The manuscript by Mozzanino and collaborators report the identification and the characterization of a transmembrane transporter involved in the decrease of iron (Fe) accumulation under elevated CO2 concentration (eCO2). The authors have used a GWAS to identify candidate genes controling the variation of Fe accumulation under eCO2. Using a knockout line, they demonstrate that MIC, a gene annotated as Fe « dicitrate transport system permease », is necessary for the decrease in Fe concentration observed under eCO2. MIC is expressed in conductive tissues of roots and leaves. MIC gene expression is not altered by changes in CO2 or Fe nutrition. Iron deficiency response is also not altered in the mic KO mutant. Intriguingly, confocal imaging of plants expressing a functional MIC-GFP fusion under its endogenous promoter reveals MIC intracellular localization. Combining, pharmacological treatements and colocalization, the authors identify the localization of MIC on the Golgi apparatus, late endosome and vacuole.
The manuscript is clearly written and reports highly novel results on a transporter that had not been characterized so far. The conclusions are overall adequately supported by the data but see comments below.

Major comments
1) The title is misleading as it highlights the golgi localization of MIC, while the protein localizes to various compartments. The title should be modified to « intracellular » or « secretory pathway » instead of « Golgi-localized ».
2) The results of the GWAS need to be described in more details. Where are the SNP localized ? It looks like it could be MIC promoter but this is not stated. What are the accessions that carry the SNPs and how do they behave in terms of Fe accumulation and MIC expression level. A better description of the haplotypes is required.
3) The phylogenetic analysis of MIC family should be presented in more details. The manuscript should report the percentage of homology between the different members. It should also include a phylogenetic tree. It would be interesting to find out the origin of the gene annotation.
4) A formal demonstration of MIC transporter activity is missing. However, this might be beyond the scope of this manuscript and would deserve a throurough separate study.
5) Have the authors tested the phenotype of mic mutant under different Fe nutrition regimes ? I could not find the result in the manuscript although the mutant was grown under Fe deficiency and excess for gene expression analyses.
6) To establish the multiple localization of MIC, the authors should perform colocalization experiments with markers for the Late Endosome and tonoplast.
7) Figure 4A shows a striking decrease in MIC-GFP fluorescence intensity in root under eCO2. It looks like there is also a change in MIC localization with more signal inside the vacuole and in larger dot-like organelles under aCO2. Do the authors wish to comment on this point ? In any case, a negative control with another fusion protein would be necessary to ascertain that the effect is specific to MIC. The western blots shown in 4C are not fully convincing. Could the authors repeat the experiment and show the whole membrane in order to vizualize potential protein degradation products ?
Minor comments
- « D » is missing in figure 4.

The manuscript by Mozzanino and collaborators report the identification and the characterization of a transmembrane transporter involved in the decrease of iron (Fe) accumulation under elevated CO2 concentration (eCO2). Using GWAS, the authors identified MIC, as a candidate genes controling the variation of Fe accumulation under eCO2. They report that MIC, a gene annotated as Fe « dicitrate transport system permease », is an intracellular transporter necessary for the decrease in Fe concentration observed under eCO2. The manuscript reports highly novel results on a transporter that had not been characterized so far.
During the revisions, the authors have adequately addressed the major issues raised about the original submission. The reviewer suggests to reorganize the figures so that important results that have been added as supplemental data during the revisions are inserted in the main figures, while negative results are shown as supplemental data :
1) The results shown in supplemental figure 1 panel B and C should integrated in figure 1.
2) The results shown in supplemental figure 6 should be integrated in figure 2.
3) The results shown in figure 2G could be moved to supplemental data.