A mitochondrial pentatricopeptide repeat protein enhances cold tolerance by modulating mitochondrial superoxide in rice

Cold stress affects rice growth and productivity. Defects in the plastid-localized pseudouridine synthase OsPUS1 affect chloroplast ribosome biogenesis, leading to low-temperature albino seedlings and accumulation of reactive oxygen species (ROS). Here, we report an ospus1-1 suppressor, sop10. SOP10 encodes a mitochondria-localized pentatricopeptide repeat protein. Mutations in SOP10 impair intron splicing of the nad4 and nad5 transcripts and decrease RNA editing efficiency of the nad2, nad6, and rps4 transcripts, resulting in deficiencies in mitochondrial complex I, thus decrease ROS generation and rescuing the albino phenotype. Overexpression of different compartment-localized superoxide dismutases (SOD) genes in ospus1-1 reverses the ROS over-accumulation and albino phenotypes to various degrees, with Mn-SOD reversing the best. Mutation of SOP10 in indica rice varieties enhances cold tolerance with lower ROS levels. We find that the mitochondrial superoxide plays a key role in rice cold responses, and identify a mitochondrial superoxide modulating factor, informing efforts to improve rice cold tolerance.

The authors build upon their earlier work, published in New Phytologist last year, characterizing a cold-sensitive, albino rice mutant affected in a chloroplast pseudouridine synthase OsPUS1, which is necessary for cp rRNA maturation at lowered temperatures of 22° vs. 28°C.The cold-sensitive PUS1 phenotype is associated with increased ROS (reactive oxygen species) accumulation.Screening for genetic suppressors in an EMSmutagenized population they identify SOP10, encoding a mitochondrial pentatricopeptide repeat (PPR) protein.SOP10 suppresses the cold-sensitive phenotype but does not reconstitute rRNA processing in chloroplasts.Instead, SOP10 turns out to affect splicing of the respective first introns in mitochondrial nad4 and nad5 transcripts resulting in reduced complex I activities in the SOP10 mutants.Expressing different SODs (superoxide dismutases) can compensate the cold-sensitive phenotype and independently mutating SOP10 in three different and widely used rice indica varieties likewise lowers cold-sensitivity and ROS levels.
The manuscript summarizes solid experimental work and is clearly written.However, in my opinion the paper suffers from overstatements assuming "regulation" and intracellular "communication", already starting with the paper's title.As in similar examples in the recent scientific literature such terms often appear overused to "sell" the respective stories (see e.g., the last two sentences of the abstract).Isn't it simply that mutation of SOP10 causes a moderate defect of mitochondrial complex 1, which results in lower ROS levels, and which is able to ameliorate the PUS1 phenotype?In that context, also note the sentence starting in line 90 speaking of "a mutation in a mitochondrial protein", which is simply not the case here, where we just find a moderate reduction in mature mRNA levels for nad4 and nad5.Along these lines, a display like figure 7 postulating a messenger molecule and labeling it with a question mark is not helpful.
The above comments do not mean to suggest that the results are not fine and interesting, but I feel that the claims go simply too far and are not quite of the wide general interest that would seem fully adequate for the readership of Nature Communications.Some further comments: 1. How can a G1475A mutation result in a G492N amino acid change (line 112f.and figure 1)?The change of GGC to GAC would instead cause a G-to-D codon conversion, wouldn't it?
2. It should be mentioned that the splicing defect affects group II introns (BTW: introns would be named nad4i461g2 and nad5i230g2, respectively, following a general nomenclature proposal for plant organellar introns).
3. The authors offer a rather arbitrary selection of references on PPR proteins (their references 34 to 38, cited in line 148).Here, it would be especially interesting to let readers know about some of the other protein factors that already have previously been shown to affect splicing of exactly these two mitochondrial introns in model systems like Arabidopsis (ABO6, MISF68, OZ2), maize (DEK35, DEK55, EMP8, EMP24, EMP25, EMP602, PPR18, PPR-SMR1) or even rice (FSE5, PPR939, RL1) 4. The above is quite important in this context, and the authors may find it worth discussing whether they would expect to find phenotypes similar to the one of SOP10 for "mild" mutant alleles of those factors, too. 5. Strangely, the authors mention PLS-type variants of PPR proteins (lines 133ff.)but then rely on a rather old/outdated PPR prediction algorithm, not taking into account modern PPR HMMer profiles.Here, the alternative modern approach (see https://ppr.plantenergy.uwa.edu.au/)comes to quite different conclusions on the makeup of SOP10, not suggesting 13 P-type PPRs but rather 18 PLS-type (including SS-type) PPRs and, most notably, also terminal E1 and E2 motifs that are otherwise frequently signatures of organelle RNA editing factors.
6.In the above context it would be interesting to learn whether the authors have used the known PPR-RNA recognition code to make prognoses about where exactly SOP10 may bind to the nad4 and nad5 mRNAs.

