Coronaviruses are a large family of viruses that typically cause mild to moderate upper respiratory tract infection. SARS-CoV-2 is a zoonotic pathogen (a virus with the ability to cross species and infect multiple hosts) and is responsible for the current COVID-19 pandemic. Many details have already been ascertained about how the virus spreads and may cause respiratory failure and sepsis in vulnerable populations. Yet, important scientific questions surrounding COVID-19 remain unanswered:
- Why do some poorly oxygenated patients continue to breathe normally (silent hypoxemia) while others become severely dyspneic?
- What do patients with normally compliant lungs (L phenotype) sometimes transition to a high elastance lung (H phenotype)?
- Are there predictive markers to determine if patients will respond to medical interventions (drugs, nitric oxide treatments, prone position ventilation)?
- How to better validate novel therapies and vaccines before running complex and costly clinical trials?
Developing COVID-19 preclinical models that closely mimic the clinical progression of the disease is vital to understand the disease and how we can treat patients effectively. The challenges in developing such models are numerous. They include infecting and replicating SARS-CoV-2 in host subjects, developing clinical characteristics of COVID-19, and ensuring that the research model is reproducible. Since in science, “all models are wrong, but some models are useful”, it is therefore essential to develop the right model to answer the right question.
Another important consideration for COVID research is to BSL3 or not to BSL3 – that is the question! Given its high virulence, an Animal Bio-Safety Level-3 (ABSL-3) facility is required to work with the SARS-CoV-2 pathogen. BSL-3 laboratories require comprehensive personal protective equipment (PPE) and strict security protocols to ensure the safety of the laboratory personnel. Working directly with dangerous pathogens like SARS-CoV-2 is necessary to understand their transmission mechanisms, their replication process, and their pathophysiological mechanisms.
Many scientists who do not have access to BLS-3 facilities are eager to join in the fight against COVID-19. While scientists who work in conventional (non BSL-3) laboratories cannot directly work with SARS-CoV-2, they can turn to different modeling techniques to replicate symptoms of the disease. Surrogate models can reproduce specific elements of COVID-19 like lung injury or pneumonia, as well as various components of the immune response that render some patients more vulnerable to COVID-19.
Researchers who have access to a BSL-3 laboratory can rely on various animal models to study COVID-19:
Mice are often a preferred mammalian biomedical research subject due to the extensive reference data found in the scientific literature, the ability to genetically manipulate their genome, their low cost, and their short lifespan. Unfortunately for COVID-19 researchers, the mouse ACE2 receptor that first binds to SARS-CoV-2 during infection is significantly different than the one found in humans. Consequently, mice do not exhibit symptoms of COVID-19 unless their immune system is compromised, or an aging model is used. Scientists, therefore, rely on various techniques to humanize the mice and make them more susceptible to the disease.
- Dr. Stanley Perlman, a scientist at the University of Iowa, developed a two-step approach to infect mice with SARS-CoV-2. The mice are first exposed to an adenovirus expressing the human ACE2 receptor, then exposed to SARS-CoV-2, which binds to the human ACE2 receptor.
- Another approach developed by Dr. Perlman uses transgenic mouse models (hACE2-mice) that express the human ACE2 receptor and are vulnerable to SARS-CoV-2.
- Other facilities are developing a modified SARS-CoV-2 virus that has a greater affinity for the mouse ACE2 receptor.
It is likely that a combination of all these disease modelling techniques will be needed to address the increasing number of questions about the virus.
GOLDEN SYRIAN HAMSTER (MESOCRICETUS AURATUS)
Hamsters are rapidly becoming an important ally in the fight against COVID-19. When researchers exposed hamsters to SARS-CoV-2, the infection-induced weight loss, lethargy and changes in its breathing patterns. Furthermore, the diseased hamsters consistently infected co-housed naïve hamsters. All infected hamsters ultimately recovered from the disease and developed antibodies. The hamster model recapitulates many of the viral replication features and clinical symptoms sought by scientists and will likely become more prevalent in COVID-19 research laboratories.
LARGER ANIMAL MODELS (RATS, FERRETS, NHPS)
The fight against COVID-19 will likely involve a multitude of other research models including rats, ferrets, swine and primates.
- Rats offer limited advantages over mouse models for infectious studies as they have similar ACE2 receptors with a low affinity for the SARS-CoV-2 pathogen. One notable exception is that they are larger subjects and therefore more biological samples may be drawn during the course of an experiment.
- Ferrets, on the other hand, are a promising model as the SARS-CoV-2 virus replicates in their lungs and intestines, and they have been shown to transmit the virus to adjacent cages. At present, no overt clinical manifestations of COVID-19 have been confirmed in the ferret model despite its more human-like respiratory physiology. The ferret airway tree has more generations than mice and they can cough and sneeze, which has proven insightful for other pulmonary diseases.
