key: cord-0288051-8g0zg7gh
authors: Wang, Ying; Thaler, Melissa; Ninaber, Dennis K.; van der Does, Anne M.; Ogando, Natacha S.; Beckert, Hendrik; Taube, Christian; Salgado-Benvindo, Clarisse; Snijder, Eric J.; Bredenbeek, Peter J.; Hiemstra, Pieter S.; van Hemert, Martijn J.
title: Impact of human airway epithelial cellular composition on SARS-CoV-2 infection biology
date: 2021-07-22
journal: bioRxiv
DOI: 10.1101/2021.07.21.453304
sha: 1e9063349ac32d434072a5a029b41d8efad47376
doc_id: 288051
cord_uid: 8g0zg7gh

Infection biology and pathogenesis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of the coronavirus disease 2019 (COVID-19), are incompletely understood. Here, we assessed the impact of airway epithelial cellular composition on infection in air-liquid interface (ALI) cultures of differentiated primary human tracheal (PTEC) and bronchial epithelial cells (PBEC). We first compared SARS-CoV-2 infection kinetics, related antiviral and inflammatory responses, and viral entry factors in PTEC and PBEC. Next, the contribution of differentiation time was investigated by differentiating ALI-PTEC/PBEC for 3-5 weeks and comparing dynamics of viral replication/spread, cellular composition and epithelial responses. We observed a gradual increase in viral load with prolonged culture duration. Ciliated and goblet cells were predominantly infected in both PTEC and PBEC. Immunofluorescence analysis and RT-qPCR showed that compared to other cell types mainly ciliated and goblet cell numbers were affected by increased culture duration. An increased proportion of these two target cell types was associated with increased viral load. Furthermore, modulation of cellular composition using IL-13 and the Notch signaling inhibitor DAPT, underlined the importance of both ciliated and goblet cells for infection. DAPT treatment resulted in a lower viral load and a relative increase in ciliated cells at the expense of goblet cells, compared to IL-13 treated cultures in which both cell types were present and viral load was higher. In conclusion, our results identify cellular composition as a contributing factor to airway epithelial susceptibility to SARS-CoV-2. IMPORTANCE In this study, we determined an effect of culture duration and airway cellular composition of ALI-PBEC and ALI-PTEC cultures on SARS-CoV-2 infection. We found that SARS-CoV-2 infection was increased with prolonged cell culture time and the total percentage and proportion of ciliated and goblet cells played an important role in infection level, suggesting that airway epithelial differentiation/maturation levels may in part determine susceptibility of SARS-CoV-2 infection. The development of effective therapies either targeting virus replication or pathogenesis against SARS-CoV-2 requires robust cell culture-based infection models to test small molecules and biologicals. Therefore, it is important to identify factors that are essential for reliably modeling SARS-CoV-2-airway epithelial cell interactions. This study sheds light on virus-airway epithelial cell interactions and adds to the complexity of SARS-CoV-2 cell tropism in the airways. In addition, the effect of IL-13 on viral infection hints at a causal connection between SARS-CoV-2 infection and (allergic) asthma.

ABSTRACT Infection biology and pathogenesis of severe acute respiratory syndrome 28 coronavirus 2 (SARS-CoV-2), the causative agent of the coronavirus disease 2019 29 , are incompletely understood. Here, we assessed the impact of airway 30 epithelial cellular composition on infection in air-liquid interface (ALI) cultures of 31 differentiated primary human tracheal (PTEC) and bronchial epithelial cells (PBEC). We 32 first compared SARS-CoV-2 infection kinetics, related antiviral and inflammatory 33 responses, and viral entry factors in PTEC and PBEC. Next, the contribution of 34 differentiation time was investigated by differentiating ALI-PTEC/PBEC for 3-5 weeks 35 and comparing dynamics of viral replication/spread, cellular composition and epithelial 36 responses. We observed a gradual increase in viral load with prolonged culture duration. 37