Reviewer #3 (Remarks to the Author):
The manuscript by Zu et al. titled "Chloroplast-to-mitochondrion communication regulates reactive oxygen species production and cold responses in rice" describes a genetic screen to identify a suppressor of the low-temperature albino phenotype due to a mutation in OsPUS1, a plastid localized pseudouridine synthase coding gene involved in chloroplast biogenesis.The <i>ospus1</i> mutant has elevated ROS levels at 22°C compared to 28°C, correlating with the albino phenotype.The identified suppressor of <i>ospus1-1</i> gene SOP10 codes for a pentatricopeptide repeat (PRR) protein regulating splicing of mitochondrial genes <i>NAD4</i> and <i>NAD5</i> , leading to deficiencies in complex I and decreased ROS generation, reversing the albino phenotype at 22°C.Moreover, overexpression of SOD genes localizing to different cellular departments also reversed the albino phenotype of the ospus1 mutant, suggesting that ROS accumulation was responsible for the phenotype.Lastly, the authors show that CRISPR/Cas9 generated single mutants in SOP10 improved the cold tolerance potential of three cold sensitive rice cultivars without a yield penalty.
This manuscript is a follow-up of their 2022 paper on identifying OsPUS1 as a pseudouridine synthase required for proper chloroplast function at mild chilling temperatures.It presents an interesting story showing evidence for chloroplast-to-mitochondrion communication, which appears to be required under abiotic stress conditions to maintain ROS homeostasis.The manuscript is well written, clear and concise, and the interpretations fit the results.While it was previously shown that PPR proteins regulate RNA metabolism of plants organelles required for proper organelle biosynthesis under abiotic stress conditions, this manuscript provides evidence suggesting that chloroplast-to-mitochondrion communication regulates cold responses via ROS homeostasis even at low chilling temperatures.
To clarify a few questions I have, please address the following points: 1. Lines 164-166: The statement that "[…] results suggest that SOP10 directly binds to <i>nad4</i> and <i>nad5</i> transcripts and participates in the splicing of the first intron of their mRNAs in a temperature-independent manner" is correct based on Fig. 2 and Fig. S4.However, it seems that only primers interrogating intron 1 were used in Fig. S4.Were the other introns investigated as well?If so, please state; if not, please qualify your statements indicating that possibly other introns are retained as well in the <i>sop10</i> mutant.There is no size marker in Fig. 2c, but it looks like the entire transcript might not be spliced, therefore, please add size markers to Fig. 2 to determine the size of the unspliced transcript.Also, <i>nad5</i> has two splice products in 9311 and the larger one is reduced in the <i>sop10</i> mutant.Please address and explain the significance of the two splice forms and why abundance of the larger one, possibly containing unspliced introns, is reduced.
2. Fig. 6: legend states "allowed to recover…for 4 days" while figure labeling says, "recover for 2-d".Also note that "recovery" would be better than "recover" in the figure labeling.

Reviewer #1 (Remarks to the Author):
The studies reported in this manuscript build on previous work by these authors on the mitochondria-localised pentatricopeptide repeat (PPR) protein called pseudouridine synthase, OSPUS1-1, which binds to transcripts encoding the mitochondrial complex I subunits NAD4 and NAD5.The mutant is albino under low temperatures because of low complex I activity and altered chloroplast ribosome biogenesis.In the present study, the authors report the isolation of a suppressor of ospus1-1, which partially rescues the low temperature-sensitive leaf albino phenotype of the ospus1-1 mutants.The authors had previously reported that the ospus1-1 mutants show enhanced ROS accumulation.
1.The authors have opted for a very traditional view of the roles of ROS in plants.Current concepts consider that all ROS are potentially signaling molecules that fulfil essential roles in plant, growth and defense.Within this context, the notion that there are safe and harmful levels of ROS is inaccurate.Moreover, ROS accumulation does not only occur under stress conditions.
Response: We thank the reviewer for pointing out this issue.ROS play important roles in plants, controlling processes such as growth, development and especially responses to biotic and abiotic stress stimuli.Truly, ROS accumulate and act as signals in both plant growth (in stem cell maintaining) and development (in cell differentiation), and stress responses.We updated our description about ROS in developmental regulation and added new references accordingly (Lines 5565 in the revised manuscript).
2. Moreover, current data suggest that catalase (CAT) only acts to remove hydrogen peroxide in peroxisomes and that the broad network of antioxidant defenses accomplish this task in other organelles.While peroxidases are relevant to this task, peroxidases can also produce hydrogen peroxide, and so this is not the most relevant enzyme to measure.

