bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 A zebrafish model for COVID-19 recapitulates olfactory and cardiovascular 2 pathophysiologies caused by SARS-CoV-2 3 4 Aurora Kraus1, Elisa Casadei1, Mar Huertas2, Chunyan Ye3, Steven Bradfute3, Pierre 5 Boudinot4, Jean-Pierre Levraud5, Irene Salinas1,6 6 7 1 Center of Evolutionary and Theoretical Immunology, Biology Department, University of New 8 Mexico, New Mexico, USA 9 2 Department of Biology, Texas State University, San Marcos, Texas, USA 10 3 Center for Global Health, Department of Internal Medicine, University of New Mexico Halth 11 Sciences Center, New Mexico, USA 12 4 Université Paris-Saclay, INRAE, UVSQ, VIM, 78350 Jouy-en-Josas, France 13 5 Unité Macrophages et Développement de l'Immunité, Institut Pasteur, CNRS UMR3637, Paris, 14 France 15 6 Lead Contact 16 Correspondence: [email protected] 17

18 Summary 19 20 The COVID-19 pandemic has prompted the search for animal models that recapitulate the 21 pathophysiology observed in humans infected with SARS-CoV-2 and allow rapid and high 22 throughput testing of drugs and vaccines. Exposure of larvae to SARS-CoV-2 Spike (S) receptor 23 binding domain (RBD) recombinant protein was sufficient to elevate larval heart rate and 24 treatment with captopril, an ACE inhibitor, reverted this effect. Intranasal administration of 25 SARS-CoV-2 S RBD in adult zebrafish recombinant protein caused severe olfactory and mild 26 renal histopathology. Zebrafish intranasally treated with SARS-CoV-2 S RBD became hyposmic 27 within minutes and completely anosmic by 1 day to a broad-spectrum of odorants including bile 28 acids and food. Single cell RNA-Seq of the adult zebrafish olfactory organ indicated widespread 29 loss of expression of olfactory receptors as well as inflammatory responses in sustentacular, 30 endothelial, and myeloid cell clusters. Exposure of wildtype zebrafish larvae to SARS-CoV-2 in 31 water did not support active viral replication but caused a sustained inhibition of ace2 32 expression, triggered type 1 cytokine responses and inhibited type 2 cytokine responses. 33 Combined, our results establish adult and larval zebrafish as useful models to investigate 34 pathophysiological effects of SARS-CoV-2 and perform pre-clinical drug testing and validation 35 in an inexpensive, high throughput vertebrate model. 36 37 Keywords: SARS-CoV-2, zebrafish (Danio rerio), animal models, olfactory damage, anosmia, 38 olfactory receptor, renal pathology, cardiovascular dysfunction, captopril, inflammation, 39 cytokines, renin-angiotensin pathway 40 41 Introduction 42 43 The current COVID-19 pandemic has infected 44.8 million people worldwide causing over 1 44 million global deaths as of October 21st 2020. Consequently, development of laboratory animal 45 models that recapitulate the pathophysiology of SARS-CoV-2 infection in humans is of utmost 46 importance to accelerate drug and vaccine testing. Thus far, the few lab compatible animal

1 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

47 species that have been reported as naturally susceptible to SARS-CoV-2 include rhesus and 48 cynomolgus macaques (Munster et al., 2020; Rockx et al., 2020), ferret (Kim et al., 2020), cat 49 (Shi et al., 2020), and Syrian hamster (Chan et al., 2020). Mice, by contrast, are not 50 spontaneously permissive to the virus, but mice expressing the human ACE2 receptor provide a 51 useful animal model (Bao et al., 2020; Jia et al., 2020; Lutz et al., 2020). All these mammalian 52 models have unique advantages and disadvantages for the study of immune responses to SARS- 53 CoV-2 and other host-pathogen interactions but do not allow rapid, whole organismal, high 54 throughput, and low-cost preclinical testing of drugs and immunotherapies. 55 56 As model vertebrates, zebrafish are permissive to human viral pathogens including Influenza A 57 (Gabor et al., 2014), Herpes Simplex virus type 1 (Burgos et al., 2008), Chikungunya virus 58 (Palha et al., 2013) and human noroviruses GI and GII (Van Dycke et al., 2019). Zebrafish offer 59 many advantages over other animal models due to their high reproductive ability, rapid 60 development, low maintenance costs, and small transparent bodies. Importantly, zebrafish 61 olfactory, immune, and cardiovascular physiology share a significant degree of conservation 62 with humans (Postlethwait et al., 2020; Saraiva et al.). Genetically, more than 80% of disease 63 related genes have a zebrafish orthologue (Howe et al., 2013). The zebrafish innate immune 64 system is already developed in the transparent larval stages and members of all major groups of 65 mammalian cytokines have been identified in the zebrafish genome (Gomes and Mostowy, 2020; 66 Zou and Secombes, 2016). Academic laboratories and the pharmaceutical industry use zebrafish 67 larvae in preclinical studies for assessing efficacy and toxicity of candidate drugs for several 68 diseases (Taylor et al., 2010). Zebrafish is a proposed model for COVID-19 and has recently 69 been used in one vaccination study (Galindo-Villegas, 2020; Ventura-Fernandes et al., 2020). 70 This study aims to elucidate the physiopathology of wildtype zebrafish in response to SARS- 71 CoV-2. 72 73 SARS-CoV-2 infection causes a wide litany of symptoms, ranging from asymptomatic to mild or 74 severe disease (Menni et al., 2020). Apart from respiratory symptoms, multi-organ pathologies 75 are often reported with heterogeneous symptoms such as olfactory and taste loss, cardiac 76 dysfunction, renal pathologies, neurological damage, muscle and joint pain, gastrointestinal 77 symptoms, clotting disorders and others (Tabata et al., 2020; Wang et al., 2020a; Zhu et al., 78 2020). A longitudinal study associated disease severity and patient outcome with specific 79 patterns of innate immune response misfiring in COVID-19 patients leading to catastrophic 80 organ damage (Lucas et al., 2020). Specifically, severe disease patients display maladaptive 81 cytokine responses characterized by high levels of pro-inflammatory cytokines such as IL6, 82 TNFα and IL1β (Blanco-Melo et al., 2020; Hadjadj et al., 2020; Lucas et al., 2020; Del Valle et 83 al., 2020), suppression of type I IFN responses (Blanco-Melo et al., 2020; Hadjadj et al., 2020; 84 Mudd et al., 2020), and elevated type 2 cytokine levels (Lucas et al., 2020). Importantly, this 85 cytokine pattern is in sharp contrast to that found in patients experiencing mild or moderate 86 symptoms, who are able to control exacerbated type 1 and type 3 cytokine responses (Lucas et 87 al., 2020). 88 89 SARS-CoV-2 enters the human host cells when SARS-CoV-2 Spike (S) protein receptor binding 90 domain (RBD) binds to angiotensin-converting enzyme 2 (ACE2) on a permissive host cell, then 91 a serine protease, such as TMPRSS2, cleaves the spike protein S1/S2 site to facilitate fusion of 92 the virion with the host cell membrane (Hoffmann et al., 2020a, 2020b). ACE2 is expressed in

2 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

93 many different cell types across many organs in the human body including lung, olfactory 94 sustentacular cells, enterocytes, and endothelial cells (Albini et al., 2020; Lamers et al., 2020; 95 Ziegler et al., 2020; Zou et al., 2020), explaining the widespread organ damage observed in 96 patients infected with SARS-CoV-2 (Wadman et al., 2020). Several studies performed 97 comparative analyses of the ACE2 amino acid sequences across vertebrate species and identified 98 a high degree of conservation use used ACE2 homology to predict host susceptibility to SARS- 99 CoV-2 infection (Damas et al., 2020; Wan et al., 2020). Ectotherms such as fish are predicted to 100 have a low propensity for SARS-CoV-2 infection. However, such in silico predictions must be 101 ascertained by performing infection experiments. 102 103 Binding of the spike protein from SARS-CoV, NL63, and SARS-CoV-2 results in rapid and 104 local decreases in ACE2 protein (Garvin et al., 2020; Glowacka et al., 2010; Imai et al., 2005; 105 Kuba et al., 2005). In turn, decreased ACE2 levels trigger negative regulation of renin- 106 angiotensin system (RAS) and bradykinin pathways ultimately responsible for hypotension, 107 vascular permeability, and multiorgan damage (Garvin et al., 2020; Patel et al., 2014). Thus, 108 COVID-19 patients can experience a variety of cardiovascular presentations including 109 myocardial injury, atrial and ventricular arrhythmias and cardiogenic shock (Babapoor- 110 farrokhran et al., 2020; Guzik et al., 2020; Kochi et al., 2020; Kuck, 2020; Libby, 2020). To 111 restore balance to the RAS pathway and ameliorate cardiac symptoms associated with decreased 112 ACE2, the use of ACE inhibitors is being considered as a therapeutic intervention in COVID-19 113 patients (Lopes et al., 2020). Importantly, drugs currently used to treat the COVID-19 can be 114 pro-arrhythmic and therefore there is a need to incorporate cardiovascular damage into the list of 115 targets of therapeutic interventions in COVID-19 and for models that replicate human cardia 116 physiology (Kochi et al., 2020). 117 118 A hallmark of SARS-CoV-2 infection is acute loss of smell (Cooper et al., 2020). Viral-induced 119 anosmia is not unique to SARS-CoV-2 infections since viruses such as rhinoviruses, influenza, 120 parainfluenza and coronaviruses are known to be the main cause of olfactory deficits in humans 121 (Suzuki et al., 2007; Imam et al., 2020). In mice and humans, ace2 expression is detected in 122 sustentacular cells, olfactory stem cells known as horizontal and globose basal cells in the 123 olfactory epithelium, and vascular cells (pericytes) in the olfactory bulb (Brann et al., 2020). 124 Olfactory dysfunction in COVID-19 patients may be caused by several mechanisms such as 125 inflammation instigated by infection of cells in the olfactory epithelium altering olfactory 126 sensory neuron’s (OSN) function and environment, as well as direct viral infection of 127 sustentacular cells, olfactory stem cells, and vascular cells wreaking havoc (Cooper et al., 2020). 128 Of note, sterile induction of type I IFN in the mouse olfactory epithelium leads to widespread 129 loss of olfactory receptor expression and COVID-19 patients have increased pro-inflammatory 130 cytokine levels in their olfactory epithelium (Rodriguez et al., 2020; Torabi et al., 2020). Still, 131 experimental data is needed to elucidate the underlying mechanisms of loss of smell in COVID- 132 19 patients. 133 134 The present study reports for the first time that zebrafish larvae exposed to SARS-CoV-2 appear 135 to mount innate immune responses that resemble cytokine responses of mild COVID-19 patients. 136 Recombinant SARS-CoV-2 S RBD is sufficient to cause olfactory, renal and cardiovascular 137 pathologies in larvae and adult zebrafish. We also identify potential mechanisms of SARS-CoV- 138 2 induced anosmia by scRNA-Seq. Our findings support the use of zebrafish as a novel

3 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

139 vertebrate model to elucidate SARS-CoV-2 pathophysiology and to screen drugs and other 140 therapies targeting COVID-19. 141 142 143 144 Results 145 146 Phylogenetic analyses of ACE2 molecules in vertebrates 147 Comparative analysis of ACE2 molecules in vertebrates indicated that ACE2 molecules are well 148 conserved in vertebrates with a 72%-73% similarity and 57.5%-58% identity between zebrafish 149 ACE2 and human ACE2, respectively (Table S1). Examination of ACE2 amino acid motifs in 150 the region involved in binding SARS-CoV-2 S protein revealed zebrafish ACE2 has 50%/64% 151 sequence similarity with the corresponding human ACE2 region compared to 71%/78% in 152 macaques ACE2 or 57%/71% in ferret ACE2 (Table S2) (Liu et al., 2020). Combined, these data 153 agree with previous reports and reveal that key amino acids involved in the interactions between 154 human ACE2 and SARS-CoV-2 S protein are not conserved in zebrafish (Fu et al., 2020; 155 Postlethwait et al., 2020). 156 157 SARS-CoV-2 Spike (S) Receptor Binding Domain (SARS-CoV-2 S RBD) elevates heart rate 158 in zebrafish larvae 159 Systemic injection of recombinant SARS-CoV-2 protein into adult zebrafish has been shown to 160 induce some toxicity (Ventura-Fernandes et al., 2020). In order to determine whether 161 recombinant SARS-CoV-2 S RBD protein causes inflammatory responses in zebrafish larvae, we 162 exposed 5 dpf larvae to SARS-CoV-2 S RBD recombinant protein for 3 hours (h) and measured 163 cytokine responses by qPCR. As shown in Figure 1A, 3 h immersion with SARS-CoV-2 S RBD 164 protein induced a significant downregulation in ifnphi1 expression and significant increase in 165 expression of ccl20a.3, a pro-inflammatory chemokine. No changes in ace2, tnfα, il1b and 166 il17a/f3 expression were observed (Figure 1A). These results indicate that rapid immune 167 responses occur in zebrafish larvae exposed to SARS-CoV-2 S RBD. We next evaluated the 168 effects of SARS-CoV-2 RBD S on zebrafish larva heart function to validate zebrafish larvae as a 169 model for COVID-19 cardiac manifestations. We immersed 7- and 5-days post fertilization (dpf) 170 zebrafish larvae with SARS-CoV-2 S RBD, or with vehicle, and measured heart rate after 3 h. 171 As shown in Figure 1B, 7 dpf and 5 dpf zebrafish treated with SARS-CoV-2 S RBD had 172 significantly higher heart rates compared to vehicle treated controls. As a positive control for the 173 recombinant protein, we used animals treated with the same dose of recombinant infectious 174 hematopoietic necrosis virus (IHNV) glycoprotein (r-IHNVg), a rhabdovirus known to cause 175 severe endothelial damage in zebrafish (Ludwig et al., 2011). r-IHNVg caused a severe decrease 176 in larval zebrafish heart rate compared to control treated animals (Figure 1B-C). To determine if 177 increased heart rate induced by SARS-CoV-2 S RBD was dependent on Ace2 binding, we co- 178 incubated 5 dpf larvae with captopril and reverted SARS-CoV-2 S RBD induced heart 179 dysfunction. Captopril had no effect on r-IHNVg induced bradycardia (Figure 1B). Ventricular 180 trace analyses showed marked differences in rhythm patterns in each treatment group (Figure 181 1D). Importantly, the captopril and SARS-CoV-2 S RBD treated animals, despite having similar 182 heart rates to those of the vehicle treated controls, displayed a unique ventricular trace pattern, 183 warranting future studies regarding the potential cardioprotective role of captopril in COVID-19. 184 Combined, these results indicate that zebrafish exposed to SARS-CoV-2 S RBD protein

4 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

185 experience tachycardia and suggest that zebrafish larvae constitute a valuable pre-clinical model 186 to test the effects of drugs for COVID-19 on cardiac activity in vivo. 187