- Non-human Primates (NHPs) are invaluable for infectious diseases research as they share many of the biological pathways present in humans, making them an important model for the development of biologics and vaccines. However, the use of primates for life sciences research is restricted to a few highly specialized centers due to ethical considerations, limited availability, and cost.
Non BSL-3 Models
Surrogate models can help identify immune pathways, pinpoint drug targets and aid in evaluating novel therapeutic strategies. Scientists who work in conventional laboratories can also model comorbidities of COVID-19 (e.g. diabetes, obesity, ventilator-induced lung injury, bacterial infections) and develop insights that will guide clinical practice and save lives.
LPS MODEL OF ACUTE RESPIRATORY DISTRESS SYNDROME (ARDS)
It is estimated that 70% of COVID-19 related deaths are attributed to respiratory failure. As such, mouse models of ARDS will certainly become more prevalent and utilized in COVID-19 research. The administration of lipopolysaccharide (LPS), whether intratracheally, systemically or via the oropharyngeal route, is one of the most common models of acute lung inflammation and ARDS in rodents. Beyond the primary LPS insult, researchers may apply additional stimuli such as prolonged or injurious mechanical ventilation to exacerbate the disease and better reproduce the human pathophysiology.
Researchers use various strains of mice for LPS modelling studies, with the C57BL/6 being the most widely published in the field. The LPS model has helped shed light on the immune pathways involved in the resolution of acute lung injuries and contributed to the development of therapeutic interventional and ventilation strategies that minimize long-term damage to the lungs. By allowing scientists to combine comprehensive hemodynamic and pulmonary mechanics parameters in different knock-out mice, the LPS model will surely help unravel some of COVID-19’s most important mysteries, even without the involvement of its main actor, SARS-CoV-2.
INFLUENZA INFECTION MODELS
Coronaviruses differ considerably from influenza viruses. However, both pathogens can illicit pronounced responses in their hosts that can lead to ARDS and, on some occasions, to death. In vivo experimentation with Influenza A strain H1N1 does not require BSL-3 facilities. Influenza models of acute lung injury, coupled with a better understanding of the pathogenesis of COVID-19, may lead to novel insights about viral ARDS and how the innate and acquired immune responses and can be monitored to identify at-risk patients, and tuned to improve patient survival.
Once relevant research models are developed, the next challenge is to extract valuable biological information that can be extrapolated to humans. Measuring clinically relevant physiological outcomes such as lung function can be daunting, especially when these measurements are obtained in smaller rodents. SCIREQ offers many instruments to help non-experts acquire predictive outcomes that generate valuable information that can be disseminated to scientists and clinicians worldwide.
DISEASE MODELLING, TRANSLATIONAL OUTCOMES AND DRUG DELIVERY
The flexiVent is an advanced pulmonary research platform highly adapted to assist in COVID-19 preclinical experimentation. The flexiVent measures detailed lung function outcomes in SARS-CoV-2 infected animals, LPS-induced ARDS models, genetic knock-outs and treatment groups, allowing scientists to test their hypotheses in complex organisms that closely recapitulate human biology. Software automation standardizes experimental conditions to create the highest level of reproducibility and produces comprehensive reports to easily disseminate scientific findings to the research community.
GENERATING RELIABLE DISEASE MODELS AND THERAPEUTIC INTERVENTIONS
The flexiVent’s computer-programable ventilator design, scripting features and aerosol delivery capabilities are the cornerstones of numerous ARDS experiments. First, the flexiVent can synchronize the activation of an Aerogen nebulizer with the inspiratory cycle to precisely deliver aerosolized compounds such as drugs, LPS, elastase and bleomycin deep into the lungs of intubated subjects. This ventilator-assisted aerosol delivery offers a more reproducible and homogenous deposition profile than intratracheal instillation. The flexiVent can further apply a wide variety of injurious or protective ventilation modes (low or high tidal volumes, low PEEP, etc.) to reproduce clinical manifestations of ARDS and ventilator-induced lung injury in animals.
COMPREHENSIVE LUNG FUNCTION ASSESSMENT
Once the appropriate model is generated, correlating COVID clinical symptoms and progression is translational measurements is key. As noted above, there are emerging reports of two phenotypes for the disease, Type L and Type H, each with specific characteristics. Type L includes Low elastance, Low lung weight, and Low gas exchange, where the lungs are behaving with nearly normal distensibility, but the patients exhibit severe hypoxemia. Type H includes High elastance, High lung weight, and High recruitment, exhibiting all the hallmarks of severe Acute Respiratory Distress Syndrome (ARDS).
