Perspective https://doi.org/10.1038/s41684-023-01231-z Animal models to study the neurological manifestations of the post-COVID-19 condition Carla Usai 1,2, Lourdes Mateu3, Christian Brander4,5,6,7, Júlia Vergara-Alert1,2,9 and Joaquim Segalés 1,8,9 ✉ More than 40% of individuals infected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have experienced persistent or relapsing multi-systemic symptoms months after the onset of coronavirus disease 2019 (COVID-19). This post-COVID-19 condition (PCC) has debilitating effects on the daily life of patients and encompasses a broad spectrum of neurological and neuropsychiatric symptoms including olfactory and gustative impairment, difficulty with concentration and short-term memory, sleep disorders and depression. Animal models have been instrumental to understand acute COVID-19 and validate prophylactic and therapeutic interventions. Similarly, studies post-viral clearance in hamsters, mice and nonhuman primates inoculated with SARS-CoV-2 have been useful to unveil some of the aspects of PCC. Transcriptomic alterations in the central nervous system, persistent activation of immune cells and impaired hippocampal neurogenesis seem to have a critical role in the neurological manifestations observed in animal models infected with SARS-CoV-2. Interestingly, the proinflamma- tory transcriptomic profile observed in the central nervous system of SARS-CoV-2-inoculated mice partially overlaps with the pathological changes that affect microglia in humans during Alzheimer’s disease and aging, suggesting shared mechanisms between these conditions. None of the currently available animal models fully replicates PCC in humans; therefore, multiple models, together with the fine-tuning of experimental conditions, will probably be needed to understand the mechanisms of PCC neurological symptoms. Moreover, given that the intrinsic characteristics of the new variants of concern and the immuno- logical status of individuals might influence PCC manifestations, more studies are needed to explore the role of these factors and their combinations in PCC, adding further complexity to the design of experimental models. T he severe acute respiratory syndrome coronavirus 2 Owing to the virtually ubiquitous expression of angiotensin- (SARS-CoV-2) is a highly pathogenic betacoronavirus that converting enzyme 2 (ACE2), the main cell receptor for SARS-CoV-2 emerged in China in late 2019; its rapid spread across the world that is expressed on endothelial and smooth muscle cells, it was sus- resulted in the coronavirus disease 2019 (COVID-19) pandemic1,2. pected that the multi-organ effects of SARS-CoV-2 infection were SARS-CoV-2 shares its genomic organization with other beta- directly linked to viral presence and replication11. However, the coronaviruses, encoding four structural proteins (Spike, Envelope, mechanisms underlying SARS-CoV-2 multi-organ spread are still Membrane and Nucleoprotein), and several nonstructural proteins. unclear because viremia is not a usual outcome of the infection and The structural proteins of SARS-CoV-2 share more than 90% amino has been detected in very few severe cases12–14. As of May 2023, only acid identity with SARS-CoV, except for the Spike (S) protein, which one severe case with high and persistent SARS-CoV-2 viremia lead- displays an insertion of a polybasic furin cleavage site at the junc- ing to meningitis, probably facilitated by blood–brain barrier (BBB) tion between the S1 and S2 subunits of the receptor-binding domain injury, has been well documented15. (RBD) of SARS-CoV-2 (ref. 3). Antibodies against the RBD have As soon as the very first wave of COVID-19 in 2020 waned, it been detected in the majority of patients with COVID-19, often was observed that a proportion of patients were experiencing symp- with neutralizing activity, prompting the early development of RBD toms weeks or months after viral clearance. The most common and S-based vaccines4. post-acute symptoms reported were shortness of breath, fatigue, Since the early phases of the pandemic, it became evident that pain, increased heart rate, anosmia–dysgeusia, headache and cogni- the novel disease was a multi-systemic disorder. Besides the respi- tive and neurological impairment, which can considerably impact ratory outcome, patients affected by COVID-19 may suffer from daily activities and quality of life16–18. This condition initially extrapulmonary manifestations, including hematological, cardio- named ‘long-COVID’, and now known as ‘post-COVID-19 condi- vascular, gastrointestinal, dermatological, endocrine, ophthalmo- tion’ (PCC), was clinically defined by a Delphi consensus process logical and neurological symptoms. Initially, it was unclear whether in 2022 as the presence of symptoms in individuals with a probable all these clinical–pathological features resulted from a widespread or confirmed SARS-CoV-2 infection, 3 months from the onset of replication of the virus itself or a consequence of the virus-induced COVID-19. Such symptoms may occur after recovery from acute inflammatory and immunological responses5–7. Nowadays, both COVID-19 or persist from the initial disease and may fluctuate or hypotheses are considered valid8–10. relapse over time. Such a definition might be different for children19. 1 Unitat Mixta d’Investigació IRTA-UAB en Sanitat Animal, Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain. 2IRTA Programa de Sanitat Animal, Centre de Recerca en Sanitat Animal (CReSA), Campus de la UAB, Bellaterra, Spain. 3 Infectious Disease Service, Germans Trias i Pujol Research Institute and Hospital, Badalona, Spain. 4IrsiCaixa AIDS Research Institute, Hospital Germans Trias i Pujol, Institute for Health Science Research Germans Trias i Pujol (IGTP), Badalona, Spain. 5University of Vic-Central University of Catalonia (UVic-UCC), Vic, Spain. 6CIBERINFEC, Centro de Investigación Biomédica en Red, Instituto de Salud Carlos III, Madrid, Spain. 7ICREA, Barcelona, Spain. 8 Department de Sanitat i Anatomia Animals, Facultat de Veterinària, Campus de la UAB, Bellaterra, Spain. 9These authors contributed equally: Júlia Vergara-Alert, Joaquim Segalés. ✉e-mail: joaquim.segales@irta.cat 202 Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban LAb ANIMAl Perspective Exposure to Symptom Viral 3 months post SARS-CoV-2 onset clearance symptom onset 1 Recovery or Pre-symptomatic phase Acute COVID-19 PCC persisting symptoms 5–13 days 1–2 weeks 1–3 weeks ≥2 months Fig. 1 | Timeline of acute COVID-19 and PCC in humans. The timings stated in each section of the figure indicate the duration of the corresponding phase of the disease. After exposure to SARS-CoV-2, a pre-symptomatic phase characterized by viral replication and early viral shedding precedes the onset of symptoms. The most common symptoms of acute COVID-19 are fever, cough, headache, dyspnea and fatigue; during this phase, the virus is still replicating, and the patient is contagious. In the weeks immediately after viral clearance, symptoms can either remit or persist. The common consensus for PCC is the persistence or new onset of symptoms three months after the beginning of the acute phase, which lasts for at least two months1,19,95–97. Adapted from ref. 82, CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Differences between cohorts reported in the literature also suggest by SARS-CoV) and Middle East respiratory syndrome (MERS, the existence of two populations: (1) patients with symptoms that caused by MERS-CoV) also experienced psychological sequelae persist after the resolution of a severe acute infection, and (2) patients and chronic fatigue after the resolution of the infection, but PCC with symptoms that develop only in the post-acute phase of the seems to be more pleiotropic than post-SARS and post-MERS disease, usually after mild COVID-19 (ref. 20). PCC is also referred syndromes27,28. Whether this is actually the case, or whether this to as post-acute sequelae of COVID-19. A timeline of acute assumption is skewed by the higher number of people infected COVID-19 and PCC is represented in Fig. 1. by SARS-CoV-2, remains to be determined. Factors such as the PCC has a debilitating effect on the daily life of patients, reduc- severity of acute COVID-19, as well as comorbidities or specific ing their ability to work or engage in social and physical activities. demographic characteristics (respiratory disease, body mass index, Given the hundreds of millions of individuals who have become older age and Black or Asian ethnicity), have been associated with infected with SARS-CoV-2, the social implications of the long-term symptom persistence and intensity29. Nonetheless, PCC has been consequences of COVID-19 are of great relevance and add to the reported even in patients who had mild to moderate acute disease30. health burden caused by the pandemic. The societal and public Interestingly, PCC seems to be more frequently associated with the health impact of PCC is even more concerning given that PCC has female gender, while studies have shown that severe acute disease is also been linked to an increased incidence of noncommunicable more prevalent in male patients30–33. diseases. However, the mechanisms of PCC pathogenesis are far The mechanisms probably contributing to the pathogenesis from being satisfactorily unveiled, leaving patients suffering from of PCC fall under five main categories (Fig. 2): (1) persistence this condition without proper treatment. of the whole virus or viral fragments; (2) prolonged inflamma- Animal models have been instrumental in the understanding tion and immunological aberrations; (3) virally induced autoim- of the acute phase of SARS-CoV-2 infection; the development of munity; (4) alterations of the renin–angiotensin system (RAS); adequate models allowing the long-term follow-up of conva­lescent and/or (5) microbiome dysbiosis29,34–36. In addition to the aforemen- animals will be useful to better understand PCC pathology and to tioned mechanisms, post-intensive care syndrome (a condition of design counteracting measures. Among the several species used as impaired physical, cognitive and psychiatric state, irrespective of the animal models of SARS-CoV-2 infection21, some have proven to be critical illness suffered by the patient) is also thought to contrib- suitable for studying various aspects of PCC. How­ever, due to the ute to PCC in patients with severe COVID-19 who had required differences in lifespan and viral kinetics between humans and most intensive care unit hospitalization29,37. Similar to acute COVID-19, laboratory animals, translating the PCC concept to other species PCC encompasses extrapulmonary symptoms such as hemato- can be challenging. In general, it is considered that all animal stud- logical and immune-related, cardiovascular, renal, gastrointestinal ies extending beyond viral clearance and displaying multi-organ and hepatobiliary, endocrine, neurological, ophthalmological and damage could provide important information about the long-term dermatological disorders summarized in Table 1 (refs. 29,34,35,38). In effects of SARS-CoV-2 infection and their mechanisms. Golden this Perspective, we will focus solely on the neurological symptoms Syrian hamsters (GSH), mice (including transgenic or transduced of PCC. with human ACE2 (hACE2)) and nonhuman primates (NHP) have all been used for the study of the long-term neurological sequelae of Neurological symptoms of PCC SARS-CoV-2 infection. Approximately one-third of patients with acute COVID-19 report This Perspective aims to summarize the current knowledge on neurological symptoms including headache, dizziness, anosmia, PCC, with a particular focus on its neurological manifestations, and ageusia, anorexia, myalgia and fatigue; more severe manifestations to discuss the applicability of existing animal models to recapitulate such as stroke, encephalopathy (with or without encephalitis) and some of the symptoms observed in patients with PCC. Guillain–Barré syndrome have also been described5,39. Considering the broad spectrum of neurological manifestations of PCC, differ- Potential mechanisms of PCC ent mechanisms may be involved in the pathogenesis of specific dis- It is estimated that, worldwide, 43% of patients who had SARS- orders. ‘Brain fog’, which is one of the most common self-reported CoV-2 infection have experienced PCC, with regional prevalence symptoms in both acute and post-acute COVID-19, describes a feel- ranging from 11.5% to 75.0% (refs. 22–26). A fraction of patients ing of being mentally slow (bradyphrenia), with confusion, memory who survived severe acute respiratory syndrome (SARS, caused loss and difficulty concentrating40. This phenomenon is very similar Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban 203 Perspective LAb ANIMAl the hallmark neurological symptoms and the associated lesions and study findings between PCC and these neurodegenerative diseases. Animal models for the study of PCC In this Perspective, we considered all studies that followed animals for more than 14 days post-inoculation (dpi) as ‘long-term studies’. GSH. The GSH (Mesocricetus auratus) is a valuable small animal model PCC for the study of SARS-CoV-2 infection, because the GSH is naturally susceptible to infection52. A study has reported multi-organ damage with infiltration of inflammatory cells and necrosis in upper and lower airways, secondary lymphoid organs, the digestive tract, kidneys, adrenal gland and ovaries in this model 5–7 dpi with the SARS-CoV-2/ WH-09/human/2020/CHN isolate. These lesions were very mild in nonrespiratory tissues and could still be present by 18 dpi. In addi- tion, different levels of viral genomic RNA (detected by quantitative PCR with reverse transcription (RT–qPCR) and in situ hybridiza- RAS tion) and/or S and N proteins (detected by immunohistochemistry) were detectable up to 7 dpi in nasopharynges, trachea, lungs, kidneys, testis, vesicular gland, prostate, adrenal gland, spleen and lymph nodes53. Similarly, another study showed that microglial activation and accumulation of hyperphosphorylated tau and α-synuclein Fig. 2 | Putative mechanisms of PCC. From top left in clockwise order: persisted in the brain of SARS-CoV-2-inoculated hamsters after persistence of the whole virus or viral fragments; prolonged inflammation viral clearance (14 dpi), in the olfactory bulb (OB) and in the cor- and immunological aberrations; virally induced autoimmunity; alterations tical neurons, respectively, without any evidence of neuroinvasion of the RAS; microbiome dysbiosis; post-intensive care syndrome. (as shown by a lack of detectable S protein by immunohistochem- istry), which suggests these pathological features contribute to the long-lasting neurological signs of SARS-CoV-2 infection54. to the so-called chemo-brain, often experienced by patients with Additionally, a long-term study compared the systemic effects cancer during or after chemotherapy; this condition also known of SARS-CoV-2 (USA-WA1/2020 isolate) with those of influenza A as cancer-related or chemotherapy-related cognitive impairment is virus (IAV) pandemic strain H1N1 (A/California/04/2009 isolate) believed to be caused by