188 Intranasal delivery of SARS-CoV-2 S RBD causes olfactory and renal damage in adult 189 zebrafish 190 Anosmia is one of the earliest manifestations of SARS-CoV-2 infection in humans (Cooper et 191 al., 2020). We have previously shown that IHNV glycoprotein protein is sufficient to induce 192 rapid nasal immune responses as well as neuronal activation in teleost fish (Sepahi et al., 2019). 193 Thus, we hypothesized that recombinant SARS-CoV-2 S RBD protein may be sufficient to 194 damage the olfactory epithelium of zebrafish. In order to test this hypothesis, we delivered 195 recombinant SARS-CoV-2 S RDB to the nasal cavity of adult AB zebrafish and sampled the 196 olfactory organ 3 h, 1 day (d), 3 d and 5 d later. Negative controls consisted of animals that 197 received vehicle only (PBS) into the nasal cavity. As shown in Figure 2, SARS-CoV-2 S RBD 198 was sufficient to cause severe damage to the zebrafish olfactory organ at all time points 199 examined. Tissue damage was characterized by presence of edema and endothelial inflammation 200 in the lamina propria as early as 3 h post-treatment (Figure 2A-C and G). Presence of 201 hemorrhages at the tip and base of the olfactory lamellae was observed 1 d post-treatment 202 (Figure 2D). Loss of the epithelial mosaic structure characteristic of the teleost olfactory 203 epithelium was observed on days 1, 3 and 5 post-treatment (Figure 2E). By day 5, loss of entire 204 apical lamellar areas due to severe necrosis was observed in the olfactory lamellae of all treated 205 animals compared to controls (Figure 2F). Significant loss of olfactory cilia was recorded in all 206 animals treated with SARS-CoV-2 S RBD at all time points (Figure 2H). These results indicate 207 the SARS-CoV-2 S RBD is sufficient to cause inflammation, edema, hemorrhages, ciliary loss, 208 and necrosis in the olfactory organ of zebrafish. Hence, olfactory damage can be caused by 209 indirect mechanisms and in the absence of active SARS-CoV-2 replication in this tissue. 210 211 Toxicity effects of intranasal SARS-CoV-2 S RBD delivery were also evaluated in distant tissues 212 such as the kidney, a target organ of SARS-CoV-2. Acute kidney injury (AKI) incidence varies 213 from 0.9% to 29% in COVID-19 patients (Su et al., 2020). Renal damage, especially AKI, is also 214 common in patients with RAS dysfunction such as diabetic patients who suffer from 215 hypertension (Ribeiro-Oliveira et al., 2008; Advani, 2020). A recent study in zebrafish injected 216 with the N-terminal part of SARS-CoV-2 S protein reported inflammation and damage in several 217 tissues of adult zebrafish including kidney 7 and 14 d post-injection (Ventura-Fernandes et al., 218 2020). In the present study, histological examination of the head-kidney of zebrafish who 219 received SARS-CoV-2 S RBD intranasally revealed renal tubule pathology characteristic of AKI 220 3 h post-treatment (Figure S1). Pathology was not as severe at later time points, but vacuolation 221 of the renal tubule epithelium was still visible 5 days post-treatment (Figure S1). We did not 222 observe signs of glomerulopathology in treated animals compared to controls. Together, these 223 results indicate that intranasal delivery of SARS-CoV-2 S RBD is sufficient to cause 224 nephropathy in adult zebrafish but that pathology is not as severe as when the protein is delivered 225 by injection. 226 227 Intranasal delivery of SARS-CoV-2 S RBD causes anosmia in adult zebrafish 228 229 Adult zebrafish exposed to SARS-CoV-2 S RBD had a significant reduction of olfactory 230 responses to food extracts of ~50% of preexposure olfaction within minutes as measured by

5 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

231 electro-olfactogram (EOG), indicating an instant effect of the protein on olfactory function (Fig. 232 3A). Reduction of olfaction was sustained for least one hour of recording, but the olfactory organ 233 remained still semi functional. Zebrafish treated with PBS never lost olfaction at any time point. 234 To further quantify the degree of olfactory reduction due to SARS-CoV-2 S RBD, we took 235 advantage of the two easily accessible and isolated olfactory chambers present in zebrafish. We 236 exposed one naris to the SARS-CoV-2 S RBD protein and the other naris, from the same animal, 237 to PBS and waited 3 h or 1d before measuring olfaction by EOG. At 3 h we observed a 37-70% 238 reduction in food and bile olfactory responses between the treated and untreated naris and a 239 complete loss of olfactory function to both odorants 1 d post-treatment (Fig. 3B). The reduction 240 of olfactory sensitivity for food extract was smaller than that found for bile, probably due to the 241 lower number of OSNs involved in bile acid detection compared to amino acids found in food 242 (Hansen et al., 2003). Our results indicate that SARS-CoV-2 S RBD-induced-anosmia is not 243 specific for a subset of OSNs, since both food extracts and bile olfactory signals were suppressed 244 in SARS-CoV-2 S RBD treated zebrafish. This fact, together with the 1d time to develop 245 complete anosmia and disrupt olfactory epithelial structure, support the hypothesis that SARS- 246 CoV-2 S RBD damage may occur first on sustentacular cells, with subsequent impacts on OSN 247 viability and function.

248 249 Single-cell analysis of the zebrafish olfactory organ 250 251 To understand the impact of SARS-CoV-2 S RBD on zebrafish OO, we performed single cell 252 RNA-Seq (scRNA-Seq) of adult zebrafish olfactory organ from control or SARS-CoV-2 S RBD 253 intranasally-treated fish. scRNA-Seq was performed after 3 h or 3 d of intranasal administration 254 with SARS-CoV-2 S RBD. These timepoints were selected based on the histological findings of 255 Figure 2. Data from the treated time point was integrated with that from vehicle treated controls 256 in Seurat and then clustered using 30 significant principal components (Figure 4A-B)(Butler et 257 al., 2018). We identified a total of 8 neuronal cell clusters (NC), 5 sustentacular cell (SC) 258 clusters, 3 endothelial cell (EC) clusters and 7 leucocyte (lymphoid and myeloid) clusters (Figure 259 4A-B). 260 Of the 8 NCs, Neuron1 and Neuron2 corresponded to mature OSNs. Neuron1 expressed markers 261 of ciliated OSNs (ompa and ompb) in addition to several olfactory receptor (OR) genes 262 (Buiakova et al., 1996). Neuron2, on the other hand, expressed markers of microvillus OSNs 263 (trpc2b, s100z, and gnao) and many vomeronasal receptors (VR) such as v2rh32 as well as OR 264 gene (Figure S2) (Kraemer et al., 2008). Neuronal clusters 3-8 included OSNs in different 265 differentiation stages, from progenitor-like neurons to differentiating intermediate neuronal 266 precursors (Yu and Wu, 2017). For instance, expression of multiple high mobility group 267 transcripts (hmga1a, and hmgb2b2) typically found in neuronal precursor cells is inverse to 268 expression of the transcription factor family of sox genes (sox4 and sox11) associated with 269 committing to differentiation to mature neurons across the neuronal subsets (Bergsland et al., 270 2006; Giancotti et al., 2018). The Neuron3 cluster expressed both ompa and OR genes as well as 271 markers of neuronal differentiation and plasticity such as neurod1, neurod4, gap43, sox11, and 272 sox4 suggesting commitment to becoming mature ciliated OSNs. The cluster Neuron4, only 273 found at 3 h, was transcriptionally similar to the Neuron3 cluster expressing hmga1a, and 274 hmgb2b2, but lacked markers of differentiation (gap43 and sox11) indicating a closer neuronal 275 progenitor state. The clusters Neuron5 and Neuron6 were the closest to the progenitor state with 276 expression of basic helix-loop-helix transcription factors (i.e. bhlhe23), associated with exit from

6 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

277 the cell cycle and progenitor state, as well as cell cycle genes (Borders et al., 2007). Clusters 278 Neuron7 and Neuron8 expressed cell cycle and early neuronal progenitor markers as well as 279 tmprss13b and tmprss4a, however the majority of cluster identifying genes in these two clusters 280 are undescribed (Figure S2). 281 SCs are supporting cells that exist in the neuroepithelium around OSNs and in humans, they 282 express ACE2 (Bryche et al., 2020). SCs in the olfactory epithelium can directly arise from 283 horizontal basal cells (HBCs) (Yu and Wu, 2017). We found 5 clusters of SCs in our datasets. 284 Sustentacular clusters 1 and 2 had similar expressing markers to those of HBCs (krt5) and other 285 basal cells (lamc2 and col1a1b) and therefore may represent olfactory basal cells (Chen et al., 286 2017; Peterson et al., 2019). Further, these clusters expressed immune genes such as ptgs and 287 ccl25, and HBCs are known to express immune factors in models of olfactory epithelium 288 damage (Chen et al., 2019). Interestingly, in addition to sox2 expression, Sustentacular3 cluster 289 expressed atoh1, a basic helix loop helix gene important in hair cell regeneration (Bermingham- 290 McDonogh and Reh, 2011). Cluster Sustentacular4 expressed cytokines il15 and il17a/f3 as well 291 as ace. The last SC cluster was only found in the 3 d time point and markers suggest it is a 292 subpopulation closely related to the Sustentacular4 cluster (Figure S2). 293 We identified three clusters of ECs that all express tmprss13 and tmprss4. While we did not 294 detect ace2 expression in any cell clusters, we detected ace2 mRNA in adult zebrafish olfactory 295 organ, and at low levels in the olfactory bulb (Figure S3). ace2 expression levels have previously 296 shown to be low in neuronal tissues and therefore may be hard to detect by scRNA-Seq (Song et 297 al., 2020). Endothelial1 and 2 clusters expressed the endothelial markers sox7 and tmp4a (Yao et 298 al., 2019). All three clusters broadly expressed genes associated with tight junctions (tjp3, jupa, 299 ppl, cldne, and cgnl1) as well as many keratin genes (Figure S2). Interestingly, Endothelial3 300 cluster also expressed the calcium channel trpv6 and a slew of non-annotated genes. 301 There are copious amounts of immune cells in the teleost olfactory organ (Mellert et al., 1992; 302 Sepahi et al., 2016), and although our method of generating single cell suspensions for scRNA- 303 Seq was optimized for neuronal cell viability, we captured small groups of both myeloid and 304 lymphoid cells in our samples. Lymphocyte1 cells expressed classical T cell markers (cd8, cd4, 305 and zap70) as well as ccr9, a mucosal homing chemokine receptor, and novel immune type 306 receptor genes (nitrs) thought to be natural killer cell receptors (Papadakis et al., 2003; Wei et 307 al., 2007). Therefore, we propose that lymphocyte1 are a type of innate-like NKT cells. 308 Lymphocyte2 cluster cells expressed the transcription factor foxp3b found in regulatory T cells 309 (Tregs). The 3 d time point revealed a third lymphocyte cluster with classical T cell identity. The 310 myeloid cell clusters 1-3 expressed MHC II complex genes as well as mpeg1.1, a marker of 311 zebrafish macrophages (Ellett et al., 2011). Myeloid4 cluster expressed mpx, a marker for 312 neutrophils, and no antigen presentation genes (Yoo and Huttenlocher, 2011). Each macrophage- 313 like myeloid cluster expressed unique cytokines contributing to their separate clustering -- 314 Myeloid1 expressed il12 and il22, Myeloid2 expressed il1fma and il10ra, and Myeloid3 315 expressed il6r and il1b. Together, these data indicate that the zebrafish olfactory organ is a 316 heterogeneous neuroepithelium with diverse immune cell subsets. 317 318 Intranasal delivery of SARS-CoV-2 S RBD induces inflammatory responses and 319 widespread loss of olfactory receptor expression in adult zebrafish olfactory organ 320 321 The cellular landscape of the zebrafish olfactory epithelium was affected by SARS-CoV-2 S 322 RBD treatment and time (Figure 4A-D). This was especially evident in the proportions of

7 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

323 neuronal cell types 3 d post-treatment when the proportion of mature, omp+ ciliated OSN was 324 much lower compared to controls and the 3 h treated group. In contrast, neuronal progenitors 325 expressing cell cycle markers (aubk, ecrg4, and mki67) and neuronal differentiation and 326 plasticity markers (neurod1, neurod4, gap43, sox11, and sox4) were expanded 3 d post-treatment 327 (Figure 4B-C). Further, we detected a noticeable decrease in the proportion of cells belonging to 328 the Lymphocyte2 cluster, a cluster that expressed markers of Treg cells (foxp3b) 3 h post 329 intranasal delivery of recombinant SARS-CoV-2 S RBD but this change was not noticeable at 330 day 3 (Figure 4C). At 3 d, we observed a third lymphocyte cluster, not detected at 3 h, highly 331 expressing the TCR subunit zap70 as well as plac8 onzin related protein (ponzr1), a molecule 332 that has immunoregulatory roles during Th1 type immune responses in mammals (Figure S2) 333 (Slade et al., 2020). These results indicate important cellular responses in neuronal and immune 334 cell subsets of the zebrafish olfactory organ in response to SARS-CoV-2 S RBD. 335 336 The transcriptional landscape of the zebrafish olfactory epithelium was affected by SARS-CoV-2 337 S RBD treatment (Figure 4E-F). By 3 h post intranasal delivery of recombinant SARS-CoV-2 S 338 RBD, SCs and ECs showed significant increased expression of cytokine genes (il15, il4r, and 339 ccl19b) and prostaglandin I2 synthase (ptgis), while myeloid clusters upregulated expression of 340 the interferon gene irf1b (Figure 4E). Strikingly, NF-kB inhibitor alpha (nfkbiaa) was 341 differentially expressed across almost all cell types including progenitor-like neuronal clusters at 342 3 h relative to vehicle control (Figure 4E). By 3 d post-treatment with recombinant SARS-CoV-2 343 S RBD, myeloid clusters expressed inflammatory markers (il1b and tnfa) and showed a 344 significant differential upregulation of the chemokine receptor cxcr4b suggesting perhaps 345 trafficking into the olfactory organ from other body sites (Figure 4F). At 3 h post intranasal 346 delivery of SARS-CoV-2 S RBD, differential expression analysis showed changes in genes 347 related to endothelial integrity including angiopoietin-like 4 (angptl4), angiomotin (amotl2b), 348 and Wiskott-Aldrich syndrome protein (wasb) as well as other genes associated with 349 vasoconstriction (dusp5 and vaspa) (Figure 4E) (Li et al., 2015; Wagener et al., 2016; Zheng et 350 al., 2009). By 3 d after intranasal SARS-CoV-2 S RBD, endothelial clusters showed significantly 351 increased expression of clotting factors (f3b and f3a) relative to vehicle treated control and 352 continued upregulation of angptl4 expression was detected in several EC and SC clusters (Figure 353 4F). Thus, this data suggested that inflammatory responses and endothelial disruption are both 354 hallmarks of SARS-CoV-2 induced damage in the olfactory epithelium. 355 Differential expression analysis revealed that while some changes in ORs expression occurred 3 356 h post-treatment in neuronal progenitor clusters, many ORs and TAARS genes were significantly 357 differentially downregulated in mature neuronal clusters 1 and 2 compared to vehicle treated 358 control (Figure 5A-B). In accord, markers of mature ciliated and microvillus OSNs (ompb and 359 v2rh32, respectively) were significantly downregulated at both 3 h and 3 days after intranasal 360 SARS-CoV-2 S RBD delivery (Figure 4A-D). Gene ontology (GO) analyses of significantly 361 down or up regulated genes from all neuronal clusters at 3 h or 3 d after intranasal SARS-CoV-2 362 S RBD compared to vehicle treated controls was done using ShinyGO v0.61 and Metascape to 363 gather an image of how OSNs as a collective are responding to the viral moiety (Figure 5E-H) 364 (Ge et al., 2020; Zhou et al., 2019). At 3 h after intranasal SARS-CoV-2 S RBD delivery, 365 upregulated genes in the olfactory neuronal clusters were enriched in regulation of cell 366 differentiation and sensory perception of smell (Figure 5E). By 3 d, upregulated genes identified 367 in olfactory neuronal clusters were enriched in processes such as neuron differentiation, sensory 368 system development, and sensory organ morphogenesis, whereas downregulated genes belonged