The flexiVent uses the forced oscillation technique (FOT) to quantitatively measure the mechanical properties of the lungs and airways in a wide range of subjects (mice, hamsters, rats, ferrets). It offers numerous clinically relevant respiratory mechanics outcomes to phenotype COVID-19 disease models:
- Static compliance and work of breathing outcomes probe the distensibility of parenchyma, which can be impacted by foreign pathogens and edema. Static compliance and work of breathing can help discern between the normally compliant lung (L phenotype) and the high elastance phenotype (H phenotype) present in COVID-19
- Broadband frequency measurements can link the pathophysiological changes caused by the virus to the central and peripheral lung, helping scientists contextualize their observations. As edema builds in the parenchyma, these outcomes will be disproportionately affected when compared to their conducting airway counterparts.
- Measurements of lung resistance identify lung obstructions and airway constriction which make it difficult for infected subjects to displace air in and out of their lungs.
- Airway hyperresponsiveness (AHR) outcomes track susceptibility to antigens and exaggerated responses of the immune system.
- Lung volume maneuvers and spirometry outcomes provide additional insights into physiological changes, including shortness of breath, and the impacts of treatments.
- Blood oxygenation (SPO2), a parameter tracked closely by clinicians and health care personnel are seamlessly integrated into the software, using monitoring devices during data collection to track and record changes over time.
- Clinicians measure patient lung weight using a CT scanner and the same is possible in animal subjects. Although the acquisition of high-resolution images is technically more challenging due to the small size of the structures and the constant motion of the respiratory system which produces artifacts. The programmable aspect of the flexiVent allows advanced gating techniques by enabling image capture during breath-hold, which greatly increases image contrast and resolution.
The combination of these outcomes provides a 360 view of the subject’s respiratory system that can help scientists develop a mechanistic understanding of the disease progression, evaluate the impact of specific genotypes and endotypes, as test novel therapeutic strategies.
- Aeffner, F., Bolon, B. and Davis, I. C. (2015) ‘Mouse Models of Acute Respiratory Distress Syndrome’, Toxicologic Pathology. SAGE PublicationsSage CA: Los Angeles, CA, 43(8), pp. 1074–1092.
- Szabari, M. V. et al. (2019) ‘Relation between Respiratory Mechanics, Inflammation, and Survival in Experimental Mechanical Ventilation’, American Journal of Respiratory Cell and Molecular Biology. American Thoracic Society, 60(2), pp. 179–188. doi: 10.1165/rcmb.2018-0100OC.
- Cai, Y. et al. (2016) ‘FOXF1 maintains endothelial barrier function and prevents edema after lung injury.’, Science Signaling, 9(424), p. ra40.
- Yang Let al. Three-Dimensional Quantitative Co-Mapping of Pulmonary Morphology and Nanoparticle Distribution with Cellular Resolution in Nondissected Murine Lungs. ACS Nano. 2019 Feb 26;13(2):1029-1041. doi: 10.1021/acsnano.8b07524
- Moskowitzova K et al. Mitochondrial transplantation enhances murine lung viability and recovery after ischemia-reperfusion injury. Am J Physiol Lung Cell Mol Physiol. 2020 Jan 1;318(1):L78-L88. doi: 10.1152/ajplung.00221.2019.
- Umstead TM et al. Lower respiratory tract delivery, airway clearance, and preclinical efficacy of inhaled GM-CSF in a postinfluenza pneumococcal pneumonia model. Am J Physiol Lung Cell Mol Physiol. 2020 Apr 1;318(4):L571-L579. doi: 10.1152/ajplung.00296.2019.
REPRODUCIBLE IN VIVO INHALATION
In COVID-19, the SARS-CoV-2 virus commonly infects the nasal epithelium before spreading to the peripheral lung, where it infects pulmonary epithelial cells and impairs the lung function of the host. The inExpose is a computer-controlled inhalation platform that enables automated, precise, and repeatable aerosol exposure in vivo and in vitro. The system integrates all the necessary components to simplify and automate traditionally complex inhalation studies, including aerosol generators (e.g. nebulizers, dry powder generators), atmospheric monitoring options and in vivo/in vitro exposure sites (e.g. nose only exposure, whole-body exposure, air/liquid exposure models). The small internal volume of the system reduces ramp-up times and minimizes the amount of test article required for preclinical studies aerosol exposure in vivo and in vitro.
The system integrates all the necessary components to simplify and automate traditionally complex inhalation studies, including aerosol generators (e.g. nebulizers, dry powder generators), atmospheric monitoring options, and in vivo/in vitro exposure sites (e.g. nose only exposure, whole-body exposure, air/liquid exposure models). The small internal volume of the system reduces ramp-up times and minimizes the amount of test article required for preclinical studies.