Ciliated and goblet cells were predominantly infected in both PTEC and PBEC. 38 Immunofluorescence analysis and RT-qPCR showed that compared to other cell types 39 mainly ciliated and goblet cell numbers were affected by increased culture duration. An 40 increased proportion of these two target cell types was associated with increased viral 41 load. Furthermore, modulation of cellular composition using IL-13 and the Notch 42 signaling inhibitor DAPT, underlined the importance of both ciliated and goblet cells for 43 infection. DAPT treatment resulted in a lower viral load and a relative increase in ciliated 44 cells at the expense of goblet cells, compared to IL-13 treated cultures in which both cell 45 types were present and viral load was higher. 46

In conclusion, our results identify cellular composition as a contributing factor to airway epithelial susceptibility to 49 IMPORTANCE In this study, we determined an effect of culture duration and airway 50 cellular composition of ALI-PBEC and ALI-PTEC cultures on SARS-CoV-2 infection. 51

We found that SARS-CoV-2 infection was increased with prolonged cell culture time and 52 the total percentage and proportion of ciliated and goblet cells played an important role in 53 infection level, suggesting that airway epithelial differentiation/maturation levels may in 54 part determine susceptibility of SARS-CoV-2 infection. 55

The development of effective therapies either targeting virus replication or pathogenesis 56 against SARS-CoV-2 requires robust cell culture-based infection models to test small 57 molecules and biologicals. Therefore, it is important to identify factors that are essential 58 for reliably modeling SARS-CoV-2-airway epithelial cell interactions. This study sheds 59 light on virus-airway epithelial cell interactions and adds to the complexity of SARS-60

CoV-2 cell tropism in the airways. In addition, the effect of IL-13 on viral infection hints 61 at a causal connection between SARS-CoV-2 infection and (allergic) asthma. 62

KEYWORDS SARS-CoV-2, human airway epithelial cells, infection biology, 64 differentiation, cellular composition 65 INTRODUCTION SARS-CoV-2. Viral load was analyzed at 72 hpi (Fig. 2) . In both ALI-PBEC and ALI-158 PTEC, an increase in intra-and extracellular viral RNA as well as infectious virus 159 particles was observed with longer culture time, with the highest viral load observed at 5 160 weeks after ALI ( Fig. 2A-2C) . A gradual 1-2 log increase in SARS-CoV-2 progeny 161 production was observed from 3 to 5 weeks of differentiation when three independent 162 cultures of different PTEC and PBEC donor mixes were followed over time (Fig. S1) . 163

This shows the robustness of our model and the reproducibility of our results, 164 independent of the ALI-PBEC or ALI-PTEC donor mix used. Immunofluorescence 165 staining of these cultures for the viral nucleocapsid protein also demonstrated an increase 166 in the number of infected cells with increasing culture duration (Fig. 2D) . A significantly 167 higher viral load was observed in ALI-PBEC than in ALI-PTEC, with an average 10-fold 168 difference in extracellular SARS-CoV-2 RNA copies ( Fig. 2A ) and infectious progeny 169 ( Fig. 2B ), in particular in 5-week differentiated cultures. 170

In conclusion, our results show that the anatomical origin of the epithelial cells and the 171 culture duration have an effect on the levels of infection, suggesting that cellular 172 composition is a contributing factor. To further investigate this possibility, we identified 173 the cell types that were primarily infected by SARS-CoV-2 and assessed their abundance 174 with respect to culture duration. 175

and tracheal epithelial cell cultures. Previous studies have indicated that within the 178 human respiratory tract, predominantly ciliated cells, but also goblet cells of the airway 179 epithelium can be infected with SARS-CoV-2 (14), besides alveolar epithelial cells (26, 180 27) . To assess if ciliated and goblet cells were also the target cells in our cultures, we 181 investigated the colocalization of the SARS-CoV-2 nucleocapsid protein with either 182 acetylated α-tubulin as a marker for ciliated cells, or MUC5AC as a marker for goblet 183