Response:
We thank the reviewer for the constructive comments.In the previous version, we did not explain well the difference between NBT and DAB staining.After re-analyzing our original data and repeating this experiment, we confirmed the NBT and DAB staining patterns in ospus1.As the reviewer suggested, we indeed found that the NBT staining signal appears in the whole leaf blade, which is consistent with the albino phenotype of ospus1.By contrast, the DAB staining signal only accumulates at the leaf tip.The albino phenotype was related to the accumulated O2 •-in the leaf blade.This difference suggest that the O2 •-is more relevant/important to the albino phenotype.We summarized the relationship between ROS staining and the albino phenotype as a schematic drawing, which was added in the revised manuscript (Fig. 4b).We believe that assessing the hydrogen peroxide scavenging system may not be helpful for fully understanding the albino phenotype.We, therefore, removed the related data (original Fig. 4c) in the revised manuscript.
3. Crucial to data interpretation within this context is that superoxide dismutases (SOD) merely convert one weakly reactive ROS form (superoxide) to another more reactive ROS form (hydrogen peroxide).

Response:
We thank the reviewer for the suggestion.As we know, hydrogen peroxide (H2O2) is more stable than superoxide (O2 •-), it can diffuse or be transported within a cell or between cells.We detected the H2O2 levels in leaves of ospus1-1 overexpressing different SOD isoforms.We found that the levels of H2O2 decreased in SOD OE/ospus1-1 transgenic rice as shown by DAB staining.Therefore, overexpression of SOD in ospus1-1 decreased ROS levels including both O2 •-and H2O2.Our data was consistent with a previous concept that overexpression of SOD does not simply lead to increased H2O2, because the presence of very efficient enzymatic pathways designed to keep H2O2 levels low, including catalase, ascorbate peroxidase (APX), and glutathione peroxidases (GPX) (MacMillan-Crow and Crow, 2011).We added the new data into the revised Fig. 5c. in the revised manuscript.
4. The premise, as stated in lines 217-218 that it may be possible to decrease ROS levels by overexpressing SOD in ospus1-1, in poorly informed.I wonder therefore if data interpretation has been biased by the above conceptual ideas.
Response: As aforementioned, the ROS staining data indicated a close link between O2 •-accumulation and the albino phenotype in ospus1-1 (Fig. 4a, b in revised manuscript).To test whether the low-temperature sensitive albino phenotype in ospus1-1 is caused by accumulated O2 •-, we overexpressed SOD which scavenges O2 •-in ospus1-1 (see Lines 228-231 in revised manuscript).

I have some concerns about the description of the data and data interpretation as outlined below:
The authors expressed Mn-SOD, Fe-SOD, and Cu/Zn-SOD in the mutants and presented evidence that all four SOD isoforms were significantly higher in the transformed ospus1-1 mutants than the wild type and other lines.I find the data in Figure 4 rather difficult to understand.Crucially, the patterns of NBT and DAB staining in the leaves.While the NBT staining appears in the leaf blade the DAB stain only accumulates at the leaf tip.These details are not discussed in the text.Since rice is a monocot species, the leaf cells develop from base to tip.Hence, hydrogen peroxide only accumulates in the oldest cells of the leaf tip.This is evident in the leaves from the ospus1-1 mutant but it is not discussed.The findings are interesting finding because they suggest that OsPUS1 regulates only the senescence -induced accumulation of hydrogen peroxide.(The reviewer suggested to describe and discuss the data of NBT and DAB staining patterns in Fig. 4)

Response:
We thank the reviewer for the constructive suggestions.We first confirmed the NBT and DAB staining patterns in 9311 and ospus1-1 plants growing at 22℃, by repeating the ROS staining experiments in Fig. 4. The NBT staining signal reproducibly appears in the leaf blade and the DAB staining reproducibly only accumulates at the leaf tip in ospus1-1 mutants at 22℃.Combining those observations with the observation that the albino phenotype occurs in the leaf blade but not only the leaf tip, we hypothesized that the albino phenotype is caused by O2 •-accumulation.By contrast, the H2O2 accumulation at the leaf tip may be related to senescence, since the cells at leaf tip are developmentally oldest.We described the data in the Result part (Lines 204212 in the revised manuscript) and interpreted the data in Discussion (Lines 321329 in the revised manuscript) 6.Similarly, the data shown in Figure 5 concerning how the overexpression of SOD genes suppresses ROS accumulation is equally intriguing but not discussed appropriately.Unlike the NBT data shown in Figure 4, the data in Figure 5, show that NBT staining occurs mainly in or around the leaf tip, again suggesting that the overexpression of MnSOD and CuZn SOD but not FeSOD protects against the senescence-induced accumulation of hydrogen peroxide at the leaf tip.(The reviewer suggested to discuss the NBT staining data in Fig. 5) Response: We thank the reviewer for the suggestions.As shown in revised Fig. 5c, overexpression of SOD genes alleviated the O2 •-accumulation in ospus1 to different extents.Mitochondrial Mn-SOD overexpression suppressed the O2 •- accumulation in the leaf blade and the albino phenotype best, indicating that the albino phenotype is caused mainly by the mitochondrial superoxide.We also examined the H2O2 accumulation by DAB staining and found again that the H2O2 accumulation occurs at the leaf tip.We discussed the ROS staining data in SOD overexpression plants in the revised manuscript (Lines 330345 in the revised manuscript) 7. The authors should also discuss why the pattern of NBT staining in the wild type and mutants is different in Figure 4 and Figure 5.The data in Figure 5B clearly show that the albino leaf phenotype observed in the ospus1-1 mutants is prevented by overexpression of overexpression of any of the SOD forms.

