Reader Comments

Post a new comment on this article

Referee comments: Referee 1

Posted by PLOS_ONE_Group on 11 Jan 2008 at 16:46 GMT

Referee 1's review:

The Allhorn et al manuscript presents additional studies of an interesting IgG glycan hydolase, EndoS, an enzyme produced by group A streptococcus. Previous studies have shown the specificity of endoS for cleavage of IgG N linked glycans. A role for this enzyme in the virulence of Group A streptococcus was also previously suggested by it destruction of opsonic antibody directed at M protein. This work extends those previous experiments by demonstrating that endoS acts on all four subclasses of IgG, releases IgG bound to FcγRIIa &b and alters the capacity of IgG to bind to these receptors. They also show that an enzymatically inactive form of endoS, E235Q, forms stable complexes with IgG. They argue the utility of this enzyme for treatment of various immune complex diseases and for use as a biochemical reagent.

This is a thorough and careful study of the interactions between endoS and IgG, however most of the results are predictable and incremental in nature. For example one would expect that endoS rapidly dissociates from its substrate after hydrolysis and that mutation of the active site would likely stabilize the complex.

This reviewer is not familiar with molecular interactions between IgG and Fc receptors, but it seems somewhat surprising that hydrolysis of the glycan of immunoglobulin bound to receptors would release IgG from the receptor. Those results suggest that the glycan is the primary determinant of receptor binding and residence time on the various receptors.
It was curious that endoS treatment increased binding of IgG2 to FcγRIIa &b, the opposite of what was observed for the other subclasses. This anomaly was not discussed, nor were explanations provided.

It was suggested that glycan hydrolysis abrogates activation of phagocytes; however, experiments is support of this conclusion were absent.

The discussion goes overboard in promotion of endoS as a possible therapeutic agent, and reagent for research. A clear picture and discussion of how and where the enzyme functions during infection would be more interesting and instructive.

N.B. These are the general comments made by the reviewer when reviewing this paper in light of which the manuscript was revised. Specific points addressed during revision of the paper are not shown.

RE: Referee comments: Referee 1

mariaallhorn replied to PLOS_ONE_Group on 14 Jan 2008 at 09:28 GMT

1. We agree with the reviewers that some, but definitely not most, of our
findings are predictable and incremental! For instance, the reviewer states that
the finding that EndoS rapidly dissociates from its substrate and mutation of the
active site would likely stabilize the complex is predictable. By studying the
interaction between both the active and inactive form of EndoS and IgG we
reveal that EndoS does not interact with IgG lacking most of its glycan. This
together with our previous findings that EndoS only interacts with native IgG
indicates a complex interaction between EndoS and IgG that requires both
protein-carbohydrate interactions and protein-protein interactions.

2. The reviewer is surprised that EndoS releases IgG from Fc-receptors (FcR).
So were we, since to our knowledge this is the first described example of IgG
release from FcRs by enzymatic glycan hydrolysis! On the other hand it is not
totally surprising since others have shown that the chitobiose core, that is
hydrolyzed by EndoS, is essential for IgG/FcR interactions.