The compact form factor of the inExpose has enabled its use in BSL3 facilities and in standard laboratory environments. For instance, researchers in Utah relied on the inExpose nose-only towers to deliver aerosolized Rift Valley Fever Virus MP-12 (RVFV) in a BSL-3 facility. The use of the Aerogen nebulizer, coupled with automated bias flow profiles, produced consistent exposure atmospheres which resulted in uniform lung deposition and tightly-controlled mortality rates in hamsters 5-7 days post infection (p.i.). On day 2 p.i., the RVFV virus was detected in the systemic circulation of the animals, indicating that the virus was able to reach the blood-gas barrier in the distal airways. The researchers also reported that the nose-only exposure method accurately delivered viral doses and permitted better estimates of the total viral load delivered when compared to alternative approaches.
- Leyva-Grado, V.G et al. (2017). Aerosol administration increases the efficacy of oseltamivir for the treatment of mice infected with influenza viruses. Antiviral Research, 142: 12-15
- Hickerson et al. (2018) Pathogenesis of Rift Valley Fever Virus Aerosol Infection in STAT2 Knockout Hamsters. Viruses. 10(11)
- Wang Q, Sundar I, Li D, et al. E-cigarette-Induced Pulmonary Inflammation and Dysregulated Repair are Mediated by nAChR α7 Receptor: Role of nAChR α7 in ACE2 Covid-19 receptor regulation. Research Square; 2020. DOI: 10.21203/rs.2.23829/v2.
CONSCIOUS VENTILATORY PARAMETERS
Dyspnea is a hallmark of COVID-19, especially for patients whose clinical course of the disease worsens. For these patients, shortness of breath typically sets in between the fourth and eight day of the illness and is sometimes accompanied by a loss of their sense of smell. Changes in breathing patterns have also been reported in studies involving hamsters, transgenic mice and other species infected with coronaviruses.
Breathing patterns can be quantitatively measured in preclinical subjects using instruments called plethysmographs. Plethysmographs enable the study of conscious, freely moving animals uninfluenced by anaesthetic. Plethysmographs are therefore well suited to study the respiratory drive, which depends on both the lungs and nervous system. Coupled with sophisticated breathing pattern analysis software, plethysmographs provide several respiratory outcomes such as breathing frequency, estimates of tidal volumes and other indexes that correlate with unnatural breathing (e.g. penh).
Conscious measurements are inherently more variable than invasive experiments (i.e. flexiVent) since experimental conditions cannot be standardized to the same degree. Plethysmograph-based analyses are often followed-up with more mechanistic, but invasive measurements of respiratory mechanics. Non-invasive and invasive techniques are therefore highly complimentary and can be combined to characterize the respiratory drive and mechanical properties of the lungs.
- Humbered-Smith, J. (2012). Nebulized live-attenuated influenza vaccine provides protection in ferrets at a reduced dose. 30(19).
- Park et al. (2018). Vaccination by microneedle patch with inactivated respiratory syncytial virus and monophosphoryl lipid A enhances the protective efficacy and diminishes inflammatory disease after challenge PLOSone. 13(10).
IN VITRO INHALATION EXPOSURE
Many pharmacologists are looking towards inhaled formulations to treat COVID-19 patients with drugs, vaccines and antibodies. The respiratory route confers beneficial characteristics such as a deposition of unaltered drugs at the primary site of infection to maximize its efficiency, a reduced systemic exposure to avoid undesired side effects in other organs, and a non-invasive delivery. Unfortunately, the development of inhaled drugs is fraught with modelling challenges and high development costs which have limited a more widespread adoption of respiratory drugs.
SCIREQ and the Fraunhofer ITEM have joined forces to develop better research instruments for inhalation scientists. The ExpoCube allows researchers to study the impacts of drugs and other airborne particles on various cells (e.g. A549, Calu-3, primary human bronchial epithelial cells) and tissues (e.g. PCLS) cultured at the air-liquid interface. The innovative design of the ExpoCube permits aerosol deposition profiles that are highly efficient, reproducible and translational. Researchers can therefore explore the efficacity and toxicity of respiratory compounds and establish in vitro/in vivo correlations in a physiologically relevant model.
- Ritter et al. (2020) In vitro inhalation cytotoxicity testing of therapeutic nanosystems for pulmonary infection. Toxicology in vitro. Volume 63
- Wronski et al. (2020) Development of alternative in vitro and ex vivo models for testing of inhalable antibiotics. Fraunhofer Models of Lung Disese Conference poster presentation
- Hiemstra et al. (2018) – Human lung epithelial cell cultures for analysis of inhaled toxicants: Lessons learned and future directions. Toxicology in vitro. Volume 47