Response:
We thank the reviewer for the comments.We repeated the experiment in Fig. 5c, NBT and DAB staining experiments in wild type and ospus1-1 at 22℃.We examined at least 12 leaves for each condition, and used the representative pattern for wild type and mutants.We found that the representative pattern of NBT staining in wild type and the mutant is consistent in Fig. 4 and Fig. 5.We replaced the pictures in original Fig. 5c with new representative ones (see revised Fig. 5c).8.In contrast, I find the data shown in Figure 6 regarding decreased ROS accumulation in the SOP10 knockouts in three indica varieties less convincing particularly because the pattern of NBT staining is not consistently changed between the lines.

Response:
We also repeated the experiment in Fig. 6c with additional DAB staining in SOP10 knockouts in three indica varieties under cold treatment (revised Fig. 6c).We analyzed the staining data statistically and found that the NBT staining signal accumulated in the entire leaf blade in all the wild-type plants of three indica varieties at 8℃.In the SOP10 knockouts, the two examined lines of each variety showed statistically consistent NBT staining patterns, with staining signal appearing mainly in or around the leaf tip (revised Fig. 6c).We replaced the NBT staining pictures in Fig. 6c with their representative ones.9. Based on the above comments I am not convinced that changes in ROS accumulation link provide any evidence for their contribution to mitochondrial and chloroplast communication as discussed by the authors.The data demonstrate that the lack of the mitochondrial PPR protein SOP10 rescues the albino phenotype.However, the conclusion that this involves chloroplast-to-mitochondrion (CTM) communication that regulates cellular ROS levels at low temperatures is far too general to be accurate.As presented, the discussion section is weak and uninformative.It does not provide a critical evaluation of the data or the mechanisms involved and that the overexpression of SOD protects against the senescence -induced accumulation of hydrogen peroxide.
Response: Thank you for the comments.Our previous work showed that a dysfunction in chloroplast results in ROS accumulation and albino phenotype under low temperatures (Wang et al., 2022).Loss of function in a mitochondrial protein SOP10 suppressed the phenotype caused by chloroplast mutation (ospus1).This is very similar to the previous report on Arabidopsis mod1 (mosaic cell death) mutant, in which a deficiency in fatty acid biosynthesis in chloroplasts results in mtROS accumulation and cell death, which are suppressed by a mutation in a mitochondrial protein, suggesting CTM communication in mod1-mediated programmed cell death (Wu et al., 2015).Further study revealed malate mediated CTM communication regulating programmed cell death: the deficiency in mod1 chloroplasts produced elevated malate in chloroplasts.The malate is then transported into mitochondria and oxidized to generate NADH, which serves as the electron donor for the mtETC to generate ROS, triggering programmed cell death.Mutations in any component of the malate shuttle pathway between chloroplasts to mitochondria block the formation of mtROS and suppress mod1 phenotypes (Zhao et al., 2018).One of the main authors of this manuscript (Dr.Lilan Luo) is also the major contributor to the work in Arabidopsis.
To test whether the malate shuttle is involved in the ospus1 albino phenotype, we created loss-of-function mutants of a malate shuttle component in the ospus1 background (Fig. 1 for reviewers only).Malate dehydrogenases (MDHs) and malate transporters in chloroplasts and mitochondria compose the malate shuttle (Fig. 1a for reviewers only).The rice genome encodes 12 and 11 annotated MDHs and malate transporters/translocators, respectively (Heng et al., 2018;Liu et al., 2017), but only two MDHs and two malate transporters are experimentally identified (Teng et al., 2019) (Tables 1 and 2 for reviewers only).OsMDH1 (LOC_Os01g61380) is the solely identified plastid MDH, whose knockout mutants show drastically decreased MDH activity in chloroplast protein extracts (Nan et al., 2020).We generated ospus1 osmdh1 mutant plants (Fig. 2 for reviewers only), and found it rescued the ospus1 phenotype (Fig. 1b for review only).