8 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

369 to sensory perception of smell, detection of chemical stimulus and GPCR signaling pathway 370 (Figure 5F). Functional enrichment analyses in Metascape showed that the top non-redundant 371 enriched clusters in both upregulated and downregulated genes in zebrafish OSNs 3 h post- 372 treatment was sensory perception of smell (Figure 5G). The same was true 3 days post-treatment, 373 but processes such as regeneration, neuron development, neuron fate commitment and the p53 374 signaling pathway were also enriched within the upregulated genes (Figure 5H). Combined, 375 these results suggest that presence of SARS-CoV-2 S RBD in the olfactory organ instigates 376 harmful effects on OSNs within hours and that the magnitude of the OSN damage increases by 3 377 days post-treatment. Further, these analyses indicate that neuronal regeneration and 378 differentiation processes were initiated by day 3 in order to begin repair of olfactory damage. 379 380 Intranasal delivery of SARS-CoV-2 S RBD induces responses in sustentacular cells and 381 endothelial cell clusters 382 383 Co-expression of ACE2 and TMPRSS2 by olfactory SCs and basal cells in the human olfactory 384 epithelium and subsequent onset of damage and inflammation of these cell subsets may explain 385 olfactory loss in COVID-19 patients (Brann et al., 2020; Cooper et al., 2020). Additionally, 386 ACE2 is expressed in ECs and these cells are infected by SARS-CoV-2 (Bryche et al., 2020). 387 Our study allowed us to dissect how each cell type in the zebrafish olfactory organ responds to 388 SARS-CoV-2 S RBD. Our results indicated unique responses by SC clusters and EC clusters to 389 treatment (Figures 4 and 6). At 3 h, we detected increased expression of apoeb, of transcription 390 factors foxq1a and id2b, the transcriptional regulator nfil3-5, two tumor necrosis factor receptor 391 superfamily members (tnfrsf11b and tnfrsf9a) as well as tcima (transcriptional and immune 392 response regulator), whose mammalian ortholog pcim regulates immune responses as well as 393 endothelial cell activation and expression of inflammatory genes (Figure 6A) (Kim et al., 2009). 394 Further, at 3 h we observed downregulation of the pro-inflammatory cytokine il17af/3 as well as 395 glutathione peroxidase gpx1b, the transcription factor notch1b, basal cell adhesion molecule 396 bcam, guanine nucleotide-binding protein subunit gamma gng8, and the calcium binding s100z 397 in SC and EC from SARS-CoV-2 S RBD treated olfactory organs relative to vehicle treated. At 3 398 d post-treatment, we observed significant increased expression of the gene that encodes brain 399 natriuretic peptide (nppc), a vasodilating hormone, the pro-inflammatory chemokine ccl19a.2, 400 the M2 macrophage marker arg2, the transcription factors foxq1a and sox11a, tubulin beta 5 401 tubb5, and the epithelial mitogen epgn, among others (Figure 6B). Downregulated genes at 3 d 402 post-treatment included hsp70.3, apoeb, the osteoblast specific factor b postb, the desmosomal 403 component periplakin (ppl), the vasoconstricting endothelin 2 (edn2), the heparin binding 404 molecule latexin (ltx) involved in pain and inflammation, and cd74b, a part of the MHC-II 405 complex (Figure 6B). Combined, these data indicated immune regulatory responses in SC and 406 EC clusters early after SARS-CoV-2 S RBD treatment, followed by transcriptional changes with 407 potential vasoactive effects by day 3. 408 409 GO and enrichment set analyses indicated that SC and EC clusters initially undergo 410 transcriptional changes enriched in metabolic responses, response to stress, and cell 411 differentiation (Figure 6C). Later on, at day 3, SC and EC responses were enriched in genes 412 involved not only in the stress response but also in immune responses and responses to wounding 413 (Figure 6D). Similar results were identified using Metascape, which showed that the 414 inflammatory response to wounding was moderately enriched in the downregulated genes at 3 h,

9 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

415 whereas by day 3, response to wounding became the top enriched set among the upregulated 416 genes (Figure 6E-F). 417 418 Exposure of wildtype zebrafish larvae with SARS-CoV-2 does not support viral replication 419 420 Zebrafish larvae have been used as models to investigate several human viruses. Infecting 421 zebrafish larvae in a BSL-3 laboratory by immersion in contaminated water is comparable to 422 infecting a cell line. We first checked the stability of SARS-CoV-2 in zebrafish water overtime, 423 in the absence of any animals. We found that SARS-CoV-2 viral loads in the water remained 424 stable throughout the experiment (Figure 7A-B). We exposed wildtype AB zebrafish larvae to 425 live SARS-CoV2 and examined viral mRNA abundance over time to determine if zebrafish 426 larvae can support viral replication. We detected no increases in the viral N copy numbers over 427 time and a steady decline in E gene copy numbers in both water from wells containing larvae and 428 virus as well as in the larval tissue (Figure 7C-F). These results indicate that wild-type zebrafish 429 larvae cannot support efficient SARS-CoV-2 replication as suggested by the in silico 430 comparative sequence analyses of the zebrafish ace2 molecule. 431 432 Exposure of zebrafish larvae to SARS-CoV-2 decreases ace2 expression and triggers pro- 433 inflammatory cytokine responses 434 435 In order to determine whether exposure of zebrafish larvae with live SARS-CoV-2 causes 436 changes in Ai, we measured ace2 mRNA levels in control and SARS-CoV-2 exposed larvae over 437 time. ace2 expression was significantly downregulated as early as 6 h post-infection. ace2 438 expression inhibition was sustained over the time course of the experiment with the greatest 439 decrease occurring 2 days post-infection (dpi) (Figure 8A). We next evaluated changes in 440 expression of cytokine and chemokine genes to establish whether zebrafish mount inflammatory 441 responses that resemble the patterns of mild or severe SARS-CoV-2 infection. Il1β expression 442 was significantly upregulated at 6h, 1 dpi (2-3 fold) and 2 dpi (10 fold) and significantly 443 downregulated at 4 dpi (Figure 8B). We detected a significant increase in tnfa expression in 444 SARS-CoV-2 exposed larvae 2 dpi (Figure 8C). ifnphi1 and ifnphi3 are the two main type I IFN 445 genes involved in larval zebrafish antiviral responses (Levraud et al., 2019). We detected a 446 significant up-regulation of ifnphi1 at 1 and 2 dpi, whereas expression was inhibited at 4 dpi. 447 Interestingly, ifnphi3 expression followed a very different pattern compared to ifnphi1, which 448 was significantly downregulated 2 dpi but significantly upregulated at 4 dpi (Figure 8D-E). Mxa 449 expression was significantly downregulated at all time points (Figure 8F). il17af/3 expression 450 was significantly elevated 1, 2 and 4 dpi (Figure 8G). Expression levels of il22, a member of the 451 IL10 family, were downregulated 6 hpi and 1 dpi (Figure 8H) whereas the type II cytokine 452 il4/il13b was downregulated at 6 hpi, 1 dpi and 2 dpi (Figure 8I). Further, a significant increase 453 in the expression of the chemokine ccl20a.3 was detected in infected larvae at 1 and 2 dpi 454 compared to controls (Figure 8J). A moderate increase in ccl19a.1 expression was observed at 2 455 dpi followed by a strong down-regulation (15 fold) at 4 dpi (Figure 8K). Taken together, these 456 data indicate that exposure to SARS-CoV-2 induces a significant antiviral and pro-inflammatory 457 immune response in wildtype zebrafish larvae. This response involved type I IFN, tnfa, il1b, il17 458 and ccl20, reminiscent of COVID-19 patients with mild disease. 459

460 Discussion

10 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

461 The current COVID-19 pandemic has propelled the investigation of SARS-CoV-2 and the 462 development of animal models that help identify therapeutic interventions and vaccines for 463 COVID-19. Thus far, all animal models reported are mammals, and therefore breeding, genetic 464 manipulation, and animal housing in BSL-3 laboratories make these models costly and not 465 readily available in large numbers. Zebrafish can overcome many of the limitations of 466 mammalian models thanks to their transparent bodies, short life-span, low maintenance costs and 467 production of large numbers of embryos. We therefore performed the simplest infection 468 procedure, where SARS-CoV-2 was added to the water of zebrafish larvae. In this manner, BSL- 469 3 trained personnel with no experience in zebrafish microinjection can readily expose larvae to 470 SARS-CoV-2 without the need of animal protocols in a similar fashion to in vitro cell culture 471 infections. Exposure of wildtype zebrafish larvae to SARS-CoV-2 in the water did not however 472 result in any detectable viral replication. This finding supports that, despite the general 473 conservation of ACE2 molecules in vertebrates, the amino acid differences identified in 474 zebrafish Ace2, are sufficient to prevent infectivity. Whether infection via microinjection as it 475 was previously done in Influenza A infection models (Gabor et al., 2014) would result in SARS- 476 CoV-2 replication remains to be elucidated. Further, our findings support that, similar to mice, 477 the generation of transgenic larvae expressing human ACE2 may be the best approach in order to 478 support active replication. 479 480 Despite the absence of active viral replication, our model provided important insights into the 481 transcriptional innate immune responses to SARS-CoV-2. Pathological cytokine responses are 482 detected in severe disease COVID-19 patients highlighting the need to identify interventions that 483 control such maladaptive inflammatory responses (Blanco-Melo et al., 2020; Hadjadj et al., 484 2020; Lucas et al., 2020; Del Valle et al., 2020). Mild disease, on the other hand, is characterized 485 by controlled inflammatory and antiviral responses. Zebrafish larvae are known to mount innate 486 immune responses that are largely homologous to those of humans (Zou and Secombes, 2016). 487 In particular, zebrafish possess type I IFNs with similar function and structure to those of 488 mammals and a core ancestral repertoire of IFN-stimulated genes (ISG) has been identified in 489 zebrafish and human (Hamming et al., 2011; Levraud et al., 2019). Zebrafish also possess 490 orthologues to the main inflammatory cytokines (IL1β, TNFα, IL-6, IL-17) and chemokines 491 identified in the response to SARS-CoV-2 in humans (Zou and Secombes, 2016). The present 492 study identified pro-inflammatory cytokine and chemokine signatures in zebrafish larvae 493 exposed to SARS-CoV-2 that largely resemble those described in humans with mild disease 494 symptoms. Interestingly, we detected onset of type 1 pro-inflammatory responses with elevated 495 expression of the cytokines il1b and tnfa, sustained inhibition of type 2 responses (il4/il13b), 496 which would further favor a pro-inflammatory state, and mixed type 3 responses with up- 497 regulation of il17a/f but downregulation of il22. 498 499 Coronaviruses deploy various mechanisms to evade type I IFN responses within infected cells 500 (Park and Iwasaki, 2020; Sa Ribero et al., 2020; De Wit et al., 2016). Interrogation of the main 501 type I IFN genes in zebrafish larvae after co-incubation with live SARS-CoV-2 indicated 502 complex type I IFN responses; significant up-regulation of ifnphi1 in response to SARS-CoV-2 503 exposure during the first two days, while ifnphi3 was downregulated. In contrast, at 4dpi, ifnphi1 504 was down-regulated but ifnphi3 expression had increased. Interestingly, Mxa, an ISG, was 505 downregulated in response to SARS-CoV-2 exposure. Combined, these data suggest the SARS- 506 CoV-2 induces some type I IFN responses in zebrafish larvae while inhibits others. Future

11 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

507 studies are clearly needed to ascertain the role of teleost type I IFN in the anti SARS-CoV-2 508 immune response. 509 510 S protein is a structural protein of SARS-CoV-2 and therefore the target of several vaccine trials. 511 Therefore, we exposed zebrafish larvae to SARS-CoV-2 S RBD protein and investigated 512 transcriptional and physiological responses. Rapid changes in gene expression were detected in 513 treated larvae, including up-regulation of the chemokine ccl20a.3 and the down-regulation of 514 ifnphi3. The ccl19/ccl20 axis appears to be critical in teleost antiviral innate responses, as 515 previous studies have shown very rapid responses in larvae exposed to the rhabdovirus SVCV 516 (Sepahi et al., 2019). This change was also detected in the larvae that were exposed to the live 517 SARS-CoV-2 virus in the present study. We further detected a significant down-regulation of 518 type I IFN ifnphi1 gene in larvae exposed to SARS-CoV-2 S RBD protein. 519 520 Examination of ace2 transcriptional changes in zebrafish larvae exposed to SARS-CoV-2 521 revealed a consistent down-regulation in expression throughout the course of infection. 522 Interestingly, we did not observe any changes in ace2 expression after 3h immersion with SARS- 523 CoV-2 S RBD protein. Recently, enterocytes were found to be the main cell type expressing 524 Ace2 in 5 dpf-old zebrafish larvae (Postlethwait et al., 2020); and therefore it is possible that the 525 down-regulation in ace2 expression observed in our experiments was the result of enterocyte 526 responses to SARS-CoV-2. However, an olfactory epithelial cell cluster was not identified in this 527 dataset, probably because these cells constitute too small a fraction of the cells of an entire larva. 528 Importantly, we exposed larvae to the virus at 3 dpf, when the olfactory pit is already sampling 529 the surrounding water, while the gut fully opens only at 4 dpf. Thus, changes in ace2 expression 530 levels in the olfactory pit of the zebrafish larvae cannot be ruled out at this point. 531 532 Previous work has shown that ACE2 knockdown in mice protects from SARS-CoV infection 533 (Kuba et al., 2005). Thus, down-regulation of zebrafish Ace2 expression may have protected 534 larvae from SARS-CoV-2 infection in our experiments. Our data agree with studies in mouse 535 lungs, where suppression of ace2 gene expression was consistently observed following SARS- 536 CoV-2 infection (Chen et al., 2020). Interestingly, changes in ACE2 levels can occur in response 537 viruses that do not require ACE2 for host entry (Chen et al., 2020). Thus, although further 538 studies are warranted, our data suggest that Ace2 is involved in antiviral SARS-CoV-2 responses 539 in zebrafish. 540 541 We took advantage of the zebrafish fish model which allows for easy live imaging of heart beats 542 in transparent larvae. We detected in vivo cardiac/heart responses in larval zebrafish exposed to 543 SARS-CoV-2 S RBD protein characterized by tachycardia. Cardiac arrhythmia is a common 544 symptom among COVID-19 patients and current research efforts aim to understand how SARS- 545 CoV-2 infection impacts cardiovascular function (Libby, 2020). Our findings underscore that 546 SARS-CoV-2 S RBD is able to cause tachycardia in the zebrafish larval model and that this 547 model can be used for rapid evaluation of drug treatments for COVID-19. As a proof of concept, 548 we used captopril, an ACE inhibitor currently being evaluated in human clinical trials 549 (NCT04355429). Captopril treatment ameliorated tachycardia in zebrafish larvae exposed to 550 SARS-CoV-2 S RBD recombinant protein. Our studies therefore suggest the beneficial use of 551 captopril in COVID-19 patients undergoing cardiac arrhythmia, but clearly further studies are

