Trends in Immunology

Review

Mechanisms of SARS-CoV-2 Transmission

and Pathogenesis
Andrew G. Harrison,1,2 Tao Lin,1,2 and Penghua Wang1,*

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Highlights

marks the third highly pathogenic coronavirus to spill over into the human The emergence of SARS-CoV-2 from
population. SARS-CoV-2 is highly transmissible with a broad tissue tropism that is China and the rapidity of a worldwide
pandemic has promoted global
likely perpetuating the pandemic. However, important questions remain regarding
collaboration, built on a body of work
its transmissibility and pathogenesis. In this review, we summarize current SARS- established from previous SARS-CoV
CoV-2 research, with an emphasis on transmission, tissue tropism, viral pathogen- and MERS-CoV outbreaks. These past
esis, and immune antagonism. We further present advances in animal models that experiences have aided the swiftness
by which the research community has
are important for understanding the pathogenesis of SARS-CoV-2, vaccine devel- responded with an astonishing body of
opment, and therapeutic testing. When necessary, comparisons are made from work.
studies with SARS to provide further perspectives on coronavirus infectious
disease 2019 (COVID-19), as well as draw inferences for future investigations. SARS-CoV-2 is a novel virus in the
Betacoronavirus genus and exhibits
similarities to SARS-CoV in genome
The Emergence of a Third, Novel Coronavirus structure, tissue tropism, and viral
Current State of the COVID-19 Pandemic pathogenesis. Yet, SARS-CoV-2 ap-
pears to be more transmissible and
CoVs have caused three large-scale outbreaks over the past two decades: severe acute respira-
the diversity of immune responses
tory syndrome (SARS), Middle Eastern respiratory syndrome (MERS), and now COVID-19. The are poorly understood.
origin of the COVID-19 pandemic was traced back to a cluster of pneumonia cases connected
to a wet seafood market in Wuhan City, Hubei Province, China [1]. Following the likely spillover Highly pathogenic coronaviruses display
potent interferon (IFN) antagonism,
of a zoonotic disease (see Glossary), further work confirmed the etiological agent to be a
which is evident in cases of severe
novel Betacoronavirus related to SARS-CoV [1,2]. The first patients developed symptoms on COVID-19 with reduced IFN signaling,
December 1, 2019 after which rapid human-to-human transmission and intercontinental spread and an overaggressive immune re-
later ensued, being declared a pandemic by the WHO in March 2020 [3]. Since then, ~35 million sponse compounded by heightened
cytokines/chemokines.
people have been infected with SARS-CoV-2, with >1 million deaths in 235 countries, areas, or
territories [4]. Although SARS-CoV-2 appears to be less lethal than SARS-CoV or MERS-CoV, Animal models for SARS-CoV-2 reca-
its transmissibility is higher. To find solutions to contain this raging pandemic, global research pitulate important aspects of human
efforts have been quickly mobilized, each day resulting in new advances in basic and clinical COVID-19 that are essential for evalu-
ating current and prospective antiviral
research, therapy, diagnosis, vaccine and drug development, as well as epidemiology. Here, therapeutics and vaccine candidates.
we conduct a comprehensive review of the current state of COVID-19 research, with a principal
focus on the mechanisms of transmission and pathogenesis of SARS-CoV-2 stemming from
clinical and animal studies.

SARS-CoV-2 Characteristics 1
Department of Immunology, School of
SARS-CoV-2 Genome and Structure Medicine, University of Connecticut
CoVs of the family Coronaviridae are enveloped, positive-sense single-stranded RNA viruses [5]. Health Center, Farmington, CT 06030,
USA
All of the highly pathogenic CoVs, including SARS-CoV-2, belong to the Betacoronavirus genus, 2
These authors contributed equally.
group 2 [5]. The SARS-CoV-2 genome sequence shares ~80% sequence identity with SARS-
CoV and ~50% with MERS-CoV [1,6]. Its genome comprises 14 open reading frames (ORFs),
two-thirds of which encode 16 nonstructural proteins (nsp 1–16) that make up the replicase com-
plex [6,7]. The remaining one-third encodes nine accessory proteins (ORF) and four structural
proteins: spike (S), envelope (E), membrane (M), and nucleocapsid (N), of which Spike mediates *Correspondence:
SARS-CoV entry into host cells [8]. However, the S gene of SARS-CoV-2 is highly variable from [email protected] (P. Wang).

1100 Trends in Immunology, December 2020, Vol. 41, No. 12 https://doi.org/10.1016/j.it.2020.10.004

© 2020 Elsevier Ltd. All rights reserved.
Trends in Immunology

SARS-CoV, sharing <75% nucleotide identity [1,6,9]. Spike has a receptor-binding domain (RBD) Glossary
that mediates direct contact with a cellular receptor, angiotensin-converting enzyme 2 Aerosol: suspension of fine solid or
(ACE2), and an S1/S2 polybasic cleavage site that is proteolytically cleaved by cellular cathepsin liquid droplets in the air (or a gas
L and the transmembrane protease serine 2 (TMPRSS2) (Figure 1) [1,9,10]. TMPRSS2 facilitates medium), such as dusts, mists, or fumes.
Anamnestic immune response:
viral entry at the plasma membrane surface, whereas cathepsin L activates SARS-CoV-2 Spike in
memory immune response to a
endosomes and can compensate for entry into cells that lack TMPRSS2 (Figure 1) [10]. Once the previously encountered antigen.
genome is released into the host cytosol, ORF1a and ORF1b are translated into viral replicase Angiotensin-converting enzyme 2
proteins, which are cleaved into individual nsps (via host and viral proteases: PLpro); these form (ACE2): cell surface enzyme of
endothelial, epithelial, and other cells,
the RNA-dependent RNA polymerase (nsp12 derived from ORF1b) [8]. Here, the replicase with a well-defined function in
components rearrange the endoplasmic reticulum (ER) into double-membrane vesicles (DMVs) maintaining normal blood pressure.
that facilitate viral replication of genomic and subgenomic RNAs (sgRNA); the latter are translated Anosmia: partial or complete loss of the
into accessory and viral structural proteins that facilitate virus particle formation (Figure 1) [11,12]. sense of smell.
Antibody-dependent enhancement:
phenomenon by which antibodies
Tissue Tropism of SARS-CoV-2 against a virus are suboptimal to the
The establishment of viral tropism depends on the susceptibility and permissiveness of a specific virus and enhance its entry into host
host cell. During the SARS epidemic, patients often presented with respiratory-like illnesses that cells.
Correlates of protection: quantifiable
progressed to severe pneumonia, observations mirroring the disease course of COVID-19, parameters, such as antibodies,
suggesting that the lung is the primary tropism of SARS-CoV-2 [13]. Both CoVs were then indicating that a host is protected
found to bind the same entry receptor, ACE2 [1,14,15]. Of note, the key mutations in the RBD against microbial infection.
of SARS-CoV-2 Spike make additional close contacts with ACE2, correlating with higher binding Cytokine storm: severe immune
reaction in which the body releases too
affinity and perhaps increased infectivity [1,14,16]. The presence of a unique furin cleavage site at many cytokines into the blood too
the S1/S2 junction of SARS-CoV-2 Spike is also suspected to enhance human transmission quickly.
events, although this remains to be further investigated [17,18]. The currently predominant D-dimer: fibrin degradation product in
the blood after a clot is degraded by
SARS-CoV-2 isolate worldwide carries a D614G mutation that is absent from its presumptive
fibrinolysis.
common ancestor, and is more infectious, likely underlying, in part, an increased human-to- Disseminated intravascular
human transmission efficiency [19–21]. Although associated with an increased viral load in the coagulation (DIC): condition in which
upper respiratory tract (URT) of patients with COVID-19, the D614G variant does not correlate blood clots form throughout the body
and block small blood vessels, leading to
with disease severity, suggesting that pathogenesis of severe COVID-19 is linked to mechanisms multiorgan failure.
that are more than just SARS-CoV-2 infectivity [21]. Fecal-oral transmission: route of
disease transmission by which an
Once SARS-CoVs enter the host via the respiratory tract, airway and alveolar epithelial cells, infectious agent in fecal materials is
passed to the mouth of another.
vascular endothelial cells and alveolar macrophages are among their first targets of viral entry Fomite: inanimate object (clothes,
[22–24]. These cells are probably ‘ground-zero’ for early infection and subsequent replication utensil, furniture, etc.) that, when
due to their expression of ACE2 [25]. Although ACE2 mRNA is detected in human and many contaminated with an infectious agent,
can transfer the infectious agent to a
mammalian (bat, ferret, cat, dog, etc.) lung biopsies, their expression is rather low compared
new host.
with extrapulmonary tissues [26]. Thus, the permissiveness of these cells to SARS-CoVs may de- Furin: proprotein convertase that
pend on additional, unappreciated cell-intrinsic factors that aid in efficient infection. First, viral cleaves a precursor protein into a
entry may heavily depend on the expression of TMPRSS2, because nearly undetectable amounts biologically active state.
Incubation period: timeframe elapsed
of ACE2 still support SARS-CoV entry as long as TMPRSS2 is present [27]. Second, the mRNA
between when a host is first exposed to
expression of cellular genes, such as endosomal sorting complex required for transport (ESCRT) an infectious agent and when signs or
machinery gene members (including CHMP3, CHMP5, CHMP1A, and VPS37B) related to a symptoms begin to appear.
pro-SARS-CoV-2 lifecycle is higher in a small population of human type II alveolar cells with abun- Lung ground glass opacity:
nonspecific radiological description of an
dant ACE2, relative to ACE2-deficient cells [28]. This suggests that SARS-CoV-2 hijacks a small
area of increased opacity in the lung
population of type II alveolar cells with high expression of ACE2 and other proviral genes for its pro- through which vessels and bronchial
ductive replication. Third, the lung, as the main tropism of SARS-CoVs, may be contingent on the structures are still visible.
regulation of ACE2 at the transcriptional and protein levels [24,25,29–31]. For example, in human Neutralizing antibody (NAb):
antibody that binds a pathogen with high
airway epithelial cells, ACE2 gene expression is upregulated by type I and II interferons (IFNs) affinity and prevents the latter from
[25,31] during viral infection. Lastly, compared with other SARS-CoVs, SARS-CoV-2 Spike con- exerting its biological effect.
tains a unique insertion of RRAR at the S1/S2 cleavage site [17,18]. This site can be precleaved Neutrophil extracellular traps
by furin, thus reducing the dependence of SARS-CoV-2 on target cell proteases (TMPRSS2/ (NETs): networks of extracellular fibers,