We then investigated the subcellular localization of the 10 annotated MDHs that were not reported previously, and found two of them are mitochondrialocalized.We tried to knock out the two mitochondria-localized putative MDHs (OsMDH2.1/LOC_Os02g01510, and OsMDH6.1/LOC_Os06g01590) in ospus1 background, however, we did not obtain the mdh double knockouts in ospus1.The data indicate a possibility that communication between these two organelles is involved in the albino formation in ospus1.Therefore, we discussed such a possibility and its potential biological significance in our revised manuscript.(Lines 290313) 10.The authors present data showing that the levels of glutathione and CAT and POD activities were also significantly higher in the transformed ospus1-1 plants than other lines.The data for enzyme activity are expressed as units per mg fresh weight but the authors do not provide the definition of a Unit in terms of activity.This information should be provided.Similarly, the authors only provide data on reduced glutathione (GSH).It would be more informative to provide data on both GSH and glutathione disulphide.This would allow a more accurate understanding of changes in the total glutathione pool and its oxidation state.
Response: Thank you for the comment.Since the albino phenotype of ospus1 is consistent with the NBT but not DAB staining pattern (Fig. 4a, b in the revised manuscript), the albino phenotype was likely related to the accumulated O2 •-in the leaf blade.Therefore, assessing the hydrogen peroxide scavenging system may not be helpful for understanding the albino mechanism.We therefore removed the data (original Fig. 4c) regarding hydrogen peroxide scavenging system.
Reviewer #2 (Remarks to the Author): Comments on "Chloroplast-to-mitochondrion communication regulates reactive oxygen species production and cold responses in rice" by Zu et al., submitted to Nat.Comm.
The authors build upon their earlier work, published in New Phytologist last year, characterizing a cold-sensitive, albino rice mutant affected in a chloroplast pseudouridine synthase OsPUS1, which is necessary for cp rRNA maturation at lowered temperatures of 22° vs. 28°C.The cold-sensitive PUS1 phenotype is associated with increased ROS (reactive oxygen species) accumulation.Screening for genetic suppressors in an EMS-mutagenized population they identify SOP10, encoding a mitochondrial pentatricopeptide repeat (PPR) protein.SOP10 suppresses the coldsensitive phenotype but does not reconstitute rRNA processing in chloroplasts.Instead, SOP10 turns out to affect splicing of the respective first introns in mitochondrial nad4 and nad5 transcripts resulting in reduced complex I activities in the SOP10 mutants.Expressing different SODs (superoxide dismutases) can compensate the cold-sensitive phenotype and independently mutating SOP10 in three different and widely used rice indica varieties likewise lowers cold-sensitivity and ROS levels.
The manuscript summarizes solid experimental work and is clearly written.However, in my opinion the paper suffers from overstatements assuming "regulation" and intracellular "communication", already starting with the paper's title.As in similar examples in the recent scientific literature such terms often appear overused to "sell" the respective stories (see e.g., the last two sentences of the abstract).Isn't it simply that mutation of SOP10 causes a moderate defect of mitochondrial complex 1, which results in lower ROS levels, and which is able to ameliorate the PUS1 phenotype?
Response: We thank the reviewer for the suggestion.We changed our paper's title to "A mitochondrial pentatricopeptide repeat protein enhances cold tolerance by modulating mitochondrial superoxide in rice".In addition, we modified the sentences that might be overstated throughout the manuscript; all the revised sentences are highlighted in yellow.
Our observation is very similar to a previous report on the Arabidopsis mod1 mutant, in which malate was identified as the chloroplast-to-mitochondria communication chemical, to regulate programmed cell death (Wu et al., 2015;Zhao et al., 2018).We knocked out plastid MDH (OsMDH1) in ospus1, and found that the ospus1 osmdh1 suppressed the albino phenotype in ospus1 (see Fig. 1 for reviewers only).The data indicate a possibility that the malate mediated communication between these two organelles worked in albino formation in ospus1.Therefore, we discussed such a possibility and its potential biological significance in our revised manuscript.(Lines 290313) See detailed response to review 1 question 9. 4. The above is quite important in this context, and the authors may find it worth discussing whether they would expect to find phenotypes similar to the one of SOP10 for "mild" mutant alleles of those factors, too.