neuroinflammation41,42. in GSH for up to 31 dpi (a timepoint when infectious particles for So far, evidence of SARS-CoV-2 infection in the central nervous either infection were no longer detectable in any of the investigated system (CNS) is limited to post-mortem samples from severe cases; tissues)55. This study found that the histological alterations induced CNS SARS-CoV-2 infection is also supported by transcriptomic by SARS-CoV-2 in the lungs (peribronchiolar metaplasia and mono- data reporting ACE2 expression in both neuronal and nonneuronal cyte and neutrophil infiltration in the alveolar space) and kidneys cells of the human brain43,44. Although not confirmed in humans or (areas of tubular atrophy and presence of proteinaceous fluid in experimental animal models following SARS-CoV-2 natural infec- the interstitial space) were more severe than those induced by IAV tion, studies have shown that the soluble monomeric S1 subunit of and persisted for a longer period after the resolution of infection. the Spike protein can cross the mouse BBB in vivo after both intra- Moreover, the transcriptomic analysis of the olfactory tissue and of venous and intranasal administration, and spread to different areas different areas of the CNS revealed that, while both viruses caused a of the brain45. Given that coronavirus S1 subunits are often shed similar pattern of perturbations in the CNS that persisted for over a from the virions by the action of host proteases located on cell sur- month, SARS-CoV-2 induced a unique profile in the olfactory tissue face, it cannot be ruled out that soluble SARS-CoV-2 Spike S1 trig- of GSH. Thirty-one days after SARS-CoV-2 infection, genes associ- gers inflammation and tissue damage, even in the absence of intact ated with microglial activation and macrophage infiltration (these viral particles. increases in gene expression were confirmed by immunostaining), On the other hand, the presence of the virus itself has seldom T cell recruitment and activation, type I interferon and chemokine been demonstrated in the cerebrospinal fluid (CSF) of patients; an response were upregulated, while genes involved in sensory per- altered CSF cytokine profile (with increased levels of IL-6, IL-8 and ception and olfactory capabilities were downregulated. At 26 dpi, IL-10) has also been reported in one case of SARS-CoV-2-associated SARS-CoV-2-infected GSH also showed a reduction in burying encephalitis46–48. Furthermore, a sizable single-cell transcriptomic activity when subjected to the marble-burying assay compared with study analyzing the brains of patients who succumbed to COVID-19 IAV-infected and noninfected hamsters, demonstrating a change in did not find any molecular traces of SARS-CoV-2 in the brain, but did this spontaneous rodent behavior following SARS-CoV-2 infection55. identify the upregulation of inflammation-related genes (predomi- Interestingly, injury and chronic inflammation of the OB have been nantly in astrocytes), antiviral defense markers (interferon-induced previously linked to neurobehavioral disorders because OB function transmembrane protein 3 among others) and T cell infiltration49. The can impact sensory, emotional and cognitive processes given the transcriptomic perturbations identified in the study, in particular proximity of the OB to the limbic system responsible for such reac- those suggesting compromised neurotransmission and information tions56. Therefore, the behavioral changes observed in SARS-CoV- processing, share features with pathological cell states of neuro- 2-infected GSH suggest that long-term OB inflammation causes degenerative disease. Moreover, Alzheimer’s-like features such as neurodegenerative changes that are compatible with PCC neurologi- excessive levels of reactive oxygen species (ROS), activation of the cal symptoms. This observation is in line with evidence of gray mat- TGF-β pathway and increased tau phosphorylation have been found ter loss in the limbic and olfactory cortical areas of patients who have in the brains of these patients. It is interesting to note that olfactory recovered from COVID-19, even with mild acute manifestations55,57. dysfunctions are also common symptoms of early Alzheimer’s In addition, SARS-CoV-2 infection in GSH has been linked to and Parkinson’s diseases, further suggesting a shared mechanism mechanical hypersensitivity during both the acute and post-acute between PCC and neurodegenerative diseases50,51. Table 2 compares phases (28 dpi) of SARS-CoV-2 infection, which resembles the 204 Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban LAb ANIMAl Perspective Table 1 | Most common manifestations of both acute and Table 1 | Most common manifestations of both acute and persistent COVID-19 in humans persistent COVID-19 in humans (Continued) Organ or system Lesions and symptoms Organ or system Lesions and symptoms Respiratory • Dyspnea Multi-system • Fever • Decreased exercise capacity inflammatory syndrome • Abdominal pain • Chronic cough in children (MIS-C) • Vomiting • Exacerbation of asthma • Diarrhea Cardiovascular • Palpitation • Skin rash • Chest pain and tightness • Mucocutaneous lesions • Arrhythmias • Hypotension • Postural orthostatic tachycardia • Cardiovascular and neurological symptoms • Orthostatic hypotension • Vasovagal syncope The information listed in this table was extracted from refs. 29,34,35,38. • Myocarditis and/or pericarditis • Myocardial ischemia • Ventricular dysfunction somatosensory abnormalities observed in patients with PCC. This • Venous thromboembolic events prolonged hypersensitivity coincided with the differential expres- Cerebrovascular • Ischemic stroke sion of several genes associated with neuro-oncological and neu- Renal • Impaired renal function rodegenerative disorders (including glioblastoma, Alzheimer’s and • Severe acute kidney injury Parkinson’s diseases and neurilemmoma) compared with nonin- • Microhematuria fected hamsters. The same transcriptomic analysis predicted the Endocrine • New onset or worsening of existing diabetes contribution of macrophages and Schwann cells, already known mellitus to be involved in neuropathic pain and hypersensitivity, to the • Glycemic abnormalities SARS-CoV-2-induced proinflammatory state. Interestingly, these • Diabetic ketoacidosis results were obtained long after viral clearance, suggesting that the • Hormonal abnormalities presence of viral transcript (but not infectious particles) and type I • Subacute thyroiditis interferon response in the peripheral nervous system (dorsal root • Autoimmune thyroiditis • Bone demineralization ganglia and spinal cord) during the first 24 h post-inoculation was • Lipid abnormalities sufficient to induce persistent maladaptive neuronal responses58. • Hot and cold sensation Gastrointestinal and • Abdominal pain Mice. Given the substantial differences in amino acid sequences hepatobiliary • Nausea between human and mouse ACE2, the latter is not a functional • Loss of appetite receptor for the ancestral SARS-CoV-2 strain. hACE2-transgenic • Altered bowel motility (hACE2-tg) mice, transient hACE2 expression and virus adaptation to • Irritable bowel syndrome the mouse are among the strategies employed to circumvent this • Dysphagia limitation21. In addition, different humanized mice have been used as • Hepatitis models of PCC, with a particular focus on the lungs and the respiratory Dermatological • Hair loss tract. MISTRG6 humanized mice, reconstituted with human immune • Skin rash cells and transiently expressing hACE2 via an adeno-associated • Nail alterations • Papulo-squamous eruptions (pernio- or vector (AAV-hACE2), have been used to reproduce the human innate chilblain-like) and adaptive responses to SARS-CoV-2 infection. Results from this Musculoskeletal • Myalgia model have suggested that human immune cells contribute to the • Arthralgia pathological inflammatory response in the lungs and to viral RNA • Asthenia persistence, which was detected until 35 dpi, either as a result of Neuropsychiatric • Migraine-like headache direct infection of ACE2-expressing myeloid cells or as a residue of • Olfactory and gustatory impairments phagocyted infected epithelial cells59. Similar results have been • Difficulties with concentration and short-term obtained in DRAGA (HLA-A2.HLA-DR4.Rag1KO.IL2RγcKO.NOD) memory humanized mice, which when infused with human hematopoietic • Dizziness stem cells from cord blood reconstitute both a functional human • Imbalance immune system and hACE2-expressing human epithelial and endo- • Encephalitis • Myelopathy thelial cells in the lung and upper respiratory airways. Inoculated • Neuropathy animals were followed for 25 days, which revealed mild to severe • Depressive symptoms lung pathology with inflammatory infiltrate and alveolar damage • Anxiety until the end of the study60. • Insomnia and sleep disorders A longer-term study monitored neuroinflammation up to 7 weeks • Mood disorders post-infection (wpi) with SARS-CoV-2 (isolate USA-WA1/2020) in • Post-traumatic stress disorder wild-type CD1 mice transduced with AAV-hACE2 in the trachea Miscellaneous • Fever and lungs by intratracheal injection61. This model simulates a mild • Hearing loss respiratory infection with no to moderate lung pathology and no • Tinnitus viral neuroinvasion. Increased protein levels of proinflammatory • Red eyes cytokines (IFNγ, IL-6, TNF, CXCL10, CCL7, CCL2, CCL11, BAFF • Sore throat and GMCSF) were observed both in the serum and CSF of infected • Light and sound sensitivity mice as early as 7 dpi. Interestingly, CSF levels of C–C motif chemo- • Blurry vision kine 11 (CCL11), known to limit neurogenesis and contribute to • Excessive bruising cognitive impairment, further increased until at least 7 wpi, while Continued they normalized in the serum at the same timepoint. This increase Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban 205 Perspective LAb ANIMAl model lacking mature lymphocytes). Histological analysis of aged Table 2 | Shared neurological symptoms and lesions between Rag–/– brain samples revealed prominent lymphocyte perivascular PCC, Parkinson’s and Alzheimer’s diseases cuffing, indicating inflammation of blood vessels at 30-days post inoculation64. Long-term studies using the mouse adapted MA10 Condition strain in immunocompetent mice are needed to confirm these find- PCC Parkinson’s Alzheimer’s ings and to determine the extent of brain damage and neurological disease disease impairment caused by SARS-CoV-2. Symptoms Headache ✓ – – NHPs. NHPs have been successfully used for the study of human brain disorders, viral infections and respiratory diseases, taking Dizziness ✓ – – advantage of their phylogenetic proximity and anatomical and Anosmia ✓ ✓ ✓ physiological similarity to humans65–67. Over the past 3 years, NHPs Ageusia ✓ – – have been of great interest in the study of both acute and post-acute SARS-CoV-2 infection21,68. Anorexia ✓ – – A study following rhesus macaques and African green monkeys Myalgia ✓ ✓ – for 28 days after infection with the SARS-CoV-2 USA-WA1/2020 Fatigue ✓ – – virus strain revealed neuroinflammation, as well as brain hypoxia, Bradyphrenia ✓ – ✓ microhemorrhages and lesions similar to autopsy findings from patients who died after SARS-CoV-2 infection69. Neuroinflammation Confusion ✓ – ✓ signatures, determined by the expression of the pan-microglial Memory loss ✓ – ✓ marker IBA1 and glial fibrillary acidic protein (GFAP), were greater Impaired concentration ✓ – ✓ in infected animals than in mock-infected control animals, and Sleep disorders ✓ ✓ – revealed more pronounced astrocytic hypertrophy in these animals. In addition, the inoculated animals in this study displayed neuronal Depression ✓ ✓ – morphological changes in the cerebellum and the brainstem, such Lesions and study findings as degenerated Purkinje neurons exhibiting cellular blebs and cyto- Altered CSF cytokine profile ✓ ✓ ✓ plasmic vacuoles, and adjacent pyknotic glial cells with condensed Microglia activation ✓ – ✓ nuclei. These lesions were found concomitantly with cleaved cas- pase 3 staining, indicating that apoptosis was one of the mechanisms Astrocyte activation ✓ – ✓ underlying such neuronal degeneration. Very low levels of viral T cell infiltration ✓ – – S RNA were detected in the brain of some of the infected animals; Excessive ROS ✓ ✓ ✓ S expression was always limited to vascular endothelial cells, with Activated TGF-β pathway ✓ – ✓ no involvement of parenchymal cells. The virus was not detected in the CSF of infected NHPs, which might suggest hematological Tau hyper-phosphorylation ✓ – ✓ dissemination of the virus to the brain. Interestingly, hypoxic– α-Synuclein hyper-phosphorylation ✓ ✓ ✓ ischemic injuries, with areas of intense positivity for the hypoxia- The information listed in this table was extracted from refs. 5,39,40,46,47,49–51,61,92,93. inducible factor 1α (HIF1α) were detected within and around blood vessels in the brain of infected animals69. Given that the expression of this protein can be upregulated by proinflammatory cytokines in CCL11 CSF levels was associated with increased microglial acti- such as TNF — a factor abundant in the serum of patients with vation in the subcortical white matter (as indicated by increased severe COVID-19 — and by the SARS-CoV-2 protein translated expression of allograft inflammatory factor 1 (AIF1; also known from ORF3 (ref. 10), at least in vitro, hypoxia could be both a direct as IBA1) marker in CD68+ cells), which persisted during the entire and an indirect effect of SARS-CoV-2 infection. It is worth men- study period. Moreover, transcriptomic analysis at 7 dpi and 7 wpi tioning that signs of hypoxia were also detected in animals that revealed an increase in the abundance of a chemokine-expressing did not develop severe respiratory disease, suggesting that it could subpopulation of the microglia as well as an altered profile in the be one of the causes of neurological disorder even after mild homeostatic microglia subpopulation, with upregulation of genes COVID-19, as observed in patients70,71. associated with cytotoxicity, inflammation, antigen processing and The advanced age (13–21 years) of the NHPs used for this study presentation. Interestingly, this gene signature partially overlaps with might have acted as a confounding factor: all morphological changes the expression profiles of pathological cell states such as disease- and inflammatory features were also observed in mock-infected associatedmicrogliaobservedinAlzheimer’sdisease,andwhite-matter- animals, although at a lower level. It is therefore probable that the associated microglia observed in aging. SARS-CoV-2-infected mice infection contributed to the worsening of pre-existing, age-related also exhibited a persistent decrease in neurogenesis in the hippo- lesions rather than being their triggering cause. Moreover, none of campus and a decrease in myelinated axon density during the entire the 12 infected animals displayed clinical signs of neurological dis- study period, as well as a decrease in mature oligodendrocytes orders, except for one female African green monkey that was found (ASPA+ CC1+ population) at 7 dpi, which recovered by 7 wpi. The