PREreview of Deep mutational scanning reveals functional constraints and antigenic variability of Lassa virus glycoprotein complex

In this study, the authors performed deep mutational scanning with lentiviruses pseudotyped with Lassa virus GPC to evaluate its mutational effects on cell entry and antibody escape. Separate assays with select mutations validated results from the deep mutational scanning. Mutational tolerance for cell entry correlated with the frequency observed in natural Lassa viruses. In addition, the antibody escape map demonstrated Lassa virus strains that possibly escape currently available antibodies. The antibody-escaping mutations were further validated within the context of the natural GPC.

Despite its high annual cases and possible profound impact on global public health, the Lassa virus has been poorly studied. This study offers a valuable resource for Lassa virus studies and therapeutics. Deep mutational scanning and validation assays are carefully designed with appropriate controls, and the results are thoroughly described and interpreted. Listed below are some minor comments on the study.

Minor comments

1. What are the criteria for choosing mutations for further validation in Figures 2D and 3D? They should be described in the manuscript. 

2. How do you estimate the extent of mutations that influence cell entry or antibody escape through the impact on the cellular expression level, not the amino-acid sequence, like nonsense mutations? It would be helpful for some readers to interpret the results. For example, mutations within the same amino-acid codon would be compared in the entry or escape scores, expanding Figure S6A.

3. Given that this group has recently achieved deep mutational scanning for HIV and SARS-CoV-2 surface protein, how different are the results of the three viruses, particularly regarding the degree of mutational tolerance? Although the hallmark of the Lassa virus is its high variance between strains, it remains unknown whether the high variance is attributed to the Lassa viruses' sequence or other evolutionary processes. The authors may mention this in the Discussion section.

4. The time interval between monoclonal therapy and the emergence of inactivating strain is an important therapeutic measure. Can this dataset give any insights into this? For example, can the time of observation of Lassa strains with an escape mutation be compared with the time of collection of the monoclonal antibody that the mutation escapes?

5. Although the authors mention in the Results section that their experiments do not assess cell-type-specific factors, this should also be mentioned in the Limitation of the Study section.

6. The global map for antibody escape mutations, shown on the website (https://dms-vep.org/LASV_Josiah_GP_DMS/htmls/89F_mut_effect.html), shows that the vast majority are non-measured mutations. To help readers understand the comprehensiveness of antibody-escape mutation maps, the number of mutations with actual measurement and the reason why many non-measured mutations should be described.

7. The “Pseudovirus deep mutational scanning of GPC” part in the Result section describes that the mutant library “covered ~99% of all 9,820 possible amino-acid mutations to GPC.” It might be helpful for some readers to state that these are all possible single amino-acid mutations.

Competing interests

The authors declare that they have no competing interests.