key: cord-0839690-8ucweqsf
authors: Brechbühl, Julien; Wood, Dean; Bouteiller, Sofiane; Lopes, Ana Catarina; Verdumo, Chantal; Broillet, Marie-Christine
title: Age-dependent appearance of SARS-CoV-2 entry cells in mouse chemosensory systems reflects COVID-19 anosmia and ageusia symptoms
date: 2021-03-29
journal: bioRxiv
DOI: 10.1101/2021.03.29.437530
sha: fed8110cbe6c7431e5fa7c9aa208f7202c9d095d
doc_id: 839690
cord_uid: 8ucweqsf

COVID-19 pandemic has given rise to a collective scientific effort to study its viral causing agent SARS-CoV-2. Research is focusing in particular on its infection mechanisms and on the associated-disease symptoms. Interestingly, this environmental pathogen directly affects the human chemosensory systems leading to anosmia and ageusia. Evidence for the presence of the cellular entry sites of the virus, the ACE2/TMPRSS2 proteins, has been reported in non-chemosensory cells in the rodent’s nose and mouth, missing a direct correlation between the symptoms reported in patients and the observed direct viral infection in human sensory cells. Here, mapping the gene and protein expression of ACE2/TMPRSS2 in the mouse olfactory and gustatory cells, we precisely identified the virus target cells to be of basal and sensory origin and we revealed their age-dependent appearance. Our results not only clarify human viral-induced sensory symptoms but also propose new investigative perspectives based on ACE2-humanized mouse models.

The Corona Virus Disease 2019 (COVID-19) has federated worldwide scientific efforts for understanding the viral epidemiological mechanisms of the coronavirus 2 (SARS-CoV-2) that causes this severe acute respiratory syndrome. In humans, the viral syndrome is characterized by an increased mortality rate in aged and/or comorbidity patients associated with the upper respiratory infection symptoms, such as severe respiratory distress [1] [2] [3] . In addition to its major impact, COVID-19 is associated by its direct alteration of human olfaction and gustation, in absence of substantial nasal inflammation or coryzal signs, resulting to anosmia and ageusia in up to 77% of the patients [4] [5] [6] [7] . While these sensory symptoms are well established and intensely affect everyday behaviors 8, 9 , the precise related mechanisms remain elusive 10 .

The target cells of the virus share a molecular signature: the concomitant cellular expression of the angiotensin-converting enzyme 2 (ACE2) and of its facilitating transmembrane serine protease 2 (TMPRSS2), which plays a crucial role for the interaction of viral spike proteins with the host cell [11] [12] [13] . Paradoxically, these entry sites seem to be lacking in sensory cells [14] [15] [16] [17] [18] , while a direct SARS-CoV-2 contamination has been observed both in humans and rodents 19, 20 , requesting further investigations to explain the sensory-associated symptoms [21] [22] [23] [24] . Therefore, the characterization of the animal model is necessary prior to its use to understand the causality underling the viral-induced sensory symptoms.

The use of mice is indeed limited for epidemiological studies due to their absence of hands, which, with aerosols, are the foremost passages of inter-individual viral transmission 25 , as well as their published lack of SARS-CoV-2 ACE2-spike protein affinity 26, 27 . Nevertheless, the ease of production of genetically-modified mice and their scientific availability, as well as their well-studied and specialized chemosensory systems [28] [29] [30] , make them a valuable ally for the development of potential prophylactic and protective treatments related to these sensory symptoms.