recumbent and marginally responsive to stimuli at 8 dpi (ref. 69). magnitude of myelin loss was strikingly similar to that observed after chemotherapy exposure, confirming the similarity between Animal models with CNS involvement during the acute phase ‘COVID-19 brain fog’ and ‘chemo fog’61. of COVID-19. Considering the small number of models currently Studies using a mouse-adapted SARS-CoV-2 strain (MA10) available for the study of the long-term neurological manifestations have reported productive infection with severe lung injury and of SARS-CoV-2 infection, it is worth considering all preclinical mortality in standard laboratory mice62,63. Although no virus was studies that have shown either neuroinvasion or neuroinflamma- detected in the brain of intranasally infected animals, this model tion during the acute phase of the infection. showed an upregulation of proinflammatory cytokine messenger In one study, infectious viral particles were isolated from differ- RNAs, altered BBB permeability and an increase in IBA1+ microg- ent areas of the brain of SARS-CoV-2-infected hamsters 4 dpi, indi- lial cells in the cortical region of 1-year-old Rag–/– mice (a mouse cating that the virus can reach and productively replicate in the CNS 206 Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban LAb ANIMAl Perspective Table 3 | List of potential PCC animal models Model Similarities to human Differences with the human Neurological manifestations Technical considerations References disease disease and laboratory findings GSH • Naturally susceptible to • Rapid humoral neutralizing • Accumulation of hyper • Easy handling in the 21,52–55, SARS-CoV-2 infection immune response -phosphorylated tau and laboratory 72,86 • Multi-organ damage • Severe disease α-synuclein in cortical • Lack of hamster- • No evidence of manifestations are limited neurons specific research neuroinvasion by quick spontaneous • Transcriptomic tools for the study of • Spontaneous viral viral clearance perturbations in the CNS immune activation clearance • Behavioral changes Wild-type • Susceptible to the Alpha, • Not susceptible to • Neuroinflammation • Easy handling in the 21,64,84–86 laboratory mice Beta, and Omicron SARS-CoV-2 ancestral (observed with laboratory SARS-CoV-2 variants strain infection SARS-CoV-2 mouse • Wide availability of with mild disease course adapted strain; studies research tools limited to the acute phase) K18-hACE2-tg • Susceptible to • Massive neuroinvasion • Nonsuppurative • Easy handling in the 21,43,73, mice SARS-CoV-2 infection and neuroinflammation meningoencephalitis laboratory 74,76,78 (multiple variants tested) • Lethal infection without • Diffuse astrogliosis and • Wide availability of viral clearance by 7 dpi microgliosis research tools • Apathy, depression, trembling and seizures hACE2-KI mice • Susceptible to • Minimal lung pathology • Low amounts of viral RNA • Easy handling in the 78,94 SARS-CoV-2 infection occasionally detected in laboratory brain samples • Wide availability of • Studies limited to the acute research tools phase AAV-hACE2 • Mild to moderate • Targeted expression of • Increased CSF levels of • Variability in results 59,61 transduced mice respiratory disease hACE2 determines which CCL11 might depend on the (multiple strains) • No neuroinvasion organs and tissues are • Increased microglial AAV batch. infected by SARS-CoV-2 activation in the subcortical • If used in different • Transient hACE2 white matter mouse strains (such expression, dependent on • Decreased neurogenesis in as different humanized the turnover rate of the the hippocampus mice), the model transduced cells • Myelin loss allows for the study of specific pathways or immune cell types NHPs • Naturally susceptible to • Different NHP species • Neuroinflammation • Ethical limitations 68,69,80 SARS-CoV-2 infection display a heterogeneous • Brain hypoxia • Handling in the • Neuroinflammation spectrum of COVID-19 • Neuronal morphological laboratory may be observed without evidence symptoms changes in cerebellum and difficult depending on of neuroinvasion brainstem the species of choice • Spontaneous viral clearance AAV, adeno-associated virus. of this animal model. Viral particles were also detected in the same between K18-hACE2 mice and humans, it is possible to achieve a animals' olfactory neuroepithelium, along with infiltrating myeloid nonfatal disease via a fine modulation of the experimental inoculation cells and upregulated expression of proinflammatory markers72. (that is, use of a sublethal dose or a less pathological variant, and use On the other hand, K18-hACE2-tg mice — mice carrying the of a more natural administration route such as aerosolization), which hACE2 sequence under the control of the cytokeratin-18 (K18) might extend the survival of the infected animals sufficiently to allow gene promoter — are susceptible to SARS-CoV-2 infection, develop the study of long-term consequences of SARS-CoV-2 infection78. severe lesions including massive brain invasion, neuronal death and Less extensively used than the K18-hACE2 mice, hACE2 neuroinflammation, and succumb by 5–7 dpi when given a dose of knock-in (KI) mice are considered a more physiologically relevant either 103 TCID50 SARS-CoV-2 or 105 plaque forming unit intrana- model, because the expression of the human receptor is under the sally73–75. This model has been successfully used for testing vaccines control of the endogenous promoter of mACE2, thereby prevent- and antivirals but does not properly replicate human COVID-19. ing ectopic expression. One short-term study using hACE2-KI mice However, a short-term study in 2022 has now shown that the adminis­ has shown very low levels of SARS-CoV-2 RNA in the brain and tration of aerosolized SARS-CoV-2 to K18-hACE2 mice leads to OB of some of the inoculated mice at 7 dpi (ref. 79). The presence of efficient respiratory infection and anosmia without fatal neuroinvasion, replicating particles and the long-term effect of viral presence in the even at the same dose76. Similarly, a study showed that K18-hACE2 brain have not been investigated in these animals. mice intranasally inoculated with 103 plaque forming unit of the An early study on rhesus macaques infected with SARS-CoV-2 BA.1 (Omicron) variant develop mild respiratory pathology without (isolate KMS1/2020) also showed neuroinvasion. Viral RNA was neuro­invasion up to 7 dpi (ref. 77). Notwithstanding the differences in detected by RT–qPCR in the brain and spinal cord of the NHPs on ACE2 expression and in the severity of SARS-CoV-2-induced disease day 9 post-intranasal infection, while it was detected as soon as 2 dpi Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban 207 Perspective LAb ANIMAl and during the entire study period in the lungs and trachea. The ani- condition observed in nonvaccinated naïve patients remains a ques- mals in the study had transient viremia, suggesting systemic spread tion to be answered87. All these factors might limit the translatability of the virus through the bloodstream. Histological analysis of the of the currently available animal models to the entire affected popu- CNS also revealed hyperemia and edema in some of the samples80, lation and call for the development of models exploring the influ- but those findings might not be specific to SARS-CoV-2 infection. ence of different vaccination status and VOCs on PCC; however, Interestingly, ferrets (Mustela putorius furo) are susceptible to such models might require longer follow-up time and increase the SARS-CoV-2 infection, but they develop only minor clinical signs