Trends in Immunology, December 2020, Vol. 41, No. 12 1101

Trends in Immunology

cathepsin L) for entry [17,18] and potentially extending its cellular tropism, given that proteolytically comprising primarily DNA from
neutrophils due to chromatin
active furin is abundantly expressed in human bronchial epithelial cells [32,33]. decondensation, which can ‘trap’
extracellular pathogens.
One of the distinctions between SARS-CoV and SARS-CoV-2 is the ability of the latter to Pattern recognition receptor:
efficiently infect the URT, such as nasopharyngeal (NP) and/or oropharyngeal (OP) tissues, germline-encoded host sensor that
recognizes a signature pattern in
possibly due to its higher affinity for ACE2, which is expressed in human nasal and oral tissues microbial molecules.
[23,25,34–36]. The readily detectable titers of SARS-CoV-2 in the URT mucus of patients with Prodromal period: time immediately
COVID-19 during prodromal periods might help explain the more rapid and effective transmis- following the incubation period of a
sibility of SARS-CoV-2 relative to SARS-CoV [37]. microbial infection in which a host begins
to experience symptoms or changes in
behavior/functioning.
Human CoVs often cause enteric infections, with variable degrees of pathogenicity [38]. Indeed, Prothrombin time: measurement of
ACE2 and TMPRSS2 are abundantly expressed within the human and many other mammalian the extrinsic pathway of coagulation.
Reproductive number (R0): expected
intestinal tracts, specifically the brush border of intestinal enterocytes [23,25,26,39]. Accordingly,
number of new disease cases generated
gastrointestinal illness has been frequently reported in patients with COVID-19 [40,41], consistent by one case. R0 >1 indicates that the
with the recovery of SARS-CoV from stool samples of patients with SARS [42], suggesting a outbreak will expand; R0 <1 indicates
potential fecal–oral route of transmission for these two CoVs. Of note, ~20% of patients with that the outbreak will die out.
Respiratory droplet: small aqueous
COVID-19 examined have had detectable SARS-CoV-2 RNA in feces, even after respiratory
droplet produced by exhalation,
symptoms subsided, suggesting that SARS-CoV-2 titers may be prolonged in the intestinal comprising saliva or mucus and other
tract [41]. Although further testing is warranted, these data suggest the possibility that fecal– matter derived from respiratory tract
oral transmission of SARS-CoV-2 occurs. Evidently, robust epidemiological studies are surfaces.
Zoonotic disease: infectious disease
needed to conclusively demonstrate whether patients with COVID-19 recovering from respiratory caused by a pathogen that has crossed
illness are able to spread SARS-CoV-2. a species barrier from animals to
humans.
Transmission Dynamics of SARS-CoV-2
Human CoVs are transmitted primarily through respiratory droplets, but aerosol, direct
contact with contaminated surfaces, and fecal–oral transmission were also reported during the
SARS epidemic [43–45]. Early reports of patients with cough, lung ground glass opacities,
and symptom progression to severe pneumonia, suggested communicability of SARS-CoV-2
via the respiratory route (Figure 2) [1–3]. Direct transmission by respiratory droplets is reinforced
by productive SARS-CoV-2 replication in both the URT and lower respiratory tract (LRT), and the
increasing number of reports indicating human-to-human spread among close contacts
exhibiting active coughing (Figure 2) [35,46–48]. So far, the basic reproduction number (R0)
is ~2.2, based on early case tracking during the beginning of the pandemic, with a doubling
time of 5 days [47,49]. Furthermore, there is now evidence for nonsymptomatic/presymptomatic
spread of SARS-CoV-2, which is in contrast to the transmission dynamics of SARS-CoV [50].
This finding underscores the ability of SARS-CoV-2 to colonize and replicate in the throat during
early infection [37,51,52]. Based on these apparent disparities in virus transmission, one study
modeled the transmission dynamics of SARS-CoV-2 in presymptomatic individuals, and
indicated that the presymptomatic R0 has approached the threshold for sustaining an outbreak
on its own (R0 >1); by contrast, the corresponding estimates for SARS-CoV were approximately
zero [49]. Similarly, asymptomatic spread of SARS-CoV-2 has been documented throughout
the course of the pandemic [48,51,53–56]. Understanding the relative importance of cryptic
transmission to the current COVID-19 pandemic is essential for public health authorities to
make the most comprehensive and effective disease control measures, which include mask-
wearing, contact tracing, and physical isolation.

For SARS-CoV-2, various modes of transmission have been proposed, including aerosol, surface
contamination, and the fecal–oral route, representing confounding factors in the current COVID-19
pandemic; thus, their relative importance is still being investigated (Figure 2) [57]. Aerosol trans-
mission (spread >1 m) was implicated in the Amoy Gardens outbreak during the SARS

1102 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

Trends in Immunology

Figure 1. The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Lifecycle. The SARS-related coronavirus (SARS-CoV and SARS-CoV-2)
lifecycle commences by binding of the envelope Spike protein to its cognate receptor, angiotensin-converting enzyme 2 (ACE2). Efficient host cell entry then depends
on: (i) cleavage of the S1/S2 site by the surface transmembrane protease serine 2 (TMPRSS2); and/or (ii) endolysosomal cathepsin L, which mediate virus–cell
membrane fusion at the cell surface and endosomal compartments, respectively. Through either entry mechanism, the RNA genome is released into the cytosol, where
it is translated into the replicase proteins (open reading frame 1a/b: ORF1a/b). The polyproteins (pp1a and pp1b) are cleaved by a virus-encoded protease into
individual replicase complex nonstructural proteins (nsps) (including the RNA-dependent RNA polymerase: RdRp). Replication begins in virus-induced double-
membrane vesicles (DMVs) derived from the endoplasmic reticulum (ER), which ultimately integrate to form elaborate webs of convoluted membranes. Here, the
incoming positive-strand genome then serves as a template for full-length negative-strand RNA and subgenomic (sg)RNA. sgRNA translation results in both structural
proteins and accessory proteins (simplified here as N, S, M, and E) that are inserted into the ER–Golgi intermediate compartment (ERGIC) for virion assembly. Finally,
subsequent positive-sense RNA genomes are incorporated into newly synthesized virions, which are secreted from the plasma membrane [6,8,11,12].
Figure generated with BioRender.

epidemic, but the inconsistency of these findings in other settings suggested that SARS-CoV
was an opportunistic airborne infection [43,58]. Similarly, no infectious SARS-CoV-2 virions
have been isolated, although viral RNA was detectable in the air of COVID-19 hospital wards [59].
Generation of experimental aerosols carrying SARS-CoV-2 (comparable to those that might be
generated by humans) have offered the plausibility of airborne transmission, but the aerodynamic
characteristics of SARS-CoV-2 during a natural course of infection is still an area of intense inquiry
[60]. Nonetheless, deposition of virus-laden aerosols might contaminate objects (e.g., fomites)
and contribute to human transmission events [59,61]. Finally, fecal–oral transmission has also been
considered as a potential route of human spread, but remains an enigma despite evidence of
RNA-laden aerosols being found nearby toilet bowls, along with detectable SARS-CoV-2 RNA in
rectal swabs during the precursor epidemic of COVID-19 in China [41,59,62].

