key: cord-0480272-2vuiafv0
authors: Schifanella, L.; Anderson, J. L.; Galli, M.; Corbellino, M.; Lai, A.; Wieking, G.; Grzywacz, B.; Klatt, N. R.; Haase, A. T.; Schacker, T. W.
title: Massive viral replication and cytopathic effects in early COVID-19 pneumonia
date: 2020-04-30
journal: nan
DOI: nan
sha: ec19ee2b25f9c4ac8b6852fde1cf4bfa52a88687
doc_id: 480272
cord_uid: 2vuiafv0

SARS-CoV-2 is the cause of COVID-19 acute respiratory illness that like its predecessors, MERS and SARS, can be severe and fatal 1-4. By April of 2020, COVID-19 infections had become a worldwide pandemic with nearly 3 million infections and over 200,000 deaths. The relative contributions of virus replication and cytopathic effects or immunopathological host responses to the severe and fatal outcomes of COVID-19 lung infections have as yet to be determined. Here we show that SARS-CoV-2 replication and cytopathic effects in type II alveolar pneumocytes causes focal lung injury in an individual with no history of pulmonary symptoms. These findings point to the potential benefit of early effective antiviral treatment to prevent progression to severe and fatal COVID-19 pneumonia.

symptoms. These findings point to the potential benefit of early effective antiviral treatment to prevent progression to severe and fatal COVID-19 pneumonia.

Understanding the mechanisms responsible for the COVID-19 fatal lung infections is important to design potentially life-saving treatments. The virus itself, an immunopathological response to infection or a combination of these direct and indirect mechanisms could be responsible for lung injury. The distinction is important, because antiviral drugs and antibodies would be appropriate to treat lung injury largely due to virus replication and cytopathic effects if direct virus insult is the major driver of lung pathology. Clinical trials underway to test antiviral drugs are designed to determine if the inhibition of SARS-CoV-2 replication can prevent or mitigate the pathological consequences of lung infection and thus far these trials are aimed at people with symptomatic and often advanced disease. On the other hand, if severe pneumonia is due to an immunopathological response to the virus, then moderating that response, e.g., by inhibiting IL-6 with the monoclonal antibody tocilizumab or other immunomodulatory approach, would be appropriate, but with the risk, as in any immunosuppressive treatment, of impairing the immune response to infection.

Ground glass opacities or patchy infiltrates in the lungs in CT images of asymptomatic SARS-CoV-2 infected individuals [5] [6] [7] suggest that lung infection and associated tissue injury may be detectable in individuals who did not have severe pneumonia or respiratory failure 7 . Studies of tissues at this stage might therefore provide a glimpse of the relationship between viral replication and cytopathic effects to lung injury before that relationship becomes blurred by progression to end stage pneumonia. We had an opportunity to test this hypothesis in analysis of lung specimens obtained postmortem from an 88 year-old Italian woman who succumbed to a heart attack while quarantined in a hotel after testing positive for SARS-CoV-2. She is described as afebrile and in good health without cough, shortness of breath or any other symptoms of clinically manifest SARS-CoV-2 infection. We describe findings in the lung of this individual at a stage of infection without manifest pneumonia or respiratory failure that support direct virusmediated tissue destruction in the pathogenesis of COVID-19 pneumonia.

We received paraffin embedded lung specimens from the SARS-CoV-2 infected subject of this report from the Department of Pathology of Fatebenefratelli Sacco Hospital, Lombardy, Milan, Italy. H&E stained lung tissue showed heterogeneous organ damage (Extended Data Fig.1 ). Alveolar structure was partially conserved, although emphysematous changes were clearly recognizable in the region defined by the blue box and arrow (Extended Data Fig.1 ). In the region defined by the red box and arrow, there is a very well delimited area of striking consolidation. At higher magnification there is evidence of diffuse alveolar damage with exudative characteristics that include hyaline membranes, inflammatory cell infiltrates (mainly lymphocytes), desquamated type II pneumocytes, microvascular thrombosis, and vascular endothelial cell damage with red blood cells in the alveolar spaces. We focused our analysis on this region of lung injury. We used RNAscope in situ hybridization (ISH) with anti-sense SN probes to detect SARS-CoV-2-specific Spike (S) region genomic sequences, and nucleocapsid (N) region genomic sequences shared with SARS-and MERS-CoV, to detect viral genomic RNA+ (vRNA+) in infected cells. We first show images consistent with spread of infection into the lung by infection of the bronchial epithelium (Fig. 1) . Degenerating vRNA+ cells line a small and larger bronchiole (Fig. 1A ,B, short arrows) and a grape-like cluster of vRNA+ hyperplastic epithelium we later show to be type II pneumocytes is adjacent to infected bronchiolar epithelium, consistent with this mode of spread into alveoli. There are also vRNA+ clusters adjacent to the alveolar space (Fig. 1A, long arrows) , vRNA released from lysed type II pneumocytes (Fig. 1C , red arrow) and CD68+ macrophages that appear to have phagocytosed vRNA+ cells and debris (Fig. 1C brown and red arrows) . Detection of SARS-CoV-2 RNA was specific because 1) there was no signal when the SN probe was hybridized to normal lung tissue (Extended Data Fig.2A) ; 2) a negative control probe did not detect SARS-CoV-2 RNA 

We identified type II pneumocytes as the principal cells in which SARS-CoV-2 replicates and produces virus. To do so, we combined ISH with Napsin A staining to detect vRNA in Napsin A+ type II pneumocytes and ISH with TSA/ELF amplification 8 to render RNA in virions visible for light microscopy to score for virus producing cells. With these two methods, we could then examine the spread of infection and cytopathic effects in the lung. In the numbered regions in the overview of infection and spread (Fig. 2) , there are foci or collections of vRNA+/Napsin A+ type II pneumocytes in close spatial proximity to one another and to vRNA negative Napsin A+ type II pneumocytes (Fig. 2, regions 1-5) . These images suggest a chain of transmission events from an infected cell to adjacent susceptible cells. We will return to the role of macrophages in phagocytosis of infected cells and virus released from dying cells but note here that a RNA+/Napsin A-non epithelial cell overlies extracellular viral RNA+ derived from lysis of infected cells (Fig.2 region 6 ) and a conjugate of viral RNA+/Napsin A+ and RNA+/Napsin A-non epithelial cell (Fig.2. region 7) . Both images are consistent with subsequent documentation of macrophage acquisition of viral RNA by phagocytosis.

