The intervening domain is required for DNA-binding and functional identity of plant MADS transcription factors

The MADS transcription factors (TF) are an ancient eukaryotic protein family. In plants, the family is divided into two main lineages. Here, we demonstrate that DNA binding in both lineages absolutely requires a short amino acid sequence C-terminal to the MADS domain (M domain) called the Intervening domain (I domain) that was previously defined only in type II lineage MADS. Structural elucidation of the MI domains from the floral regulator, SEPALLATA3 (SEP3), shows a conserved fold with the I domain acting to stabilise the M domain. Using the floral organ identity MADS TFs, SEP3, APETALA1 (AP1) and AGAMOUS (AG), domain swapping demonstrate that the I domain alters genome-wide DNA-binding specificity and dimerisation specificity. Introducing AG carrying the I domain of AP1 in the Arabidopsis ap1 mutant resulted in strong complementation and restoration of first and second whorl organs. Taken together, these data demonstrate that the I domain acts as an integral part of the DNA-binding domain and significantly contributes to the functional identity of the MADS TF.

In the abstract, the sentence "Here, we demonstrate that DNA binding in both lineages absolutely requires a short amino acid sequence C-terminal to the M domain called the Intervening domain (I domain) in type II MADS." May cause ambiguity, I suggest to change as "Here, we demonstrate that DNA binding in both lineages absolutely requires a short amino acid sequence C-terminal to the M domain called the Intervening domain (I domain) that was previously defined in type II MADS." Some minor points: 1) The mutant protein expression and characterization may need (not absolutely required) to be checked by CD (circular dichroism) spectrometry to ensure their structural integrity; 2) Crystallographic data and structure determination: 2.1A is a rather high resolution, but in the highest resolution shell (2.16-2.10) 130% is too high for Rmerge, please also list I/sig(I) for judgement of overall signal to noise ratio. Statistics will look better by cutting off the resolution a little, and will not lose any information. The structural determination and refinement have been described too simple or even omitted, please add anything necessary to validify the structural determination and analysis.
3) The structural labeling is also much too simplified, particularly for untrained eyes, please label names for helices and strands, particularly please label N-and C-termini. Please also label domain names, e.g. Idomain, where is it? 4) Without a DNA-TF complex structure, it's rather difficult to explain why the I-domain is so much important for DNA binding (and binding specificities), and PPIs, particularly as the authors claimed the Idomain is not even touching the DNA. Some vague arguments about allosteric change are rather weak in this case. Although this is a non-structure focused paper, some DNA-protein modeling including Idomain may be necessary. The following is some complex structural examples, mentioned by the authors in the paper.
(A PDF file containing the figures will be attached ...)

MADS review: June 15-22, 2021
The Manuscript NCOMMS-21-20994-T entitled "The Intervening Domain Is Required For DNA-binding and Functional Identity of Plant MADS Transcription Factors" by Drs. Lai … et al. and Zubieta has clearly identified and confirmed in depth the I-domain (Intervening Domain) that has been defined previously in type II plant MADS transcription factors (TF) exists also in type I plant MADS, and the I-domain is important for its function as DNA-binding and protein-protein interactions (PPI) by a wide range of techniques including biochemical and structural (crystallography) methods; HTP sequencing Seq-DAP-seq, and CHiP-seq; and Yeast 2-hybrid experiments; particularly genetic domain-swapping of I-domains among different MADS TFs SEP3-AG and SEP3-AGIAP1 in planta. These work and results are of importance for deep understanding of the MADS TF family which is a conserved TF superfamily in eukaryotic organisms ranging from yeasts, plant and human. Thus, it deserves publication in a good journal such as NC.
Good words said as above, I do have some critics for some content and technique details: In the abstract, the sentence "Here, we demonstrate that DNA binding in both lineages absolutely requires a short amino acid sequence C-terminal to the M domain called the Intervening domain (I domain) in type II MADS." May cause ambiguity, I suggest to change as "Here, we demonstrate that DNA binding in both lineages absolutely requires a short amino acid sequence C-terminal to the M domain called the Intervening domain (I domain) that was previously defined in type II MADS." As suggested by the reviewer, we have made this change. We have also rewritten the abstract to be 165 words, as required.
Some minor points: 1) The mutant protein expression and characterization may need (not absolutely required) to be checked by CD (circular dichroism) spectrometry to ensure their structural integrity; We used thermal shift assays to provide an overview of structural integrity and indeed it is highly variable depending on the mutation in the I domain. This is expected as the secondary structural elements exhibit many different interactions-hydrophobic, salt bridges, hydrogen bonding as described in the structure. CD would likely show a similar trend, although we are most interested in the quaternary structure which is efficiently monitored by thermal shift assay.
2) Crystallographic data and structure determination: 2.1A is a rather high resolution, but in the highest resolution shell (2.16-2.10) 130% is too high for Rmerge, please also list I/sig(I) for judgement of overall signal to noise ratio. Statistics will look better by cutting off the resolution a little, and will not lose any information. The structural determination and refinement have been described too simple or even omitted, please add anything necessary to validify the structural determination and analysis.
We used CC1/2 as the main criteria for determining resolution limits as this criterion is replacing Rmerge for high resolution limit determination. See, for example, Karplus and Diedrichs, "Assessing and maximizing data quality in macromolecualr crystallography," Current Opinion in Structural Biology. 2015 October;34:60-68. doi:10.1016/j.sbi.2015 We have added details to the materials and methods section as to the data analysis, molecular replacement program and the refinement programs used for the structure building and analysis.
3) The structural labeling is also much too simplified, particularly for untrained eyes, please label names for helices and strands, particularly please label N-and C-termini. Please also label domain names, e.g. I-domain, where is it?
We have updated the Figure 1 accordingly with N and C termini labelled and the I domain circled in order to clarify the structural presentation. The figure legend has also been clarified as per the reviewer's suggestions. 4) Without a DNA-TF complex structure, it's rather difficult to explain why the I-domain is so much important for DNA binding (and binding specificities), and PPIs, particularly as the authors claimed the I-domain is not even touching the DNA. Some vague arguments about allosteric change are rather weak in this case. Although this is a non-structure focused paper, some DNA-protein modeling including I-domain may be necessary. The following is some complex structural examples, mentioned by the authors in the paper.