoV-2, Vero E6 cells will be pre-treated with effective concentrations of MFQ-NPs, equivalent concentration of molecular MFQ, or unloaded PGC-NPs/DMSO as controls. Next, cells will be inoculated with SARS-CoV-2 Spike pseudo-virus at MOI 0.5 at 37 oC for 1 h. Cells will then be washed with PBS to remove unbound virus, media containing treatments will be reintroduced, and the pseudoviral reporter fluorescence will be measured 18-24 h after inoculation.
To test whether mefloquine inhibits TMPRSS2-mediated fusion during SARS-CoV-2 infection we will use TMPRS2 expressing Vero E6-TMPRSS2-T2A-ACE2 and Calu-3 cells. Cells will be pre-treated with Leupeptin/Pepstatin (inhibitors of endolysosomal proteases), camostat mesylate (serine protease inhibitor), MFQ-NPs, molecular MFQ, unloaded PCG-NPs, or DMSO as control for 1 h, and then inoculated with SARS-CoV-2 Spike pseudo-virus (MOI = 0.5). Cells will then be washed with PBS to remove unbound virus, media containing treatments will be reintroduced, and the pseudoviral reporter fluorescence will be measured 18-24 h after inoculation.
Minor comments from Reviewer #1
_
“Further discussion in the introduction as to the potential mechanism of action of MFQ should be included. I would also suggest the authors read the work by Elizabeth Campbell and Bruno Canard concerning the potential difficulties in designing direct acting antivirals for coronaviruses.”“The figure legends lack detail concerning the number of replicates and statistical comparisons. For viral infections MOIs used are also absent.”_
Response:
We have included a brief summary describing what the field knows of the mechanism of action of MFQ. Namely that MFQ does not directly inhibit the virus/cell membrane attachment process, rather MFQ somehow inhibits viral entry after attachment. We speculate a few mechanisms by which this may occur, which include: (1) inhibiting viral membrane fusion with the cell membrane or endolysosomal membrane, (2) inhibiting proteases responsible for processing SARS-CoV-2 S protein and exposing the fusion peptide, (3) modulating expression levels of ACE2, TMPRSS2, and/or cathepsin, or (4) promoting exocytosis of SARS-CoV-2 particles after uptake.
While it is not the primary goal of the manuscript to determine this mechanism of action, we have evidence currently that MFQ inhibits endolysosomal proteolysis (e.g., mechanism 2 above). Through viral attachment assays and evaluation of receptor expression levels we will also probe mechanism 3 in the revision process.
We have updated the figure legends and methods section to include further details concerning replicates and statistical comparisons. For viral infections we had previously listed the MOIs used in the Materials & Methods section but have also included them in the figure legends.
In reference to comments from Reviewer #2:
“On the one hand, both the introduction and the abstract contain too many elements that give a biased view of the extremely controversial literature. For example, the activity of Hydroxychloroquine and its toxicity have been explored most extensively on the "C19Early" website, which reports on trials carried out in a multitude of countries, including more than 300 trials with Hydroxychloroquine, and it is unreasonable in a scientific paper, designed to last, to report on the major beliefs at a given moment in order to develop a work that has nothing to do with this debate. The same applies to the efficacy of the vaccine. It is difficult to say that the vaccine was poorly distributed, with 20 billion doses, making it the most widely distributed vaccine in the history of mankind in such a short space of time, with results that were not as spectacular as the studies predicted, since the epidemic continued at a comparable level. I suggest that the authors concentrate on their work rather than getting involved in the controversies that are developing around treatments and vaccination.”
Response:
When possible, we have tried to highlight the mixed/controversial results, both positive and negative, of pre-existing therapeutics targeting SARS-CoV-2 and COVID-19 and cited them accordingly. Conjectural phrases such as “selective pressure may lead to 3CLpro mutations conferring nirmatrelvir resistance to new viral mutants” or “there is a growing concern that Molnupiravir, especially when administered at sub therapeutic doses, may result in the creation of more virulent SARS-CoV-2 mutants” are supported by observations in the literature and have been proposed by other experts in the field. Otherwise, we have revised the text to remove any unsupported speculative phrases.
We believe it is worth noting the existing therapeutic strategies in the field to provide context and rationale for our differentiated approach. We have extensively explored the C19Early site as well, and although it does a fantastic job of compiling relevant literature, this site also appears to have a biased view. Throughout preparation of this manuscript, we have considered FDA recommendations and clinical practice in prophylactic protection/treatment of COVID-19 paramount in guiding our introduction and discussion.
We have removed phrasing regarding the limited distribution of vaccines.
In reference to comments from Reviewer #1:
_“The authors claim that MFQ loaded nanoparticles have reduced cytotoxicity compared to 'free MFQ' dissolved in DMSO, however with the NP data and free MFQ data not plotted with the same units this conclusion is hard to reach (Fig.2). While I appreciate the molar units are presented in the text - the reduction in cytotoxicity with MFQ-NP appears to be relatively minor in the both the Calu3 and Vero cell models (doubling in IC50 in both instances). This may indicate quite a narrow therapeutic window for antiviral efficacy without unwanted cytotoxicity. Can the authors replot the data on scales using either molar or ug/ml and use the same dose range for all treatments to enable statistical determination as to whether MFQ-NP significantly reduce cytotoxicity.