Trends in Immunology, December 2020, Vol. 41, No. 12 1103

Trends in Immunology

Trends in Immunology

Figure 2. Proposed Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Transmission Routes.
The ongoing COVID-19 pandemic has resulted in numerous accounts of different transmission routes between humans.
Droplet transmission (>5 μm) is the most pronounced and heavily implicated mode of transmission reported during the
pandemic. Direct contact spread from one infected individual to a second, naïve person has also been considered a driver
of human-to-human transmission, especially in households with close interactions between family members. The
contagiousness of SARS-CoV-2 after disposition on fomites (e.g., door handles) is under investigation, but is likely a
compounding factor for transmission events, albeit less frequently than droplet or contact-driven transmission. Both airborne
and fecal–oral human-to-human transmission events were reported in the precursor SARS-CoV epidemic but have yet to be
observed in the current crises. Solid arrows show confirmed viral transfer from one infected person to another, with a declin-
ing gradient in arrow width denoting the relative contributions of each transmission route. Dashed lines show the plausibility of
transmission types that have yet to be confirmed. SARS-CoV-2 symbol in ‘infected patient’ indicates where RNA/infectious
virus has been detected [43,44,47–49,57,59,60]. Figure generated with BioRender.

SARS-CoV-2 Pathogenesis
Clinical Presentation of COVID-19
In general, common cold CoVs tend to cause mild URT symptoms and occasional gastrointestinal
involvement (Figure 3). By contrast, infection with highly pathogenic CoVs, including SARS-CoV-2,
causes severe ‘flu’-like symptoms that can progress to acute respiratory distress (ARDS),
pneumonia, renal failure, and death [46,48,63,64]. The most common symptoms are fever,
cough, and dyspnea, accounting for 83%, 82%, and 31% of patients with COVID-19 (N = 99),
respectively, in one epidemiological study [65]. The incubation period in COVID-19 is rapid:
~5–6 days versus 2–11 days in SARS-CoV infections [38,47,48]. As the pandemic is progressing,
it has become increasingly clear that COVID-19 encompasses not only rapid respiratory/
gastrointestinal illnesses, but can also have long-term ramifications, such as myocardial inflamma-
tion [66]. Furthermore, severe COVID-19 is not restricted to the aged population as initially
reported; children and young adults are also at risk [67]. From a diagnostic perspective, COVID-19
presents with certain ‘hallmark’ laboratory and radiological indices, which can be helpful
in assessing disease progression (Table 1). Together, COVID-19 initially presents with ‘flu’-like
symptoms and can later progress to life-threatening systemic inflammation and multiorgan
dysfunction.

1104 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

Trends in Immunology

Figure 3. Clinical Symptoms of Coronavirus Infectious Disease 2019 (COVID-19). COVID-19 manifestations in
humans have been described to incorporate multiple body systems with varying degrees of onset and severity. Both the
upper respiratory tract and lower respiratory tract manifestations are often the most noticeable if a patient is not
asymptomatic, in addition to systemic symptoms that are the most frequently reported regardless of disease severity. Red-
highlighted signs/symptoms tend to be over-represented in severe patients, but common symptoms are also present in
more advanced COVID-19. A severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus symbol denotes where
a live virus and/or viral RNA has been isolated. Abbreviation: ARDS: acute respiratory distress syndrome [37,46,48,66,139].
Figure generated with BioRender.

Age-Associated COVID-19 Severity

It is widely accepted that the aging process predisposes individuals to certain infectious dis-
eases [68]. In the case of COVID-19, older age is associated with greater COVID-19 morbidity,
admittance to the intensive care unit (ICU), progression to ARDS, higher fevers, and greater
mortality rates [69–71]. Moreover, lymphocytopenia, neutrophilia, elevated inflammation-
related indices, and coagulation-related indicators have been consistently reported in older
(≥65-years old) relative to young and middle-aged patients with COVID-19 (Table 1)
[46,65,71–75]. At the cellular level, a lower capacity of CD4+ and CD8+ T-cells to produce
IFN-γ and IL-2, as well as an impairment in T-cell activation from dendritic cells (DCs) in
patients with acute COVID-19 (≥55-years old) could compromise an optimal adaptive immune
response [76]. Based on examples from mice, a productive CD4+ T-cell response relies
heavily on lung-resident DCs (rDCs) and abates SARS-CoV infection [77,78]. However,
whether a reduction in the DC population in the lungs of older patients with more severe
COVID-19 causes suboptimal T-cell activation during SARS-CoV-2 infection remains to be ro-
bustly investigated.

Trends in Immunology, December 2020, Vol. 41, No. 12 1105

Trends in Immunology

Table 1. Common Laboratory Indices and Radiological Findings from Patients with COVID-19
Laboratory findings COVID-19 studya
Refs [65] [140] [141] [137] [73] [138]
Patient population Middle-aged, Middle-aged, Elderly, Middle-aged, Elderly, Young, mild
hospitalized; hospitalized/ICU; mild-severe hospitalized adults; deceased; disease;
N = 99 N = 1099 N = 71 N = 140 N = 82 N = 46
Ground-glass 14 56.4b 100 99.3 -- 63
opacities
Pneumonia 100 91.1 73 -- 93.9 --
(unilateral or bilateral)c
Lymphocytopenia 35 83.2 37 (<1200/μl) 75.4 89.2 63 (<1.5×109/l)
(1.1–3.2 ×109/ml) (<1500/mm3) (<1.0 ×109/l)
Leukopenia 9 33.7 21 (<4000/μl) 19.6 21.7 (<4×109/l)
(3.5–9.5x109/ml) (<4000/mm3) (3.5–9.5×109/ml)
Thrombocytopenia 12 36.2 10 (<15×104/μl) -- 24.3 21.7 (<150 ×109/l)
(125–350×109/ml) (<150 000/mm3) (<100 ×109/l)
Neutrophilia 38 -- -- -- 74.3 --
(1.8–6.3×109/ml) (>6.3 ×109/l)
↑d C-reactive protein 86 (0–5) 60.7 (≥10) 59 (>10–30) 91.9 (0–3) 100 (>10) 19.6 (≥10)
(mg/l)
↑ Alanine 28 (9–50) 21.3 (>40) 18 (>45) -- 30.6 (>40) 15.2 (>40)
aminotransferase (U/l)
↑ D-dimer (μg/l) 36 (0–1.5) 46.4 (≥500) -- 43.2 (0–243) 97.1 (>550) 15.2 (≥500)
↑ Prothrombin time (s) 5 (10.5–13.5) -- -- -- 100 (12.3–14.3) e
--

a
Values for each laboratory manifestation represent the percentage of patients with that clinical finding above or below the normal range (listed below in brackets); dashed
lines indicate measurements not taken during referenced study.
b
Percentage of patients with lung ground-glass opacity on a chest CT scan (technique specifically denoted in original study).
c
Pneumonia was not always definitively mentioned in studies, albeit lung manifestations were commonly recorded.
d
↑ denotes an elevation of measured indices above reference value for those percentage of patients
e
Over the past 24 h, leading up to death, all 13 patients who were included for this metric had a prothrombin time of >12.1 s.

Higher proportions of proinflammatory macrophages and neutrophils have also been observed
in the bronchoalveolar lavage fluid (BALF) of COVID-19 patients with severe symptoms
compared with those exhibiting mild symptoms (Figure 4, Key Figure) [79]. Accordingly, pro-
inflammatory cytokines (e.g., IL-6 and IL-8) are elevated in the BALF of patients with severe
COVID-19, along with higher expression of inflammatory chemokines (e.g., CCL2) in macro-
phages relative to patients with nonsevere COVID-19 [79–82]. Indeed, similar inflammatory
milieux have been associated with severe lung pathology in patients with SARS, along with the
notable ‘cytokine storm’ that can present in patients critically ill with COVID-19 [71,83–87].
These proinflammatory mediators can, in turn, perpetuate lung disease by elevating C-reactive
protein (CRP) from the liver (Table 1) through signal transducer and activator of transcription
3 (STAT3)-IL-6 signaling [88]. Therefore, an increase in CRP concentrations can correlate with
elevated serum IL-6 production observed in patients with COVID-19 [79,80,88]. From another
angle, formation of neutrophil extracellular traps (NETs) inside microvessels is pronounced
in patients with severe relative to mild COVID-19, implicating NETs as possible potentiators of
COVID-19 pathogenesis [89]. The recruitment of these activated neutrophils and monocytes
may be driven by pulmonary endothelial cell dysfunction through vascular leakage, tissue
edema, endotheliitis, and possibly, disseminated intravascular coagulation (DIC) pathways;
indeed, a recent study demonstrated direct SARS-CoV-2 infection of vascular endothelial cells
with concomitant accumulation of inflammatory mononuclear cells (e.g., neutrophils) in multiple
organs (lung, heart, kidney, small bowel, and liver) in patients with severe COVID-19 (Figure 4)

1106 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

Key Figure
A Brief Overview of Lung Pathology in Patients with Coronavirus Infectious Disease 2019 (COVID-19)

Trends in Immunology

Figure 4. Following inhalation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) into the respiratory tract, the virus traverses deep into the lower lung,
where it infects a range of cells, including alveolar airway epithelial cells, vascular endothelial cells, and alveolar macrophages. Upon entry, SARS-CoV-2 is likely
detected by cytosolic innate immune sensors, as well as endosomal toll-like receptors (TLRs) that signal downstream to produce type-I/III interferons (IFNs) and
proinflammatory mediators. The high concentration of inflammatory cytokines/chemokines amplifies the destructive tissue damage via endothelial dysfunction and
vasodilation, allowing the recruitment of immune cells, in this case, macrophages and neutrophils. Vascular leakage and compromised barrier function promote
endotheliitis and lung edema, limiting gas exchange that then facilitates a hypoxic environment, leading to respiratory/organ failure. The inflammatory milieu induces
endothelial cells to upregulate leukocyte adhesion molecules, thereby promoting the accumulation of immune cells that may also contribute to the rapid progression of
respiratory failure. Hyperinflammation in the lung further induces transcriptional changes in macrophages and neutrophils that perpetuate tissue damage that ultimately
leads to irreversible lung damage. Recent evidence suggests that systemic inflammation induces long-term sequela in heart tissues [66,79,80,82,84,87,90,95].
Abbreviations: BALF, bronchoalveolar lavage fluid; IRF3, interferon regulatory factor 3; NF-κB, nuclear factor-κB; RIG-I, retinoic acid-inducible gene I; STAT1/2, signal
transducer and activator of transcription 1/2; STING, Stimulator of interferon genes. Figure generated with BioRender.

[90]. In fact, many patients with COVID-19 have met the DIC case definition based on elevated
serum D-dimer amounts and prolonged prothrombin times [91,92]. Together, it is reasonable
to assume that direct viral insult and immune cell recruitment escalate endothelial contractility and
the loosening of gap junctions, thus promoting vascular leakage and the systemic impairment of
the circulatory system in this pathology.

Trends in Immunology, December 2020, Vol. 41, No. 12 1107

Trends in Immunology

SARS-CoV-2 Innate Immune Evasion Strategies: Examples from other Betacoronavirus Infections
The recognition of virus infection begins with the detection of viral nucleic acid by host cell pattern
recognition receptors (PRRs), which signal downstream via recruited adaptor proteins, ubiqui-
tin ligases, and kinases, culminating in transcription factors and ultimate expression of immune
genes, including IFNs, cytokines, and chemokines (Figure 5). The IFN pathway is often a primary
target of evasion due to its rapidity and potency in eliminating viral infection. CoVs have evolved
multiple mechanisms to target the signaling components of several PRR-IFN pathways to survive

Trends in Immunology

Figure 5. Evasion of the Pattern Recognition Receptor-Type I Interferon (PRR-IFN-I) Pathways by Coronaviruses (CoVs). A simplified schematic of the
canonical IFN response after sensing RNA viruses. Viral nucleic acid is first recognized by PRRs (e.g., retinoic acid-inducible gene I; RIG-I) that perpetuate signal transduc-
tion through an adaptor complex on the mitochondrial (mitochondrial antiviral-signaling protein; MAVS) or endoplasmic reticulum (Stimulator of interferon genes; STING)
membrane surface. Here, the PRR–adaptor interactions recruit kinases that converge into a large complex, leading to phosphorylation of interferon regulatory factor
3/7 (IRF3/7) and nuclear factor-κB (NF-κB), transcription factors that enter the nucleus and transcribe IFN genes. Type-I and type-III IFNs then signal in an autocrine or
paracrine manner through the Janus kinase 1 (JAK1)/signal transducer and activator of transcription 1 and 2 (STAT1/2) pathway, culminating in antiviral IFN-stimulated
gene (ISG) transcription. Listed here are SARS-CoV (CoV), SARS-CoV-2 (CoV-2), and MERS-CoV (M-CoV) IFN-I antagonists, which render these viruses resistant to
IFN responses. IFN-III is also implicated in exhibiting potent antiviral effects in lung/intestinal tissues, but the underlying evasion strategies of this pathway for these viruses
are currently unknown. SARS-CoV proteins are highlighted in blue, while functions of SARS-CoV-2 and MERS-CoV proteins are highlighted in red and green, respectively.
? denotes that a SARS-CoV-2 protein bound a member of that signaling pathway in [122], but further work is necessary to confirm its immunological mechanism. SARS-
CoV-2 proteins with * denotes functional conservation with SARS-CoV [93–96,98,100–107]. Figure generated with BioRender.

1108 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

in host cells (Figure 5). CoVs are highly sensitive to IFN and, therefore, act at several levels in these
pathways to antagonize mammalian immune recognition, interfering with downstream signaling,
or inhibiting specific IFN-stimulated gene (ISG) products [93]. Specifically, CoVs can avoid
immune sensing by: (i) the formation of DMVs that sequester viral nucleic acid from being recog-
nized by PRRs; and (ii) direct ablation of the functionality of immune signaling molecules by viral
proteins [11,94]. The structural and functional conservation of these proteins across the
Betacoronavirus and in nsps between SARS-CoV and SARS-CoV-2 suggest that some of
these suppressive mechanisms are used by SARS-CoV-2 (see later) [1]. Indeed, patients with
severe COVID-19 have reported an imbalanced immune response with high concentrations of
inflammatory cytokines/chemokines, but little circulating IFN-β or IFN-λ, resulting in persistent
viremia [95]. Of note, among several respiratory viruses tested, SARS-CoV-2 has demonstrated
to most potently suppress type I and type III IFN expression in both human bronchial epithelial
cells and ferrets [81]. Thus, evasion of IFN signaling by SARS-CoV-2 and impaired IFN production
in human peripheral blood immune cells might contribute to the productive viral replication, trans-
mission, and severe pathogenesis during COVID-19, although further testing is warranted to fully
dissect these putative evasion pathways [95].

With regard to functional conservation of viral proteins, SARS-CoV and MERS-CoV nsps and
accessory proteins circumvent viral RNA-sensing pathways at multiple stages [e.g., retinoic acid-
inducible gene I (RIG-I) and melanoma differentiation-associated protein 5; MDA-5) through
proteasomal degradation and/or prevention of protein activation (Figure 5) [94]. Functional conserva-
tion between SARS-CoV and MERS-CoV PLpro (encoded by nsp3) proteins has been reported,
where these proteins target the initial PRR signaling cascade at multiple levels of the pathway includ-
ing, but not limited to, RIG-I, mitochondrial antiviral-signaling protein (MAVS), TANK-binding kinase 1
(TBK1), interferon regulatory factor 3 (IRF3), and nuclear factor (NF)-κB (Figure 5) [96–98]. The SARS-
CoV PLpro also targets the DNA-sensing pathway at Stimulator of IFN genes (STING) (Figure 5); an-
tagonizing this pathway might be important because mitochondrial stress during dengue virus infec-
tion triggers IFN-β production that is dependent on STING activation [99,100]. Recent evidence
suggests the SARS-CoV-2 PLpro also inhibits IFN-I expression in human kidney epithelial cells, yet
the mechanisms remain to be defined [101]. Moreover, nsp1 of highly pathogenic HCoVs, including
SARS-CoV and MERS-CoV, displays a pleiotropic effect, targeting several components of IFN-I sig-
naling (Figure 5) [102,103]. This potent suppressive function of nsp1 also appears to be maintained in
SARS-CoV-2, primarily through shutdown of translational machinery and prevention of immune gene
expression [101,104,105]. Furthermore, because there are only five accessory genes in the MERS-
CoV genome compared with eight and seven in the SARS-CoV and SARS-CoV-2 genomes, respec-
tively, similar immunosuppressive mechanisms may exist but appear to be mediated via different pro-
teins [106,107]. For example, SARS-CoVs ORF6 can inhibit IRF3 activation and STAT1 nuclear
translocation, whereas this same effect is obtained by ORF4a/b and ORF5 of MERS-CoV
(Figure 5) [106,107]. Coincidently, the apparent loss of these proteins may provide evidence
for why MERS-CoV is more sensitive to IFN treatment than are SARS-CoVs in primary and con-
tinuous cells of the human airways [110]. The SARS-CoV-2 proteins appear to have stronger
inhibitory effects than their counterparts in highly pathogenic SARS- and MERS-CoV [105].
In light of these findings, SARS-CoV-2 has replicated more efficiently than SARS-CoV in
ex vivo human lung explants, possibly through the greater suppression of IFN-I/III cytokines
[111]; further work will be needed to discern whether the suppressive nature of SARS-CoV-2
can impact virus transmission during early phases of COVD-19, when IFNs are typically impor-
tant for virus control. The ‘common-cold’ CoVs (e.g., HCoV-229E) and murine hepatitis virus
(MHV) also compensate for the loss of many supplementary immunosuppressive proteins
through capping viral mRNAs via nsp16 2′-O-methyltransferase (2′-O-MTase), and mutants
lacking this activity exhibit diminished replication and dissemination in mice [112]. Thus, further

Trends in Immunology, December 2020, Vol. 41, No. 12 1109

Trends in Immunology

investigation is warranted to determine whether these evasion genes account for the increased
virulence observed in individuals infected with SARS-CoV-2 (see Outstanding Questions).