Response:
We thank the reviewer for the suggestion.We discuss this in the revised manuscript (Lines 346364).5. Strangely, the authors mention PLS-type variants of PPR proteins (lines 133ff.)but then rely on a rather old/outdated PPR prediction algorithm, not taking into account modern PPR HMMer profiles.Here, the alternative modern approach (see https://ppr.plantenergy.uwa.edu.au/)comes to quite different conclusions on the makeup of SOP10, not suggesting 13 P-type PPRs but rather 18 PLS-type (including SS-type) PPRs and, most notably, also terminal E1 and E2 motifs that are otherwise frequently signatures of organelle RNA editing factors.
Response: Thank you for the suggestion.We used the recommended prediction algorithm (https://ppr.plantenergy.uwa.edu.au/) and found the SOP10 is indeed composed of 18 PLS motif with C-terminal E1 and E2 motifs.We replaced the corresponding description in the text (Line 133) and in Fig. 1d.We performed a new experiment (STS-PCR-seq) to examine whether SOP10 is involved in mitochondrial RNA editing, and the results are added in the Results (Lines 169182) and as Fig. S6 in the revised manuscript.
6.In the above context it would be interesting to learn whether the authors have used the known PPR-RNA recognition code to make prognoses about where exactly SOP10 may bind to the nad4 and nad5 mRNAs.
Response: Thank you for the constructive suggestion.We used the PPR-RNA recognition code to predict the binding sequences where SOP10 may interact with nad4 and nad5 mRNAs.nad5 contains 2 cis-spliced and 2 trans-spliced introns, since the nucleotide sequences of the two trans-spliced introns are unavailable, we used the sequences including the CDS and the two cis-spliced introns (intron1 and intron 4) for prediction, and the predicted SOP10 binding region located in nad5 intron1 and exon 2 mRNA.However, for nad4, the predicted SOP10 binding sequences reside in intron 3 of nad4, see detailed information in Fig. 3 for reviewers only.
Reviewer #3 (Remarks to the Author): The manuscript by Zu et al. titled "Chloroplast-to-mitochondrion communication regulates reactive oxygen species production and cold responses in rice" describes a genetic screen to identify a suppressor of the low-temperature albino phenotype due to a mutation in OsPUS1, a plastid localized pseudouridine synthase coding gene involved in chloroplast biogenesis.The ospus1 mutant has elevated ROS levels at 22°C compared to 28°C, correlating with the albino phenotype.The identified suppressor of ospus1-1 gene SOP10 codes for a pentatricopeptide repeat (PRR) protein regulating splicing of mitochondrial genes NAD4 and NAD5 , leading to deficiencies in complex I and decreased ROS generation, reversing the albino phenotype at 22°C.Moreover, overexpression of SOD genes localizing to different cellular departments also reversed the albino phenotype of the ospus1 mutant, suggesting that ROS accumulation was responsible for the phenotype.Lastly, the authors show that CRISPR/Cas9 generated single mutants in SOP10 improved the cold tolerance potential of three cold sensitive rice cultivars without a yield penalty.This manuscript is a follow-up of their 2022 paper on identifying OsPUS1 as a pseudouridine synthase required for proper chloroplast function at mild chilling temperatures.It presents an interesting story showing evidence for chloroplast-tomitochondrion communication, which appears to be required under abiotic stress conditions to maintain ROS homeostasis.The manuscript is well written, clear and concise, and the interpretations fit the results.While it was previously shown that PPR proteins regulate RNA metabolism of plants organelles required for proper organelle biosynthesis under abiotic stress conditions, this manuscript provides evidence suggesting that chloroplast-to-mitochondrion communication regulates cold responses via ROS homeostasis even at low chilling temperatures.

Response:
We appreciate your positive comments.Thank you.
To clarify a few questions I have, please address the following points: 1. Lines 164-166: The statement that "[…] results suggest that SOP10 directly binds to nad4 and nad5 transcripts and participates in the splicing of the first intron of their mRNAs in a temperature-independent manner" is correct based on Fig. 2 and Fig. S4.However, it seems that only primers interrogating intron 1 were used in Fig. S4.Were the other introns investigated as well?If so, please state; if not, please qualify your statements indicating that possibly other introns are retained as well in the sop10 mutant.There is no size marker in Fig. 2c, but it looks like the entire transcript might not be spliced, therefore, please add size markers to Fig. 2 to determine the size of the unspliced transcript.Also, nad5 has two splice products in 9311 and the larger one is reduced in the sop10 mutant.Please address and explain the significance of the two splice forms and why abundance of the larger one, possibly containing un-spliced introns, is reduced.
Response: Thank you for your constructive suggestion.We performed a new experiment to test whether other introns are retained in nad4 and nad5.The data showed that only intron1 of nad4 and nad5 was retained, the other introns were not affected by sop10 mutation.We added the new data as Fig. S4 and Fig. S5 in the revised manuscript.
In addition, we repeated the experiment in Fig. 2c and added size markers to the blot.We found that nad5 has only one spliced product corresponding the intact mature nad5 transcript (about 2 kb) in 9311, the absence of the intron1-retained nad5 spliced form in the blot may be due to its low abundance.We replaced Fig. 2c in the revised version.
2. Fig. 6: legend states "allowed to recover…for 4 days" while figure labeling says, "recover for 2-d".Also note that "recovery" would be better than "recover" in the figure labeling.
Response: Thank you for pointing it out and for the suggestion.We corrected the figure labeling as "Recovery for 4 d" in the revised manuscript (Line 768).Barkan et al., 2012;Yin et al., 2013;and Gully et al., 2015).Each PPR code, which is composed of the 5 th and the last amino acids of each PPR repeat, is indicated.Nucleotides matching the amino acid combination are indicated in red."?" indicates an unidentified nucleotide.

