SCCM Resource Library-Developing a Novel, More Accessible Model of COVID-19 Acute Respiratory Distress Syndrome

Video Transcription

Hi there, my name is Alvin Thieu and I'm currently an MD-PhD candidate at the University of Ottawa in Canada.  My thesis combines clinical research methodology with basic science techniques to study a novel stem cell product in acute lung injury.  However, during the COVID-19 pandemic, as much of us know, non-COVID-related projects were halted,  and hence I decided to spearhead a new project with hopes of developing a novel animal model of COVID-19 ARDS.  I'm honoured to be funded as a Bagnier Scholar as well as a Canadian Vascular Network Scholar for my doctoral thesis,  and I'd like to thank the conference organizers for inviting me here to speak today as a star research presenter during the Critical Care Congress.  So the title of my project is Developing a Novel and More Accessible Model of COVID-19 Acute Respiratory Distress Syndrome, or ARDS.  COVID-19 has led to over 400 million cases worldwide with nearly 6 million deaths. As we know, ARDS is the leading cause of mortality in COVID-19,  and hence there's this need to develop an animal model of COVID-19 ARDS to hopefully evaluate novel therapeutics,  vaccines, as well as study the pathogenesis of this lethal lung injury. However, current mouse models are limited by two very important factors.  First being that working with a live SARS coronavirus requires biosafety level 3 facilities, which are only found at limited institutes around the world.  And secondly, SARS-CoV-2 does not infect wild-type animals very well. Hence, we aim to develop a novel and widely accessible animal model of COVID-19 acute lung injury, or ALI,  which is the preclinical correlate of ARDS, to allow for more researchers around the world to contribute to COVID-19 research and combat this devastating illness.  So how did we do this? Well, first we needed to generate a pseudovirus for widespread use and accessibility.  As we know, looking at the diagram on the left here, SARS coronavirus expresses spike protein labeled in green  to actually allow for binding to human ACE2 receptors in the body, and specifically in the lungs, to cause viral infection.  Hence, we expressed this spike protein on vesiculovirus, or VSV, which is a less infective, less pathogenic virus as compared to SARS-CoV-2.  And this allows us to use our pseudovirus in biosafety level 2 facilities, which are found at a variety and numerous different institutes around the world,  to allow for that increased accessibility. The hope was that by expressing spike protein, this pseudovirus can still bind to human ACE2 receptors and cause infection.  What we did was injected this pseudovirus intratracheally into mice, and specifically to transgenic mice expressing human ACE2,  and assessed for various outcomes of lung injury, including lung histology, vascular permeability, as well as inflammation. So what did we find?  Well, first, we didn't know whether or not our pseudovirus would actually induce any significant lung injury response at all.  So we need to first conduct a dose response analysis using doses from 1E5 to 2.4E7 PFU of our pseudovirus. Again, we injected this virus intratracheally into transgenic mice.  And looking at the diagram on the left here at disease severity score, we can see that by day 5, there's a significant increase in disease severity and decline in mice wellness,  in mice receiving the two highest viral doses of our pseudovirus.  Correspondingly, we saw a 9% drop in both weights and temperature by day 5 in those mice that also received the two highest viral doses.  Hence, phenotypically, at least, we see that a pseudovirus is actually able to induce clinical responses that may be indicating lung injury in our transgenic mice.  We next want to look at various molecular markers of lung injury,  first being the bronchoalveolar lavage fluid, or BALP, protein concentration, which is a marker for lung vascular permeability.  As we see, again, the two highest viral doses labeled in red and orange here led to significant increases in BALP protein levels,  and hence, a significant lung injury response represented by lung vascular hyperpermeability.  In the middle here, we assessed for BALP IL-6 concentrations, which is a pro-inflammatory cytokine that's often used as a marker clinically to assess for the severity of lung injury in COVID-19.  And again, what we saw was that the two highest viral doses led to significant increases in IL-6 concentration in lungs.  Lastly, on the right here, we looked at plasma concentrations of IL-6,  and interestingly, what we found was that there was negligible changes between our viral doses and our control treatment,  showing that our pseudovirus intratracheal injection actually creates a localized lung injury response,  which is our goal here to actually create a model of COVID-19 ARDS for modeling.  So our next question here was whether or not our pseudovirus was inducing disease specific with the human ACE2 receptor expression,  or whether or not the virus can actually induce disease in any animals by itself.  To answer this question, what we did was injected our pseudovirus intratracheally again to transgenic mice expressing human ACE2,  as well as wild-type mice that don't have human ACE2 expression. And then we would sacrifice the mice at various time points and compare between the groups for different outcomes.  What we found again on the left here, looking at disease severity,  interestingly, in the green again, the transgenic mice by day six led to a significant and exponential increase in disease severity from our pseudovirus,  and correspondingly saw a marked drop by 30% in weights and temperature by day six in our transgenic mice.  