12 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

552 required to fully translate these findings to the clinic and to determine the duration and timing of 553 captopril treatment in COVID-19 patients. 554 555 A recent report in zebrafish adults indicated that injection of recombinant SARS-CoV-2 S N 556 terminal protein caused histopathology of several tissues including the liver, kidney, brain and 557 ovary (Ventura-Fernandes et al., 2020). Additionally, some animals succumbed to injection with 558 the recombinant protein. We did not detect any mortalities neither in larvae nor in adults in any 559 of our experiments, perhaps suggesting that mortalities were due to the injection procedure rather 560 than the protein treatment itself. Of note, the dose used in the present study was considerably 561 lower than the dose delivered in the Ventura-Fernadez study, perhaps explaining the differences 562 in toxicity between both studies. We observed histological damage following a single intranasal 563 delivery of SARS-CoV-2 protein, specifically at the site of delivery, the olfactory organ, whereas 564 more transient and moderate damage was detected in the renal tubules. Renal damage may have 565 occurred by direct uptake of SARS-CoV-2 S RBD by the kidney once the protein reached the 566 bloodstream following intranasal administration or, alternatively, by activation of RAS or 567 inflammatory cascades at the olfactory organ. Thus, toxicity of SARS-CoV-2 S protein in adult 568 zebrafish may be less severe when delivered intranasally than by injection, and future studies 569 should evaluate whether current vaccine candidates also exert similar effects and whether 570 different administration routes cause the same side-effects or not. This is particularly important 571 as the intranasal route appears promising for some vaccine candidates (Ku et al., 2020). 572 573 SARS-CoV-2 infection causes anosmia and ageusia in 41-64% of patients (Dell’Era et al., 2020; 574 Spinato et al., 2020). Several mechanisms for the impaired chemosensory perception in COVID- 575 19 patients have been proposed based on the observation that, in humans, sustentacular cells and 576 basal cells but not OSNs co-express ace2 and tmprss2 (Brann et al., 2020). Initial infection of 577 SCs may indirectly cause damage to OSNs, notably via lack of metabolic support or via 578 inflammatory-derived damage (Brann et al., 2020). Animal models of SARS-CoV-2 infection 579 have provided some insights into the mechanisms of olfactory loss driven by SARS-CoV-2. In 580 golden Syrian hamster, intranasal SARS-CoV-2 infection results in drastic damage of the 581 olfactory epithelium including ciliary loss. In this animal model, while a study reported infection 582 of sustentacular cells but not OSNs, another study claimed direct infection of both mature and 583 immature OSNs (Zhang et al., 2020). In the Syrian hamster model, studies found infiltration of 584 inflammatory cells into the olfactory lamina propria (Bryche et al., 2020). Our results indicate 585 that intranasal delivery of SARS-CoV-2 S RBD is sufficient to cause severe olfactory damage 586 including loss of cilia, edema, hemorrhages and necrosis in adult zebrafish. However, no 587 inflammatory infiltrates were observed in our model compared to the hamster model, likely due 588 to the absence of an active viral infection. scRNA-Seq of the zebrafish olfactory organ identified 589 dramatic losses in OR expression in most OSN subsets. As a result, intranasal delivery of SARS- 590 CoV-2 S RBD caused changes in OSN populations characterized by loss of ciliated and 591 microvillous mature OSNs and increased proportions of cycling immature OSNs 3 d post- 592 treatment. Further, inflammatory gene signatures could be detected in SCs, ECs and myeloid cell 593 subsets in animals treated with SARS-CoV-2 S RBD. The broad loss in OR, VR and TAAR 594 expression in the zebrafish olfactory organ was paralleled by sudden hyposmia and anosmia 595 according to functional EOG recordings. This finding largely recapitulates previous results in 596 K18-hACE2 mice, which develop anosmia soon after infection (Zhen et al., 2020). The present 597 study did not determine when zebrafish recover olfactory function following SARS-CoV-2 S

13 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

598 RBD intranasal treatment, but based on our histopathological observations, the olfactory organ 599 was still severely damaged 5 days after treatment, suggesting that recovery of olfactory function 600 may take several weeks in our model. Our findings therefore indicate that similar to humans, 601 zebrafish suffer from olfactory pathology and loss of smell in response to SARS-CoV-2 S RBD 602 protein. Thus, olfactory pathophysiology appears to occur even in the absence of viral replication 603 raising the possibility that nasal vaccines for COVID-19 may also cause transient anosmia in 604 humans. 605 606 In conclusion, the present study reports that both adult and larval wild-type zebrafish can be 607 useful models to advance our understanding of COVID-19 pathophysiology, particularly of mild 608 disease. The present study is the foundation for the use of zebrafish as a novel vertebrate model 609 that can accelerate the investigation of SARS-CoV-2 pathophysiology, decipher the dynamics of 610 host innate immune responses and test drugs in a fast, inexpensive and high-throughput manner. 611 Identification of molecules with antiviral effects against SARS-CoV-2 is typically done using in 612 vitro cell cultures (Wang et al., 2020b). However, such in vitro approaches do not provide organ 613 specific information or any physiologically relevant readouts. Thus, the zebrafish model here 614 established can serve as a platform to screen and develop drug interventions for COVID-19 615 where whole-organismal in vivo readouts can be generated in very short periods of time and in an 616 inexpensive manner. Further, zebrafish could be used in high throughput screens in which not 617 only drugs but also genetic and environmental risk factors known to impact COVID-19 severity 618 can be incorporated during drug validation and testing. 619 620 Acknowledgments 621 Authors thank Dr. F. Krammer for kindly sharing the SARS-CoV-2 spike RBD recombinant 622 protein. Thank you to Dr. Ryan Wong for providing some of the AB adult animals used in this 623 study. This study was funded by NSF award # #1755348. 624 625 Author Contributions 626 Conceptualization, A.K., I.S., P.B., and JP.L., Methodology, A.K., I.S., M.H., S.B, JP.L.; 627 Investigation, A.K., I.S., M.H., S.B, JP.L.; Writing –Original Draft, A.K and I.S.; Writing – 628 Review & Editing, A.K., I.S., M.H.,W.L., S.B., P.B., and JP.L.; Funding Acquisition, I.S., M.H., 629 and S.B.; Resources, I.S., M.H., S.B. JP.L.; Supervision, I.S. 630 631 Declaration of Interests 632 Authors declare no competing interests 633 634 Figure legends 635 636 Figure 1: SARS-CoV-2 S RBD induces immune responses and elevates heart rate of larval 637 zebrafish. 638 (A) Changes in gene expression of ace2, tnf-α, il1β, il17a/f3, ccl20a.3, and ifnphi1 in 5 dpf 639 zebrafish larvae in responses to SARS-CoV-2 S RBD. Animals were exposed to SARS-CoV-2 S 640 RBD protein (r-Spike, 2 ng/ml) for 3 h at 28.5°C or vehicle. Changes in gene expression were 641 measured by RT-qPCR using rps11 as the house-keeping gene. Each data point represents a pool 642 of 4 larvae/well. Data are expressed fold-change compared to vehicle controls using the Pffafl 643 method.

14 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

644 (B) Average heart beat per minute of 7 dpf (n=6) zebrafish larvae after 3 h of incubation with 645 vehicle, 2 ng/ml r-Spike, or 2 ng/ml r-IHNVg. 646 (C) Average zebrafish heart beats per minute in 5 dpf zebrafish larvae (n=8) after 3 h of 647 incubation with vehicle, 2 ng/ml r-Spike, or 2 ng/ml r-IHNVg with and without treatment with 648 12mM of captopril. Heart beats were recorded for 3 min at 20 under a Nikon Ti microscope 649 and videos were processed with ImageJ Time Series Analyzer V3. Data were analyzed by one- 650 way ANOVA with Tukey’s multiple comparisons post hoc. Results are representative of two 651 independent experiments. 652 (D) Representative ventricular traces show the average intensity in region of interest (heart 653 ventricle) in 5 dpf zebrafish treated for 3 h with vehicle, or 2 ng/ml r-Spike, 12 mM captopril or 654 12mM captopril and 2 ng/ml r-Spike. Traces are representative of one animal per treatment (n=8 655 per treatment group). * P-value<0.05; ** P-value<0.01 *** P-value<0.001. 656 657 Figure 2: Intranasal delivery of SARS-CoV-2 S RDB protein causes severe olfactory damage in 658 adult zebrafish. H&E stains of zebrafish olfactory organ paraffin sections. (A) Control vehicle 659 treated olfactory organ. (B -C) 3 h; (D) 1 d; (E) 3 d; (F) 5 days post intranasal delivery of SARS- 660 CoV-2 S RDB protein (r-Spike). Images are representative of n=10 sections per animal and n=3 661 fish per experimental group. Asterisk denotes edemic spaces in the olfactory lamellae, mostly in 662 the lamina propria. He: hemorrhages. Arrow heads: apical loss of olfactory sensory neurons. 663 Black Arrow: Necrotic lamella. Image analyses in G and H were performed in ImageJ. 664 Differences were considered statistically significant when P-value<0.05. 665 666 Figure 3: Intranasal delivery of SARS-CoV-2 S RDB protein causes anosmia in adult zebrafish. 667 (A) Representative EOG of adult zebrafish showing olfactory responses to food over time before 668 and after addition of SARS-CoV-2 S RDB protein to the naris. (B) Mean percentage loss of 669 olfactory detection of food or bile salts measured by EOG recording. Adult zebrafish (n=4) 670 treated with SARS-CoV-2 S RDB protein in one naris or vehicle control in the other naris for 3 h 671 or 1 d after which EOG recorded as animal exposed to food odorant or bile in both nares. Percent 672 reduction calculated as percent of the SARS-CoV-2 S RDB treated naris’ EOG response to 673 vehicle treated naris’ EOG response. Statistical analyses are ANOVA with Tukey’s post hoc 674 multiple comparison **** P-value < 0.0001. 675 676 Figure 4: ScRNA-Seq of zebrafish olfactory organ reveals changes in cellular landscape due to 677 intranasal SARS-CoV-2 S RBD protein delivery. 678 (A-B) Cellular composition of the zebrafish olfactory organ visualized using t-distributed 679 stochastic neighbor embedding (tSNE) of vehicle treated, 3 h and 3 d post-intranasal delivery of 680 SARS-CoV-2 S RBD (r-Spike). Individual single-cell transcriptomes were colored according to 681 cluster identity. 682 tSNE plots were generated in Seurat and identified 20 and 22 distinct cell clusters, respectively. 683 (C) tSNE plots of zebrafish olfactory organ cell suspensions colored according to treatment. 684 (D) Bar plots showing the proportion of cells belonging to each cluster in each treatment group. 685 (E-F) Violin plots showing expression of selected markers of inflammation and vasculature 686 remodeling in the 3 h SARS-CoV-2 S RBD (orange) versus vehicle treated (blue) and 3 d SARS- 687 CoV-2 S RBD (green) versus vehicle treated (blue) zebrafish olfactory organ obtained by 688 scRNA-Seq. All represented genes were differentially regulated between treatments (adjusted P- 689 value <0.05).

15 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

690 691 Figure 5. Intranasal SARS-CoV-2 S RBD protein delivery causes widespread losses of olfactory 692 receptor expression in adult zebrafish. 693 (A-B) Bar plots showing fold-change in expression of olfactory receptor genes in the zebrafish 694 olfactory organ 3 h and 3 d post intranasal SARS-CoV-2 S RBD protein compared to vehicle 695 controls identified by scRNA-Seq. 696 (C-D) Violin plots showing changes in ompb expression, a marker of mature ciliated OSNs, in 697 zebrafish olfactory organ cell suspensions, across all olfactory neuronal cell clusters in 3 h 698 SARS-CoV-2 S RBD (orange) and 3 d SARS-CoV-2 S RBD (green) versus vehicle treated 699 (blue). 700 (E-F) Summary webs of gene ontology analyses performed in ShinyGO v0.61 from the scRNA- 701 Seq data showing the number of significantly up and down regulated genes across olfactory 702 neuronal clusters in zebrafish olfactory organ 3 h and 3 d post-intranasal SARS-CoV-2 S RBD 703 protein versus vehicle treated controls. 704 (G-H) Gene set enrichment analysis using Metascape of genes from all neuronal clusters that 705 were either significantly up or down regulated in zebrafish olfactory organ 3 h and 3 d post- 706 intranasal SARS-CoV-2 S RBD protein compared to vehicle treated controls (adjusted P-value< 707 0.05). Enriched terms are colored by P-value. 708 709 Figure 6: Intranasal SARS-CoV-2 S RBD protein delivery causes transcriptional changes in 710 zebrafish sustentacular and endothelial clusters. 711 (A-B) Bar plots showing fold-change in expression of selected genes in sustentacular and 712 endothelial cell clusters in the zebrafish olfactory organ 3 h and 3 d post intranasal SARS-CoV-2 713 S RBD protein compared to vehicle controls identified by scRNA-Seq. 714 (C-D) Summary webs of gene ontology analysis performed in ShinyGO v0.61 from the scRNA- 715 Seq data showing the number of genes significantly up and down regulated in sustentacular and 716 endothelial cell clusters in the olfactory organ of zebrafish 3 h and 3 days post-intranasal SARS- 717 CoV-2 S RBD protein compared to vehicle treated controls. 718 (E-F) Gene set enrichment analysis using Metascape of genes from sustentacular and endothelial 719 clusters that were either significantly up or down regulated in zebrafish olfactory organ 3 h and 3 720 d post-intranasal SARS-CoV-2 S RBD protein compared to vehicle treated controls (adjusted P- 721 value< 0.05). Significantly enriched terms are colored by P-value. 722 723 Figure 7: Wildtype zebrafish larvae do not support replication of SARS-CoV-2 virus. 724 (A-B) Mean viral loads quantified as log of SARS-CoV-2 N gene and SARS-CoV-2 E gene copy 725 numbers in control tank water and tank water + 104 pfu of SARS-CoV-2 for 6 h, 1 d, 2 d and 4 d. 726 (C-D) Mean viral loads quantified as log of SARS-CoV-2 N gene and SARS-CoV-2 E gene copy 727 numbers in control supernatants from well with larvae not exposed to virus, and supernatants 728 from wells with larvae that were exposed to 104 pfu of SARS-CoV-2 for 6 h, 1 d, 2 d and 4 d. 729 Each sample represents the supernatant of one well containing 12 larvae. 730 (E-F) Mean viral loads quantified as log of SARS-CoV-2 N gene and SARS-CoV-2 E gene copy 731 numbers in control larvae and larvae exposed to 104 pfu of SARS-CoV-2 for 6 h, 1 d, 2 d and 4 732 d. Each sample point represents one well containing 12 larvae. 733 Larval infections began at 2 dpf after mechanical dechorionation at 1 dpf. * P-value<0.05; ** P- 734 value<0.01 *** P-value<0.001. Results are representative of two independent experiments. 735