complexity of the studies. Translating the neurological and cogni- even in the presence of mild histological lesions in the lower respi- tive symptoms experienced by patients with COVID-19 and PCC ratory tract. In a study aimed at determining the effect of age in into objective and measurable outcomes for animal studies is a nec- SARS-CoV-2-induced pathology and immunity, viral RNA was essary step, which adds another level of complexity to the preclinical detected in extra respiratory tissues including the brain of both studies on PCC. young and adult animals up to 5 dpi, but the production of infec- A practical aspect to be considered when carrying out long-term tious virus was not demonstrated81. Moreover, no evidence of neu- in vivo studies of SARS-CoV-2 infection is the requirement for bio- rological clinical signs was observed. safety level 3 animal facilities and adequately trained personnel. Pseudoviruses have been used to develop in vivo models of infec- Conclusion and future perspectives tion by highly pathogenic agents including SARS-CoV-2 (refs. 88–90), Currently available animal models for SARS-CoV-2 infection should but to our knowledge, this technology has not been used yet to study be considered useful tools for understanding the long-term conse- the long-term effect of COVID-19. Pseudoviruses are a promising quences of SARS-CoV-2 infection82 (Table 3). In particular, studies alternative to the use of infectious virus, which would also circum- in hamster and mouse models have confirmed that some of the neu- vent the biosafety level 3 requirements. rological manifestations of PCC share molecular mechanisms with The heterogeneity of PCC manifestations observed in patients neurodegenerative disease characterized by neuroinflammation and the finding that PCC is apparently independent of the sever- and proteinopathy53,54,61. Promising results have also been obtained ity of the prior acute infection makes it unlikely that a single ani- from long-term studies in NHPs, which share similar infection mal model will be able to fully recapitulate the human disease. This kinetics with humans. However, data on neurological signs in NHPs problem could be overcome by challenging the animals with a range following SARS-CoV-2 infection are still scarce, and more studies of viral doses to induce a wider spectrum of clinical signs. In addi- are needed to determine the actual usefulness of this model69. On tion, animal models and humans do not always perfectly match in the other hand, ferrets seem not to be an adequate model for the terms of genetic background, lifespan and aging processes, among study of the long-term consequences of SARS-CoV-2 infection, other factors, which might affect responses to SARS-CoV-2. As because they do not develop any clinical signs, and active viral repli­ already pointed out by other authors, a useful strategy to experi- cation has not been clearly demonstrated in lung samples81. Given mentally reproduce such variability would be to introduce system- that none of the models presented in this Perspective seems to com- atic heterogeneity in the study design91. prehensively phenocopy PCC in humans, it is reasonable to foresee Given that none of the currently available animal models that multiple animal models will be necessary to unveil the several seems to fully replicate PCC, it seems sensible to use multiple models aspects of this multi-systemic condition. to study different aspects of this condition. In particular, we think When focusing on the neurological manifestations of PCC, it is that GSHs are suitable for profiling neuroinflammation signature important to consider that productive SARS-CoV-2 infection of the after SARS-CoV-2 inoculation, while mouse models would be the CNS might not be necessary for the development of cognitive and best tool to test sublethal doses of SARS-CoV-2 and less virulent behavioral signs, because these manifestations might be triggered VOCs as a mean to extend animal survival and observe the potential by inflammatory signals originating in other tissues. This is particu- long-term effects of the infection. larly relevant for K18-hACE2-tg mice, one of the most used small animal models for SARS-CoV-2 infection so far, which develop Received: 11 January 2023; Accepted: 14 July 2023; massive lethal neuroinvasion upon intranasal SARS-CoV-2 inocu- Published online: 24 August 2023 lation; this limitation has been overcome by the use of an aerosol- ized inoculum instead of direct intranasal inoculation73,74,76. References All studies published so far have analyzed the long-term con- 1. Huang, C. et al. Clinical features of patients infected with 2019 novel sequences of SARS-CoV-2 infections in naïve animals using the coronavirus in Wuhan, China. Lancet 395, 497–506 (2020). ancestral strain. However, after the first wave of the COVID-19 2. WHO Director-General’s Opening Remarks at the Media Briefing on COVID-19 - 11 March 2020 (World Health Organization, 2020). pandemic, several variants of concern (VOCs) have appeared 3. Hu, B., Guo, H., Zhou, P. & Shi, Z. L. Characteristics of SARS-CoV-2 and with different characteristics (such as transmissibility, replication COVID-19. Nat. Rev. Microbiol. 19, 141–154 (2020). abilities and lesion severity)83 that may influence the intensity and/ 4. Moore, J. P. & Klasse, P. J. COVID-19 vaccines: ‘warp speed’ needs mind or duration of possible PCC manifestations. In this regard, it is melds, not warped minds. J. Virol. 94, e01083-20 (2020). worth mentioning that the B.1.1.7 (Alpha), B.1.351 (Beta) and BA.1.1 5. Gupta, A. et al. Extrapulmonary manifestations of COVID-19. Nat. Med. 26, 1017–1032 (2020). (Omicron) variants can infect wild-type mice causing a mild disease 6. Johnson, K. D. et al. Pulmonary and extra-pulmonary clinical manifestations without neurological signs during the acute phase84–86. This points of COVID-19. Front. Med. 7, 526 (2020). to the possibility of developing mouse models to study the infection 7. Song, W.-C. & FitzGerald, G. A. COVID-19, microangiopathy, hemostatic by specific VOCs of SARS-CoV-2, which would be extremely help- activation, and complement. J. Clin. Invest. 130, 3950–3953 (2020). ful for characterizing the long-term consequences of SARS-CoV-2 8. Stein, S. R. et al. SARS-CoV-2 infection and persistence in the human body and brain at autopsy. Nature 612, 758–763 (2022). in people infected after the first wave of the pandemic. The use of 9. Torge, D., Bernardi, S., Arcangeli, M. & Bianchi, S. Histopathological features such models might help to dissect the molecular pathways associ- of SARS-CoV-2 in extrapulmonary organ infection: a systematic review of ated with different PCC symptoms when they are observed specifi- literature. Pathogens 11, 867 (2022). cally with certain VOCs. 10. Tian, M. et al. HIF-1α promotes SARS-CoV-2 infection and aggravates Moreover, the global COVID-19 vaccine rollout has reduced the inflammatory responses to COVID-19. Signal Transduct. Target. Ther. 6, 308 (2021). percentage of the immunologically naïve population. An interesting 11. Hamming, I. et al. Tissue distribution of ACE2 protein, the functional point to consider is that PCC occurs also after vaccine breakthrough receptor for SARS coronavirus. A first step in understanding SARS infections; whether PCC would then substantially differ from the pathogenesis. J. Pathol. 203, 631–637 (2004). 208 Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban LAb ANIMAl Perspective 12. Wang, W. et al. Detection of SARS-CoV-2 in different types of clinical 44. Chen, R. et al. The spatial and cell-type distribution of SARS-CoV-2 receptor specimens. J. Am. Med. Assoc. 