(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



Distinct spatial arrangements of ACE2 and TMPRSS2 expression in Syrian hamster lung lobes dictates SARS-CoV-2 infection patterns

['Ilhan Tomris', 'Department Of Chemical Biology', 'Drug Discovery', 'Utrecht Institute For Pharmaceutical Sciences', 'Utrecht University', 'Utrecht', 'The Netherlands', 'Kim M. Bouwman', 'Youri Adolfs', 'Department Of Translational Neuroscience']

Date: 2022-05

Abstract SARS-CoV-2 attaches to angiotensin-converting enzyme 2 (ACE2) to gain entry into cells after which the spike protein is cleaved by the transmembrane serine protease 2 (TMPRSS2) to facilitate viral-host membrane fusion. ACE2 and TMPRSS2 expression profiles have been analyzed at the genomic, transcriptomic, and single-cell RNAseq levels. However, transcriptomic data and actual protein validation convey conflicting information regarding the distribution of the biologically relevant protein receptor in whole tissues. To describe the organ-level architecture of receptor expression, related to the ability of ACE2 and TMPRSS2 to mediate infectivity, we performed a volumetric analysis of whole Syrian hamster lung lobes. Lung tissue of infected and control animals was stained using antibodies against ACE2 and TMPRSS2, combined with SARS-CoV-2 nucleoprotein staining. This was followed by light-sheet microscopy imaging to visualize their expression and related infection patterns. The data demonstrate that infection is restricted to sites containing both ACE2 and TMPRSS2, the latter is expressed in the primary and secondary bronchi whereas ACE2 is predominantly observed in the bronchioles and alveoli. Conversely, infection completely overlaps where ACE2 and TMPRSS2 co-localize in the tertiary bronchi, bronchioles, and alveoli.

Author summary The ongoing COVID19 pandemic necessitates additional tools to study SARS-CoV infection dynamics. This increases our understanding of where these viruses cause infections in the lung, and how essential different receptor molecules are. Here we applied tissue optical clearing and 3D immunofluorescence of whole organs from our best animal model, the Syrian hamster. We describe the organ-level architecture of receptor expression profiles, to visualize how their concerted action mediates SARS-CoV-2 infection patterns.

Citation: Tomris I, Bouwman KM, Adolfs Y, Noack D, van der Woude R, Kerster G, et al. (2022) Distinct spatial arrangements of ACE2 and TMPRSS2 expression in Syrian hamster lung lobes dictates SARS-CoV-2 infection patterns. PLoS Pathog 18(3): e1010340. https://doi.org/10.1371/journal.ppat.1010340 Editor: Amy L. Hartman, University of Pittsburgh, UNITED STATES Received: August 20, 2021; Accepted: February 4, 2022; Published: March 7, 2022 Copyright: © 2022 Tomris et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data are within the manuscript and its Supporting Information files. Funding: R.P.dV is a recipient of an ERC Starting Grant from the European Commission (802780) and a Beijerinck Premium of the Royal Dutch Academy of Sciences. SH was funded by NIH/NIAID (contract number HHSN272201400008C). R.W.S. is funded by the Netherlands Organization for Scientific Research (NWO) with a Vici grant, by the Bill & Melinda Gates Foundation, Collaboration for AIDS Vaccine Discovery (CAVD) grant INV-002022. M.J.vG is funded by a Amsterdam UMC AMC Fellowship a Bill & Melinda Gates Foundation, COVID-19 Wave 2 mAbs grant INV-024617. The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.