16 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

736 Figure 8: Exposure to SARS-CoV-2 down-regulates ace2 expression and induces pro- 737 inflammatory immune responses in zebrafish larvae. 738 (A) Changes in ace2 expression levels in control and SARS-CoV-2 exposed zebrafish. (B) 739 Changes in il1b expression in control and infected zebrafish. (C) Changes in tnfa expression in 740 control and infected zebrafish. (D) Changes in ifnphi1 expression in control and infected 741 zebrafish. (E) Changes in ifnphi3 expression in control and infected zebrafish. (F) Changes in 742 mxa expression in control and infected zebrafish. (G) Changes in il17a/f3 expression in control 743 and infected zebrafish. (H) Changes in il22expression in control and infected zebrafish. (I) 744 Changes in il4/il13 expression in control and infected zebrafish. (J) Changes in ccl20a.3 745 expression in control and infected zebrafish. (K) Changes in ccl19a.1 expression in control and 746 infected zebrafish. Zebrafish larvae were exposed to 104 pfu of SARS-CoV-2 for 6 h, 1 d, 2 d 747 and 4 d. Larvae incubated at 30. Infections began at 2 dpf after mechanical dechorionation at 1 748 dpf. Each sample represents one pool of 12 larvae. Changes in gene expression were calculated 749 as the fold-change compared to vehicle controls using the Pffafl method. Data were normalized 750 using rps11 expression which was used as the house-keeping gene. * P-value<0.05; ** P- 751 value<0.01 *** P-value<0.001. 752 753

754 755 EXPERIMENTAL MODEL AND SUBJECT DETAILS 756 Ethics statement 757 All animal protocols were done in accordance with American Veterinary Medical Association 758 guidelines under a protocol approved by the University of New Mexico’s Institutional Animal 759 Care and Use Committee (Protocol Number 19-200863-MC), the University of New Mexico 760 Health Sciences Center Institutional Biosafety Committee (Protocol Number 17-11-01R) and the 761 Texas State University Institutional Animal Care and Use Committee (Protocol Number 43). 762 763 Zebrafish Husbandry 764 For all experiments, wild type AB zebrafish were obtained from ZIRC (Oregon, USA). For the 765 intranasal delivery of SARS-CoV-2 S RBD protein into adult zebrafish, female and male adult 766 zebrafish were obtained from Dr. Wong’s laboratory at the University of Nebraska due to 767 lockdown of ZIRC during the pandemic. All fish were maintained in a filtered aquarium system 768 at 28℃ with a 14 h light and 10 h dark cycle at the University of New Mexico Aquatics Animal 769 Facility. All experiments with adults utilized a mix of male and female animals, and the larvae 770 sex is indeterminable. Animals were fed ad libitum Gemma complete nutrition (Skretting). AB 771 larvae were obtained by batch-crossing AB adults allowing for natural fertilization. The morning 772 of fertilization, larvae were collected at n=50 per Petri dish and kept in E3 medium containing 773 0.002% methylene blue. In the afternoon, larvae were placed in fresh E3 medium without 774 methylene blue and non-surviving embryos were removed. Larvae were maintained at 28.5℃ in 775 E3 medium until 4dpf when they are slowly changed to system water. 776 777 SARS-CoV-2 778 The SARS-CoV-2 isolate, a CDC isolate from a US patient (USA-WA1/2020), was obtained 779 from BEI Resources. The strain was grown at a low MOI to minimize generation of 780 noninfectious particles and low passaged virus was used in all experiments described. The virus

17 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

781 was propagated on Vero E6 cells and viral loads quantitated by RT-qPCR and by plaque forming 782 assays as we have described previously (Bradfute et al, 2020). 783

784 METHOD DETAILS 785 Intranasal delivery of SARS-CoV-2 S RBD recombinant protein to adult zebrafish 786 Adult zebrafish were anesthetized for 1 min in 0.04 mg/ml Tricaine-S (Syndel) solution and then 787 moved to an absorbent boat where their gills were still covered with anesthetic solution for 788 administration of solutions to nares. Using a microloader tip (Eppendorf, 930001007), 5 μl of 20 789 ng/μl SARS-CoV-2 S RBD (kindly provided by Dr. F. Krammer) was directly pipetted into each 790 naris, while 5 μl of sterile PBS was applied in control fish. After inoculation, animals were 791 recovered in a separate tank supplemented with O2 before returning to their rearing tank until the 792 end of the experiment. Euthanasia was performed on ice to ensure rapid death without perturbing 793 the olfactory system. 794 795 Histology and image analysis 796 Adult zebrafish heads (n=3) were fixed in 4% formaldehyde solution buffered to pH 7.4 with 60 797 mM HEPES overnight at 4℃ on a low speed rocker. Heads were then rinsed three times with 798 PBS for 30 min at room temperature (RT) on a slow rocker and decalcified in 10% EDTA 799 solution (pH 7.4) at RT for 24 h. After rinsing in PBS, samples were placed in 70% ethanol for 800 24 h at 4℃ and then processed for embedding in paraffin according to ZIRC protocols. Paraffin 801 blocks were sectioned at 5-μm thickness and stained by hematoxylin and eosin (H&E) staining. 802 Sections were observed under a Nikon Ti inverted microscope (brightfield), and images were 803 acquired with the Nikon Elements Advanced Research software and analyzed using ImageJ. 804 805 Isolation of olfactory organ cells from adult zebrafish, scRNA-Seq, scRNA-Seq data 806 analysis and Gene Ontology 807 To obtain single cell suspensions, adult AB zebrafish olfactory rosettes were dissected and 808 placed directly into 5 ml of Neurobasal medium (Gibco, 21103049) supplemented with 10% heat 809 inactivated fetal bovine serum (FBS, Peak Scientific, Colorado, USA), 100 U/ml penicillin- 810 streptomycin (Gibco, 15140122), and 0.04% filtered bovine serum albumin (Sigma) and rocked 811 at 4℃. Every 20 min, the supernatant from each sample was collected and replaced by fresh 812 medium for a total of 4 times. Supernatants were then filtered through a cell strained and kept on 813 ice. Next, the rosettes were placed in 10 ml of a DTT solution (1.3 mM EDTA, 864 μM DTT in 814 PBS) at RT for 30 min. Suspensions were filtered through a cell strainer and combined with the 815 supernatants obtained from the first steps. Rosettes were then rinsed off the DTT solution and 816 incubated in Hank’s buffer containing 10% FBS, 0.04% BSA, 100U/ml penicillin-streptomycin 817 and collagenase type IV (Gibco, 0.37 mg/ml) for 1.5 h at RT. The supernatants were filtered 818 through a cell strainer and combined with those obtained from the previous steps while the 819 rosettes were mashed against the cell strainer and washed with neurobasal medium. The 820 combined supernatants were centrifuged for 10 min at 400g in supplemented neurobasal medium 821 and cells were counted with a hemocytometer. Viability was estimated by trypan blue staining. 822 Cells were then strained twice through Flowmi 10 μm strainers and loaded onto the chromium 823 controller with a viability of > 85%. Cell libraries were generated according to 10x Genomics 824 protocols at the University of New Mexico Cancer Center Genomics core facility and sequenced 825 on an Illumina NovaSeq 6000 at the University of Colorado Genomics and Microarray core

18 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

826 facility. Sequencing depth and statistics of the scRNA-Seq run are shown in Figure S2. SRAs for 827 this project can be found on NCBI under bioproject #PRJNA668529. 828 Fastqs were run through the Cell Ranger v3.0 pipeline with default settings using the GRCz11 829 zebrafish genome. Output matrices were loaded to R (v1.2.5001) as a Seurat object (Package 830 Seurat v3.1.1). First, cells with less than 200 or greater than 2500 features, and greater than 5% 831 mitochondrial features were removed. After counts were normalized using the “lognormalize” 832 method and a scale factor of 10000, 2000 variable genes were selected using the ‘vst” method. 833 Data was scaled, and PCA dimensional reduction was run. Jackstraw analysis determined the 834 vehicle control to have 38 significant principal components (PCs) and the treated samples to 835 have 40 significant PCs which were used for clustering analysis. SARS-CoV-2 S RBD treated 836 samples were integrated with the vehicle treated sample and clustered together using 30 837 significant PCs and a resolution of 0.5. Cluster markers were identified with “FindAllMarkers” 838 in Seurat and exported for cluster identification. Differential expression analysis was done with 839 seurat “FindMarkers” in default settings for each cluster and exported for gene ontology 840 analysis. 841 Gene ontology (GO) analysis was done with web-based GUIs Metascape and ShinyGO v0.61 842 which draw multiple currently maintained databases (ensembl, ENTRZ, KEGG among others). 843 Biological process webs were created using biological process output from ShinyGO v0.61 in 844 Prism GraphPad. Biological processes bar graphs were produced by Metascape. 845 846 Electro-olfactogram recordings 847 Adult AB zebrafish were anesthetized and received 2 µl of recombinant SARS-CoV-2 S RBD 848 protein (50 ng/µl) in PBS or PBS alone. After 3 h or 1 day, zebrafish were anaesthetized (0.1 g 849 MS222/L), placed in a v-shape stand and supplied with aerated water containing MS222 850 anesthetic (0.05 g/L). The nasal flap was removed with sterile fine forceps to expose the 851 olfactory rosettes to a continuous tank water source. Olfactory responses to zebrafish food 852 extract or goldfish bile were measured by electrical recordings as detailed in Sepahi et al., 2019. 853 The food extract was prepared as a filtered solution of 1L tap water and 0.1 g of dry food pellets. 854 Water food extracts were separated in 200 ml aliquots and kept frozen until the recording day. A 855 0.5 ml mix of bile fluid from 50 adult goldfish was aliquoted in 10 µl and kept frozen until the 856 day of the recording. Before each recording, bile aliquots were diluted 1:1000 in water from the 857 EOG system. There were no significant differences in olfactory responses between males and 858 females, hence responses of both sexes were averaged together. The percentage reduction in 859 olfactory activity was calculated by dividing the amplitude of the olfactory signal at time x by 860 amplitude of the olfactory signal at time 0 *100. Percentage of olfactory signal reduction 861 between control and treated naris was calculated as follows (Amplitude response to odorant in 862 control naris (mV) - Amplitude response to odorant in treated naris (mV))/Amplitude response to 863 odorant in control naris (mV)) * 100.

864 Exposure of zebrafish larvae to SARS-CoV-2 S RBB recombinant protein and captopril 865 treatment 866 5 dpf or 7 dpf zebrafish (n=5/well, total of n=7-8 replicate wells) were placed in a 12 flat- 867 bottomed well plate (FALCON) with 2 ml of tank water per well. Wells were then treated with 868 either 2 ng/ml of recombinant SARS-CoV-2 S RBD, 2 ng/ml recombinant IHNVg protein 869 (Sepahi et al., 2019), with or without 12 µM captopril (Sigma-Aldrich, C4042). The 870 corresponding vehicle controls consisted of sterile PBS for recombinant proteins and water for

19 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

871 captopril. Animals were incubated at 28.5℃ for 3 h, mounted in 1% low melting point agarose 872 (Sigma) at 42℃, stabilized for 5 min at RT and then imaged to record heart-beat activity. 873 874 Zebrafish larvae heart-beat recordings and analysis 875 As the agarose solidified, animals were adjusted to the microscope stage (approx. 5 min) then 876 hearts were recorded using brightfield AVI for 3 min at 16.667 frames/s at RT. AVI images were 877 then opened in ImageJ with the Time series analyzer V3 plugin. Circular ROIs were drawn in 878 either the atrium or ventricle and average intensity was extracted. The maximum average 879 intensity peaks were identified and counted per 60s as bpm. Data were analyzed by one-way 880 ANOVA with Tukey’s post hoc. 881 882 Infection of zebrafish fish larvae with SARS-CoV-2 883 Ten animals were placed in each well in 12-well plates containing 2 ml of tank water and 884 transferred to BSL3 facility the day before infection. Infection was performed by adding with 104 885 plaque forming units (pfu)/well in a volume of approximately 2 µl. Larvae were collected 6 hpi, 886 1 dpi, 2 dpi, or 4 dpi in Trizol for RNA extraction. Water (supernatants) from all wells was 887 collected in AVL buffer (Qiaamp viral RNA mini kit, Qiagen) with carrier RNA at 6 hpi, 1 dpi, 2 888 dpi, or 4 dpi. 889 890 Quantification of SARS-CoV-2 viral loads 891 RNA from tank water was extracted using the Qiaamp viral RNA mini kit (Qiagen, 52904) 892 according to manufacturer’s instruction, with one modification: after the AVL buffer was frozen 893 and thawed, the solution was the heated to 50℃ for 5 min and allowed to come to RT before use. 894 RT-qPCR of the N and E genes of SARS-CoV-2 was done using research use only primer probes 895 sets from IDT (10006713 (N) and 10006888, 10006890, and 10006892 (E)) with plasmid 896 positive controls for a standard curve (10006625 and 10006896, respectively). 5 µl of RNA 897 eluted from the Qiaamp columns was added for each working reaction with Taqman Fast Virus 898 one-step master mix (4444432) and appropriate working concentrations of each primer/probe set 899 (N: 500 nM primers and 125 nM probe; E: 400 nM primers and 200 nM probe). Plates were run 900 on BioRad C1000 touch and CEX real time systems according to Taqman master mix’s protocol 901 and analyzed using Biorad CFX qPCR software. 902 903 Gene expression analyses by RT-qPCR 904 Whole larvae RNA was extracted using trizol. For tissue homogenization bead beater tubes are 905 preloaded with 1.5 g 1.0 mm dia Zirconia beads, 1.5 g 2.0 mm Zirconia beads and 500 µl Trizol. 906 Samples were loaded into the tubes and bead beat at 4350 rpm for 45 s. Tubes were then 907 centrifuged at 7,000 rpm for 7 min. The homogenate/lysates were transferred to clean 1.5 ml 908 microfuge tubes and spun at 7,000 rpm for 7 min to pellet debris. Supernatants were then 909 processed to extract the total RNA using a standard chloroform/phenol extraction protocol. RNA 910 was quantified by Nanodrop and samples were normalized and 1 µg of RNA was used to 911 synthesize cDNA using the Superscript III first strand system (Thermofisher, 18080051). qPCR 912 was performed using SSOadvanced supermix (Biorad, 1725270) and primers listed in Table S3 913 (Supplemental Methods). Gene expression changes were quantified using the Pfaffl method 914 (Pfaffl, 2001). 915 916 QUANTIFICATION AND STATISTICAL ANALYSIS

20 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

917 918 Statistical analysis 919 Statistics were done in GraphPad Prism V7.0d. Unpaired t-tests or one-way ANOVA followed 920 by a Tukey’s post hoc comparison was used to detect differences among treatments. All data 921 were checked for normal distribution prior to performing statistical analyses using F test or 922 Bartlett’s. The P-values are as follows: * P-value < 0.05, ** P-value < 0.01, *** P-value < 923 0.001, and **** P-value < 0.0001. 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 Supplemental Information titles and legends 940 941 Table S1: Phylogenetic sequence analyses of vertebrate ACE2 amino acid sequences showing 942 the percent similarity and homology. Analyses were performed in MATGAT. Homo sapiens 943 (human), Macaca mulatta (macaque), Mus musculus (mouse), Rattus norvegicus (rat), Mustela 944 putorius furo (ferret), Manis javanica (pangolin), Rhinolophus sinicus (bat), Camelus 945 dromedaries (camel), Canis familiaris (dog), Gallus gallus (chicken), Alligator mississippiensis 946 (alligator), Xenopus tropicalis (frog), Danio rerio (zebrafish). 947 948 949 Table S2: Sequence analyses of SARS-CoV-2 S protein-ACE2 binding amino acids in different 950 vertebrates. Amino acids in red denote non-conserved amino acids. In green, amino acid 951 substitutions that represent non-functional changes. 952 953 Table S3: List of primers used in the present study. 954 955 Figure S1: Intranasal delivery of SARS-CoV-2 S RDB protein causes renal injury in adult 956 zebrafish. H&E stains of zebrafish head-kidney paraffin sections. (A) Control vehicle treated 957 head-kidney. (B) 3 h; (C) 1 day; (D) 5 days post intranasal delivery of SARS-CoV-2 S RDB 958 protein. Arrows indicate vacuolization of tubule epithelial cells. Images representative n=10 959 sections per animal and n=3 per experimental group. PT: proximal tubule. DT: distal tubule. 960 961 Figure S2: scRNA-Seq identification of cell clusters in zebrafish olfactory organ based on 962 previously established homologous markers.