Thus, we aimed here at characterizing the potential viral entry sites across mouse sensory systems. We found SARS-CoV-2 entry cells to be of different origins depending on the sensory systems. In summary, the virus could target cells involved in tissue regulation such as the supporting cells of the olfactory receptor neurons and the regenerative basal cells but also, specifically, the chemosensory cells of both gustatory and olfactory systems. We finally revealed that the emergence of viral entry sites in sensory and basal cells only occurs with age, which could explain both, the observed COVID-19 long-lasting effects and the age-dependent sensorysymptomatology in human.

systems. Focusing on the mouse, where chemosensing takes place in different sensory systems (Fig. 1a) , we took advantage of our previous studies [31] [32] [33] [34] [35] to first assess their overall expression of the major SARS-CoV-2 entry sites. Interestingly, we found a differential expression of the Ace2 and Tmprss2 transcripts (Fig. 1b) . Ace2 is strongly expressed in a specific area of the main olfactory epithelium (MOE), the dorsal part (MOED) which is directly exposed to the environment and specialized in sensing volatile chemical cues 14, 15 . Moreover, we observed a previously unreported expression of Ace2, in the most rostral sensory subsystem, the Grueneberg ganglion (GG), mostly implicated in volatile danger cues detection [34] [35] [36] and in the circumvallate taste papilla (CV) involved in water-soluble tastant perception 29 . Interestingly, this specific pattern of Ace2 expression seems to be correlated with the mode of viral dissemination (volatile suspensions of viral droplets 25 ), as only a limited expression of Ace2 is found in the vomeronasal organ (VNO) and in the septal organ of Masera (SO), considered to be implicated in pheromonal and retronasal communications via indirect and only limited access to the environment 30 . Concerning the associated facilitating cleavage protease, we found that the Tmprss2 transcript was expressed in all olfactory subsystems and to a limited extent at the CV level (Fig. 1b) . Thus, this apparent disparity of expression requires further investigations to identify the precise cells expressing these entry sites in the different mouse subsystems.

Taking advantage of a genetically modified mouse model in which, the olfactory marker protein 37 (OMP) drives the expression of the green fluorescent protein (GFP) in all mature olfactory neurons 38,39 , we first examined the MOED sensory epithelium (Fig. 2) which is continuously exposed to inhaled air (Fig. 2a) . Remarkably, we found by immunohistochemical stainings that the ACE2 protein was not only expressed at the apical surface and in the Bowman's glands [14] [15] [16] 21 but also in the basal layer (Fig. 2b) .

Moreover, we found that, in addition to being irregularly localized in different regions of the neuroepithelium 15 ( Supplementary Fig. 1 ), TMPRSS2 was mostly co-expressed with ACE2 in these OMP-negative basal cells (Fig. 2c, d) . Focusing on the identification of these ACE2-expressing cells (Fig. 3) , we first confirmed that the cytokeratin 18 (CK18)-positive sustentacular cells 16, 21 , operating as supporting cells for the olfactory neurons, were indeed harboring, at their luminal surface, the observed ACE2 protein (Fig. 3a) . We next established the multipotency characteristics of the ACE2-expressing basal cells, as they also expressed two proteins specific to their identity of stem cells, the perinuclear cytokeratin 5 (CK5; Fig. 3b ) and the nuclear transcription factor sex determining region Y-box 2 (SOX2; Fig. 3c ), markers of horizontal basal cells (HBCs) or of the pear-shaped globose basal cells (GBCs) respectively. These two populations of basal cells are both involved in the regeneration of the neuroepithelium by acting as short-and long-lasting reservoir cells 14, 40, 41 . Taken together, we confirmed our initial RT-PCR results (Fig. 1b) by precisely profiling the ACE2 and TMPRSS2 expression in the MOED neuroepithelium. We found that the cell candidates for viral entry are ACE2expressing cells of non-neuronal and multipotent origin.

Wondering about the apparent limited level of Tmprss2 transcript in the CV (Fig. 1b) , we next decided to exploit our histological methodology (Figs. 2, 3) to precisely localize the potential SARS-CoV-2 entry sites in sensory cells of the CV taste buds (Fig. 4) . For that, we first used Ga gustucin (GUST) as a marker of sensory cells in the tongue 32 , to identify taste buds in close contact with the oral cavity (Fig. 4a) . Surprisingly, using a double immunostaining approach, we distinctly observed ACE2 expression, not only in apical keratinocytes 17, 42 , but also in sensory cells (Fig. 4b) . Moreover, ACE2 is predominantly expressed in the microvilli of sensory cells, gathered in the so-called taste pore, which are directly exposed to substances entering the oral cavity 29 , such as viral droplets ( Fig. 4b) . Furthermore, we observed a restricted localization of TMPRSS2 in the taste pore region, strikingly co-expressed with ACE2 (Fig. 4c , d) which might also explain the observed low-level signal for the Tmprss2 transcript (Fig. 1b) . We next confirmed the mature status of these ACE2-expressing sensory cells as they preferentially expressed the CK18 marker 43 (Fig. 5a) . Moreover, and contrary to the MOED, we found only sporadic ACE2 expression in basal CK5-positive cells (Fig. 5b ) and SOX2-positive cells (Fig. 5c, d) . In summary, we identified here the co-expression of the main SARS-CoV-2 entry sites in mature taste sensory cells.

In the mouse, there is an ancestral sensory subsystem, the GG (Fig. 6 ), which has the particularity of displaying both olfactory and gustatory traits 31, 32, 44 . Morphologically, GG neurons are covered with a keratinized epithelium permeable to volatile water-soluble molecules coming from the nasal cavity 35 (Fig. 6a) . Molecularly, it expresses both main sensory markers, OMP and GUST 32 and uses common sensory signals such as olfactory and taste receptors 32 to respond to both danger-associated odorants and tastants 32, 35, 36 .

Performing histological approaches, we first noticed that the apical keratinocyte cell layer was, as in the CV, positive for ACE2 expression (Fig. 6b) . But more surprising, we next found ACE2 expression in the GG cells (Fig. 6b ). Interestingly and contrary to the MOED, this expression was distinctly restricted to the sensory GG neurons and not found in the supporting S100 calcium-binding protein b (S100B)-expressing glial cells 35 ( Supplementary Fig. 2a) . Remarkably, we furthermore localized TMPRSS2 both in apical keratinocytes and in GG neurons indicating that the two main SARS-CoV-2 entry sites are indeed found in the GG sensory subsystem (Fig. 6c, d) . Looking for a precise characterization of ACE2-expressing cells in the GG, we next discovered that OMPpositive GG neurons were also CK18-positive ( Fig. 7a ) giving the GG a double cellular affiliation (neuronal and fibroblastic) that also highlights the discovery of CK18 as a reliable marker of ACE2 expressing cells, the putative target cells of viral infection 45 .

We next observed that in the apical keratinocyte layer of the GG, ACE2-expressing cells are mostly mature as they rarely co-expressed CK5 and SOX2 ( Supplementary   Fig. 2b , c), a characteristic shared with taste tissue (Fig. 5) . Interestingly, and as previously shown, no regeneration happened throughout the lifespan in the GG subsystem 31,46 as basal cells are absent which was further confirmed here by a lack of SOX2 staining ( Supplementary Fig. 2c ). However, we observed in GG neurons, a punctiform-like staining for ACE2 and CK5 ( Supplementary Fig. 2b) indicating a coexpression in a precise cytoskeleton structure, as for the taste buds (Fig. 4b) . We found that these cytoskeleton regions were associated to GG primary cilia (Fig. 7b, c) , an organelle implicated in chemosensing 33, 36 . Thanks to the particulate guanylyl cyclase G (pGCG; Fig. 7b ), a marker of the axonemes of the GG-cilia 33 , we observed that ACE2 was located in the so-called basal body structures (Fig. 7c) , where it was also co-expressed with the gamma-tubulin marker 35 (g-TUB; Fig. 7c ). In summary, we found that the GG sensory subsystem possesses both the viral target cells and the specific protein profiles of expression displayed by both the olfactory and taste systems.

Moreover, we bring here evidences of the presence of viral entry sites in chemosensory neurons.