The visualization of virus production, spread and cytopathic effects in the lung confirmed the predominant focal nature of spread (Fig. 3A) in clusters or chains of susceptible cells in close spatial proximity (Fig. 3B) . The image at higher magnification (Fig. 3C ) of individual virions, virus aggregates and inclusion bodies in and around dying cells captures the massive production of virus and associated cytopathic effects. This extraordinary level of virus production, fusion and cell lysis leaves a visible record in the lung of vast swaths of lysed virus + cells surrounding airways, which are themselves lined by fused virus + cells, and around rare still recognizable virus producing cells (Fig. 3D) as well as mats of virions from necrotic type II pneumocytes (Fig. 3E) . These images thus support the mode of spread into and within the lung and underscore the extent of tissue injury directly attributable to SARS-CoV-2 replication and cytopathic effects.

With these images of the massive cytopathic effects and cell lysis in mind, we return to the question of whether vRNA+ macrophages represent cells in which SARS-CoV-2 is replicating or macrophages that acquire vRNA by phagocytosis. We have already shown images that suggested that vRNA+ macrophages could acquire vRNA by phagocytosis of infected type II pneumocytes or vRNA released from lysed cells (Fig. 1, 2) , and now show further evidence in support of that conclusion. In the region enclosed by a rectangle adjacent to vRNA+ lysed epithelium (Fig. 4A) shown at higher magnification (Fig. 4B) , a CD68+ vRNA+ contacts lysed vRNA+ lysed epithelium. The white line outlines the RNA+ positive area of the macrophage that is in contact with the viral RNA in the necrotic cells, suggesting phagocytic acquisition of vRNA. Small CD68+vRNA-negative cells are also in cell-to-cell contact with CD68-negative vRNA+ type II pneumocytes, consistent with a sequence in which macrophages will acquire RNA from dying infected cells. Moreover, macrophages with little vRNA also overly sheets of lysed vRNA+ cells consistent with acquisition of vRNA from infected pneumocytes at an advanced stage of infection in these cells (Fig. 4C, D) . This conclusion is further supported by the co-localization of vRNA inclusions in LAMP1+ lysosomes in CD68+ cells in some cases within CD68+ extensions of a macrophage (Fig. 4E, F) .

We investigated what local production of IL-6 might contribute to the inflammatory component of lung injury by staining tissues for IL-6. The morphology and geographical distribution of the IL-6 + cells suggested that the IL-6+ cells were hyperplastic infected type II pneumocytes (Extended Data Fig.3A) , and we confirmed their identity by showing that the IL-6+ cells were vRNA+ and Napsin A+ (Extended Data Fig.3B) . While IL-6 production by infected cells could contribute to immunopathology, local production could also be involved in macrophage recruitment and epithelial reparative processes 9-13 .

We show in a SARS-CoV-2 infected individual with ostensibly early infection, who had no known history of pulmonary symptoms, that there was a region of the lung with focal pneumonia in which massive SARS-CoV-2 replication and cytopathic effects in type II alveolar pneumocytes directly contributes to lung pathology. The extent of diffuse alveolar damage from the lytic effects of infection of type II pneumocytes could explain the current observation that many people testing positive for SARS-CoV-2 have significant hypoxia often in the absence of overt symptoms of pneumonia. It might therefore be prudent in this stealth phase of SARS-CoV-2 lung infection to use pulse oximetry to measure blood oxygen levels as a guide to early supplemental oxygen. Our findings further point to the potential therapeutic benefit of early treatment to inhibit viral replication to prevent progression to severe and fatal pneumonia. 

The RNAscope 2.5 Brown detection kit used the same method as above, except the addition of the Pretreat 1 (H2O2) step. Following the amp 6 step, Opal 570 (Perkin Elmer) was added following company instructions. IL6/CD68 and Napsin A (Biocare Medical) were added overnight. The following day, Donkey anti-rabbit Alexa Fluor 488 and Donkey anti-mouse Alexa Fluor 647 was added for 45 minutes before adding DAPI and mounting in Aqua polymount.

Antigen retrieval was performed using 0.05% citraconic anhydride for 30 sec at 122 degrees using a declocker (Biocare Medical). Slides with 5 micron sections were cooled, and sections were circled with a hydrophobic pen (vector laboratories ImmPact DAB (Vector Laboratories) was added before counterstaining with CAT Hematoxlyin (Biocare Medical) and bluing in TBST. Sides were dehydrated and mounted in permount. 

Forty years with coronaviruses

Identification of a novel coronavirus in patients with severe acute respiratory syndrome

Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia

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

Asymptomatic and Presymptomatic SARS-CoV-2 Infections in Residents of a Long-Term Care Skilled Nursing Facility

Estimating the asymptomatic proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship

Clinical characteristics of 24 asymptomatic infections with COVID-19 screened among close contacts in Nanjing

Defining HIV and SIV Reservoirs in Lymphoid Tissues

Roles of substrate availability and infection of resting and activated

CD4+ T cells in transmission and acute simian immunodeficiency virus infection

The authors declare no competing interests.

Material and Correspondence should be addressed to: