Review 1: "β-Coronaviruses use lysosomal organelles for cellular egress"

RR:C19 Evidence Scale rating by reviewer:

• Misleading. Serious flaws and errors in the methods and data render the study conclusions misinformative. The results and conclusions of the ideal study are at least as likely to conclude the opposite of its results and conclusions than agree. Decision-makers should not consider this evidence in any decision.

***************************************

Review:

This is a review of Figure 1 of β-Coronaviruses use lysosomal organelles for cellular egress S Ghosh, TA Dellibovi-Ragheb, E Pak, Q Qiu, M Fisher, PM Takvorian, C Bleck, V Hsu, AR Fehr, S Perlman, SR Achar, MR Straus, GR Whittaker, CAM de Haan, G Altan- Bonnet, N Altan-Bonnet bioRxiv 2020.07.25.192310; doi: https://doi.org/10.1101/2020.07.25.192310 

I understand that it is unusual to review only a single figure of a “preliminary report that has not been peer-reviewed,” but I believe that all will agree that review of this one figure will be a sufficient evaluation of the paper. 

Here is the description of Figure 1A: “(A) Kinetics of MHV replication and release from HeLa-mCC1a cells. Viral genomic RNA quantified in cell lysates and extracellular medium with QPCR and plotted as fold increase over mock-infection. Experiments done in triplicates. Error bars are SD.” 

The numbers on the axes are peculiar, and there are no error bars in Figure 1A. 

Here is the description of Figure 1B: “(B) Impact of Brefeldin A (5μgr/ml) treatment between 8hr and 14hr pi on MHV release (black bar) or Gaussia Luciferase release (pink bar). Extracellular viral genomic RNA was quantified in extracellular medium with QPCR and plotted as fold increase over mock-infection. Experiments done in triplicates. Error bars are SD.” Please note that there is no mention of how the luciferase was measured in the legend or anywhere else in the article or what “au” (“arbitrary units?”—when did that become acceptable?) means. Most importantly, the data on “extracellular genomic vRNA” are represented on a logarithmic scale, whereas the data on “Extracellular Gaussia” (can’t the authors be bothered to use a proper descriptive label?) are on a linear scale. This is an unacceptable and fraudulent practice that completely misrepresents the data. These are the “results” that prompt all of the later experiments in the article. There are, in fact, no results that justify the statement “Remarkably, the quantity of virus released in the presence of BFA during egress was almost identical to that in its absence ... On the other hand, the secretion of Gaussia Luciferase ... was completely blocked when BFA was present during infection (Figure 1B pink bars). 

Here is the description of Figure 1D: “(D) HeLa-mCC1a cells infected with MHV, fixed at 6,9,12hr pi and coimmunostained with anti-LAMP1 (green) and anti-MHV (J1.3) (red) antibodies. Arrows point to lysosomes containing M. Scale bar 5μm.” 

In fact, the arrows in the magnified “12hr pi” image (it is not easy to see how the image corresponds to the boxed region of the unmagnified image from which it is supposed to originate) appear to point to random overlap of the (mostly distinct) red and green signals. If these are the data that underly the analysis shown in Figure 1G (“(G) Frequency of LAMP1 positive organelles in MHV-infected HeLa-mCC1a cells (n=16) at 12hr pi. In each cell, lysosomes within a 100μm2 region of interest were scored for presence or absence of MHV and plotted. Error bars are SE.”—no information on how the “scoring” was conducted is presented in the article), then the analysis is worthless. 

The article is misleading and should be rejected. An investigation of how sixteen authors allowed this to be published as a preprint is warranted.