During the course of our experiments, we noticed interindividual variations of ACE2 expression not only in its intensity but also in its general arrangement (e.g., apical versus basal ACE2 expression in the MOED; (Fig. 2b, c) . Moreover, this observation was not only done for the MOED (Figs. 2, 3 ), but also for the CV (Figs. 4, 5) as well as for the GG (Figs. 6, 7) . We considered a potential age-dependent correlation 47-50 and thus we challenged the expression of ACE2 through the different sensory subsystems using mice of chosen ages. (Fig. 8) . We saw first an intense variation of ACE2 expression in the MOED (Fig. 8a) . Indeed, in young mice, this expression was restricted to the apical region, and then gradually intensifies with age ( Fig. 8a) . At the basal level, it appears to be absent first and only shows up in adult mice (Fig. 8a ). In the sensory cells of the CV, ACE2 appears to be absent in young mice and only emerges with age ( Fig. 8b) . Interestingly, ACE2 expression initiates in the microvilli, at around 5 months of age, and then gradually extents throughout the cell body with aging (Fig. 8b) .

Concerning the GG, we first noticed that its apical keratinocyte cell layer constitutively expressed ACE2, while its expression was distinctly age-dependent in the sensory neurons (Fig. 8c) . Indeed, and as for the sensory cells of the CV (Fig. 8b) , a subcellular expression was first observed which spread into the soma with time (Fig. 8c) . Overall, we demonstrated here a striking age-dependent effect on ACE2 expression across mouse sensory systems, correlating with an increase of viral entry sites.

Our ability to perceive and interact with our environment is directly linked to our senses.

From an evolutionary point of view, our sensory performance is often associated with the way we communicate with our surroundings. In humans, although olfaction and taste have lost importance in comparison with mice, they remain essential not only to our well-being but also to protect us from dangers such as intoxication 10 SARS-CoV-2 seems to have found a mechanism to thwart our sensory defenses.

Indeed, we have shown here that, in mice, these very protective cells express viral entry sites. Although the direct action by the virus on these cells remains to be demonstrated, the underlying inflammatory 22, 24, 54 or cytopathic destruction mechanisms 23 would strongly impact the senses of smell and taste. It therefore seems obvious that their targeted impairment can directly contribute to long-term anosmia and ageusia 7, 55 . Moreover, we have demonstrated here that the sensory cells themselves also express the SARS-CoV-2 entry sites which could contribute to both the sensory symptoms observed on the short term 14, 15, 19, 21 and the ability of coronaviruses to directly infect human sensory cells 20 .

From a clinical point of view, anosmia and ageusia have a low prevalence in infected children and increase with the age of the COVID-19 patient 50, 56 , which seems to be consistent with the protein expression of ACE2 that we have observed in the mice olfactory system. Indeed, after we first confirmed the age-dependent trend made of ACE2 expression in sustentacular cells 47 , we focused on the dorsal part of the MOE and found that the basal cells of the neuronal epithelium also expressed ACE2 as a function of age. Interestingly, TMPRSS2 is also expressed in these SARS-CoV-2 target cells. However, further studies, using viral inoculation for example, are still necessary to link this specific protein expression with viral sensitivity. Moreover, we also observed this age-effect in the taste system. As an absence of expression of the viral entry sites is found in the sensory cells of young mice 17 , but an increased expression of ACE2 and TMPRSS2 appears in sensory cells when the mouse advances in age. Thus, SARS-CoV-2 could therefore also directly target these sensory cells in aged mice and, possibly, in aged humans 42 , altering them and consequently disabling the taste sensory ability leading to ageusia symptoms. Thus, the use of mice sensitized to the virus 45,57-60 appears to be a promising strategy but the age of the mice has to be carefully considered and older mice should preferentially be used (from 9 months of age). Interestingly, based on our results, additional studies performed on the lower respiratory airways could show whether a similar increased ACE2 expression as a function of age occurs, not only in respiratory ciliated cells 49,61 but also in basal cells which could therefore contribute to the differences in pulmonary infectiousness observed between young and older people 62 .