Animal Models of SARS-CoV-2

Mouse Models
Given that SARS-CoV-2 uses the same ACE2 entry receptor as SARS-CoV, rapid deployment of
mouse models for pathogenesis studies was well underway within weeks of the inception of the
pandemic. However, various impediments remain for SARS-CoV-2 in productively infecting mice
in these models, because it is unable to bind mouse ACE2 (mACE2) [113]. To overcome these
prerequisites, several mouse models have been developed that recapitulate certain components
of human COVID-19. One of these strategies is to genetically modify mice to express human
ACE2 (hACE2) (humanized mice) under the epithelial cell-specific cytokeratin-18 (Krt18) promoter
[114], a universal chicken beta-actin promoter [115], or the endogenous mACE2 promoter [113].
All these mice are susceptible to SARS-CoV-2 infection, but phenotypic disease varies because
of differential hACE2 tissue expression [113–115]. For instance, Krt18-hACE2 and beta-actin-
hACE2-transgenic mice rapidly succumb to SARS-CoV-2 infection with lung infiltration of
inflammatory immune cells inducing severe pulmonary disease, accompanied by evident throm-
bosis and anosmia, which partially recapitulate human COVID-19 [116,117]. Given that the
onset of severe histopathological changes occurs days after peak virus infection, these models
recapture the delayed morbidity seen in patients with COVID-19 as a result of inflammatory cell
infiltration [117]. Therefore, using humanized mouse models of severe SARS-CoV-2 infection
might be useful for testing the efficacy of antiviral drugs, vaccines, and immune therapeutics
that ablate hyperinflammation [116]. However, the broad expression of hACE2 in these models
significantly expands SARS-CoV-2 tissue tropisms and might alter its pathogenic mechanisms
[116,117]. For example, both SARS-CoV and SARS-CoV-2 infections lead to encephalitis in
these mouse models, which is not common in patients with COVID-19 [115,117,118]. Given
that most human SARS-CoV-2 infections are asymptomatic or mild, mice originally bearing
mACE2 that is replaced by hACE2 may be more appropriate for assessing pathogenesis and
tissue tropism [113]. This model develops mild lung pathology, with SARS-CoV-2 infection
being restricted to the lung and intestine [113]. In addition to the transgenic modification, mice
can also be sensitized to SARS-CoV-2 infection via transient transduction of adenovirus (Ad5)-
or adeno-associated virus (AAV)-expressing hACE2 in respiratory tissues, akin to the approach
used for MERS-CoV infection [108,109]. These mice develop viral pneumonia, weight loss, se-
vere pulmonary pathology, and a high viral load in the lung, consistent with human COVID-19
[109]. This approach might be quickly adapted to many genetically modified mouse strains that
could provide mechanisms of SARS-CoV-2 pathogenesis and protective immune responses.
However, this model is limited by the transient ectopic expression of hACE2 from the Ad5/AAV
vector that can induce mild bronchial inflammation and expand cell tropism of SARS-CoV-2 and,
thus, alter disease pathogenesis [119].

Rather than genetic modification in host animals, viruses can also be genetically modified and
used in model animals [120,121]. For instance, in one study, serial passaging of SARS-CoV-2
in mice led to enrichment of a N501Y viral mutant that elicited interstitial pneumonia and
inflammatory responses in both young and aged wild-type BALB/c mice [122]. Another
mouse-adapted SARS-CoV-2 strain (MA10) carrying three mutations in the RBD of Spike protein
caused severe lung pathology and ARDS in mice, characteristic of severe COVID-19 [123].
Despite the three mutations in the RBD of the mouse-adapted Spike, vaccination with full-length
SARS-CoV-2 Spike elicited robust neutralizing antibody titers and complete protection against
a secondary challenge with MA10 [123]; these findings suggest that this strain is applicable to
pathogenesis studies, as well as antiviral drug and vaccine testing in rodents.

1110 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

Nonhuman Primate Models

The role of nonhuman primates (NHP) in evaluating coronavirus pathogenesis cannot be understated.
Depending on the NHP model utilized, clinical signs/symptoms may be mild or absent entirely
[124–126]. In rhesus macaques, several studies noted reduced appetite, transient fevers
(1 day post infection: dpi) and mild weight loss without overt signs of respiratory distress or
mortality [124–126]. By contrast, cynomolgus macaques did not display any observational
signs of disease in another study [125]. Although certain NHPs appear to only mimic mild dis-
ease (if any), rhesus macaques have exhibited high viral loads in nasal swabs, throat samples,
and BALF early post inoculation, and viral RNA was still measurable by quantitative (q)PCR in
the trachea and lung 21 dpi, highlighting the apparent tropism of SARS-CoV-2 for the URT
and lingering viral nucleic acid in respiratory tissues after resolution of disease [51,124].
SARS-CoV-2 has also been detected in nasal swabs at 10 dpi in NHPs, consistent with the
prolonged URT shedding of virus in patients with COVID-19 at ~9 dpi [51,124,125,127]. The
tropism of SARS-CoV-2 for the LRT in NHPs has also been recapitulated by the development
of multifocal lesions and interstitial pneumonia, supporting the hypothesis that lung injury is
driven by increased infiltration of neutrophils and macrophages into the lung following viral
infection [124–126].

However, additional hallmarks of severe disease are absent in NHPs, particularly the characteris-
tic systemic ‘cytokine storm’ present in patients with COVID-19; indeed, only transient elevations
of serum inflammatory cytokines have been observed in NHPs, and have been reported to
decline rapidly by 2 dpi [124]. Overall, these NHP models display mild disease accompanied by
viral dissemination in the URT and LRT, leading to localized lung inflammation, but devoid of
the sustained systemic inflammatory response that has been noted in patients with COVID-19.
Thus, NHP models might be useful for studying mild COVID-19 characteristics, but provide little
information on the pathogenic mechanism(s) of severe COVID-19. To partially overcome this
issue, aged rhesus macaques (15-years old) have been tested following SARS-CoV-2 infection
and have demonstrated shedding of the virus for longer periods of time (14 days), as well as
increased radiological and histopathological changes, such as thickened alveolar septum and
diffuse severe interstitial pneumonia, compared with young macaques (3–5-years old) [128].
Therefore, these studies highlight the importance of also considering age, as an additional variable,
when selecting animal models that might closely, or accurately, recapitulate human disease.

Evaluating efficacious vaccine candidates in NHPs will also be important for understanding
correlates of protection against SARS-CoV-2. Accordingly, reports of antibody-dependent
enhancement, as well as of non-neutralizing humoral responses to the conserved regions of
SARS-CoV-2, raise concerns on our future ability to effectively administer an immunogen without
inducing immunopathology [129,130]. Furthermore, upon viral challenge, lymphocytes have
expanded in rhesus macaque models around 5 dpi with complementary B cell responses
against SARS-CoV-2 Spike appearing 10–15 dpi in blood samples [124]; expansion of these
adaptive immune compartments was analogous to those observed in patients with COVID-19
[37,124,131–133]. Subsequent rechallenged rhesus macaque have presented a rapid anamnestic
immune response characterized by significantly higher neutralizing antibody (NAb) titers
compared with the primary infection macaques [126]. Thus, protective efficacy appears to depend
primarily on NAb titers, at least in NHPs; so far, T-cell numbers have not substantially increased in
the serum following rechallenge of these animals and, in a secondary study, CD4+ and CD8+
cytokine (e.g., IFN- γ) responses did not correlate with immune protection from DNA vaccines with
different components of the SARS-CoV-2 Spike protein [126,134]. Although these animals have
failed to manifest overt signs of infection and respiratory compromise, NHPs still represent the
‘gold standard’ for evaluating the protective efficacy of human-bound SARS-CoV-2 vaccines