We observed positive staining of the viral nucleocapsid protein in acetylated α-tubulin- hypothesized that changes in cellular composition might account for these differences. 194 Therefore, we compared cellular composition between 2, 3, 4 and 5 week-differentiated 195 cultures, for the same donor mixes that were used for infection. The results showed that 196 the cultures were well-differentiated at all time points, showing expression of markers 197 related to all cell types (ciliated, goblet, club and basal cells) in . However, there were differences in the proportions of goblet and ciliated 199 cells. Using FOXJ1 and acetylated α-tubulin as markers for ciliated cells, we observed 200 that the percentage of FOXJ1-positive cells was significantly higher in ALI-PBEC after 5 201 weeks ALI culture compared to 3-week cultures. The percentage of ciliated cells was 202 significantly higher in ALI-PBEC than in ALI-PTEC at all culture durations ( Fig. 4A and 203 4B). The change in the percentage of MUC5AC-positive goblet cells was not significant 204 over time in ALI-PBEC and ALI-PTEC ( Fig. 4A and 4B ). The sum of the percentage of 205 ciliated and goblet cells was higher in week 5 ALI-PBEC cultures compared to week 3 206 cultures, and it was also higher in ALI-PBEC than ALI-PTEC when 5-week cultures 207 were compared (Fig. 4B ). These differences in the total proportion of both cell types 208 between PTEC and PBEC and between cultures of different duration correlated with 209 differences in virus levels (Fig. 2) . Furthermore, mRNA levels of FOXJ1 were 210 significantly increased in 4 week ALI-PBEC compared to 3 week cultures, however they 211 did not further increase in 5 week cultures (Fig. 4C ). In addition, FOXJ1 mRNA was 212 higher in 4/5-week ALI-PBEC cultures compared to 4/5-week PTEC cultures (Fig. 4C) . 213

In line with these findings, MUC5AC mRNA levels were higher at week 5 in ALI-PTEC 214 cultures compared to week 3, and also higher than in week 5 ALI-PBEC (Fig. 4C ). In 215 contrast, there was no significant difference in the expression of SCGB1A1 (club cell 216 marker) and TP63 (basal cell marker) (Fig. S3) . These results suggest that despite the 217 early presence of transcripts which are specific for certain cell types, maturation of these 218 cell types (which also requires expression at the protein level) continues for several 219 weeks in cultures at ALI. In addition to culture duration, difference of ciliated and goblet 220 cell markers between ALI-PBEC and ALI-PTEC indicated that the origin of the epithelial 221 cells (tracheal versus bronchial) also has an impact on epithelial cellular composition. 7C and 7D). When 5-week PBEC were (long-term) treated with DAPT, we observed 294 lower SARS-CoV-2-induced antiviral responses than in untreated and IL-13-treated 295 cultures ( Fig. 7E and 7F ). All these findings correlate with the observed differences in the 296 number of infected cells and viral load resulting from culture duration and DAPT and IL-297 13 treatment (Fig. 5) . This suggests that antiviral responses were likely not affected by 298 differentiation time or long-term DAPT/IL-13 treatment, but rather by a direct response 299 to the virus load. 300

Here we describe and characterize the use of well-differentiated human primary airway 304 epithelial cell cultures to investigate the effect of cell composition on susceptibility to 305 SARS-CoV-2 infection. Our key finding is that culture duration at the air-liquid interface, 306 which is needed to achieve mucociliary differentiation, is an important contributor to 307 SARS-CoV-2 infection kinetics. Specifically, the percentage of goblet and ciliated cells 308