Comparatively speaking, looking at, in blue here, wild-type mice, we actually saw negligible changes in mildness.  They actually increased in weight over time, as well as negligible changes in temperature, despite a high dose of our pseudovirus.  Hence, phenotypically showing that pseudovirus administration requires the expression of human ACE2 to actually induce a lung injury and response.  Looking at molecular markers, first looking at lung vascular permeability here on the left,  again, we see that hyperpermeability and lung vascular leakiness is only seen in transgenic mice with our pseudovirus, whereas negligible changes are seen in the wild-type mice.  In the middle here, again, we see that BALF cell concentration or inflammatory cell infiltration into the lung space is as negligible in wild-type mice, but markedly increased in transgenic mice.  Hence, looking at totality of the phenotypic markers and molecular markers, it's no surprise to see that from a survival analysis, 100% mortality is seen in transgenic mice receiving our pseudovirus intratracheal administration,  whereas wild-type mice actually all survived to day 10 with no indication clinically or molecularly of a lung injury.  Showing here, again, that there is a specificity for our pseudovirus to induce disease with human ACE2 expression, which is similarly seen from the original SARS-CoV-2 as well.  To further confirm whether or not we actually do see any even minor changes in lung injury, we looked at the histology of wild-type mice and human ACE2 transgenic mice receiving the pseudovirus.  On the left here, wild-type mice, we can see grossly and as well as microscopically that lung architecture is very healthy, thin alveolar membranes, minimal protein deposition as well as minimal cell infiltration,  whereas when we look grossly as well as microscopically in transgenic mice with our pseudovirus, we see a very inflamed lung, significant increase in thickness of the alveolar membrane, and marked increase in inflammatory cell infiltration as well as protein deposition.  When we assess these histology in a blinded fashion and using standardized methods of lung injury scoring, we see that the transgenic mice actually has a markedly elevated lung injury score as compared to wild-type mice after our pseudovirus administration.  Interestingly, when we actually looked at our human ACE2 transgenic mice, there seemed to be histopathological findings that are very similar to what is seen clinically in COVID-19 patients with ARDS.  For example, first looking at panels A and B here at the top, we see that our pseudovirus actually induces microthrombi formation in the lung vasculature of our mice, as well as looking at panel C here, specifically C1 and C2, we also see significant peribronchiolar as well as perivascular inflammatory cell infiltration as well as vascular occlusion from our pseudovirus administration.  Which are markers and indicators of very severe and weak lung injury response.  And lastly, looking at panel D, we also see hallmark features of clinical ARDS, including hyaline membrane formation as well as fibrin protein deposition.  All of this indicating that there is a very severe ALI being induced in our mouse model from pseudovirus that is clinically relevant with clinical features.  And lastly here, we want to assess whether or not our pseudovirus is actually traveling to other tissues and actually inducing disease and inflammation in other tissues as well, and specifically you're looking at the brain, as it's well known that VSV actually induces neuroinflammation, neuroinjury that can cause mortality.  Specifically, we looked at the cortex olfactory bulb and the hippocampus and compared between the wild type mice and our transgenic mice, both of which being again the pseudovirus.  And what we found again, looking at these histological images, there is no brain injury or inflammation that was induced from our pseudovirus.  And there was negligible differences between the wild type and transgenic mice, showing that our pseudovirus may actually be creating a localized inflammatory model in our mice.  Again, looking at molecular markers of apoptosis here, looking at cleaved caspase 3 again, we see minimal differences between our wild type and transgenic mice in cleaved caspase 3 expression throughout the brain in the cortex, olfactory bulb and hippocampus,  showing again that our pseudovirus administration through the intratracheal route creates acute lung injury without dissemination to the brains for any neuroinflammation or infection.  So in conclusion, similar to SARS coronavirus, our pseudovirus induces very clinically relevant features of acute lung injury that is specific and dependent on human ACE2 expression.  Our findings, interestingly, are similar to published effects of live SARS-CoV-2 delivery.  However, we are actually now able to conduct our studies in a standard biosafety level 2 facility and create a model that is very specific for COVID-19 ARDS.  Hence, this mouse model may actually facilitate more widespread investigation around the world into the pathogenesis, as well as studying novel therapeutics to combat the lethal lung injury of COVID-19 acute respiratory distress syndrome.  With that, I'd like to thank you for listening. I'd like to thank my supervisors, my lab members, and my collaborators for making this work possible, as well as all the funding agency.  I'd be happy to take any questions today or even after our presentation, and feel free to email me or reach out to me through Twitter or my email here at the top.  I'd also like to thank the conference organizers for, again, selecting me as a star research presenter and allowing me to present some of my thesis project here at the Critical Care Congress.  And thank you for listening.