Trends in Immunology, December 2020, Vol. 41, No. 12 1111

Trends in Immunology

based on parallels to humans in terms of viral tropism, immunopathology, and correlates of protec- Outstanding Questions
tion [126]. Further research is urgently needed to explore the durability of immune responses Which animal(s) serves as the natural
to SARS-CoV-2, considering reports of waning immunity to other CoVs and the detection of reservoir of SARS-CoV-2?
pre-existing cross-reactive ’common-cold‘ CoV T-cells with SARS-CoV-2 in naïve humans
Does active replication of SARS-CoV-
(see Outstanding Questions) [135,136]. 2 in the URT contribute to enhanced
transmissibility in humans?
Concluding Remarks
Is intestinal SARS-CoV-2 infection a
The emergence of SARS-CoV-2 as the most recent example of zoonotic virus spillovers into source of virus transmission?
humans underscores the fundamental need for well-funded surveillance organizations. The
unrivaled spread of SARS-CoV-2 urgently demands that the global science community acts in Which SARS-CoV-2 proteins antag-
onize innate and adaptive immune
harmony to disseminate accurate and stipulatory knowledge, with an immediate potential to
responses? Do the SARS-CoV-2 pro-
influence policy and public health strategies/interventions at the national and local levels. Studies teins with more potent antagonistic
stemming from previous CoVs have jumpstarted our basic understanding of SARS-CoV-2 immune functions increase virulence in
biology and facilitated the rapid deployment of vaccine candidates into clinical trials. Although humans compared with other HCoVs?
Why do some recovered patients fail
clinical symptoms of COVID-19 resemble some aspects of SARS and MERS, respectively, distinct
to develop neutralizing antibodies?
and significant disparities exist in the transmissibility and immune responses to SARS-CoV-2. To
modify therapeutics that protect against the full spectrum of disease (see Outstanding Questions), What are the host and/or viral factors
additional research will need to define the molecular mechanisms contributing to severe driving inflammatory imbalances in
severe COVID-19 cases?
immunopathology in some patients, whereas others are completely protected from clinical
symptoms. Prior work involving other CoVs, together with current and future studies of SARS- What are the underlying mechanisms
CoV-2, should provide society with the foundational knowledge to prepare for expected seasonal contributing to an inadequate IFN
response to SARS-CoV-2?
resurgences of SARS-CoV-2, as well as any potential spillover of additional CoVs into human
populations; hopefully, current investigations will make important advances and contribute to What are the correlates of immune
increased knowledge and preparedness in this regard. protection for SARS-CoV2 and will
they provide sterilizing immunity?
Acknowledgment
Will candidate vaccines against SARS-
This work was supported by a National Institutes of Health grant R01AI132526 to P.W.
CoV2 also be effective in older subpop-
ulations (with or without comorbidities)?
References
1. Zhou, P. et al. (2020) A pneumonia outbreak associated with a 13. Peiris, J. et al. (2003) Clinical progression and viral load in a
new coronavirus of probable bat origin. Nature 579, 270–273 community outbreak of coronavirus-associated SARS pneu-
2. Peiris, J. et al. (2003) Coronavirus as a possible cause of monia: a prospective study. Lancet 361, 1767–1772
severe acute respiratory syndrome. Lancet 361, 1319–1325 14. Wrapp, D. et al. (2020) Cryo-EM structure of the 2019-nCoV
3. Huang, C. et al. (2020) Clinical features of patients infected spike in the prefusion conformation. Science 367, 1260–1263
with 2019 novel coronavirus in Wuhan, China. Lancet 395, 15. Li, W. et al. (2003) Angiotensin-converting enzyme 2 is a func-
497–506 tional receptor for the SARS coronavirus. Nature 426, 450–454
4. WHO (2020) Coronavirus Disease (COVID-19) Pandemic, 16. Yan, R. et al. (2020) Structural basis for the recognition of
WHO SARS-CoV-2 by full-length human ACE2. Science 367,
5. Gorbalenya, A.E. et al. (2020) The species Severe acute respi- 1444–1448
ratory syndrome-related coronavirus: classifying 2019-nCoV 17. Shang, J. et al. (2020) Cell entry mechanisms of SARS-CoV-2.
and naming it SARS-CoV-2. Nat. Microbiol. 5, 536–544 Proc. Natl. Acad. Sci. U. S. A. 117, 11727–11734
6. Lu, R. et al. (2020) Genomic characterisation and epidemiol- 18. Hoffmann, M. et al. (2020) A multibasic cleavage site in the
ogy of 2019 novel coronavirus: implications for virus origins spike protein of SARS-CoV-2 is essential for infection of
and receptor binding. Lancet 395, 565–574 human lung cells. Mol. Cell 78, 779–784
7. Zhang, Y.Z. and Holmes, E.C. (2020) A genomic perspective 19. Yurkovetskiy, L. et al. (2020) Structural and functional analysis
on the origin and emergence of SARS-CoV-2. Cell 181, of the D614G SARS-CoV-2 spike protein variant. Cell.
223–227 Published online September 15, 2020. https://doi.org/
8. Perlman, S. and Netland, J. (2009) Coronaviruses post-SARS: 10.1016/j.cell.2020.09.032
update on replication and pathogenesis. Nat. Rev. Microbiol. 20. Li, Q. et al. (2020) The impact of mutations in SARS-CoV-2
7, 439–450 spike on viral infectivity and antigenicity. Cell 182, 1284–1294
9. Wu, F. et al. (2020) A new coronavirus associated with human 21. Korber, B. et al. (2020) Tracking changes in SARS-CoV-2
respiratory disease in China. Nature 579, 265–269 Spike: evidence that D614G increases infectivity of the
10. Hoffmann, M. et al. (2020) SARS-CoV-2 cell entry depends on COVID-19 virus. Cell 182, 812–827
ACE2 and TMPRSS2 and is blocked by a clinically proven 22. Jia, H.P. et al. (2005) ACE2 receptor expression and severe
protease inhibitor. Cell 181, 271–280 acute respiratory syndrome coronavirus infection depend
11. Snijder, E.J. et al. (2006) Ultrastructure and origin of membrane on differentiation of human airway epithelia. J. Virol. 79,
vesicles associated with the severe acute respiratory syndrome 14614–14621
coronavirus replication complex. J. Virol. 80, 5927–5940 23. Hamming, I. et al. (2004) Tissue distribution of ACE2
12. Wu, H.-Y. and Brian, D.A. (2010) Subgenomic messenger RNA protein, the functional receptor for SARS coronavirus. A
amplification in coronaviruses. Proc. Natl. Acad. Sci. U. S. A. first step in understanding SARS pathogenesis. J. Pathol.
107, 12257–12262 203, 631–637

