1. Reviewed by eLife

    Reviewer #2 (Public Review):

    Overall this is a solid and technically sound manuscript, and I have only two relatively minor suggestions for improvement:

    1. Tetramer versus dimer

    The particles that were analyzed by cryoEM were composed of four THO-Sub2 protomers, yet the authors argue that a dimer is the functional unit. Why? The tetramer versus dimer organization needs to be better discussed, also in light of the observation that the human complex can also form a tetramer.

    1. Sub2 activation mechanism

    The authors should more carefully discuss how THO 'activates' Sub2 (and how the 'semi-open state' leads to activation) and indicate the RNA binding surface of Sub2 in their model.

    Read the original source
    Was this evaluation helpful?
  2. Reviewed by eLife

    Reviewer #1 (Public Review):

    1. I found the initial description of the overall structure confusing. At first the authors say the complex is a tetramer, which is not what was seen by the Conti lab and then follow that with a confusing discussion leading to the conclusion that the dimer with a rigid subunit and a flexible one is the functional unit. It rather feels like they arrive at this conclusion because that's what Conti's lab saw, rather than any other reason. Since the human complex is a tetramer, perhaps the tetrameric complex observed here is one possible form and that possibility should be considered more carefully. Please state whether there is any similarity in the arrangements between the human tetramer and the tetramer observed here. I found the figure 2 supp 1C was not easy to follow. Coloring each of the four protomers differently would make things clearer.

    2. The authors previously determined the structure of yra1C domain bound to sub2 and several labs have shown this interaction activates Sub2 atpase activity. Are those interaction observed previously between Yra1 and Sub2 compatible with this new structure? If so, perhaps the authors could provide a model showing how Yra1 fits into this larger complex. Also, could Yra1 C domain and Gbp2 bind simultaneously to a single THO-Sub2 protomer or would one protomer bind Yra1 and perhaps another bind Gbp2? This is worth considering because this would strengthen the concept that TREX acts as a general platform engaging with multiple export factors to drive recruitment of multiple Mex67 molecules and eventual export of the Mex67:mRNP complex. In the human system, the SR proteins and Alyref have an overlapping binding site on Nxf1, suggesting they may not act together to recruit a single Nxf1, but rather they recruit different Nxf1 molecules perhaps to the same mRNP via a single multimeric THO platform.

    Read the original source
    Was this evaluation helpful?
  3. Reviewed by eLife

    Evaluation Summary:

    This is an interesting paper describing the structure of the yeast THO:Sub2 complex and how it interacts with the SR like protein Gbp2. The paper extends what we have learned from two recently published Tho:Sub2 complex structures by the Conti and Plaschka groups in two ways. Firstly, it shows how Gbp2 interacts with the THO complex. Secondly, it reveals a substantially different orientation between THO:Sub2 protomers compared with the earlier structure, so provides more information on the flexibility and range of movements that the two protomers might make with respect to each other. The structural inferences are supported by some biochemical experiments but mechanistically the work has limitations, similar to other recent cryo-EM structures of this complex. However, this is an important structure of wide interest to people working on gene expression in eukaryotes and it undoubtedly advances the field.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers opted to remain anonymous to the authors.)

    Read the original source
    Was this evaluation helpful?