323, 1843–1844 (2020). ACE2 in the human and mouse brains. Front. Neurol. 11, 573095 (2020). 13. Bwire, G. M., Majigo, M. V., Njiro, B. J. & Mawazo, A. Detection profile of 45. Rhea, E. M. et al. The S1 protein of SARS-CoV-2 crosses the blood–brain SARS-CoV-2 using RT-PCR in different types of clinical specimens: a barrier in mice. Nat. Neurosci. 24, 368–378 (2021). systematic review and meta-analysis. J. Med. Virol. 93, 719–725 (2021). 46. Al Saiegh, F. et al. Status of SARS-CoV-2 in cerebrospinal fluid of 14. Roy-Vallejo, E. et al. SARS-CoV-2 viremia precedes an IL6 response in severe patients with COVID-19 and stroke. J. Neurol. Neurosurg. Psychiatry 91, COVID-19 patients: results of a longitudinal prospective cohort. Front. Med. 846–848 (2020). 9, 855639 (2022). 47. Benameur, K. et al. Encephalopathy and encephalitis associated with 15. D'Abramo, A. et al. The first case of meningitis associated to SARS- cerebrospinal fluid cytokine alterations and coronavirus disease, Atlanta, Coronavirus-2 BA.2 variant infection with persistent viremia. Int. J. Infect. Dis. Georgia, USA, 2020. Emerg. Infect. Dis. 26, 2016–2021 (2020). 124, 38–40 (2022). 48. Huang, Y. H., Jiang, D. & Huang, J. T. SARS-CoV-2 detected in cerebrospinal 16. Mantovani, A. et al. Long Covid: where we stand and challenges ahead. fluid by PCR in a case of COVID-19 encephalitis. Brain. Behav. Immun. 87, Cell Death Differ. 29, 1891–1900 (2022). 149 (2020). 17. de Oliveira, J. F. et al. Persistent symptoms, quality of life and risk factors 49. Yang, A. C. et al. Dysregulation of brain and choroid plexus cell types in in long COVID: a cross-sectional study of hospitalized patients in Brazil. severe COVID-19. Nature 595, 565–571 (2021). Int. J. Infect. Dis. 122, 1044–1051 (2022). 50. Reiken, S. et al. Alzheimer’s-like signaling in brains of COVID-19 patients. 18. Wurz, A. et al. ‘I feel like my body is broken’: exploring the experiences of Alzheimers Dement. J. Alzheimers Assoc. 18, 955–965 (2022). people living with long COVID. Qual. Life Res. 31, 3339–3354 (2022). 51. Kay, L. M. COVID-19 and olfactory dysfunction: a looming wave of 19. Soriano, J. B., Murthy, S., Marshall, J. C., Relan, P. & Diaz, J. V. A clinical case dementia? J. Neurophysiol. 128, 436–444 (2022). definition of post-COVID-19 condition by a Delphi consensus. Lancet Infect. 52. Chan, J. F. W. et al. Simulation of the clinical and pathological manifestations Dis. 22, e102–e107 (2022). of coronavirus disease 2019 (COVID-19) in a Golden Syrian hamster model: 20. Bahmer, T. et al. Severity, predictors and clinical correlates of post-COVID implications for disease pathogenesis and transmissibility. Clin. Infect. Dis. 71, syndrome (PCS) in Germany: a prospective, multi-centre, population-based 2428–2446 (2020). cohort study. EClinicalMedicine 51, 101549 (2022). 53. Song, Z. et al. SARS-CoV-2 causes a systemically multiple organs damages 21. Muñoz-Fontela, C. et al. Animal models for COVID-19. Nature 586, and dissemination in hamsters. Front. Microbiol. 11, 618891 (2021). 509–515 (2020). 54. Käufer, C. et al. Microgliosis and neuronal proteinopathy in brain persist 22. Chen, C. et al. Global prevalence of post-coronavirus disease 2019 beyond viral clearance in SARS-CoV-2 hamster model. EBioMedicine 79, (COVID-19) condition or long COVID: a meta-analysis and systematic 103999–103999 (2022). review. J. Infect. Dis. 226, 1593–1607 (2022). 55. Frere, J. J. et al. SARS-CoV-2 infection in hamsters and humans results in 23. Moreno-Pérez, O. et al. Post-acute COVID-19 syndrome. Incidence and risk lasting and unique systemic perturbations post recovery. Sci. Transl Med. 14, factors: a Mediterranean cohort study. J. Infect. 82, 378–383 (2021). eabq3059 (2022). 24. Huang, C. et al. 6-Month consequences of COVID-19 in patients discharged 56. Hellweg, R., Zueger, M., Fink, K., Hörtnagl, H. & Gass, P. Olfactory from hospital: a cohort study. Lancet 401, e21–e33 (2023). bulbectomy in mice leads to increased BDNF levels and decreased serotonin 25. Ballering, A. V., van Zon, S. K. R., olde Hartman, T. C. & Rosmalen, J. G. M. turnover in depression-related brain areas. Neurobiol. Dis. 25, 1–7 (2007). Persistence of somatic symptoms after COVID-19 in the Netherlands: an 57. Douaud, G. et al. SARS-CoV-2 is associated with changes in brain structure observational cohort study. Lancet 400, 452–461 (2022). in UK Biobank. Nature 604, 697–707 (2022). 26. Perlis, R. H. et al. Prevalence and correlates of Long COVID symptoms 58. Serafini, R. A. et al. SARS-CoV-2 airway infection results in the development among US adults. JAMA Netw. Open 5, e2238804 (2022). of somatosensory abnormalities in a hamster model. Sci. Signal. 16, 27. Lam, M. H.-B. et al. Mental morbidities and chronic fatigue in severe acute eade4984 (2023). respiratory syndrome survivors: long-term follow-up. Arch. Intern. Med. 169, 59. Sefik, E. et al. A humanized mouse model of chronic COVID-19. 2142–2147 (2009). Nat. Biotechnol. 40, 906–920 (2021). 28. Lee, S. H. et al. Depression as a mediator of chronic fatigue and post- 60. Brumeanu, T.-D. et al. Human-Immune-System (HIS) humanized mouse traumatic stress symptoms in Middle East respiratory syndrome survivors. model (DRAGA: HLA-A2.HLA-DR4.Rag1KO.IL-2RγcKO.NOD) for Psychiatry Investig. 16, 59–64 (2019). COVID-19. Hum. Vaccines Immunother. 18, 2048622 (2022). 29. Nalbandian, A. et al. Post-acute COVID-19 syndrome. Nat. Med. 27, 61. Fernández-Castañeda, A. et al. Mild respiratory COVID can cause 601–615 (2021). multi-lineage neural cell and myelin dysregulation. Cell 185, 2452–2468. 30. Blomberg, B. et al. Long COVID in a prospective cohort of home-isolated e16 (2022). patients. Nat. Med. 27, 1607–1613 (2021). 62. Gu, H. et al. Adaptation of SARS-CoV-2 in BALB/c mice for testing vaccine 31. Sudre, C. H. et al. Attributes and predictors of long COVID. Nat. Med. 27, efficacy. Science 369, 1603–1607 (2020). 626–631 (2021). 63. Leist, S. R. et al. A mouse-adapted SARS-CoV-2 induces acute lung injury 32. Fumagalli, C. et al. Factors associated with persistence of symptoms 1 year and mortality in standard laboratory mice. Cell 183, 1070–1085.e12 (2020). after COVID-19: a longitudinal, prospective phone-based interview follow-up 64. Amruta, N. et al. Mouse adapted SARS-CoV-2 (MA10) viral infection cohort study. Eur. J. Intern. Med. 97, 36–41 (2022). induces neuroinflammation in standard laboratory mice. Viruses 15, 33. Dorjee, K., Kim, H., Bonomo, E. & Dolma, R. Prevalence and predictors of 114 (2022). death and severe disease in patients hospitalized due to COVID-19: A 65. Buffalo, E. A., Movshon, J. A. & Wurtz, R. H. From basic brain research comprehensive systematic review and meta-analysis of 77 studies and 38,000 to treating human brain disorders. Proc. Natl Acad. Sci. USA 116, patients. PLoS ONE 15, e0243191 (2020). 26167–26172 (2019). 34. Mehandru, S. & Merad, M. Pathological sequelae of long-haul COVID. 66. Estes, J. D., Wong, S. W. & Brenchley, J. M. Nonhuman primate models of Nat. Immunol. 23, 194–202 (2022). human viral infections. Nat. Rev. Immunol. 18, 390–404 (2018). 35. Ramakrishnan, R. K., Kashour, T., Hamid, Q., Halwani, R. & Tleyjeh, I. M. 67. Miller, L. A., Royer, C. M., Pinkerton, K. E. & Schelegle, E. S. Nonhuman Unraveling the mystery surrounding post-acute sequelae of COVID-19. primate models of respiratory disease: past, present, and future. ILAR J. 58, Front. Immunol. 12, 686029 (2021). 269–280 (2017). 