To test the antiviral efficacy of the MFQ-NP the authors adopt 2 infection systems, either treating cells pre or post infection. The authors either use fluorescently tagged reporter viruses (OC-43/MHV) or immunofluorescence to visualise and quantify viral infection in cell-line models. Given a key aim of this paper is to determine whether MFQ-NP rather than free MFQ is a superior treatment option, it is challenging to assess this with the data presented in figures 5-6. The units of treatment between NP, MFQ-NP and free MFQ again differ, and the molar dose range of MFQ-NP and free NP is not the same. This makes it very hard to conclude whether MFQ-NPs are more effective then free MFQ. Formal dose response curves with the same dose of empty-NPs, MFQ-NPs and free MFQ are needed here, preferably with match cell viability data using the same assay as figure 2.”_
Response:
We will include additional plots with axis scaling for MFQ-NPs as concentration of MFQ in µM rather than concentration of NPs in µg/mL for further comparison to the free MFQ group. We chose concentration of NPs to match with the equivalent unloaded NP controls. Unfortunately, it would not be possible to create a formal dose response curve with the same dose of empty-NPs and free MFQ as those entities only co-exist in the MFQ-NPs treatment group.
We agree with the observation that the therapeutic window is likely narrow in vitro. This is observed in the viral inhibition and cytotoxicity experiments, where the most efficacious dosing of NPs (e.g., 50 – 100 µg/mL) in inhibiting viral infection is similar to the IC50 value (e.g., 54 µg/mL in Calu-3) at 24 h. Similarly, we see a biphasic dose response in our protease activity assay, suggesting that low doses actually increase endolysosomal protease activity which may promote viral infection.
We speculate this dose limiting toxicity and narrow therapeutic window will likely be improved more drastically in vivo (ongoing continuation of this study), however in vitro we still see at least a minor improvement in MFQ tolerability.
In reference to comments from Reviewer #1:
“Finally while animal experiments are likely beyond the scope of this study, use of air-liquid interface cultures of lung epithelial cells would be a significant improvement to the work and provide further support to their conclusions in a physiologically relevant system.”
Response:
While we appreciate the suggestion of ALI models and animal models to further assess MFQ-NPs in a more physiologically relevant system, we agree that these studies would be beyond the scope of this study. This investigation is planned as a continuation of the current study.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
This work is interesting but poses two problems.
On the one hand, both the introduction and the abstract contain too many elements that give a biased view of the extremely controversial literature. For example, the activity of Hydroxychloroquine and its toxicity have been explored most extensively on the "C19Early" website, which reports on trials carried out in a multitude of countries, including more than 300 trials with Hydroxychloroquine, and it is unreasonable in a scientific paper, designed to last, to report on the major beliefs at a given moment in order to develop a work that has nothing to do with this debate. The same …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
This work is interesting but poses two problems.
On the one hand, both the introduction and the abstract contain too many elements that give a biased view of the extremely controversial literature. For example, the activity of Hydroxychloroquine and its toxicity have been explored most extensively on the "C19Early" website, which reports on trials carried out in a multitude of countries, including more than 300 trials with Hydroxychloroquine, and it is unreasonable in a scientific paper, designed to last, to report on the major beliefs at a given moment in order to develop a work that has nothing to do with this debate. The same applies to the efficacy of the vaccine. It is difficult to say that the vaccine was poorly distributed, with 20 billion doses, making it the most widely distributed vaccine in the history of mankind in such a short space of time, with results that were not as spectacular as the studies predicted, since the epidemic continued at a comparable level. I suggest that the authors concentrate on their work rather than getting involved in the controversies that are developing around treatments and vaccination.
The substance of the work is very interesting, but experience shows that the idea that by testing one isolate you can generalize about all other viruses is not accurate. Viruses mutate. SARS-CoV-2 has shown an exceptional capacity for mutation, and the mechanism by which viruses enter cells by endocytosis or membrane fusion plays a role in their exposure to products concentrated in the lysosome. Viruses that enter by membrane fusion (including certain isolates of SARS-CoV-2) should not a priori be as sensitive because they are not subject to phagolysosome fusion.In this sense, it is important to evaluate efficacy on several viral strains and not on a single strain, as has long been the case for acute viral infections. As with HIV, viruses can present natural or acquired resistances linked to evolution under selection pressure or not.
In practice, we cannot rely on testing two viruses, one that has disappeared (Wuhan) and the first virus of the Omicron generation (B1), to assess the therapeutic capacity of a new strategy.Significance
This work is interesting but poses two problems.