21 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

963 (A) Table of metrics from the scRNA-Seq. Cells were captured with 10x chromium platform and 964 libraries sequenced on an Illumina NovaSeq 6000. 965 (B) Heatmap showing markers for each of the 20 cell clusters present in the olfactory organ of 966 adult zebrafish following vehicle treatment and 3 h post intranasal delivery of SARS-CoV-2 S 967 RDB. 968 (C) Heatmap of markers for each of the 22 cell clusters present in the olfactory organ of adult 969 zebrafish following vehicle treatment and 3 d post intranasal delivery of SARS-CoV-2 S RDB. 970 971 Figure S3: Basal expression levels of ace2 in wildtype adult zebrafish olfactory organ (OO) and 972 olfactory bulb (OB) (n=3) quantified by RT-qPCR. Data were normalized to rps11 as the house 973 keeping gene. 974

975 976 977 978 979 980 981 982 983 984 985 References 986 987 Advani, A. (2020). Acute Kidney Injury: A Bona Fide Complication of Diabetes. Diabetes 69 988 (11), 2229-2237. 989 990 Albini, A., Di Guardo, G., Noonan, D.M.C., and Lombardo, M. (2020). The SARS-CoV-2 991 receptor, ACE-2, is expressed on many different cell types: implications for ACE-inhibitor- and 992 angiotensin II receptor blocker-based cardiovascular therapies. Intern. Emerg. Med. 15, 759–766. 993 994 Babapoor-farrokhran, S., Gill, D., Walker, J., Rasekhi, R.T., Bozorgniaa, B., and Amanullaha, A. 995 (2020). Myocardial injury and COVID-19: Possible mechanisms Savalan. Life Sci. 253, 117723. 996 997 Bao, L., Deng, W., Huang, B., Gao, H., Liu, J., Ren, L., Wei, Q., Yu, P., Xu, Y., Qi, F., et al. 998 (2020). The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice. Nature 583, 830–833. 999 1000 Bergsland, M., Werme, M., Malewicz, M., Perlmann, T., and Muhr, J. (2006). The establishment 1001 of neuronal properties is controlled by Sox4 and Sox11. Genes Dev. 20, 3475–3486. 1002 1003 Bermingham-McDonogh, O., and Reh, T.A. (2011). Regulated reprogramming in the 1004 regeneration of sensory receptor cells. Neuron 71, 389–405. 1005 1006 Blanco-Melo, D., Nilsson-Payant, B.E., Liu, W.-C., Uhl, S., Hoagland, D., Møller, R., Jordan, 1007 T.X., Oishi, K., Panis, M., Sachs, D., et al. (2020). Imbalanced Host Response to SARS-CoV-2 1008 Drives Development of COVID-19. Cell 181, 1036–1045.

22 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1009 Borders, A.S., Hersh, M.A., Getchell, M.L., Van Rooijen, N., Cohen, D.A., Stromberg, A.J., and 1010 Getchell, T. V. (2007). Macrophage-mediated neuroprotection and neurogenesis in the olfactory 1011 epithelium. Physiol. Genomics 31, 531–543. 1012 1013 Brann, D.H., Tsukahara, T., Weinreb, C., Lipovsek, M., Van Den Berge, K., Gong, B., Chance, 1014 R., Macaulay, I.C., Chou, H.J., Fletcher, R.B., et al. (2020). Non-neuronal expression of SARS- 1015 CoV-2 entry genes in the olfactory system suggests mechanisms underlying COVID-19- 1016 associated anosmia. Sci. Adv. 6, 1–20. 1017 1018 Bryche, B., St Albin, A., Murri, S., Lacôte, S., Pulido, C., Ar Gouilh, M., Lesellier, S., Servat, 1019 A., Wasniewski, M., Picard-Meyer, E., et al. (2020). Massive transient damage of the olfactory 1020 epithelium associated with infection of sustentacular cells by SARS-CoV-2 in golden Syrian 1021 hamsters. Brain. Behav. Immun. 89, 579–586. 1022 1023 Buiakova, O.I., Baker, H., Scott, J.W., Farbman, A., Kream, R., Grillo, M., Franzen, L., 1024 Richman, M., Davis, L.M., Abbondanzo, S., et al. (1996). Olfactory marker protein (OMP) gene 1025 deletion causes altered physiological activity of olfactory sensory neurons. Proc. Natl. Acad. Sci. 1026 U. S. A. 93, 9858–9863. 1027 1028 Burgos, J.S., Ripoll-Gomez, J., Alfaro, J.M., Sastre, I., and Fernando Valdivieso (2008). 1029 Zebrafish as a New Model for Herpes Simplex Virus Type 1 Infection. Zebrafish 5, 323–333. 1030 1031 Butler, A., Hoffman, P., Smibert, P., Papalexi, E., and Satija, R. (2018). Integrating single-cell 1032 transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36, 1033 411–420. 1034 1035 Chan, J.F.W., Zhang, A.J., Yuan, S., Poon, V.K.M., Chan, C.C.S., Lee, A.C.Y., Chan, W.M., 1036 Fan, Z., Tsoi, H.W., Wen, L., et al. (2020). Simulation of the clinical and pathological 1037 manifestations of Coronavirus Disease 2019 (COVID-19) in golden Syrian hamster model: 1038 implications for disease pathogenesis and transmissibility. Clin. Infect. Dis. 2019. ciaa325. 1039 1040 Chen, J., Jiang, Q., Xia, X., Liu, K., Yu, Z., Tao, W., Gong, W., and Han, J.D.J. (2020). 1041 Individual variation of the SARS-CoV-2 receptor ACE2 gene expression and regulation. Aging 1042 Cell 19, 1–12. 1043 1044 Chen, M., Reed, R.R., and Lane, A.P. (2017). Acute inflammation regulates neuroregeneration 1045 through the NF-κB pathway in olfactory epithelium. Proc. Natl. Acad. Sci. 114, 8089–8094. 1046 1047 Chen, M., Reed, R.R., and Lane, A.P. (2019). Chronic Inflammation Directs an Olfactory Stem 1048 Cell Functional Switch from Neuroregeneration to Immune Defense. Cell Stem Cell 25, 501- 1049 513.e5. 1050 1051 Cooper, K.W., Brann, D.H., Farruggia, M.C., Bhutani, S., Pellegrino, R., Tsukahara, T., 1052 Weinreb, C., Joseph, P. V., Larson, E.D., Parma, V., et al. (2020). COVID-19 and the Chemical 1053 Senses: Supporting Players Take Center Stage Keiland. Neuron 107, 219–233. 1054

23 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1055 Damas, J., Hughes, G.M., Keough, K.C., Painter, C.A., Persky, N.S., Corbo, M., Hiller, M., 1056 Koepfli, K.P., Pfenning, A.R., Zhao, H., et al. (2020). Broad host range of SARS-CoV-2 1057 predicted by comparative and structural analysis of ACE2 in vertebrates. Proc. Natl. Acad. Sci. 1058 U. S. A. 117, 22311–22322. 1059 1060 Dell’Era, V., Farri, F., Garzaro, G., Gatto, M., Aluffi Valletti, P., and Garzaro, M. (2020). Smell 1061 and taste disorders during COVID-19 outbreak: Cross-sectional study on 355 patients. Head 1062 Neck 42, 1591– 1596. 1063 1064 Del Valle, D.M., Kim-Schulze, S., Huang, H.H., Beckmann, N.D., Nirenberg, S., Wang, B., 1065 Lavin, Y., Swartz, T.H., Madduri, D., Stock, A., et al. (2020). An inflammatory cytokine 1066 signature predicts COVID-19 severity and survival. Nat. Med. 26, 1636–1643. 1067 1068 Ellett, F., Pase, L., Hayman, J.W., Andrianopoulos, A., and Lieschke, G.J. (2011). Mpeg1 1069 Promoter Transgenes Direct Macrophage-Lineage Expression in Zebrafish. Blood 117, 49–56. 1070 1071 Fu, J., Zhou, B., Zhang, L., Balaji, K.S., Wei, C., Liu, X., Chen, H., Peng, J., and Fu, J. (2020). 1072 Expressions and significances of the angiotensin-converting enzyme 2 gene, the receptor of 1073 SARS-CoV-2 for COVID-19. Mol. Biol. Rep. 47, 4383–4392. 1074 1075 Gabor, K.A., Goody, M.F., Witten, P.E., Mowel, W.K., Breitbach, M.E., Kim, C.H., and 1076 Gratacap, R.L. (2014). Influenza A virus infection in zebrafish recapitulates mammalian 1077 infection and sensitivity to anti-influenza drug treatment. Dis. Model. Mech. 7, 1227–1237. 1078 1079 Galindo-Villegas, J. (2020). COVID-19: Real-time dissemination of scientific information to 1080 fight a public health emergency of international concern. Front. Pharmacol. 11, 10–11. 1081 1082 Garvin, M.R., Alvarez, C., Miller, J.I., Prates, E.T., Walker, A.M., Amos, B.K., Mast, A.E., 1083 Justice, A., Aronow, B., and Jacobson, D. (2020). A mechanistic model and therapeutic 1084 interventions for COVID-19 involving a RAS-mediated bradykinin storm. eLife 2020;9:e59177. 1085 1086 Ge, S.X., Jung, D., and Yao, R. (2020). ShinyGO: a graphical gene-set enrichment tool for 1087 animals and plants. Bioinformatics 36, 2628–2629. 1088 1089 Giancotti, V., Bergamin, N., Cataldi, P., and Rizzi, C. (2018). Epigenetic Contribution of High- 1090 Mobility Group A Proteins to Stem Cell Properties. Int. J. Cell Biol. 2018, 3698078. 1091 1092 Glowacka, I., Bertram, S., Herzog, P., Pfefferle, S., Steffen, I., Muench, M.O., Simmons, G., 1093 Hofmann, H., Kuri, T., Weber, F., et al. (2010). Differential Downregulation of ACE2 by the 1094 Spike Proteins of Severe Acute Respiratory Syndrome Coronavirus and Human Coronavirus 1095 NL63. J. Virol. 84, 1198–1205. 1096 1097 Gomes, M.C., and Mostowy, S. (2020). The Case for Modeling Human Infection in Zebrafish. 1098 Trends Microbiol. 28, 10–18. 1099 1100 Guzik, T.J., Mohiddin, S.A., Dimarco, A., Patel, V., Savvatis, K., Marelli-Berg, F.M., Madhur,

24 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1101 M.S., Tomaszewski, M., Maffia, P., D’Acquisto, F., et al. (2020). COVID-19 and the 1102 cardiovascular system: Implications for risk assessment, diagnosis, and treatment options. 1103 Cardiovasc. Res. 116, 1666–1687. 1104 1105 Hadjadj, J., Yatim, N., Barnabei, L., Corneau, A., Boussier, J., Pere, H., Charbit, B., Bondet, V., 1106 Chenevier-Gobeaux, C., Breillat, P., et al. (2020). Impaired type I interferon activity and 1107 exacerbated inflammatory responses in severe Covid-19 patients. Science 369, 718-724. 1108 1109 Hamming, O.J., Lutfalla, G., Levraud, J.-P., and Hartmann, R. (2011). Crystal Structure of 1110 Zebrafish Interferons I and II Reveals Conservation of Type I Interferon Structure in Vertebrates. 1111 J. Virol. 85, 8181–8187. 1112 1113 Hansen, A., Rolen, S.H., Anderson, K., Morita, Y., Caprio, J., and Finger, T.E. (2003). 1114 Correlation between Olfactory Receptor Cell Type and Function in the Channel Catfish. J. 1115 Neurosci. 23, 9328–9339. 1116 1117 Hoffmann, M., Kleine-Weber, H., and Pöhlmann, S. (2020a). A Multibasic Cleavage Site in the 1118 Spike Protein of SARS-CoV-2 Is Essential for Infection of Human Lung Cells. Mol. Cell 78, 1119 779-784.e5. 1120 1121 Hoffmann, M., Kleine-Weber, H., Schroeder, S., Krüger, N., Herrler, T., Erichsen, S., 1122 Schiergens, T.S., Herrler, G., Wu, N.H., Nitsche, A., et al. (2020b). SARS-CoV-2 Cell Entry 1123 Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor. Cell 1124 181, 271-280.e8. 1125 1126 Howe, K., Clark, M.D., Torroja, C.F., Torrance, J., Berthelot, C., Muffato, M., Collins, J.E., 1127 Humphray, S., McLaren, K., Matthews, L., et al. (2013). The zebrafish reference genome 1128 sequence and its relationship to the human genome. Nature 496, 498–503. 1129 1130 Imai, Y., Kuba, K., Rao, S., Huan, Y., Guo, F., Guan, B., Yang, P., Sarao, R., Wada, T., Leong- 1131 Poi, H., et al. (2005). Angiotensin-converting enzyme 2 protects from severe acute lung failure. 1132 Nature 436, 112–116. 1133 1134 Imam, S.A., Lao, W.P., Reddy, P., Nguyen, S.A., Schlosser, R.J. (2020). Is SARS-CoV-2 1135 (COVID-19) postviral olfactory dysfunction (PVOD) different from other PVOD? World J. 1136 Otorhinolaryngol. Head Neck Surg. 2020;10.1016/j.wjorl.2020.05.004. 1137 doi:10.1016/j.wjorl.2020.05.004 1138 1139 Jia, H., Yue, X., and Lazartigues, E. (2020). ACE2 mouse models: a toolbox for cardiovascular 1140 and pulmonary research. Nat. Commun. 11, 5165. 1141 1142 Kim, J., Kim, Y., Kim, H.-T., Kim, D.W., Ha, Y., Kim, J., Kim, C.-H., Lee, I., and Song, K. 1143 (2009). TC1(C8orf4) Is a Novel Endothelial Inflammatory Regulator Enhancing NF-κB Activity. 1144 J. Immunol. 183, 3996–4002. 1145 1146 Kim, Y. Il, Kim, S.G., Kim, S.M., Kim, E.H., Park, S.J., Yu, K.M., Chang, J.H., Kim, E.J., Lee,