Video Summary

Alvin Thieu, an MD-PhD candidate at the University of Ottawa, discusses his project on developing a novel animal model of COVID-19 acute respiratory distress syndrome (ARDS). Current mouse models are limited due to the need for biosafety level 3 facilities and the lack of SARS-CoV-2 infection in wild-type animals. Thieu and his team developed a pseudovirus that expresses the spike protein of SARS-CoV-2 and injected it into transgenic mice expressing human ACE2 receptors. The pseudovirus induced lung injury responses, including increased disease severity, inflammation, and vascular permeability. The model may facilitate research on COVID-19 pathogenesis and therapeutics in more accessible facilities.

Asset Subtitle

Pulmonary, Infection, 2022

Asset Caption

INTRODUCTION: Acute respiratory distress syndrome (ARDS) is the leading cause of mortality in COVID-19. A pertinent preclinical model would greatly facilitate our understanding of COVID-19 ARDS and the development of novel therapies. However, current models using live SARS-CoV2 are limited by costly biosafety level (BSL) 3 facilities which are not widely available. This study aimed to create an animal model of COVID-19 ARDS that is inexpensive, broadly accessible and unrestricted by biohazard requirements. METHODS: Since rodent ACE2 does not bind to SARS-CoV2 spike protein, K18 human (h) ACE2 transgenic mice were intratracheally inoculated with vesiculovirus harboring the SARS-CoV2 spike protein (VSV-CoV2-S) to induce acute lung injury (ALI, preclinical ARDS). Disease severity score (0-30), temperature and weight were recorded daily. Clinically relevant features of ALI including lung permeability, inflammation and histological injury score (0-1) were analyzed. Data are shown as mean±SEM. RESULTS: In a dose ranging study, 1E7 and 2.4E7 (but not 1E5 or 1E6) pfu of VSV-CoV2-S induced clinical, molecular and histological evidence of ALI in K18-hACE2 mice by day 5. To ascertain if ARDS requires hACE2 expression, we inoculated transgenic and wildtype mice with VSV-CoV2-S (1E7 pfu). In K18-hACE2 mice, 100% mortality was observed by day 6 with a 10-fold greater disease severity score (17.0±1.2; P < 0.001) and marked declines in weight/temperature (~25%, P < 0.001). This was associated with increased immune cell infiltration, inflammatory cytokine levels and alveolar permeability (P < 0.01). Histological lung injury score was 4-fold higher in K18-hACE2 mice (0.73±0.04, P < 0.001) and revealed diffuse alveolar damage, pneumocyte hyperplasia and pulmonary hemorrhage. In contrast, all wildtype mice survived until day 10 with no evidence of ALI, which is consistent with the requirement of hACE2 expression for VSV-CoV2-S to induce lethal lung injury. CONCLUSIONS: Our findings indicate that intratracheal VSV-CoV2-S peudovirus causes a severe ALI/ARDS in hACE2 transgenic mice, mirroring the reported effects of live SARS-CoV2. Thus, we have established a novel mouse model of COVID-19 ARDS that can be used in standard BSL2 facilities, thereby enabling more widespread research into its pathogenesis and the development of novel therapies.

Meta Tag

Content Type Presentation

Knowledge Area Pulmonary

Knowledge Area Infection

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Acute Respiratory Distress Syndrome ARDS

Tag Acute Lung Injury

Tag Respiratory Failure

Tag COVID-19

Year 2022

Keywords

animal model

COVID-19

ARDS

pseudovirus

transgenic mice

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.