Accumulation Dynamics of Defective Genomes during Experimental Evolution of Two Betacoronaviruses

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, Hillung and Olmo-Uceda et al. investigate the dynamics of defective viral genomes (DVGs) in betacoronaviruses. By performing "long-term" experimental evolution of human coronavirus OC43 (HCoV-OC43) and murine hepatitis virus (MHV) in tissue culture at different MOI, the researchers aimed to understand the accumulation dynamics of DVGs. Using RNA-seq, they observed that DVGs evolved to show greater diversity and abundance, with deletions and insertions being the most common types. Interestingly, certain deletions were transient, while others became prevalent over time, suggesting a dynamic interplay between viral replication and DVG accumulation.

In this context, RNA-seq was used to reconstruct DVGs and analyze their accumulation dynamics over time. The study revealed variations in DVG abundance and size distribution between different MOI conditions and virus types, suggesting that DVG formation during betacoronavirus evolution is driven by different factors, including MOI and virus-specific characteristics. In addition, hotspot regions for deletions were identified, particularly within cistrons encoding structural and accessory proteins, shedding light on the potential impact of DVGs on viral replication and evolution.

I believe this paper paper not only meets but exceeds the scope and standards for publication in Viruses.

I have only minor, truly minor, comments:

 

Part of the results could be only in the discussion section, for example:

3.1. Experimental evolution of betacoronavirus DVGs populations

 

An overall decrease in viral load in most cases could be explained by two alternative hypotheses. Firstly, it could be attributed to the progressive accumulation of DVGs that interfere with the replication of wild-type viruses [6,14,43]. This effect would be stronger at high MOI, which seems not to be a general case in Figure 1. Secondly, it could be a consequence of strong transmission bottlenecks that turns on Muller’s ratchet, leading to 265 the fixation of deleterious mutations and, consequently, fitness declines, as well established for other RNA viruses [44-4818].

 

There are some titles of the results section that could be improved, for example by stating the main message of each section, if there is a clear one: 

3.2. Evolution of DVG richness and abundance + "is cell type and virus dependent"?

Author Response

Q.1.1. Part of the results could be only in the discussion section, for example:

3.1. Experimental evolution of betacoronavirus DVGs populations

An overall decrease in viral load in most cases could be explained by two alternative hypotheses. Firstly, it could be attributed to the progressive accumulation of DVGs that interfere with the replication of wild-type viruses [6,14,43]. This effect would be stronger at high MOI, which seems not to be a general case in Figure 1. Secondly, it could be a consequence of strong transmission bottlenecks that turns on Muller’s ratchet, leading to 265 the fixation of deleterious mutations and, consequently, fitness declines, as well established for other RNA viruses [44-48].

R.1.1.  We have moved this paragraph to the Discussion as recommended and placed it into the broader discussion about the role of DVGs and drift in determining the fate of viral alleles.

 

Q.1.2. There are some titles of the results section that could be improved, for example by stating the main message of each section, if there is a clear one:

3.2. Evolution of DVG richness and abundance + "is cell type and virus dependent"?

R.1.2.  Thank you for this recommendation.  We have modified all the subsection titles to make them more informative.

Reviewer 2 Report

Comments and Suggestions for Authors

This is a clear, complete, concise, elegant and well conducted study, on the dynamics of defective viral genomes (DVGs) of two betacoronaviruses (OC43 and MHV) in cell culture using high and low m.o.i. passages. Deleted genomes exhibited differences in size distribution both between low and high MOIs and among different viruses. The manuscript is well-written and clearly presented, and the discussion is very interesting.

Comments:

OC43 is cultured in BHK-21 and HCT-8, while MHV in CCL-9.1 murine liver cells. Is it possible that the cell line used might be involved in the differences of DVGs observed between OC43 and CCL-9.1?

Frequently, a high MOI is obtained when one cell is infected by more than one virus, typically 10/100 virus per cell. Maybe I have missed something, but I don’t understand why authors are talking about High MOI for MHV and they are using 6.96x10^-3 or 1x10^-2 which it looks like there is less than 1 virus per cell (Table S1). Could the authors please clarify this point?

Page 6 Section 3.1 Lines 1-2: “Viral load was evaluated each passage and dilutions adjusted to ensure that,” should be corrected by “Viral load was evaluated at each passage, and dilutions were adjusted to ensure that,”.

Author Response

Q.2.1. OC43 is cultured in BHK-21 and HCT-8, while MHV in CCL-9.1 murine liver cells. Is it possible that the cell line used might be involved in the differences of DVGs observed between OC43 and CCL-9.1?

R.2.1.  This is absolutely the case!  Indeed, this is one of the take home messages of our manuscript, as highlighted in several places along the manuscript, including the final remark in the Abstract.  To further stress out the importance of cell type in the accumulation patterns of DVGs, we have reworded the title of section 3.2 and expanded the conclusions in section 5.

 

Q.2.2. Frequently, a high MOI is obtained when one cell is infected by more than one virus, typically 10/100 virus per cell. Maybe I have missed something, but I don’t understand why authors are talking about High MOI for MHV and they are using 6.96x10-3 or 1x10-2 which it looks like there is less than 1 virus per cell (Table S1). Could the authors please clarify this point?

R.2.2.  We fully understand this concern.

High MOI commonly refers to values of > 10 particles/cell, as correctly pointed by the reviewer.  In our case, “high” and “low” are relative to each other.  As indicated in section 3.1, we refer to “high MOI” to values that are between 10- to 20-fold larger than what we refer as “low MOI”.  Particularly in the case of MHV in CCL-9.1 cells, our “high MOI” regime started infections with one particle every 100 – 1000 cells, values that many readers would qualify as low MOI.  However, we would like to stress out that since the relevant comparison is with the “low MOI” treatment for the same virus and cell type, the magnitude of the difference between both treatments is still in the order of 10-fold higher for the “high MOI” treatment.  Notice this is consistent across all three experimental treatments.

The first paragraph of section 3.1 has been slightly rewritten to make this point clearer.

 

Q.2.3. Page 6 Section 3.1 Lines 1-2: “Viral load was evaluated each passage and dilutions adjusted to ensure that,” should be corrected by “Viral load was evaluated at each passage, and dilutions were adjusted to ensure that,”.

R.2.3.  We have rewritten as suggested.  Thank you.