25 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1147 S., Casel, M.A.B., et al. (2020). Infection and Rapid Transmission of SARS-CoV-2 in Ferrets. 1148 Cell Host Microbe 27, 704-709.e2. 1149 1150 Kochi, A.N., Tagliari, A.P., Forleo, G.B., Fassini, G.M., and Tondo, C. (2020). Cardiac and 1151 arrhythmic complications in patients with COVID-19. J. Cardiovasc. Electrophysiol. 31, 1003– 1152 1008. 1153 1154 Kraemer, A.M., Saraiva, L.R., and Korsching, S.I. (2008). Structural and functional 1155 diversification in the teleost S100 family of calcium-binding proteins. BMC Evol. Biol. 8, 1–23. 1156 1157 Ku, M.-W., Bourgine, M., Authié, P., Lopez, J., Nemirov, K., Moncoq, F., Noirat, A., Vesin, B., 1158 Nevo, F., Blanc, C., et al. (2020). Intranasal Vaccination with a Lentiviral Vector Strongly 1159 Protects against SARS-CoV-2 in Mouse and Golden Hamster Preclinical Models. BioRxiv. doi: 1160 https://doi.org/10.1101/2020.07.21.214049 1161 1162 Kuba, K., Imai, Y., Rao, S., Gao, H., Guo, F., Guan, B., Huan, Y., Yang, P., Zhang, Y., Deng, 1163 W., et al. (2005). A crucial role of angiotensin converting enzyme 2 (ACE2) in SARS 1164 coronavirus-induced lung injury. Nat. Med. 11, 875–879. 1165 1166 Kuck, K.H. (2020). Arrhythmias and sudden cardiac death in the COVID-19 pandemic. Herz 45, 1167 325–326. 1168 1169 Lamers, M.M., Beumer, J., Vaart, J. Van Der, Knoops, K., Puschhof, J., Breugem, T.I., Ravelli, 1170 R.B.G., Schayck, J.P. Van, Mykytyn, A.Z., Duimel, H.Q., et al. (2020). SARS-CoV-2 1171 productively infects human gut enterocytes. Science 369, 50–54. 1172 1173 Levraud, J.-P., Jouneau, L., Briolat, V., Laghi, V., and Boudinot, P. (2019). IFN-Stimulated 1174 Genes in Zebrafish and Humans Define an Ancient Arsenal of Antiviral Immunity. J. Immunol. 1175 203(12), 3361-3373 1176 1177 Li, L., Chong, H.C., Ng, S.Y., Kwok, K.W., Teo, Z., Tan, E.H.P., Choo, C.C., Seet, J.E., Choi, 1178 H.W., Buist, M.L., et al. (2015). Angiopoietin-like 4 increases pulmonary tissue leakiness and 1179 damage during influenza pneumonia. Cell Rep. 10, 654–663. 1180 1181 Libby, P. (2020). The Heart in COVID-19: Primary Target or Secondary Bystander? JACC Basic 1182 to Transl. Sci. 5, 537–542. 1183 1184 Liu, Z., Xiao, X., Wei, X., Li, J., Yang, J., Tan, H., Zhu, J., Zhang, Q., Wu, J., and Liu, L. 1185 (2020). Composition and divergence of coronavirus spike proteins and host ACE2 receptors 1186 predict potential intermediate hosts of SARS-CoV-2. J. Med. Virol. 92, 595–601. 1187 1188 Lopes, R.D., Macedo, A.V.S., Silva, P.G.M. de B. e, JunqueiraMoll-Bernardes, R., Feldman, A., 1189 Arruda, G.D.S., Souza, A.S. de, Albuquerque, D.C. de, Mazza, L., Santos, M.F., et al. (2020). 1190 Continuing versus suspending angiotensin- converting enzyme inhibitors and angiotensin 1191 receptor blockers: Impact on adverse outcomes in hospitalized patients with severe acute 1192 respiratory syndrome coronavirus 2 ( SARS-CoV-2 ) – The BRACE CORONA Trial. Am. Heart

26 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1193 J. 226, 50–59. 1194 1195 Lucas, C., Wong, P., Klein, J., Castro, T.B.R., Silva, J., Sundaram, M., Ellingson, M.K., Mao, T., 1196 Oh, J.E., Israelow, B., et al. (2020). Longitudinal analyses reveal immunological misfiring in 1197 severe COVID-19. Nature 584, 463–469. 1198 1199 Ludwig, M., Palha, N.¤, Torhy, C., Rie Briolat, V., Colucci-Guyon, E., Bré Mont, M., Herbomel, 1200 P., Boudinot, P., and Levraud, J.-P. (2011). Whole-Body Analysis of a Viral Infection: Vascular 1201 Endothelium is a Primary Target of Infectious Hematopoietic Necrosis Virus in Zebrafish 1202 Larvae. PLoS Pathog 7(2):e1001269. 1203 1204 Lutz, C., Maher, L., Lee, C., and Kang, W. (2020). COVID-19 preclinical models: Human 1205 angiotensin-converting enzyme 2 transgenic mice. Hum. Genomics 14, 1–11. 1206 1207 Mellert, T.K., Getchell, M.L., Sparks, L., and Getchell, T. V (1992). Characterization of the 1208 immune barrier in human olfactory mucosa. Otolaryngol Head Neck Surg. 106, 181–188. 1209 1210 Menni, C., Valdes, A.M., Freidin, M.B., Sudre, C.H., Nguyen, L.H., Drew, D.A., Ganesh, S., 1211 Varsavsky, T., Cardoso, M.J., El-Sayed Moustafa, J.S., et al. (2020). Real-time tracking of self- 1212 reported symptoms to predict potential COVID-19. Nat. Med. 26, 1037–1040. 1213 1214 Mudd, P.A., Crawford, J.C., Turner, J.S., Souquette, A., Reynolds, D., Bender, D., Bosanquet, 1215 J.P., Anand, N.J., Striker, D.A., Martin, R.S., et al. (2020). Targeted Immunosuppression 1216 Distinguishes COVID-19 from Influenza in Moderate and Severe Disease. MedRxiv. 1217 doi: 10.1101/2020.05.28.20115667. 1218 1219 Munster, V.J., Feldmann, F., Williamson, B.N., van Doremalen, N., Pérez-Pérez, L., Schulz, J., 1220 Meade-White, K., Okumura, A., Callison, J., Brumbaugh, B., et al. (2020). Respiratory disease 1221 in rhesus macaques inoculated with SARS-CoV-2. Nature 585, 268–272. 1222 1223 Palha, N., Guivel-Benhassine, F., Briolat, V., Lutfalla, G., Sourisseau, M., Ellett, F., Wang, C.H., 1224 Lieschke, G.J., Herbomel, P., Schwartz, O., et al. (2013). Real-Time Whole-Body Visualization 1225 of Chikungunya Virus Infection and Host Interferon Response in Zebrafish. PLoS Pathog. 9(9), 1226 e1003619. 1227 1228 Papadakis, K.A., Landers, C., Prehn, J., Kouroumalis, E.A., Moreno, S.T., Gutierrez-Ramos, J.- 1229 C., Hodge, M.R., and Targan, S.R. (2003). CC Chemokine Receptor 9 Expression Defines a 1230 Subset of Peripheral Blood Lymphocytes with Mucosal T Cell Phenotype and Th1 or T- 1231 Regulatory 1 Cytokine Profile. J. Immunol. 171, 159–165. 1232 1233 Park, A., and Iwasaki, A. (2020). Type I and Type III Interferons – Induction, Signaling, 1234 Evasion, and Application to Combat COVID-19. Cell Host Microbe 27, 870–878. 1235 1236 Patel, V.B., Clarke, N., Wang, Z., Fan, D., Parajuli, N., Basu, R., Putko, B., Kassiri, Z., Turner, 1237 A.J., and Oudit, G.Y. (2014). Angiotensin II induced proteolytic cleavage of myocardial ACE2 is 1238 mediated by TACE/ADAM-17: A positive feedback mechanism in the RAS. J. Mol. Cell.

27 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1239 Cardiol. 66, 167–176. 1240 1241 Peterson, J., Lin, B., Barrios-Camacho, C.M., Herrick, D.B., Holbrook, E.H., Jang, W., Coleman, 1242 J.H., and Schwob, J.E. (2019). Activating a Reserve Neural Stem Cell Population In Vitro 1243 Enables Engraftment and Multipotency after Transplantation. Stem Cell Reports 12, 680–695. 1244 1245 Pfaffl, M.W. (2001). A new mathematical model for relative quantification in real-time RT– 1246 PCR. Nucleic Acids Res. 29, e45. 1247 1248 Postlethwait, J.H., Farnsworth, D.R., and Miller, A.C. (2020). An intestinal cell type in zebrafish 1249 is the nexus for the SARS-CoV-2 receptor and the Renin Angiotensin-Aldosterone System that 1250 contributes to COVID-19 comorbidities. BioRxiv. doi: 10.1101/2020.09.01.278366. 1251 1252 Ribeiro-Oliveira, A., Nogueira, A.I., Pereira, R.M., Vilas Boas, W.W., Souza dos Santos, R.A., 1253 and Simões e Silva, A.C. (2008). The renin-angiotensin system and diabetes: An update. Vasc. 1254 Health Risk Manag. 4, 787–803. 1255 1256 Rockx, B., Kuiken, T., Herfst, S., Bestebroer, T., Lamers, M.M., Munnink, B.B.O., De Meulder, 1257 D., Van Amerongen, G., Van Den Brand, J., Okba, N.M.A., et al. (2020). Comparative 1258 pathogenesis of COVID-19, MERS, and SARS in a nonhuman primate model. Science 368, 1259 1012–1015. 1260 1261 Rodriguez, S., Cao, L., Rickenbacher, G.T., Benz, E.G., Magdamo, C., Gomez, L.R., Holbrook, 1262 E.H., Albers, A.D., Gallagher, R., Westover, M.B., et al. (2020). Innate immune signaling in the 1263 olfactory epithelium reduces odorant receptor levels: modeling transient smell loss in COVID-19 1264 patients. MedRxiv. doi: 10.1101/2020.06.14.20131128. 1265 1266 Sa Ribero, M., Jouvenet, N., Dreux, M., and Nisole, S. (2020). Interplay between SARS-CoV-2 1267 and the type I interferon response. PLoS Pathog. 16, 1–22. 1268 1269 Saraiva, L.R., Ahuja, G., Ivandic, I., Syed, A.S., Marioni, J.C., Korsching, S.I., and Logan, D.W. 1270 Molecular and neuronal homology between the olfactory systems of zebrafish and mouse. Sci. 1271 Rep. 5, 11487. 1272 1273 Sepahi, A., Casadei, E., Tacchi, L., Muñoz, P., LaPatra, S.E., and Salinas, I. (2016). Tissue 1274 Microenvironments in the Nasal Epithelium of Rainbow Trout ( Oncorhynchus mykiss ) Define 1275 Two Distinct CD8α + Cell Populations and Establish Regional Immunity. J. Immunol. 197, 1276 4453–4463. 1277 1278 Sepahi, A., Kraus, A., Casadei, E., Johnston, C.A., Galindo-villegas, J., Kelly, C., García- 1279 Morenoc, D., Muñoze, P., Mulero, V., Huertas, M., et al. (2019). Olfactory sensory neurons 1280 mediate ultrarapid antiviral immune responses in a TrkA-dependent manner. Proc. Natl. Acad. 1281 Sci. U. S. A. 116, 12428–12436. 1282 1283 Shi, J., Wen, Z., Zhong, G., Yang, H., Wang, C., Huang, B., Liu, R., He, X., Shuai, L., Sun, Z., 1284 et al. (2020). Susceptibility of ferrets, cats, dogs, and other domesticated animals to SARS-

28 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1285 coronavirus 2. Science 368, 1016–1020. 1286 1287 Slade, C.D., Reagin, K.L., Lakshmanan, H.G., Klonowski, K.D., and Watford, W.T. (2020). 1288 Placenta-specific 8 limits IFNγ production by CD4 T cells in vitro and promotes establishment of 1289 influenza-specific CD8 T cells in vivo. PLoS One 15, 1–18. 1290 1291 Song, E., Zhang, C., Israelow, B., Lu-Culligan, A., Prado, A.V., Skriabine, S., Lu, P., Weizman, 1292 O.-E., Liu, F., Dai, Y., et al. (2020). Neuroinvasion of SARS-CoV-2 in human and mouse brain. 1293 BioRxiv. doi: https://doi.org/10.1101/2020.06.25.169946. 1294 1295 Spinato, G., Fabbris, C., Polesel, J., Cazzador, D., Borsetto, D., Hopkins, C., and Paolo Boscolo- 1296 Rizzo (2020). Alterations in Smell or Taste in Mildly Symptomatic Outpatients With SARS- 1297 CoV-2 Infection. N. Engl. J. Med. 323, 2005–2011. 1298 1299 Su, H., Yang, M., Wan, C., Yi, L.X., Tang, F., Zhu, H.Y., Yi, F., Yang, H.C., Fogo, A.B., Nie, 1300 X., et al. (2020). Renal histopathological analysis of 26 postmortem findings of patients with 1301 COVID-19 in China. Kidney Int. 98, 219–227. 1302 1303 Suzuki, M., Saito, K., Min, W.P., Vladau, C., Toida, K., Itoh, H., Murakami, S. (2007). 1304 Identification of viruses in patients with postviral olfactory dysfunction. Laryngoscope 117(2), 1305 272-277. 1306 1307 1308 Tabata, S., Imai, K., Kawano, S., Ikeda, M., Kodama, T., Miyoshi, K., Obinata, H., Mimura, S., 1309 Kodera, T., Kitagaki, M., et al. (2020). Clinical characteristics of COVID-19 in 104 people with 1310 SARS-CoV-2 infection on the Diamond Princess cruise ship: a retrospective analysis. Lancet 1311 Infect. Dis. 20, 1043–1050. 1312 1313 Taylor, K.L., Grant, N.J., Temperley, N.D., and Patton, E.E. (2010). Small molecule screening in 1314 zebrafish: An in vivo approach to identifying new chemical tools and drug leads. Cell Commun. 1315 Signal. 8, 1–14. 1316 1317 Torabi, A., Mohammadbagheri, E., Akbari Dilmaghani, N., Bayat, A.H., Fathi, M., Vakili, K., 1318 Alizadeh, R., Rezaeimirghaed, O., Hajiesmaeili, M., Ramezani, M., et al. (2020). 1319 Proinflammatory Cytokines in the Olfactory Mucosa Result in COVID-19 Induced Anosmia. 1320 ACS Chem. Neurosci. 11, 1909–1913. 1321 1322 Van Dycke, J., Ny, A., Conceição-Neto, N., Maes, J., Hosmillo, M., Cuvry, A., Goodfellow, I., 1323 Nogueira, T.C., Verbeken, E., Matthijnssens, J., et al. (2019). A robust human norovirus 1324 replication model in zebrafish larvae. PLoS Pathog. 15, 1–21. 1325 1326 Ventura-Fernandes, B.H., Feitosa, N.M., Barbosa, A.P., Bomfim, C.G., Garnique, A.M.B., 1327 Gomes, F.I.F., Nakajima, R.T., Belo, M.A.A., Eto, S.F., Fernandes, D.C., et al. (2020). Zebrafish 1328 studies on the vaccine candidate to COVID-19, the Spike protein: Production of antibody and 1329 adverse reaction. BioRxiv. doi: https://doi.org/10.1101/2020.10.20.346262. 1330