REVIEWERS' COMMENTS
Reviewer #1 (Remarks to the Author): Your manuscript has been appropriately revised to address all the points raised.
Reviewer #2 (Remarks to the Author): Revision of manuscript NCOMMS-23-02347A by Zu and colleagues, submitted to Nature Communications The manuscript has been revised quite extensively in response to comments by three reviewers focusing on different subjects ranging from ROS accumulation and metabolism to details of molecular genetics concerning mitochondrial gene expression and PPR protein makeup.Judging on the latter part, concerning my core expertise, the paper has now improved in content and quality with a better molecular characterization of the SOP10 protein and, as an immediate consequence, the additional finding of RNA editing defects on top of splicing defects (new section, lines 169-182).Accordingly, corresponding text has been added.The added discussion of splicing factors (lines 346 ff.) needs some re-checking, however.For example, intron nad5i392g2 is not present in flowering plants but unique to Lycopodiales, the (trans-splicing) nad5 "intron 2" of angiosperms in question is nad5i1455g2.Also, it may be worth noting and emphasizing that MISF68 and ABO6, like SOP10 presented here, also affect both nad4i461g2 and nad5i230g2 (and other introns) at the same time.Rice PPR939 affects nad5 introns 1, 2 and 3.
One of the major changes not addressed in the authors' response latter is that FIVE more co-authors from TWO more laboratories have now been added.In the paper, these contributions are now given for four of the new co-authors simply as "Y.W., X. Q., C. C. and B. T. helped to write the paper."This evidently is an editorial issue but from my point of view the addition of 5 more co-authors would need to be better elaborated and justified.

Reviewer #3 (Remarks to the Author):
This is a resubmission of the Zu et al. manuscript "Chloroplast-to-mitochondrion communication regulates reactive oxygen species production and cold responses in rice", now titled "A mitochondrial pentatricopeptide repeat protein enhances cold tolerance by modulating mitochondrial superoxide in rice".The authors addressed all my concerns and comments and I have further suggestions for improving the manuscript.
Dear Dr. Xu and reviewers: Thank you for your letter and for the reviewers' comments concerning our manuscript (NCOMMS-23-02347A).We have read the comments and have made corresponding revisions.Revised portions are highlighted in yellow in the paper.Our responses to the reviewers' comments are as follows: Reviewer #1 (Remarks to the Author): Your manuscript has been appropriately revised to address all the points raised.
Response: We appreciate your positive comments.Thank you.
Reviewer #2 (Remarks to the Author): Revision of manuscript NCOMMS-23-02347A by Zu and colleagues, submitted to Nature Communications The manuscript has been revised quite extensively in response to comments by three reviewers focusing on different subjects ranging from ROS accumulation and metabolism to details of molecular genetics concerning mitochondrial gene expression and PPR protein makeup.Judging on the latter part, concerning my core expertise, the paper has now improved in content and quality with a better molecular characterization of the SOP10 protein and, as an immediate consequence, the additional finding of RNA editing defects on top of splicing defects (new section, lines 169-182).Accordingly, corresponding text has been added.
The added discussion of splicing factors (lines 346 ff.) needs some re-checking, however.For example, intron nad5i392g2 is not present in flowering plants but unique to Lycopodiales, the (trans-splicing) nad5 "intron 2" of angiosperms in question is nad5i1455g2.

Fig. 3 (
Fig. 3 (for reviewers only) Binding predictions for the SOP10 proteins on the nad4 and nad5RNAs (referred toBarkan et al., 2012; Yin et al., 2013; and Gully et al., 2015).Each PPR code, which is composed of the 5 th and the last amino acids of each PPR repeat, is indicated.Nucleotides matching the amino acid combination are indicated in red."?" indicates an unidentified nucleotide.

Fig. 2 (
Fig. 2 (for reviewers only) Genotype of the osmdh1 mutants created by CRISPR-Cas9 technology.The sgRNA sequences of CRISPR targeting sites on OsMDH1 gDNA are highlighted in yellow.The predicted protein length of the osmdh1 CR mutants are shown on the right.Genotype and mutant pattern were confirmed by the Sanger sequencing.