Introduction SARS-CoV-2 has sparked a pandemic and additional means to understand the infection dynamics of this virus will facilitate counter-measures. SARS coronaviruses carry a single protruding envelope protein, called spike, that is essential for binding to and subsequent infection of host cells. The trimeric spike protein binds to angiotensin-converting enzyme 2 (ACE2), which functions as an entry receptor for SARS-CoV [1,2]. After binding and internalization, TMPRSS2 induces the spike protein into its fusogenic form allowing fusion of the viral and target membrane. Several other attachment factors and/or receptors have been reported [3–6], but it is generally accepted that ACE2 and TMPRSS2 are essential. To understand how a zoonotic coronavirus is so successful in the human population, complementary techniques are required to describe the organ-level architecture of ACE2 and TMPRSS2 expression profiles, to describe how their concerted action mediates infectivity. ACE2 and TMPRSS2 are expressed in a wide variety of tissues and have been analyzed using different genetic techniques [7–9]. High expression of these proteins was observed in extrapulmonary tissues, whereas the virus mainly infects the respiratory tract [7,8,10]. Conflicting transcriptomic data for ACE2 and TMPRSS2 expression in various tissues and cell types have been reported [7,8,10–13]. Several studies found that the majority of ACE2 was expressed in alveolar type II cells whilst others detected the highest expression in nasal epithelial cells followed by lower airway tissues [7,8,11]. In several studies, no ACE2 expression was observed in the gastrointestinal tract [14], whilst others found the highest expression in this tissue [8,10]. Even studies assessing actual biochemical expression of ACE2 portray conflicting data [10,12,13]. TMPRSS2 expression is described in fewer studies. Transcriptomic analyses indicated the highest expression in alveolar type I and II cells, followed by lung bronchus and in the nasal cavity [8]. High protein expression has been observed in bronchial epithelium and alveolar type II cells [13]. Beyond transcript concentration other factors contribute to protein expression, hence a direct correlation between mRNA and protein abundance for the same location/cell type may not reflect the actual situation [15]. Importantly, a deep proteome and transcriptome abundance atlas of 29 healthy human tissues assessing the expression and quantities of 13,640 human proteins confirmed that a majority of proteins with high mRNA expression were hardly detected in the proteome. For ACE2 and TMPRSS2 transcriptomic and proteomic data can be extracted from this publicly available dataset [10]. Initial comparison of ACE2 transcripts and ACE2 protein abundance already conveys a disparity, with low expression being observed in the appendix, lung and rectum whilst actual protein detection portray high expression in these tissues relative to actin (ACTB). TMPRSS2 transcript presence and protein abundance appeared to have a higher correlation. To determine whether two variables (transcripts and proteins) vary together the correlation coefficient can be assessed (S1 Fig). The direction and magnitude of correlation of several non-SARS-CoV-2 related genes (EIF4A3, ACTB and SYK) were determined as a baseline for ACE2 and TMPRSS2 correlation. For EIF4A3 there appeared to be no correlation (rho = -0.03) between transcript concentration and protein abundance, whilst ACTB displayed occasional correlation (rho = 0.52) and SYK showed an almost perfect correlation (rho = 0.92). ACE2 transcript and protein abundance correlation appeared to be occasional (rho = 0.45), whilst for TMPRSS2 the correlation was almost perfect (rho = 0.98). Although TMPRSS2 transcript presence and protein abundance appear to correlate better, a clear drawback of these genetic studies is that actual biochemical expression is not determined and mass-spectrometry-based assays do not retain spatial information. High-resolution mapping of three-dimensional (3D) structures in intact tissues is indispensable in many biological studies, yet rarely employed to study viral receptors in their host organs concerning viral infection [16]. The conventional method of histological sectioning followed by the imaging of individual sections is commonly used and valuable, but does not provide spatial information, hampering our advancement of understanding viral infection spatially. Recent developments in whole organ clearing, imaging, and analysis of these large datasets do now allow for the characterization of whole organs [17]. Different animal models have been employed to recapitulate SARS-CoV-2 infection in humans, which are indispensable to develop vaccines and anti-viral therapeutics [18]. Several reviews succinctly compare the advantages and disadvantages of different animal models [19,20]. The Syrian hamster is widely accepted as the small animal model of choice [21–23]. Therefore an increased understanding of receptor expression in this model is of importance to study SARS-CoV-2 infection patterns, especially, since the new Omicron variant predominantly infects the upper respiratory tract and infection in the lung was absent [24–27]. Here, we examined receptor distribution in Syrian hamster lungs and correlated this distribution pattern of ACE2 and TMPRSS2 with the location of infection. Whole lung lobes of SARS-CoV-2 infected Syrian hamsters and control animals were stained using a variety of antibodies for ACE2, TMPRSS2, and the viral nucleoprotein to detect the functional receptor, cellular protease, and location of infection. We observed for ACE2 and TMPRSS2 a variation in protein expression levels within different regions of the lung lobes, with also a variation in protein expression between lung lobes from different Syrian hamsters. The results demonstrate that ACE2 and TMPRSS2 are unequally distributed in the Syrian hamster lung and that mainly overlapping regions were infected by SARS-CoV-2.

Discussion Transcriptomic data of ACE2 and TMPRSS2 expression has been informative, although a direct correlation between mRNA transcripts and protein abundance does not necessarily convey actual biochemical expression as described with the deep proteome analyses, also demonstrated by The Human Cell/Protein Atlas consortium [10] (S1 Fig). Beyond transcript concentration other factors contribute to protein expression, including translation rates influenced by gene codon composition, translation rate modulation through non-coding RNA (microRNA), Half-life modulation, protein synthesis delay, post-translational modifications, and protein transport, also residence time should be taken into account [15]. Therefore, imaging of ACE2, TMPRSS2, and infected cells in whole Syrian hamster lung lobes greatly contribute to our understanding of SARS-CoV-2 pathogenesis. In this study, we demonstrate distinct spatial distribution, or gradient, of ACE2 and TMPRSS2 proteins in the lung lobes of Syrian hamsters, which is considered to be the most relevant small animal model for studying SARS-CoV-2 infection [19,21–23]. In most studies, whole Syrian hamster lungs have been analyzed to determine where SARS-CoV-2 infection occurs, and high viral loads have been detected in nasal and lung turbinate with similar lung histological changes as humans. Virus titers are normally determined from whole organs, and thus fail to give a detailed overview of infection dynamics in a spatial context. Furthermore, in these studies, whole lungs are subjected to tissue slicing, which may induce damage and result in a loss of spatial context when imaged in 2D. We now show the proteinaceous presence of the functional receptor ACE2, and the essential cellular protease TMPRSS2 [33–35], in concert with NP detection. We observed that in regions where ACE2 and TMPRSS2 overlap, actual infection occurs. Although other attachment factors such as sialic acids and heparan sulfate have been reported to be important for SARS-CoV-2 infection dynamics [3–6], NP always co-localizes with ACE2 in our study. Light-sheet microscopy enables imaging of whole organs or large intact tissues in three dimensions. Previous implementations have focused on studying developmental biology or neuroscience [36], whilst this imaging technique also permits the investigation of viral dynamics in a native non-destructive 3D context [32]. Even though the spatial context is lost, tissue sectioning provides valuable information and functions as a complementary technique to light-sheet imaging to assess SARS-CoV-2 infection in relation to ACE2 and TMPRSS2 expression in bronchioles and alveoli. Using whole Syrian hamster heads and lung lobes with ferret trachea and lung lobes, we here provide a blueprint for analyzing infection patterns and receptor expression organization in a whole organ section. This approach will be valuable in identifying and testing new therapeutic agents for SARS-CoV-2. It will also provide opportunities to investigate extrapulmonary infection patterns and the approaches can be extended to other pathogens [37–39]. Finally, we envisage our approach can be applied to circulating variant viruses that are more infectious [40,41]. It can test a possible hypothesis that these viruses can infect the upper respiratory tract where the expression of ACE2 is rather scarce, resulting in increased transmission currently observed for the various variant of concern (VOC) [24]. Increased transmission of Omicron has been attributed to different ACE2 binding properties, differential dependency on cellular proteases [25], and indeed seems to infect upper respiratory tissues with higher efficiency compared to previous VOCs [26,27]. Lung infection in the Syrian hamster model is almost absent [42], with our imaging approach it is possible to characterize the infection pattern of Omicron in lung and URT. This model of adaptation to upper respiratory tract for increased transmission is extremely similar to the model for human influenza A viruses [43–45].