36. Gang, J., Wang, H., Xue, X. & Zhang, S. Microbiota and COVID-19: 68. Böszörményi, K. P. et al. The post-acute phase of SARS-CoV-2 infection in long-term and complex influencing factors. Front. Microbiol. 13, two macaque species is associated with signs of ongoing virus replication and 963488 (2022). pathology in pulmonary and extrapulmonary tissues. Viruses 13, 1673 (2021). 37. Inoue, S. et al. Post-intensive care syndrome: its pathophysiology, prevention, 69. Rutkai, I. et al. Neuropathology and virus in brain of SARS-CoV-2 infected and future directions. Acute Med. Surg. 6, 233–246 (2019). non-human primates. Nat. Commun. 13, 1745–1745 (2022). 38. Merad, M., Blish, C. A., Sallusto, F. & Iwasaki, A. The immunology and 70. Østergaard, L. SARS CoV-2 related microvascular damage and symptoms immunopathology of COVID-19. Science 375, 1122–1127 (2022). during and after COVID-19: Consequences of capillary transit-time changes, 39. Xu, E., Xie, Y. & Al-Aly, Z. Long-term neurologic outcomes of COVID-19. tissue hypoxia and inflammation. Physiol. Rep. 9, e14726 (2021). Nat. Med. 28, 2406–2415 (2022). 71. Brouqui, P. et al. Asymptomatic hypoxia in COVID-19 is associated with 40. Asadi-Pooya, A. A. et al. Long COVID syndrome-associated brain fog. poor outcome. Int. J. Infect. Dis. 102, 233–238 (2021). J. Med. Virol. 94, 979–984 (2022). 72. de Melo, G. D. et al. COVID-19-related anosmia is associated with viral 41. Rao, V. et al. in Advances in Cancer Research (eds Gewirtz, D. A. & Fisher, P. B.) persistence and inflammation in human olfactory epithelium and brain vol. 155 29–76 (Academic Press, 2022). infection in hamsters. Sci. Transl. Med. 13, eabf8396 (2021). 42. Van Dyk, K. & Ganz, P. A. The inflammation complication: new evidence in 73. Moreau, G. B. et al. Evaluation of K18- hACE2 mice as a model of cancer-related cognitive impairment. Brain. Behav. Immun. 81, 6–7 (2019). SARS-CoV-2 infection. Am. J. Trop. Med. Hyg. 103, 1215–1219 (2020). 43. Song, E. et al. Neuroinvasion of SARS-CoV-2 in human and mouse brain. 74. Vidal, E. et al. Chronological brain lesions after SARS-CoV-2 infection in J. Exp. Med. 218, e20202135 (2021). hACE2-transgenic mice. Vet. Pathol. 59, 613–626 (2022). Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban 209 Perspective LAb ANIMAl 75. Villadiego, J. et al. Full protection from SARS-CoV-2 brain infection and 94. Zhou, B. et al. SARS-CoV-2 spike D614G change enhances replication and damage in susceptible transgenic mice conferred by MVA-CoV2-S vaccine transmission. Nature 592, 122–127 (2021). candidate. Nat. Neurosci. 26, 226–238 (2023). 95. Alene, M. et al. Serial interval and incubation period of COVID-19: a 76. Fumagalli, V. et al. Administration of aerosolized SARS-CoV-2 to systematic review and meta-analysis. BMC Infect. Dis. 21, 257 (2021). K18-hACE2 mice uncouples respiratory infection from fatal neuroinvasion. 96. He, F., Deng, Y. & Li, W. Coronavirus disease 2019: what we know? J. Med. Sci. Immunol. 7, eabl9929 (2022). Virol. 92, 719–725 (2020). 77. Seehusen, F. et al. Neuroinvasion and neurotropism by SARS-CoV-2 variants 97. Tay, M. Z., Poh, C. M., Rénia, L., MacAry, P. A. & Ng, L. F. P. The trinity of in the K18-hACE2 mouse. Viruses 14, 1020 (2022). COVID-19: immunity, inflammation and intervention. Nat. Rev. Immunol. 78. Dong, W. et al. The K18-Human ACE2 transgenic mouse model recapitulates 20, 363–374 (2020). non-severe and severe COVID-19 in response to an infectious dose of the SARS-CoV-2 virus. J. Virol. 96, e00964–21 (2022). 79. Winkler, E. S. et al. SARS-CoV-2 causes lung infection without severe disease Acknowledgements The authors acknowledge the funding from the European Commission EPIVINF in human ACE2 knock-in mice. J. Virol. 96, e0151121 (2022). (HORIZON-HLTH-2021-DISEASE-04, no. 101057548) and CaixaHealth EPIVIRCO 80. Zheng, H. et al. Virulence and pathogenesis of SARS-CoV-2 infection in (HR22-00681) projects. rhesus macaques: a nonhuman primate model of COVID-19 progression. PLoS Pathog. 16, e1008949 (2020). 81. van de Ven, K. et al. Pathology and immunity after SARS-CoV-2 infection in Author contributions male ferrets is affected by age and inoculation route. Front. Immunol. 12, C.U., J.V.-A. and J.S. conceptualized the article; C.U. and L.M. performed the literature 750229 (2021). research; C.U. conceptualized the figures; C.U. wrote the draft of the manuscript; L.M., 82. Jansen, E. B. et al. After the virus has cleared-Can preclinical models be C.B., J.V.-A. and J.S. critically reviewed the manuscript. All authors contributed to and employed for Long COVID research? PLoS Pathog. 18, e1010741 (2022). approved the final version of the manuscript. 83. Mahilkar, S. et al. SARS-CoV-2 variants: impact on biological and clinical outcome. Front. Med. 9, 409–424 (2022). 84. Shuai, H. et al. Emerging SARS-CoV-2 variants expand species tropism to Competing interests murines. eBioMedicine 73, 103643 (2021). The authors declare no competing interests. 85. Tarrés-Freixas, F. et al. Heterogeneous infectivity and pathogenesis of SARS-CoV-2 variants beta, delta and omicron in transgenic K18-hACE2 and wildtype mice. Front. Microbiol. 13, 840757 (2022). Additional information Correspondence should be addressed to Joaquim Segalés. 86. Halfmann, P. J. et al. SARS-CoV-2 Omicron virus causes attenuated disease in mice and hamsters. Nature 603, 687–692 (2022). Peer review information Lab Animal thanks Jacob Raber and the other, anonymous, 87. Al-Aly, Z., Bowe, B. & Xie, Y. Long COVID after breakthrough SARS-CoV-2 reviewer(s) for their contribution to the peer review of this work. infection. Nat. Med. 28, 1461–1467 (2022). Reprints and permissions information is available at www.nature.com/reprints. 88. Tseng, S.-H. et al. A novel pseudovirus‐based mouse model of SARS-CoV-2 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in infection to test COVID-19 interventions. J. Biomed. Sci. 28, 34 (2021). published maps and institutional affiliations. 89. Nie, J. et al. Nipah pseudovirus system enables evaluation of vaccines in vitro and in vivo using non-BSL-4 facilities. Emerg. Microbes Infect. 8, Open Access This article is licensed under a Creative Commons 272–281 (2019). Attribution 4.0 International License, which permits use, sharing, adap- 90. Chen, Q., Tang, K., Zhang, X., Chen, P. & Guo, Y. Establishment of tation, distribution and reproduction in any medium or format, as long pseudovirus infection mouse models for in vivo pharmacodynamics as you give appropriate credit to the original author(s) and the source, provide a link to evaluation of filovirus entry inhibitors. Acta Pharm. Sin. B 8, 200–208 (2018). the Creative Commons license, and indicate if changes were made. The images or other 91. Voelkl, B. et al. Reproducibility of animal research in light of biological third party material in this article are included in the article’s Creative Commons license, variation. Nat. Rev. Neurosci. 21, 384–393 (2020). unless indicated otherwise in a credit line to the material. If material is not included in 92. Su, M., Nizamutdinov, D., Liu, H. & Huang, J. H. Recent mechanisms of the article’s Creative Commons license and your intended use is not permitted by statu- neurodegeneration and photobiomodulation in the context of Alzheimer’s tory regulation or exceeds the permitted use, you will need to obtain permission directly disease. Int. J. Mol. Sci. 24, 9272 (2023). from the copyright holder. To view a copy of this license, visit http://creativecommons. 93. Bloem, B. R., Okun, M. S. & Klein, C. Parkinson’s disease. Lancet 397, org/licenses/by/4.0/. 2284–2303 (2021). © The Author(s) 2023 210 Lab Animal | VOL 52 | September 2023 | 202–210 | www.nature.com/laban