is pivotal as these cell types are likely the first to become infected. 309

As a highly relevant cell model, cultures of well-differentiated airway epithelial cells are 310 employed to study infection biology of SARS-CoV-2. With regard to differentiation time, 311 there is no consensus or standardized protocol for ALI-PBEC cultures and it is described 312 in literature as anywhere from 2 to 6 weeks after start of culture at the air-liquid interface 313 (14, 15, 28) . To validate our model, we decided to use up to 5 weeks differentiated ALI-314 PBEC cultures, in line with a recent study (14) . In this study, we report efficient infection 315 with the peak of viral load at 72 hpi with around 10 11 extracellular RNA copies and 10 6 316 PFU/ml, and observed virus spread across the epithelium by immunofluorescence 317 staining of infected cells. The amount of viral progeny produced is in line with other 318 studies assessing viral load levels (14). We have established our findings using cultures 319 derived from single donors and mixes derived from multiple donors, as well as compared 320 primary bronchial and tracheal cell cultures. The viral load was previously reported to 321 plateau at 72 hpi and onset of related antiviral responses as well as inflammation were 322 also delayed, and therefore we mainly focused on this time point for analyses (14, 15 ). In 323 this study, mRNA levels of IL-6 and IL-8 were significantly upregulated by SARS-CoV-324 2, which is consistent with previous findings (29) To achieve mucociliary differentiation, primary human bronchial/tracheal epithelial cells 436

were cultured at the air-liquid interface (ALI) as previously described (22) Aldrich) to half of apical washes. Tripure reagent was spiked with Equine arteritis virus 497 (EAV) to control for variation in RNA extraction efficiency and possible inhibitors of 498 RT-qPCR. Intracellular RNA was isolated by adding 500 ul of TriPure reagent directly 499 onto the insert. Samples were stored at -20 o C until RNA was isolated using the Direct-500 zol TM -96 RNA plate isolation (Zymo), 5PRIME Phase Lock Gel extraction (Quantabio) 501 or Maxwell® 16 simply RNA tissue kit (Promega, the Netherlands). The cellular 502 reference gene PGK-1 was used as a house-keeping gene for intracellular RNA. Primers 503 and probes for EAV and PGK-1 and the normalization procedure was performed as 504 described before (44). Viral RNA was quantified by RT-qPCR using the TaqMan™ Fast 505

Virus 1-Step Master Mix (Thermo Fisher Scientific). Primers and probes were used as 506 described previously (45), but with modifications as listed in Table 1 . A standard curve 507 generated by RT-qPCR on 10-fold serial dilutions of a T7 RNA polymerase-generated in 508 vitro transcript containing the target sequences was used for absolute quantification of 509 RNA copy numbers. For other gene expression analysis, RNA was reverse-transcribed 510 and cDNA was amplified by real-time qPCR (Bio-Rad, Veenendaal, the Netherlands) 511 using specific primers. Relative normalized gene expression compared to reference genes 512 Ribosomal Protein L13a (RPL13A) and ATP synthase, H + transporting, mitochondrial F1 513 complex, beta polypeptide (ATP5B) were calculated according to the standard curve 514 method. Reference genes were selected out of 8 candidate reference genes using the 515 "Genorm" software (Genorm; Primer Design, Southampton, UK). A RT-qPCR program 516 for both RNA copy numbers and host genes of 5 min at 50°C and 20 s at 95°C or direct 3 517 min at 95°C, followed by 45 cycles of 5 s at 95°C and 30 s at 60°C or 63°C (optimal 518 temperature depending on primers), was performed on a CFX384 Touch™ Real-Time 519 PCR Detection System (Bio-Rad). Primer pairs are listed in Table 1 . 520

For quantification of the number of infectious virus particles, the apical wash was serially 521 diluted and infectious titers were determined by plaque assay on Vero E6 as described 522 before (44). 523 524 Immunofluorescence staining. ALI cultures were rinsed using PBS and cells were fixed 525 by adding 3% (w/v) paraformaldehyde diluted in PBS into the basal and apical 526 compartments followed by incubation at room temperature for at least 35 min. Next, 527 inserts were washed two times with PBS and stored in PBS with 10 mM glycine at 4 o C 528 until further use. Ice-cold methanol was added for 10 min at 4 o C, and PBS containing 1% 529 (w/v) BSA, 0.3% (w/v) Triton-X-100 (PBT) was used to block non-specific binding sites 530 and permeabilize cells for 30 min at 4 o C. Membranes were excised from the insert and 531 cut into 4 pieces that were incubated overnight at 4 o C with specific antibodies at the 532 following dilutions: rabbit anti-SARS-CoV-2 N antibody (JUC3,1:500, (46) 