Methods Ethics statement Research was conducted in compliance with the Dutch legislation for the protection of animals used for scientific purposes (2014, implementing EU Directive 2010/63) and other relevant regulations. The licensed establishment where this research was conducted (Erasmus MC) has an approved OLAW Assurance # A5051-01. The research was conducted under a project license from the Dutch competent authority and the study protocol (#17–4312) was approved by the institutional Animal Welfare Body. Animals were handled in a BSL3 biocontainment laboratory. Animals were housed in groups of 2 animals in filter top cages (T3, Techniplast), in Class III isolators allowing social interactions, under controlled conditions of humidity, temperature, and light (12-hour light/12-hour dark cycles). Food and water were available ad libitum. Animals were cared for and monitored (pre-and post-infection) by qualified personnel. The animals were sedated/anesthetized for all invasive procedures. Animal procedures SARS-CoV-2 Female Syrian golden hamsters (Mesocricetus auratus; 6-week-old hamsters from Janvier, France) were anesthetized by chamber induction (5 liters 100% O 2 /min and 3 to 5% isoflurane). Animals were inoculated with 105 TCID50 of SARS-CoV-2 or PBS (mock controls) in a 100 μl volume via the intranasal route. Animals were monitored for general health status and behavior daily and were weighed regularly for the duration of the study (up to 22 days post-inoculation; d.p.i.). Animals were euthanized on day 4 after inoculation, and lung samples were removed and stored in 10% formalin for histopathology. Antibodies The following commercial antibodies were purchased and utilized (S1 Table): anti-ACE2 antibody 2.5 μg/ml (Abcam, ab272690), anti-TMPRSS2 antibody 2.5 μg/ml (Santa Cruz, sc-515727), anti-NP 2.5 μg/ml (Sino-Biological, 40143-MM05), anti-NP 2.5 μg/ml (Sino-biological, 40143-R001), anti-NP 2.5 μg/ml (Thermofisher MA1-7401), anti-K8/K18 2.5 μg/ml (progen, 90001), donkey-anti-rabbit750 5.0 μg/ml (Abcam, ab175731), donkey-anti-GP647 2.5 μg/ml (Jackson immunoresearch, 706-605-148), goat anti-rabbit647 2.5 μg/ml (Thermofisher A-21245), goat-anti-mouse647 2.5 μg/ml (Thermofisher, A-21235), goat-anti-mouse555 2.5 μg/ml (A-21422), goat-anti-human647 2.5 μg/ml (Thermofisher, A-21445), goat-anti-rabbit488 10 μg/ml (Thermofsher, A-21432), goat-anti-human555 2.5 μg/ml (Thermofisher, A-21433). COVA103-1 antibody (10 μg/ml) was used as an anti-NP stain which was obtained from a convalescent patient. Biotinylated recombinant SARS-CoV-2 N proteins were conjugated with a streptavidin fluorophore resulting in fluorescent labeled-probes to stain N-specific B cells for flow cytometry sorting to obtain N-specific monoclonal antibody. The recombinant proteins were conjugated in a 7:1 ratio to the streptavidin-conjugates AF647 (0.5 mg/ml, BioLegend) and BV421 (0.1 mg/ml BioLegend). PBMCs from a convalescent patient were stained for 30 min at 4°C with the conjugated proteins and cell surface markers, as previously described [46]. Live B cells that were double-positive for the SARS-CoV-2 N protein were single-cell sorted using index sorting into a 96-well plate containing lysis buffer, stored at -80°C for at least 1 h before performing the reverse transcriptase (RT)-PCR followed by PCR amplification of the V(D)J variable regions, as previously described [47]. The amplified variable V(D)J-region of the heavy and light chain of the antibody were cloned into correspondingly expression vectors containing the constant regions of the human IgG1 for the heavy or light chain and larger-scale expression of the monoclonal antibody was done using transient transfection of suspension HEK293F cells (Invitrogen, cat no. R79007), as previously described [46]. Tissue staining Sections of formalin-fixed, paraffin-embedded Syrian hamster lungs were obtained from the department of