1112 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

24. Kuba, K. et al. (2005) A crucial role of angiotensin converting person transmission: a study of a family cluster. Lancet 395,
enzyme 2 (ACE2) in SARS coronavirus-induced lung injury. 514–523
Nat. Med. 11, 875–879 49. Ferrett, L. et al. (2020) Quantifying SARS-CoV-2 transmission
25. Ziegler, C.G.K. et al. (2020) SARS-CoV-2 receptor ACE2 is an suggests epidemic control with digital contact tracing. Science
interferon-stimulated gene in human airway epithelial cells and 368, eabb6936
is detected in specific cell subsets across tissues. Cell 181, 50. Arons, M.M. et al. (2020) Presymptomatic SARS-CoV-2 infec-
1016–1035 tions and transmission in a skilled nursing facility. N. Engl.
26. Sun, K. et al. (2020) Atlas of ACE2 gene expression in mam- J. Med. 382, 2081–2090
mals reveals novel insights in transmission of SARS-CoV-2. 51. Zou, L. et al. (2020) SARS-CoV-2 viral load in upper respiratory
BioRxiv. Published online March 31, 2020. https://doi.org/ specimens of infected patients. N. Engl. J. Med. 382,
10.1101/2020.03.30.015644 1177–1179
27. Shulla, A. et al. (2011) A transmembrane serine protease is 52. Pan, Y. et al. (2020) Viral load of SARS-CoV-2 in clinical
linked to the severe acute respiratory syndrome coronavirus samples. Lancet Infect. Dis. 20, 411–412
receptor and activates virus entry. J. Virol. 85, 873–882 53. Mizumoto, K. et al. (2020) Estimating the asymptomatic
28. Zhao, Y. et al. (2020) Single-cell RNA expression profiling of proportion of coronavirus disease 2019 (COVID-19) cases on
ACE2, the receptor of SARS-CoV-2. Am. J. Respir. Crit. Care board the Diamond Princess cruise ship, Yokohama, Japan,
Med. 202, 756–759 2020. Eurosurveillance 25, 2000180
29. Imai, Y. et al. (2005) Angiotensin-converting enzyme 2 protects 54. Meng, Heng et al. (2020) CT imaging and clinical course of
from severe acute lung failure. Nature 436, 112–116 asymptomatic cases with COVID-19 pneumonia at admission
30. Wang, S. et al. (2008) Endocytosis of the receptor-binding in Wuhan, China. J. Infect. 81, 33–39
domain of SARS-CoV spike protein together with virus receptor 55. Rothe, C. et al. (2020) Transmission of 2019-nCoV infection
ACE2. Virus Res. 136, 8–15 from an asymptomatic contact in Germany. N. Engl. J. Med.
31. Smith, J.C. et al. (2020) Cigarette smoke exposure and in- 382, 970–971
flammatory signaling increase the expression of the SARS- 56. He, G. et al. (2020) The clinical feature of silent infections of
CoV-2 receptor ACE2 in the respiratory tract. Dev. Cell 53, novel coronavirus infection (COVID-19) in Wenzhou. J. Med.
514–529 Virol. 92, 1761–1763
32. Follis, K.E. et al. (2006) Furin cleavage of the SARS coronavirus 57. Mukhra, R. et al. (2020) Possible modes of transmission of
spike glycoprotein enhances cell-cell fusion but does not affect novel coronavirus SARS-CoV-2: a review. Acta Biomed 91,
virion entry. Virology 350, 358–369 e2020036
33. Lukassen, S. et al. (2020) SARS-CoV-2 receptor ACE2 and 58. Tomlinson, B. and Cockram, C. (2003) SARS: experience
TMPRSS2 are primarily expressed in bronchial transient secre- at Prince of Wales Hospital, Hong Kong. Lancet 361,
tory cells. EMBO J. 39, e105114 1486–1487
34. Hui, K.P.Y. et al. (2020) Tropism, replication competence, and 59. Liu, Y. et al. (2020) Aerodynamic analysis of SARS-CoV-2 in
innate immune responses of the coronavirus SARS-CoV-2 in two Wuhan hospitals. Nature 582, 557–560
human respiratory tract and conjunctiva: an analysis in 60. Doremalen, N.v. et al. (2020) Aerosol and surface stability of
ex-vivo and in-vitro cultures. Lancet Respir. Med. 8, 687–695 SARS-CoV-2 as compared with SARS-CoV-1. N. Engl.
35. COVID-19 Investigation Team (2020) Clinical and virologic J. Med. 382, 1564–1567
characteristics of the first 12 patients with coronavirus disease 61. Ong, S.W.X. et al. (2020) Air, surface environmental, and
2019 (COVID-19) in the United States. Nat. Med. 26, 861–868 personal protective equipment contamination by severe acute
36. Xu, H. et al. (2020) High expression of ACE2 receptor of respiratory syndrome coronavirus 2 (SARS-CoV-2) from a
2019-nCoV on the epithelial cells of oral mucosa. Int. symptomatic patient. JAMA 323, 1610–1612
J. Oral Sci. 12, 8 62. Xu, Y. et al. (2020) Characteristics of pediatric SARS-CoV-2
37. Wölfel, R. et al. (2020) Virological assessment of hospitalized infection and potential evidence for persistent fecal viral
patients with COVID-2019. Nature 581, 465–469 shedding. Nat. Med. 26, 502–505
38. Su, S. et al. (2016) Epidemiology, genetic recombination, and 63. Ksiazek, T.G. et al. (2003) A novel coronavirus associated with
pathogenesis of coronaviruses. Trends Microbiol. 24, 490–502 severe acute respiratory syndrome. N. Engl. J. Med. 348,
39. Zhang, H. et al. (2020) The digestive system is a potential route 1953–1966
of 2019-nCov infection: a bioinformatics analysis based on 64. Guery, B. et al. (2013) Clinical features and viral diagnosis of
single-cell transcriptomes. Gut 69, 1010–1018 two cases of infection with Middle East Respiratory Syndrome
40. Cholankeril, G. et al. (2020) High prevalence of concurrent coronavirus: a report of nosocomial transmission. Lancet 381,
gastrointestinal manifestations in patients with SARS-CoV-2: 2265–2272
early experience from California. Gastroenterology 159, 775–777 65. Chen, N. et al. (2020) Epidemiological and clinical characteris-
41. Xiao, F. et al. (2020) Evidence for gastrointestinal infection of tics of 99 cases of 2019 novel coronavirus pneumonia in
SARS-CoV-2. Gastroenterology 158, 1831–1833 Wuhan, China: a descriptive study. Lancet 395, 507–513
42. Leung, W.K. et al. (2003) Enteric involvement of severe 66. Puntmann, V.O. et al. (2020) Outcomes of cardiovascular
acute respiratory syndrome-associated coronavirus infection. magnetic resonance imaging in patients recently recovered
Gastroenterology 125, 1011–1017 from coronavirus disease 2019 (COVID-19). JAMA Cardiol.
43. Yu, I.T.S. et al. (2004) Evidence of airborne transmission of the Published online July 27, 2020. https://doi.org/10.1001/
severe acute respiratory syndrome virus. N. Engl. J. Med. 350, jamacardio.2020.3557
1731–1739 67. Chao, J.Y. et al. (2020) Clinical characteristics and outcomes of
44. Otter, J.A. et al. (2016) Transmission of SARS and MERS hospitalized and critically ill children and adolescents with coro-
coronaviruses and influenza virus in healthcare settings: the navirus disease 2019 (COVID-19) at a tertiary care medical
possible role of dry surface contamination. J. Hosp. Infect. center in New York City. J. Pediatr. 223, 14–19
92, 235–250 68. Nikolich-Zugich, J. (2018) The twilight of immunity: emerging
45. Li, Y. et al. (2004) Role of air distribution in SARS transmission concepts in aging of the immune system. Nat. Immunol. 19,
during the largest nosocomial outbreak in Hong Kong. Int. 10–19
J. Indoor Environ. Health 15, 83–95 69. Wu, J.T. et al. (2020) Estimating clinical severity of COVID-19
46. Wang, D. et al. (2020) Clinical characteristics of 138 hospital- from the transmission dynamics in Wuhan, China. Nat. Med.
ized patients with 2019 novel coronavirus-infected pneumonia 26, 506–510
in Wuhan, China. JAMA 323, 1061–1069 70. Chow, R.D. and Chen, S. (2020) The aging transcriptome and
47. Li, Q. et al. (2020) Early transmission dynamics in Wuhan, cellular landscape of the human lung in relation to SARS-CoV-2.
China, of novel coronavirus-infected pneumonia. N. Engl. BioRxiv. Published online April 15, 2020. https://doi.org/
J. Med. 382, 1199–1207 10.1101/2020.04.07.030684
48. Chan, J.F.W. et al. (2020) A familial cluster of pneumonia asso- 71. Yang, X. et al. (2020) Clinical course and outcomes of critically
ciated with the 2019 novel coronavirus indicating person-to- ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a