COVID-19 outbreak: Migration, effects on society, 562 global environment and prevention

Risk related to spread of new SARS-CoV-2 variants of concern in 564 the EU/EEA

New variant of SARS-CoV-2 in UK causes surge of COVID-19

Linked 570 Clusters of SARS-CoV-2 Variant B.1.351 -Maryland

India's massive COVID surge puzzles scientists

Structural and Functional Analysis of the D614G 578 SARS-CoV-2 Spike Protein Variant

CoV-2 genomic variations associated with mortality rate of COVID-19

Epidemiology, virology, and clinical features of severe acute 583 respiratory syndrome -coronavirus-2 (SARS-CoV-2

Coronavirus Disease-19)

A pneumonia outbreak 589 associated with a new coronavirus of probable bat origin

Early local immune defences in the 591 respiratory tract

The innate immune function of 593 airway epithelial cells in inflammatory lung disease

Suramin Inhibits 596 SARS-CoV-2 Infection in Cell Culture by Interfering with Early Steps of the 597 Replication Cycle

SARS-coronavirus-2 replication in Vero E6 cells: replication kinetics, rapid 601 adaptation and cytopathology

Morphogenesis and cytopathic effect of SARS-CoV-2 infection in human 605 airway epithelial cells

Long-Term Modeling of 607 SARS-CoV-2 Infection of In Vitro Cultured Polarized Human Airway 608 Epithelium. mBio 11

SARS-CoV-2 Infection of Airway Cells

Pöhlmann 612 S. 2020. SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is 613 Blocked by a Clinically Proven Protease Inhibitor

SARS-CoV-2 entry factors are highly expressed in nasal epithelial cells 617 together with innate immune genes

Gene expression and 622 in situ protein profiling of candidate SARS-CoV-2 receptors in human airway 623 epithelial cells and lung tissue

AXL is a candidate receptor for SARS-CoV-2 that promotes infection of 627 pulmonary and bronchial epithelial cells

Repair and Remodeling of airway epithelium after 630 injury in Chronic Obstructive Pulmonary Disease. Curr Respir Care Rep 2

Fluticasone Inhibit Rhinovirus-Induced Mucin Production via Multiple 633 Mechanisms in Differentiated Airway Epithelial Cells

Aberrant epithelial differentiation by cigarette smoke dysregulates respiratory 638 host defence

Cigarette smoke differentially affects IL-13-induced gene expression in 641 human airway epithelial cells

An Organoid-derived Bronchioalveolar Model 646 for SARS-CoV-2 Infection of Human Alveolar-type II-like Cells

Human Lung Stem Cell-Based 651

Alveolospheres Provide Insights into SARS-CoV-2-Mediated Interferon 652 Responses and Pneumocyte Dysfunction

SARS-CoV-2 Infection

Pluripotent Stem Cell-Derived Human Lung Alveolar Type 2 Cells Elicits a 658

Rapid Epithelial-Intrinsic Inflammatory Response

Expression of the SARS-CoV-2 ACE2 Receptor in the Human 661 Airway Epithelium

2020. Type I and Type III Interferons 666

Restrict SARS-CoV-2 Infection of Human Airway Epithelial Cultures

Distinct 669 infection process of SARS-CoV-2 in human bronchial epithelial cell lines

Novel dynamics of human mucociliary differentiation 674 revealed by single-cell RNA sequencing of nasal epithelial cultures

SARS-CoV-2 Receptor ACE2 Is an Interferon-Stimulated 684

Gene in Human Airway Epithelial Cells and Is Detected in Specific Cell Subsets 685 across Tissues

Host-Viral Infection Maps Reveal Signatures of Severe

2020. 692 SARS-CoV-2 receptor ACE2 and TMPRSS2 are primarily expressed in bronchial 693 transient secretory cells

Type 2 inflammation modulates ACE2 and 696 TMPRSS2 in airway epithelial cells

Anti-TNF-α agents Modulate SARS-CoV-2 698 Receptors and Increase the Risk of Infection Through Notch-1 Signaling

Asthma and COVID-19: do we finally have answers?