On the one hand, both the introduction and the abstract contain too many elements that give a biased view of the extremely controversial literature. For example, the activity of Hydroxychloroquine and its toxicity have been explored most extensively on the "C19Early" website, which reports on trials carried out in a multitude of countries, including more than 300 trials with Hydroxychloroquine, and it is unreasonable in a scientific paper, designed to last, to report on the major beliefs at a given moment in order to develop a work that has nothing to do with this debate. The same applies to the efficacy of the vaccine. It is difficult to say that the vaccine was poorly distributed, with 20 billion doses, making it the most widely distributed vaccine in the history of mankind in such a short space of time, with results that were not as spectacular as the studies predicted, since the epidemic continued at a comparable level. I suggest that the authors concentrate on their work rather than getting involved in the controversies that are developing around treatments and vaccination.
The substance of the work is very interesting, but experience shows that the idea that by testing one isolate you can generalize about all other viruses is not accurate. Viruses mutate. SARS-CoV-2 has shown an exceptional capacity for mutation, and the mechanism by which viruses enter cells by endocytosis or membrane fusion plays a role in their exposure to products concentrated in the lysosome. Viruses that enter by membrane fusion (including certain isolates of SARS-CoV-2) should not a priori be as sensitive because they are not subject to phagolysosome fusion.In this sense, it is important to evaluate efficacy on several viral strains and not on a single strain, as has long been the case for acute viral infections. As with HIV, viruses can present natural or acquired resistances linked to evolution under selection pressure or not.
In practice, we cannot rely on testing two viruses, one that has disappeared (Wuhan) and the first virus of the Omicron generation (B1), to assess the therapeutic capacity of a new strategy. -
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
Summary
Petcherski et al describe a nanoparticle delivery system to deliver mefloquine, a inhibitor of viral endocytosis, into cell line models of coronavirus replication. They describe the generation of these nanoparticles and test their antiviral efficacy against 3 different beta coronaviruses OC43, MHV and SARS-CoV-2, including the omicron variant.
Major Comments
The authors claim that MFQ loaded nanoparticles have reduced cytotoxicity compared to 'free MFQ' dissolved in DMSO, however with the NP data and free MFQ data not plotted with the same units this conclusion is hard to reach (Fig.2). While I appreciate the molar units are …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
Summary
Petcherski et al describe a nanoparticle delivery system to deliver mefloquine, a inhibitor of viral endocytosis, into cell line models of coronavirus replication. They describe the generation of these nanoparticles and test their antiviral efficacy against 3 different beta coronaviruses OC43, MHV and SARS-CoV-2, including the omicron variant.
Major Comments
The authors claim that MFQ loaded nanoparticles have reduced cytotoxicity compared to 'free MFQ' dissolved in DMSO, however with the NP data and free MFQ data not plotted with the same units this conclusion is hard to reach (Fig.2). While I appreciate the molar units are presented in the text - the reduction in cytotoxicity with MFQ-NP appears to be relatively minor in the both the Calu3 and Vero cell models (doubling in IC50 in both instances). This may indicate quite a narrow therapeutic window for antiviral efficacy without unwanted cytotoxicity. Can the authors replot the data on scales using either molar or ug/ml and use the same dose range for all treatments to enable statistical determination as to whether MFQ-NP significantly reduce cytotoxicity.
To test the antiviral efficacy of the MFQ-NP the authors adopt 2 infection systems, either treating cells pre or post infection. The authors either use fluorescently tagged reporter viruses (OC-43/MHV) or immunofluorescence to visualise and quantify viral infection in cell-line models. Given a key aim of this paper is to determine whether MFQ-NP rather than free MFQ is a superior treatment option, it is challenging to assess this with the data presented in figures 5-6. The units of treatment between NP, MFQ-NP and free MFQ again differ, and the molar dose range of MFQ-NP and free NP is not the same. This makes it very hard to conclude whether MFQ-NPs are more effective then free MFQ. Formal dose response curves with the same dose of empty-NPs, MFQ-NPs and free MFQ are needed here, preferably with match cell viability data using the same assay as figure 2.
The claim that MFQ may impact cell entry is not supported by the data in the paper. At minimum, the impact of treatment on expression of viral entry receptors for all 3 viruses should be performed, viral attachment assays (see PMID 35176124) and viral pseudoparticle assays. Further the conclusion that MFQ inhibits replication as well as entry is not fully supported by the data presented. This could be improved using a single cycle infection experiment using a synchronised infection protocol. The gold standard to determine impacts on replication would be the use of a viral replicon however I appreciate the technical difficulties in performing these experiments.
Finally while animal experiments are likely beyond the scope of this study, use of air-liquid interface cultures of lung epithelial cells would be a significant improvement to the work and provide further support to their conclusions in a physiologically relevant system.
Minor Comments
It is not clear why the authors used cell number as a measure of viability compared to the MTS assay used in figure 2.
Why did the authors adopt a different pre-treatment infection protocol for SARS-CoV-2 compared to MHV and OC43?
Further discussion in the introduction as to the potential mechanism of action of MFQ should be included. I would also suggest the authors read the work by Elizabeth Campbell an Bruno Canard concerning the potential difficulties in designing direct acting antivirals for coronaviruses.
The figure legends lack detail concerning the number of replicates and statistical comparisons. For viral infections MOIs used are also absent.
Significance
This work has good potential but is let down by the execution of the experimental design. The use of NP for antiviral drugs is a very interesting area and could greatly contribute to drug design for this viral family.
-