Trends in Immunology, December 2020, Vol. 41, No. 12 1113

Trends in Immunology

single-centered, retrospective, observational study. Lancet 97. Frieman, M. et al. (2009) Severe acute respiratory syndrome
Respir. Med. 8, 475–481 coronavirus papain-like protease ubiquitin-like domain and
72. Imai, S.T.K. et al. (2020) Clinical characteristics of COVID-19 in catalytic domain regulate antagonism of IRF3 and NF-κB
104 people with SARS-CoV-2 infection on the Diamond signaling. J. Virol. 83, 6689–6705
Princess cruise ship: a retrospective analysis. Lancet Infect. 98. Bailey-Elkin, B.A. et al. (2014) Crystal structure of the
Dis. 20, 1043–1050 Middle East respiratory syndrome coronavirus (MERS-CoV)
73. Zhang, B. et al. (2020) Clinical characteristics of 82 cases of papain-like protease bound to ubiquitin facilitates targeted
death from COVID-19. PLoS ONE 15 disruption of deubiquitinating activity to demonstrate its
74. Wu, C. et al. (2020) Risk factors associated with acute respiratory role in innate immune suppression. J. Biol. Chem. 289,
distress syndrome and death in patients with coronavirus disease 34667–34682
2019 pneumonia in Wuhan, China. JAMA Intern. Med. 180, 7 99. Aarreberg, L.D. et al. (2019) Interleukin-1β induces mtDNA
75. Liu, K. et al. (2020) Clinical features of COVID-19 in elderly release to activate innate immune signaling via cGAS-STING.
patients: a comparison with young and middle-aged patients. Mol. Cell 74, 801–815
J. Infect. 80, 14–18 100. Sun, L. et al. (2012) Coronavirus papain-like proteases nega-
76. Zhou, R. et al. (2020) Acute SARS-CoV-2 infection impairs tively regulate antiviral innate immune response through disrup-
dendritic cell and T cell responses. Immunity. Published online tion of STING-mediated signaling. PLoS ONE 7, e30802
August 4, 2020. https://doi.org/10.1016/j.immuni.2020.07.026 101. Lei, X. et al. (2020) Activation and evasion of type I interferon
77. Chen, J. et al. (2010) Cellular immune responses to severe responses by SARS-CoV-2. Nat. Commun. 11, 3810
acute respiratory syndrome coronavirus (SARS-CoV) infection 102. Lokugamage, K.G. et al. (2012) Severe acute respiratory syn-
in senescent BALB/c mice: CD4+ T cells are important in drome coronavirus protein nsp1 is a novel eukaryotic transla-
control of SARS-CoV infection. J. Virol. 84, 1289–1301 tion inhibitor that represses multiple steps of translation
78. Zhao, J. et al. (2009) Evasion by stealth: inefficient immune initiation. J. Virol. 86, 13598–13608
activation underlies poor T cell response and severe disease 103. Lokugamage, K.G. et al. (2015) Middle East Respiratory
in SARS-CoV-infected mice. PLoS Pathog. 5, e1000636 syndrome coronavirus nsp1 inhibits host gene expression
79. Liao, M. et al. (2020) Single-cell landscape of bronchoalveolar by selectively targeting mRNAs transcribed in the nucleus
immune cells in patients with COVID-19. Nat. Med. 26 while sparing mRNAs of cytoplasmic origin. J. Virol. 89,
80. Qin, C. et al. (2020) Dysregulation of immune response in 10970–109981
patients with COVID-19 in Wuhan, China. Clin. Infect. Dis. 104. Thoms, M. et al. (2020) Structural basis for translational shut-
71, 762–768 down and immune evasion by the Nsp1 protein of SARS-
81. Blanco-Melo, D. et al. (2020) Imbalanced host response to CoV-2. Science 369, 1249–1255
SARS-CoV-2 drives development of COVID-19. Cell 181, 105. Xia, H. et al. (2020) Evasion of type-I interferon by SARS-CoV-2.
1036–1045 Cell Rep. 33, 108234
82. Chen, H. et al. (2020) SARS-CoV-2 activates lung epithelia cell 106. Yang, Y. et al. (2013) The structural and accessory proteins M,
proinflammatory signaling and leads to immune dysregulation ORF 4a, ORF 4b, and ORF 5 of Middle East respiratory
in COVID-19 patients by single-cell sequencing. MedRxiv. syndrome coronavirus (MERS-CoV) are potent interferon
Published online May 13, 2020. https://doi.org/10.1101/ antagonists. Protein Cell 4, 951–961
2020.05.08.20096024 107. Kopecky-Bromberg, S.A. et al. (2007) Severe acute respiratory
83. Ding, Y. et al. (2003) The clinical pathology of severe acute syndrome coronavirus open reading frame (ORF) 3b, ORF 6,
respiratory syndrome (SARS): a report from China. J. Pathol. and nucleocapsid proteins function as interferon antagonists.
200, 282–289 J. Virol. 81, 548–557
84. Nicholls, J.M. et al. (2003) Lung pathology of fatal severe acute 108. Zhao, J. et al. (2014) Rapid generation of a mouse model
respiratory syndrome. Lancet 361, 1773–1778 for Middle East respiratory syndrome. Proc. Natl. Acad. Sci.
85. Ruan, Q. et al. Clinical predictors of mortality due to COVID-19 U. S. A. 111, 4970–4975
based on an analysis of data of 150 patients from Wuhan, 109. Sun, Jing et al. (2020) Generation of a broadly useful model for
China. Intensive Care Med. 46, 846-848 COVID-19 pathogenesis vaccination, and treatment. Cell. 182,
86. Moore, J.B. and June, C.H. (2020) Cytokine release syndrome 734–743
in severe COVID-19. Science 368, 473–474 110. Zielecki, F. et al. (2013) Human cell tropism and innate immune
87. Channappanavar, R. and Perlman, S. (2017) Pathogenic system interactions of human respiratory coronavirus EMC
human coronavirus infections: causes and consequences of compared to those of severe acute respiratory syndrome
cytokine storm and immunopathology. Semin. Immunopathol. coronavirus. J. Virol. 87, 5300–5304
39, 529–539 111. Chu, H. et al. (2020) Comparative replication and immune acti-
88. Marnell, L. et al. (2005) C-reactive protein: ligands, receptors vation profiles of SARS-CoV-2 and SARS-CoV in human lungs:
and role in inflammation. Clin. Immunol. 117, 104–111 an ex vivo study with implications for the pathogenesis of
89. Zuo, Y. et al. (2020) Neutrophil extracellular traps in COVID-19. COVID-19. Clin. Infect. Dis. 71, 1400–1409
JCI Insight 5, e138999 112. Züst, R. et al. (2011) Ribose 2′-O-methylation provides a
90. Varga, Z. et al. (2020) Endothelial cell infection and endotheliitis molecular signature for the distinction of self and non-self
in COVID-19. Lancet 395, 1417–1418 mRNA dependent on the RNA sensor Mda5. Nat. Immunol.
91. Zhou, F. et al. (2020) Clinical course and risk factors for 12, 137–143
mortality of adult inpatients with COVID-19 in Wuhan, China: 113. Bao, L. et al. (2020) The pathogenicity of SARS-CoV-2 in
a retrospective cohort study. Lancet 395, 1054–1062 hACE2 transgenic mice. Nature 583, 830–833
92. Tang, N. et al. (2020) Abnormal coagulation parameters are 114. McCray, P.B. et al. (2007) Lethal infection of K18-hACE2 mice
associated with poor prognosis in patients with novel corona- infected with severe acute respiratory syndrome coronavirus.
virus pneumonia. J. Thromb. Haemost. 18, 844–847 J. Virol. 81, 813–821
93. Totura, A.L. and Baric, R. (2012) SARS coronavirus pathogenesis: 115. Tseng, C.-T.K. et al. (2007) Severe acute respiratory syndrome
host innate immune responses and viral antagonism of interferon. coronavirus infection of mice transgenic for the human
Curr. Opin. Virol. 2, 264–275 angiotensin-converting enzyme 2 virus receptor. J. Virol. 81,
94. Kindler, E. et al. (2016) Interaction of SARS and MERS 1162–1173
coronaviruses with the antiviral interferon response. Adv. Viral 116. Zheng, J. et al. (2020) K18-hACE2 mice for studies of COVID-19
Res. 96, 219–243 treatments and pathogenesis including anosmia. BioRxiv.
95. Hadjadj, J. et al. (2020) Impaired type I interferon activity and Published online August 10, 2020. https://doi.org/10.1101/
inflammatory responses in severe COVID-19 patients. Sci. 2020.08.07.242073
Immunol. 369, 718–724 117. Winkler, E.S. et al. (2020) SARS-CoV-2 infection of human
96. Devaraj, S.G. et al. (2007) Regulation of IRF-3 dependent ACE2-transgenic mice causes severe lung inflammation and
innate immunity by the Papain-like protease domain of the impaired function. Nat. Immunol. Published online August 24,
SARS coronavirus. J. Biol. Chem. 282, 32208–32221 2020. https://doi.org/10.1038/s41590-020-0778-2

1114 Trends in Immunology, December 2020, Vol. 41, No. 12

Trends in Immunology

118. Netland, J. et al. (2008) Severe acute respiratory syndrome 130. Liu, L. et al. (2019) Anti-spike IgG causes severe acute lung
coronavirus infection causes neuronal death in the absence of injury by skewing macrophage responses during acute
encephalitis in mice transgenic for human ACE2. J. Virol. 82, SARS-CoV infection. JCI Insight 4, e123158
7264–7275 131. Zhao, J. et al. (2020) Antibody responses to SARS-CoV-2 in
119. Wold, W.S. and Toth, K. (2013) Adenovirus vectors for gene patients of novel coronavirus disease 2019. Clin. Infect. Dis.
therapy, vaccination and cancer gene therapy. Curr. Gene Published online March 28, 2020. https://doi.org/10.1093/
Ther. 13, 421–433 cid/ciaa344
120. Roberts, A. et al. (2007) A mouse-adapted SARS-coronavirus 132. Long, Q.X. et al. (2020) Antibody responses to SARS-CoV-2 in
causes disease and mortality in BALB/c mice. PLoS Pathog. patients with COVID-19. Nat. Med. 26, 845–848
3, e5 133. Mathew, D. et al. (2020) Deep immune profiling of COVID-19
121. Dinnon, K.H. et al. (2020) A mouse-adapted SARS-CoV-2 model patients reveals distinct immunotypes with therapeutic
for the evaluation of COVID-19 medical countermeasures. Nature. implications. Science 369, eabc8511
Published online August 27, 2020. https://doi.org/10.1038/ 134. Yu, J. et al. (2020) DNA vaccine protection against SARS-CoV-2
s41586-020-2708-8 in rhesus macaques. Science 369, 806–811
122. Gordon, D.E. et al. (2020) A SARS-CoV-2 protein interaction 135. Edridge, A.W.D. et al. (2020) Seasonal coronavirus protective
map reveals targets for drug repurposing. Nature 583, immunity is short-lasting. Nat. Med. Published online September
459–468 14, 2020. https://doi.org/10.1038/s41591-020-1083-1
123. Leist, S.R. et al. (2020) A mouse-adapted SARS-CoV-2 136. Grifoni, A. et al. (2020) Targets of T cell responses to SARS-
induces acute lung injury (ALI) and mortality in standard labora- CoV-2 coronavirus in humans with COVID-19 disease and
tory mice. Cell. Published online September 23, 2020. https:// unexposed individuals. Cell 181, 1489–1501
doi.org/10.1016/j.cell.2020.09.050 137. Zhang, J.-J. et al. (2020) Clinical characteristics of 140 patients
124. Munster, V.J. et al. (2020) Respiratory disease in rhesus infected with SARS-CoV-2 in Wuhan, China. Allergy 75,
macaques inoculated with SARS-CoV-2. Nature 585, 1730–1741
268–272 138. Liao, J. et al. (2020) Epidemiological and clinical characteristics
125. Rockx, B. et al. (2020) Comparative pathogenesis of COVID-19, of COVID-19 in adolescents and young adults. Innovation 1,
MERS, and SARS in a nonhuman primate model. Science 368, 100001
1012–1015 139. Rifno, N. et al. (2020) Neurologic manifestations in 1760
126. Chandrashekar, A. et al. (2020) SARS-CoV-2 infection protects COVID-19 patients admitted to Papa Giovanni XXIII Hospital,
against rechallenge in rhesus macaques. Science 369, 812–817 Bergamo, Italy. J. Neurol. Published online October 7, 2020.
127. Lawler, J.V. et al. (2006) Cynomolgus macaque as an animal https://doi.org/10.1007/s00415-020-10251-5
model for severe acute respiratory syndrome. PLoS Med. 3, 140. Guan, W.J. et al. (2020) Clinical characteristics of coronavirus
e149 disease 2019 in China. N Engl J Med 382, 1708–1720.
128. Yu, P. et al. (2020) Age-related rhesus macaque models of https://doi.org/10.1056/NEJMoa2002032
COVID-19. Anim. Models Exp. Med. 3, 93–97 141. Imai, S.T.K. et al. (2020) Clinical characteristics of COVID-19 in
129. Tseng, C.-T. et al. (2012) Immunization with SARS coronavirus 104 people with SARS-CoV-2 infection on the Diamond Prin-
vaccines leads to pulmonary immunopathology on challenge cess cruise ship: a retrospective analysis. Lancet Infectious Dis-
with the SARS virus. PLoS ONE 7, e35421 eases 20 (9), 1043–1050

Trends in Immunology, December 2020, Vol. 41, No. 12 1115