29 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1331 Wadman, M., Couzin-Frankel, J., Kaiser, J., and O, C.M. (2020). A rampage through the body. 1332 Science 368, 356–360. 1333 1334 Wagener, B.M., Hu, M., Zheng, A., Zhao, X., Che, P., Brandon, A., Anjum, N., Snapper, S., 1335 Creighton, J., Guan, J.-L., et al. (2016). Neuronal Wiskott–Aldrich syndrome protein regulates 1336 TGF-β1–mediated lung vascular permeability. FASEB J. 30, 2557–2569. 1337 1338 Wan, Y., Shang, J., Graham, R., Baric, R.S., and Li, F. (2020). Receptor Recognition by the 1339 Novel Coronavirus from Wuhan: an Analysis Based on Decade-Long Structural Studies of 1340 SARS Coronavirus. J. Virol. 94, 1–9. 1341 1342 Wang, D., Hu, B., Hu, C., Zhu, F., Liu, X., Zhang, J., Wang, B., Xiang, H., Cheng, Z., Xiong, 1343 Y., et al. (2020a). Clinical Characteristics of 138 Hospitalized Patients with 2019 Novel 1344 Coronavirus-Infected Pneumonia in Wuhan, China. JAMA - J. Am. Med. Assoc. 323, 1061– 1345 1069. 1346 1347 Wang, M., Cao, R., Zhang, L., Yang, X., Liu, J., Xu, M., Shi, Z., Hu, Z., Zhong, W., and Xiao, 1348 G. (2020b). Remdesivir and chloroquine effectively inhibit the recently emerged novel 1349 coronavirus (2019-nCoV) in vitro. Cell Res. 30, 269–271. 1350 1351 Wei, S., Zhou, J.M., Chen, X., Shah, R.N., Liu, J., Orcutt, T.M., Traver, D., Djeu, J.Y., Litman, 1352 G.W., and Yoder, J.A. (2007). The zebrafish activating immune receptor Nitr9 signals via 1353 Dap12. Immunogenetics 59, 813–821. 1354 1355 De Wit, E., Van Doremalen, N., Falzarano, D., and Munster, V.J. (2016). SARS and MERS: 1356 Recent insights into emerging coronaviruses. Nat. Rev. Microbiol. 14, 523–534. 1357 1358 Yao, Y., Yao, J., and Boström, K.I. (2019). SOX Transcription Factors in Endothelial 1359 Differentiation and Endothelial-Mesenchymal Transitions. Front. Cardiovasc. Med. 6, 1–8. 1360 1361 Yoo, S.K., and Huttenlocher, A. (2011). Spatiotemporal photolabeling of neutrophil trafficking 1362 during inflammation in live zebrafish. J. Leukoc. Biol. 89, 661–667. 1363 1364 Yu, C.R., and Wu, Y. (2017). Regeneration and rewiring of rodent olfactory sensory neurons. 1365 Exp. Neurol. 287, 395–408. 1366 1367 Zhang, A.J., Lee, A.C.-Y., Chu, H., Chan, J.F.-W., Fan, Z., Li, C., Liu, F., Chen, Y., Yuan, S., 1368 Poon, V.K.-M., et al. (2020). Severe Acute Respiratory Syndrome Coronavirus 2 Infects and 1369 Damages the Mature and Immature Olfactory Sensory Neurons of Hamsters. Clin. Infect. Dis. 1370 ciaa995. https://doi.org/10.1093/cid/ciaa995. 1371 1372 1373 Zheng, Y., Vertuani, S., Nyström, S., Audebert, S., Meijer, I., Tegnebratt, T., Borg, J.P., Uhlén, 1374 P., Majumdar, A., and Holmgren, L. (2009). Angiomotin-like protein 1 controls endothelial 1375 polarity and junction stability during sprouting angiogenesis. Circ. Res. 105, 260–270. 1376

30 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1377 Zhou, Y., Zhou, B., Pache, L., Chang, M., Khodabakhshi, A.H., Tanaseichuk, O., Benner, C., 1378 and Chanda, S.K. (2019). Metascape provides a biologist-oriented resource for the analysis of 1379 systems-level datasets. Nat. Commun. 10, 1523. 1380 1381 Zhu, J., Ji, P., Pang, J., Zhong, Z., Li, H., He, C., Zhang, J., and Zhao, C. (2020). Clinical 1382 characteristics of 3062 COVID-19 patients: A meta-analysis. J. Med. Virol. 92, 1902–1914. 1383 1384 Ziegler, C.G.K., Allon, S.J., Nyquist, S.K., Mbano, I.M., Miao, V.N., Tzouanas, C.N., Cao, Y., 1385 Yousif, A.S., Bals, J., Hauser, B.M., et al. (2020). SARS-CoV-2 Receptor ACE2 Is an 1386 Interferon-Stimulated Gene in Human Airway Epithelial Cells and Is Detected in Specific Cell 1387 Subsets across Tissues. Cell 181, 1016-1035.e19. 1388 1389 Zou, J., and Secombes, C.J. (2016). The function of fish cytokines. Biology (Basel). 5. 1390 1391 Zou, X., Chen, K., Zou, J., Han, P., Hao, J., and Han, Z. (2020). Single-cell RNA-seq data 1392 analysis on the receptor ACE2 expression reveals the potential risk of different human organs 1393 vulnerable to 2019-nCoV infection. Front. Med. 14, 185–192. 1394

31 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint Figure 3 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

A + I.N SARS-CoV-2 S RBD B

**** **** 100 100 **** **** Vehicle 3 h SARS-CoV-2 Spike RBD 1 day SARS-CoV-2 50 50 Spike RBD to food

0 treated vs untreated O 0

% of pre-exposure response -15 10 35 60 Food Bile Time (min) % Reduction olfaction in paire d bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Figure 5 A B 3 d 2 3 h 2 Neuron1 1 Neuron2 1 Neuron3

0 Neuron4 Neuron5 0 Neuron6 fold chang e fold chang e -1 2 2 Neuron7 Neuron8 -1 Lo g Lo g -2

-2 -3

v2ra1 v2rc1 ompb taar20i v2rh32 or137-4taar19cor105-1taar19sor106-9 or11or1-8111-9or132-5or108-3taar19n or125-2or11or133-11-5 or132-5or132-2or115-5or108-3or132-4or106-9or132-1or132-4or132-5or137-9or108-3or115-1or102-1or108-1taar19ptaar19uor132-1taar20wtaar19n or115-10 or132-2.1 or106-10 or115-12or132-2.1 C D

4 4

3 3 2

2 ompb ompb 1 1 0

0 Neuron1 Neuron2 Neuron3 Neuron4 Neuron5 Neuron6 Neuron7 Neuron1 Neuron2 Neuron3 Neuron5 Neuron6 Neuron7 Neuron8 E F Neuron Di erentiation Sensory system development Regulation of cell Sensory perception of smell di erentiation 19 21

10 24 12 G-protein-coupled receptor Sensory organ signaling pathway morphogenesis

Significant genes Significant genes Upregulated Detection of chemical stimulus Upregulated G protein-coupledbioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191 receptor ; this version posted November 8, 2020. The copyright holder for this preprint 42 21 (which was not certified by peer review) is the author/funder.Dow All rightsnregulated reserved. No reuse allowed without permission. involved in sensory Downregulated signaling pathway 3 perception of smell 72

48 Sensory perception 42 22 Detection of chemical stimulus of smell 11 involved in sensory perception Nervous system process Sensory perception 10 9 32 24 31 13 10 34 G H sensory perception of chemical stimulus sensory perception of smell Glutathione metabolism circadian regulation of gene expression sensory organ development Iron uptake and transport ventral spinal cord interneuron differentiation negative regulation of cell differentiation hypochord development GPCR signaling, coupled to cyclic nucleotide 2nd messenger closure of optic fissure neuron development microtubule polymerization or depolymerization regeneration Apoptosis epithelial cell differentiation Upregulated Glycerophospholipid biosynthesis response to cadmium ion Keratinization 0 5 10 15 20 25 30 35 40 neuron fate commitment p53 signaling pathway sensory perception of smell Tight junction Ca2+ pathway Keratinization Upregulated Post-translational protein phosphorylation morphogenesis of an epithelial sheet negative regulation of transcription by RNA polymerase II reactive oxygen species metabolic process mitotic DNA integrity checkpoint cell chemotaxis response to reactive oxygen species response to hypoxia GPCR signaling, coupled to cyclic nucleotide 2nd messenger Cell adhesion molecules (CAMs) -2 0 2 4 6 8 10 12 14 16 Downregulated 0 5 10 15 20 10 sensory perception of chemical stimulus -log10(P) 10-3 Keratinization GPCR signaling, coupled to cyclic nucleotide 2nd messenger -4 cilium movement involved in cell motility 10 atrioventricular valve morphogenesis -6 prostaglandin biosynthetic process 10 response to xenobiotic stimulus -10 G alpha (s) signalling events 10 Phase I - Functionalization of compounds -20 biomineral tissue development 10 Downregulated 0 10 20 30 40 50 60 -log10(P) Figure 6 A 3 h B 3 d 3 2 Sustentacular1 2 Sustentacular2 1 Sustentacular3 Sustentacular4 1 0 Sustentacular5 Endothelial1 fold chang e fold chang e 2 0 2 -1 Endothelial2 Endothelial3 Lo g Lo g -1 -2

-2 -3

ppl lxn id2b nppc arg2krt17 epgn edn2 krt17 tcima gng8 bcam rpb4 elavl4 tubb5 cd74bapoeb apoeb foxq1anfil3-5s100z gpx1b igfbp1a foxq1asox11a muc5.1 postnb igfbp1atnfrs11b tnfrsf9a il17a/f3notch1b ccl19a.2 hsp70.3

BX511021.1 Regulation of cellular Negative regulation of cellular CR318588.1 metabolic process metabolic process Developmental process C D Multicellular organism Cell chemotaxis development

13 67 33 46 Developmental 22 process 58 Cell di erentiation 6 39

52 23 61 Anatomical structure development 29 Immune system response

Significant genes Significant genes Upregulated Anatomical structure Upregulated Response to stress Downregulated development Downregulated 66 22 46 18

20 51 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint 30(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 58 Response to stress Response to wounding Regulation of cell Multicellular organism proliferation development

34 16 25 10 8 46 15 52

E F response to wounding Tight junction morphogenesis of embryonic epithelium sensory perception of smell inflammatory response to wounding response to inorganic substance Metal sequestration by antimicrobial proteins cell chemotaxis cell chemotaxis hydrogen peroxide catabolic process fin morphogenesis Notch signaling pathway epithelial tube formation sensory system development regulation of multicellular organismal development negative regulation of signal transduction microtubule depolymerization circadian regulation of gene expression extracellular matrix organization negative regulation of molecular function glial cell development cell activation sensory perception of chemical stimulus Upregulated Upregulated fin regeneration neural nucleus development Neutrophil degranulation response to xenobiotic stimulus morphogenesis of an epithelial sheet circadian regulation of gene expression Post-translational protein phosphorylation Phagosome central nervous system development 0 1 2 3 4 5 regulation of cation channel activity

0 1 2 3 4 5 6 7 8 cell proliferation cell migration Notch signaling pathway response to stimulus regeneration locomotion tube development developmental process Platelet activation, signaling and aggregation regulation of biological process lateral line system development inflammatory response to wounding growth liver development biological regulation RHO GTPases Activate NADPH Oxidases positive regulation of biological process regulation of neurogenesis signaling Keratinization localization cell chemotaxis multi-organism process 10-2 erythrocyte maturation metabolic process immune system process -3 Downregulated Downregulated lymphangiogenesis apoptotic process multicellular organismal process 10 response to reactive oxygen species 0 1 2 3 4 5 -4 G alpha (i) signalling events -log10(P) 10 regulation of cellular component organization -6 otic vesicle morphogenesis 10 0 2 4 6 8 -log10(P) bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 7

A B SARS-CoV-2 N SARS-CoV-2 E 10 **** 10 **** *** *** 8 **** 8 ****

6 6

4 4 Log(Copies) Log(Copies) 2 2 N.D. N.D. 0 0 C SARS-CoV-2 N D SARS-CoV-2 E 10 10

8 8

6 6

4 4 Log(Copies) Log(Copies) 2 2 N.D. N.D. 0 0 E SARS-CoV-2 N F SARS-CoV-2 E 10 10 * * * 8 8 *

6 6

4 4 Log(Copies) Log(Copies) 2 2 N.D. 0 0

Control Control

6 h SARS-CoV-21 d SARS-CoV-22 d SARS-CoV-24 d SARS-CoV-2 6 h SARS-CoV-21 d SARS-CoV-22 d SARS-CoV-24 d SARS-CoV-2 bioRxiv preprint doi: https://doi.org/10.1101/2020.11.06.368191; this version posted November 8, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 8

ace2 il1β A 4 B 20 C 4 tnfα * * * * 2 15 3

/ rps11 *

0 β 10 -2 2 5 * -4 * * Fold change Fold change 1 -6 0 Fold change il1 -8 -5 0 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d

D E ifnφ3 F mxa 8 ifnφ1 * 6 4 * * * * * * 4 2 4 * * 2 0

0 0 -2

-2 -4 Fold change Fold change Fold change -4 -4 -6

-8 -6 -8 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d

il22 il4 G 5 il17a/f3 H 4 I 3

* 2 * * * 4 2 * * 1

3 0 0 * * 2 -2 -1 Fold change Fold change Fold change -2 1 -4 -3

0 -6 -4 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d

ccl20a.3 ccl19a.1 J 4 K 5.0 * * 2.5 * 3 0.0 * Control 2 -2.5 -5 Fold change Fold change 1 -10 SARS-CoV-2 -15

0 -20 6 h 1 d 2 d 4 d 6 h 1 d 2 d 4 d