Viroscience, Erasmus University Medical Center, The Netherlands. Tissue sections were rehydrated in a series of alcohol from 100%, 96% to 70%, and lastly in distilled water. Tissue slides were boiled in citrate buffer pH 6.0 for 10 min at 900 kW in a microwave for antigen retrieval and washed in PBS-T three times. Endogenous peroxidase activity was blocked with 1% hydrogen peroxide for 30 min. Tissues were subsequently incubated with 3% BSA in PBS-T overnight at 4°C. The next day, primary antibodies were added to the tissues for 1 h at RT. With rigorous washing steps in between followed by secondary antibodies with either an Alexa488, Alexa555, or Alexa647 fluorescent probe. IDISCO SARS-CoV-2 infected and non-infected Syrian hamster lungs were provided in 10% formalin, for longer storage hamster lungs were kept in PBS and 0.01% sodium azide. Dehydration of the lungs was performed by washing with PBS for 1.5 hours, followed by 50% methanol for 1.5 hours, 80% methanol for 1.5 hours, and as of last 100% methanol for 1.5 hours on a tilting laboratory shaker. Bleaching was performed overnight at 4°C in 90% methanol (100% v/v) and 10% H 2 O 2 (30% v/v). Tissues were rehydrated with 100% methanol for 1 hour, followed by 100% methanol for 1 hour, 80% methanol, 50% methanol, and 1x PBS for 1 hour on a tilting shaker. Syrian hamster lungs were subsequently blocked for 24 hours at room temperature (22°C) in 1x PBS with 0.2% gelatin, 0.5% triton-x-100, and 0.01% sodium azide (PBSGT) on a tilting shaker. Hamster lungs were stained with primary/secondary antibodies possessing an alexa555, alexa647, or alexa750 dye. After blocking samples were incubated for 120 hours with the primary antibody in PBSGT with 0.1% saponin (S2149, Sigma-Aldrich) on a shaking incubator at 200 rpm, following 120 hours incubation, hamster lungs were washed with PBSGT six times for 1 hour on a tilting shaker. Post-PBSGT wash, samples stained with primary antibodies were incubated with secondary antibodies that possess alexa555, alexa647 or alexa750 dyes, the antibodies were diluted in PBSGT with 0.1% saponin and filtered with 0.45 μm filter for potential antibody aggregates. Hereafter samples were incubated for 120 hours at room temperature on a shaking incubator at 200 rpm. After staining washing was performed with PBSGT six times for 1 hour on a tilting shaker. Immunostained samples were subsequently treated with 50% methanol for 24 hours, 80% methanol for 24 hours, 100% methanol for 24 hours followed by 100% methanol for 24 hours on a tilting shaker for dehydration of the lungs. Hereafter lipid solubilization was performed with dichloromethane (Thermofisher, 402152) for 40 minutes after which refractive index matching and optical clearing were performed overnight with dibenzyl ether (108014, Sigma-Aldrich). Light-sheet microscopy Light-sheet imaging was performed using the Ultramicroscope II (LaVision BioTec) equipped with an MVX-10 Zoom Body. The laser lines 488 nm (Coherent OBIS 488–50 LX Laser 50mW), 561 nm (Coherent OBIS 561–100 LS Laser (100mW), 647 nm (Coherent OBIS 647–120 LX Laser 120mW), and 730 nm (Coherent OBIS 730–30 LX Laser 30mW) were used with 70% laser power for 488 channel, 70% laser power for the 555 channel, 70% laser power for the 647 channel and 100% laser power for the 730 nm channel. Emission filters ET525/50 (488 channel), ET615/40m (561 channel), ET676/29 (647 channel) and 716/40 (730 channel) filter was used with a Neo 5.5 sCMOS detector (2560x2160 pixels, pixel size: 6.5 x 6.5 μm2). A corrected dipping cap (CDC) (1.26x – 12.6x) was utilized with an Olympus MVPLAPO 2x objective lens. Imaging was performed with 0.63 zoom and step size of 5 μm (voxel resolution X, Y, Z: 4.79 μm, 4.79 μm, 5 μm), also with 6.3 zoom and step size of 2 μm (voxel resolution X, Y, Z: 0.48 μm, 0.48 μm, 2 μm). The exposure time was set to 200 ms and illumination for the 0.63 zoom was