Revised
Fig. 1c, d c Schematic diagram of SOP10 gene.The black triangle indicates the mutation site (G1475A; G492D) in ospus1-1 sop10-1.d Protein structure of OsSOP10 in rice, the PPR motifs were shown in different colored box.Revised Fig. 2c c Northern blot analysis to detect the nad4 and nad5 transcripts.Total RNA was extracted from seedlings of 9311, ospus1-1, four suppressors of ospus1-1, and comp plants grown 28℃ and 22℃.The specific fragments of nad4 and nad5 used as the probes.Red triangles indicate the intron retention in nad4.Methylene blue staining (MB stain) is shown as a loading control.Revised Fig. 4b Schematic drawing of relationship between albino phenotype and ROS accumulation pattern in ospus1-1 at 22℃.The albino occurred in the whole leaf blade which is consistent with O2 •-accumulation pattern.Revised Fig. 5c Accumulated ROS in ospus1-1 was suppressed by overexpressing SOD-encoding genes.The accumulation of O2 •-and H2O2 in the leaf of rice grown at 22℃ in (b) were assessed by NBT and DAB staining, respectively.The numbers on top of each photo show how often this representative staining pattern occurred, relative to total examined pictures.Scale bars, 1 cm.Revised Fig. 6 | Assessment of cold stress tolerance of sop10 single mutant plants at seedling stage.a Seedlings of LKZ1, ZJZ17, ZZ35, and their two corresponding sop10 CR mutants were grown at 28℃ for 2 weeks, shifted to 8℃ for 2 days, and allowed to recover at 28℃ for 4 days.Scale bars, 8 cm.b Survival rates of LKZ1, ZJZ17, ZZ35, and their sop10 CR seedlings after cold treatment in (a).Values are means ± S.D. (n = 3 biological replicates).The asterisks represent significant differences between mutants and wild type grown under the same conditions, determined by Student's t-test (twotailed).*, P < 0.05; **, P < 0.01; ns, no significant difference.c NBT and DAB staining were used to assess the accumulation of O2 •-and H2O2 in LKZ1, ZJZ17, ZZ35, and their sop10 CR mutant seedlings before and after cold treatment in (a).The numbers on top of each photo show how often this representative staining pattern occurred, relative to total examined pictures.Scale bars, 1 cm.d In-gel assay of NADH oxidase capacity of LKZ1, ZJZ17, ZZ35, and their sop10 CR mutants.Dihydrolipoamide dehydrogenase activity was used as a loading control.The red triangle indicates mitochondrial complex I, and the black triangle indicates dihydrolipoamide.Revised Fig. 7 | A proposed model of albino formation mediated by ospus1 under low temperature.(Left) In wild-type 9311, PUS1 is upregulated and catalyzes the pseudouridylation of pre-rRNAs under cold conditions, enabling normal chloroplast ribosome assembly and translation, thus maintaining homeostasis in chloroplast metabolism.(Middle) A deficiency in chloroplast pre-rRNA pseudouridylation in ospus1 mutants results in aberrant chloroplast ribosomes and defects in plastid translation, which then leads to an imbalance of chloroplast metabolism.The disruption of chloroplast homeostasis generated a chemical/signal that was transmitted into mitochondrion and induced O2 •-overproduction from the mitochondrial ETC (mETC) complex I.The mitochondrial O2 •-then caused the albino phenotype in ospus1 under low temperatures, by oxidizing target proteins (Oxi-PTMs), which remain to be identified.(Right) Mutations in a mitochondrial PPR protein (SOP10) that directly binds to nad4 and nad5 transcripts and regulates splicing of their first exons, decreases mETC complex I capacity and O2 •-accumulation, and suppresses the albino phenotype of ospus1, enhances rice cold tolerance in wild-type 9311.

Fig. S4
Fig. S4 The first intron of nad4 is retained in ospus1-1 sop10-1 plants.a Diagram of nad4 gene structure.Positions of the primers used for PCR are indicated by black arrows.b Intron retention occurred in nad4 transcripts in suppressors of ospus1-1 (ospus1-1 sop10-1 and three ospus1-1 sop10 CR lines) at 28℃ and 22℃ by RT-PCR.The purple triangle represents the intron-retained nad4 transcript.c The first intron of nad4 was retained in ospus1-1 sop10-1 plants.Sanger sequencing of RT-PCR products amplified in b.

Table 1 .
(for reviewers only) Summary of candidate malate dehydrogenase genes in the rice genome.

Table 2 .
(for reviewers only) Summary of candidate malate transporter genes in the rice genome.