Mucus production stimulated by IFN-AhR 704 signaling triggers hypoxia of COVID-19

SARS-CoV-2 709 infection damages airway motile cilia and impairs mucociliary clearance

Type 2 and interferon inflammation regulate 716 SARS-CoV-2 entry factor expression in the airway epithelium

Critical Role 720 of Type III Interferon in Controlling SARS-CoV-2 Infection in Human Intestinal 721 Epithelial Cells

Inhibition of SARS-CoV-2 by type I and type III 724 interferons

Impairs Vitamin D-Induced and Constitutive Airway Epithelial Host Defense 727 Mechanisms

6'-β-Fluoro-Homoaristeromycin and 6'-Fluoro-Homoneplanocin A Are 731

Potent Inhibitors of Chikungunya Virus Replication through Their Direct Effect 732 on Viral Nonstructural Protein 1

Detection of 737 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

MERS-coronavirus replication induces severe in vitro 741 cytopathology and is strongly inhibited by cyclosporin A or interferon-α 742 treatment

FIG 2. Effects of culture duration on SARS-CoV-2 infection in PTEC and PBEC

PFU per insert). (A) Extracellular viral RNA copies in the apical 778 wash or (C) intracellular copies were measured by RT-qPCR. (B) Viral infectious 779 progeny was determined by plaque assay in Vero E6 cells. Mean values ± SEM is 780 presented from 3 independent experiments using 3 different donor mixes

Significant differences are indicated by *P<0.05. (D) Cells were immunofluorescence 783 stained with rabbit polyclonal anti-SARS-CoV-2 N protein antibody (JUC3) and DAPI 784 for nuclear staining. Images shown are representative for results

from the same 3 independent experiments shown in A-C at 400 x 786 original magnification

Cell types infected by SARS-CoV-2 in ALI-PBEC

donors) cultured for 5 weeks were infected with SARS-CoV-2 (30,000 PFU per insert)

Immunofluorescence staining at 72 hpi with primary antibodies against acetylated α-790 tubulin as a ciliated cell marker (A) or MUC5AC as a goblet cell marker (B) in 791 combination with anti-SARS-CoV-2 N protein antibody (JUC3) and DAPI for nuclear 792 staining. Immunofluorescence images shown are representative for results of 3 793 independent experiments derived from the same donor mix with 630 x original 794 magnification

Effect of culture duration on epithelial differentiation markers in PTEC and 797

(mix of 5 donors) were differentiated at ALI for 3

A) using antibodies 799 against acetylated α-tubulin and FOXJ1 (ciliated cell markers) or MUC5AC (goblet cell 800 marker) in combination with DAPI for nuclear staining. Images shown are representative 801 for results of 3 independent experiments with 100x/400x (insert) original magnification

Quantification of FOXJ-positive cells and MUC5AC-positive cells was done by 803

C) mRNA levels of FOXJ1 and MUC5AC were measured by RT-804

qPCR. n=3 independent experiments derived from 3 different donor mixes same as in 805

CoV-2 RNA copies were measured by RT-qPCR and plaque assay was performed with 892 apical washes to quantify infectious virus titers. n=3 independent experiments. Data are 893 mean ± SEM. Fold change in intracellular RNA copies was compared to untreated 894 controls. Statistical analysis was performed using a paired t test. Significant differences 895 are indicated by *P<0.05. 896 897