bidirectional and 6.3 zoom unidirectional. The sheet numerical aperture was 0.033 with a sheet thickness of 7.09 μm and a sheet width of 40%. The imaging chamber was filled with dibenzyl ether and the microscope software is Inspector version 7.1. Obtained image stacks were analyzed with ImageJ v1.54f and Imaris 9.7. Stills of image slices (orthoslices) were generated in ImageJ/Imaris, whilst 3D renders were generated in Imaris. TIFF files generated by the light-sheet microscope were imported to ImageJ with “File -> Import -> Bio-formats”, the stack is viewed as “Hyperstack” color mode was set to “Composite” and “Display metadata” with “Display OME-XML metadata” was enabled to obtain voxel information. Brightness adjustments were performed for each channel with setting “Image -> Adjust -> Brightness/Contrast”. The 488 channel color was set to hex code #FFFFFF, the 647 channel was set to hex code #FF0000 and the 730 channel was set to hex code #8080FF. Snapshots of the Z-stack were saved as JPEGs and the orthoslice animations with “Save As -> Avi -> Compression JPEG and Frame Rate 30 FPS”. For the 3D renders obtained imaging data was imported into Imaris, display adjustments were made for each channel. Snapshots and animations were made with 3D renders (3D View) within Imaris. Volume mode was set to maximum intensity projection (MIP) and the rendering quality was set to highest. Frame settings were used for the outer bounding box of the 3D render, “Box”, “Tickmarks”, and “Axis labels” were enabled. The spacing of the tickmarks was set to μm 500 for the X, Y, and Z-axis. The 488 channel color was set to hex code #FFFFFF, the 647 channel was set to hex code #FF0000 and the 730 channel was set to hex code #8080FF. For the triple stains, the 488 channel color was set to hex code #FFFFFF, the 561 channel color was set to hex code #0000FF, the 647 channel was set to hex code #FF0000 and the 730 channel was set to #00FF00. The 3D animations were made with 1920 x 1080 resolution (16:9) and 360 frames, 360° horizontal turn, and exported using Adobe Media Encoder 2020 or Adobe Premiere Pro 2020. Quantification For TMPRSS2 and ACE2 quantification surfaces and masks were generated in Imaris with the “Surfaces Creation Wizard” (according to the reference manual). In this surface/mask regions of interest were selected, i.e. the branches and alveoli in the hamster lung. Regions of interest were selected in the surface tool with the corresponding source channel, whereby the 647 channel with TMPRSS2 signal was used for TMPRSS2 and ACE2 double-stained samples and 730 source channel for ACE2 and anti-NP double-stained samples. For the TMPRSS2 single stained sample, the 647 channel was used as the source channel. Thresholding (Absolute Intensity) was performed arbitrarily to obtain complete coverage of the alveoli and branches. The automatically provided value for “Sphere Diameter” was used. For the separation of two or more objects that are identified as one, the “Split touching Objects (Region Growing)” setting was utilized. The “Seed Points Diameter” value was set to 35 μm. The generated surfaces were subsequently used to mask the TMPRSS2 and ACE2 channels. The surfaces are split into multiple smaller regions within the surface, due to the setting “Seed Points Diameter”, hereafter all statistical values from these multiple smaller regions can be obtained and exported to GraphPad Prism 9 after which a grouped analysis was performed for statistical significance.

Acknowledgments The authors thank the MIND facility of the UMC Utrecht Brain Center for support with iDISCO. The authors wish to thank Gestur Vidarsson and Federica Linty of Sanquin, Amsterdam, the Netherlands for providing the SARS-CoV-2 Nucleocapsid protein. Fig 1A created with BioRender.com.

[END]

[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1010340

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.