Mice possess multiple sensory systems, separated into specialized subsystems 30 , that we have now precisely characterized for their expression of the viral entry proteins. Interestingly, our study identified the GG to express the SARS-CoV-2 entry sites on its sensory neurons. We also found that GG sensory neurons expressed CK18, confirming that the cellular expression of this protein is an excellent marker for the localization of putative viral entry cells 45 . Moreover, this observation is remarkable and gives to the GG a double cellular affiliation, namely neuronal and fibroblastic by the expression of both OMP and CK18. In this period of race for collective immunity 9 , we desperately need animal models to study this disease. Our results not only confirm the unique primary origin of the GG 44 but, also make it useful as a model of sensory systems to study, in ex-vivo and in-vivo preparations 35 , the mechanism of viral entry into sensory cells as well as the testing of potential protective treatments against viral infection, such as intranasal drugs and / or COVID-19 vaccine candidate delivery 13, 63, 64 .

Reinforcing this idea, it should be noted that another SARS-CoV-2 portal of entry, neuropilin-1 65,66 is also expressed in the GG 67 , thus increasing its infectious susceptibility. 

Clinical Characteristics of Coronavirus Disease 2019 in China

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Virological assessment of hospitalized patients with COVID-2019

Real-time tracking of self-reported symptoms to predict potential COVID-19

Alterations in Smell or Taste in Mildly Symptomatic Outpatients With SARS-CoV-2 Infection

Objective Sensory Testing Methods Reveal a Higher Prevalence of Olfactory Loss in COVID-19-Positive Patients Compared to Subjective Methods: A Systematic Review and Meta-Analysis

Smell and taste changes are early indicators of the COVID-19 pandemic and political decision effectiveness

Five reasons why COVID herd immunity is probably impossible

COVID's toll on smell and taste: what scientists do and don't know

Cleavage and activation of the severe acute respiratory syndrome coronavirus spike protein by human airway trypsin-like protease

Influenza and SARS-coronavirus activating proteases TMPRSS2 and HAT are expressed at multiple sites in human respiratory and gastrointestinal tracts

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Non-neuronal expression of SARS-CoV-2 entry genes in the olfactory system suggests mechanisms underlying COVID-19-associated anosmia

SARS-CoV-2 Receptors and Entry Genes Are Expressed in the Human Olfactory Neuroepithelium and Brain

Elevated ACE-2 expression in the olfactory neuroepithelium: implications for anosmia and upper respiratory SARS-CoV-2 entry and replication

SARS-CoV-2 Receptor ACE2 Is Enriched in a Subpopulation of Mouse Tongue Epithelial Cells in Nongustatory Papillae but Not in Taste Buds or Embryonic Oral Epithelium

Evidence of SARS-CoV2 Entry Protein ACE2 in the Human Nose and Olfactory Bulb

Pathogenesis and transmission of SARS-CoV-2 in golden hamsters

Olfactory transmucosal SARS-CoV-2 invasion as a port of central nervous system entry in individuals with COVID-19

Massive transient damage of the olfactory epithelium associated with infection of sustentacular cells by SARS-CoV-2 in golden Syrian hamsters

Taste and Smell Disorders in COVID-19 Patients: Role of Interleukin-6

Viral infection and smell loss: The case of COVID-19

Pathological inflammation in patients with COVID-19: a key role for monocytes and macrophages

COVID-19 rarely spreads through surfaces. So why are we still deep cleaning?

Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor

Structural basis of receptor recognition by SARS-CoV-2

How the olfactory system makes sense of scents

Genes and ligands for odorant, vomeronasal and taste receptors

Subsystem organization of the mammalian sense of smell

The Grueneberg ganglion controls odor-driven food choices in mice under threat

Alarm pheromone and kairomone detection via bitter taste receptors in the mouse Grueneberg ganglion

Morphological and physiological species-dependent characteristics of the rodent Grueneberg ganglion

Mouse alarm pheromone shares structural similarity with predator scents

Grueneberg ganglion cells mediate alarm pheromone detection in mice

Guanylyl cyclase-G is an alarm pheromone receptor in mice

A brain protein unique to the olfactory bulb

Visualizing an olfactory sensory map

Structure and emergence of specific olfactory glomeruli in the mouse

Deconstructing Olfactory Stem Cell Trajectories at Single-Cell Resolution

Injury Induces Endogenous Reprogramming and Dedifferentiation of Neuronal Progenitors to Multipotency

Existence of SARS-CoV-2 Entry Molecules in the Oral Cavity

Keratin 18 is associated with a subset of older taste cells in the rat

Mouse Grueneberg ganglion neurons share molecular and functional features with C. elegans amphid neurons

SARS-CoV-2 infection of human ACE2-transgenic mice causes severe lung inflammation and impaired function

Projection of the Grüneberg ganglion to the mouse olfactory bulb. The European journal of neuroscience 23

Understanding the age divide in COVID-19: why are children overwhelmingly spared?

Neonatal hyperoxia enhances age-dependent expression of SARS-CoV-2 receptors in mice

Reduced development of COVID-19 in children reveals molecular checkpoints gating pathogenesis illuminating potential therapeutics

The Sense of Smell Impacts Metabolic Health and Obesity

Depression, Olfaction, and Quality of Life: A Mutual Relationship

A Pathophysiological Model for COVID-19: Critical Importance of Transepithelial Sodium Transport upon Airway Infection. Function (Oxf) 1, zqaa024

Single-cell meta-analysis of SARS-CoV-2 entry genes across tissues and demographics

Long-term persistence of olfactory and gustatory disorders in COVID-19 patients

Age-Dependent Sensory Impairment in COVID-19 Infection and its Correlation with ACE2 Expression

The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice

A mouse-adapted model of SARS-CoV-2 to test COVID-19 countermeasures

Animal models for COVID-19

A Mouse Model of SARS-CoV-2 Infection and Pathogenesis

ACE2 localizes to the respiratory cilia and is not increased by ACE inhibitors or ARBs

Why is COVID-19 less severe in children? A review of the proposed mechanisms underlying the age-related difference in severity of SARS-CoV-2 infections

Intranasal fusion inhibitory lipopeptide prevents direct-contact SARS-CoV-2 transmission in ferrets

A Single-Dose Intranasal ChAd Vaccine Protects Upper and Lower Respiratory Tracts against SARS-CoV-2

Neuropilin-1 is a host factor for SARS-CoV-2 infection

Neuropilin-1 facilitates SARS-CoV-2 cell entry and infectivity

The wiring of Grueneberg ganglion axons is dependent on neuropilin 1

Figure 2. Expression profiles of the ACE2 and TMPRSS2 proteins in the MOED

Immunohistochemical investigations on the MOED for SARS-CoV-2 entry sites

Chemosensory cells are in direct contact with the nasal cavity. b Enlarged view of (a, dashed rectangle), showing expression of ACE2 in the apical surface, Bowman's gland and basal layer. A zoom in view of the apical surface is shown. c Co-expression profiles of TMPRSS2 and ACE2 in OMP-basal cells (white arrowhead, highlighted with zoom in view). d Control experiment (Ctrl neg) illustrating the absence of fluorescent background expression. (a-d) Nuclei are counterstained with Dapi (DAPI, in blue)

OMP+ cellular profile of ACE2 expression in the GG by immunohistochemistry. a Co-expression of ACE2 (in pink) with CK18+ GG neurons (CK18, in red). b ACE2 is expressed together with pGCG+ GG primary cilia (pGCG, in red). c Co-expression of ACE2 with gTUB+ punctiform staining in GG neurons (gTUB, in red) indicating a localization in the basal body of GG primary cilia

Representative protein expression profile obtained from heterozygote OMP-GFP mice of 11 (a) and 5 months old (b and c)

White arrow head highlights a precise wrapping glial cell where a clear absence of ACE2 staining is observed. b Co-expression of ACE2 with CK5+ punctiform staining in GG neurons (CK5, in red). Absence of co-expression in the apical keratinocyte cell layer (zoom in view, white dashed circle). c Co-expression of ACE2 and SOX2 (SOX2, in red) in apical keratinocyte cells (white arrowhead). Absence of SOX2 expression in OMP+ GG neurons

Representative protein expression profile obtained from heterozygote OMP-GFP mice of 13 (a), 6 (b) and 7 months old (c)