Polymer-coated carbon nanotube hybrids with functional peptides for gene delivery into plant mitochondria

The delivery of genetic material into plants has been historically challenging due to the cell wall barrier, which blocks the passage of many biomolecules. Carbon nanotube-based delivery has emerged as a promising solution to this problem and has been shown to effectively deliver DNA and RNA into intact plants. Mitochondria are important targets due to their influence on agronomic traits, but delivery into this organelle has been limited to low efficiencies, restricting their potential in genetic engineering. This work describes the use of a carbon nanotube-polymer hybrid modified with functional peptides to deliver DNA into intact plant mitochondria with almost 30 times higher efficiency than existing methods. Genetic integration of a folate pathway gene in the mitochondria displays enhanced plant growth rates, suggesting its applications in metabolic engineering and the establishment of stable transformation in mitochondrial genomes. Furthermore, the flexibility of the polymer layer will also allow for the conjugation of other peptides and cargo targeting other organelles for broad applications in plant bioengineering.


Results summary:
The work by Lau et al uses polymer coated carbon nanotubes modified with peptides to enable plasmid delivery into the mitochondria of plants. The team uses different reporters to validate their platform and are able to quantify protein expression with western blots and fluorescence microscopy. The concept and advance is very high in its significance for mitochondria transformation in plants, which is likely unprecedented. However, given its potential very high impact, there are several points that should be addressed by the authors prior to publication as noted below.

Main points:
• Colorimetric assays such as evans blue relies on absorbance and emission of dyes. It is known that SWNT have very broad absorption and emission spectra that overlap with those of dyes. qPCR of plant stress genes should be performed for molecular-level validation of biocompatibility for the SWNT-PM-Peptide/pDNA complexes. • Authors claim that the Cytcox peptide confers specific mitochondrial delivery while the data suggest that SWNT location is unspecific. SWNT-PM-KH9 in SFig6 are shown to be localized to mitochondria while there is no targeting peptide in this material (so localization should not be specific). Authors show that SWNT-PM-CytKH9 are able to deliver pDNA to both mitochondria and nucleus for expression. From the data they present, organelle-specific expression is given by the promoter in the pDNA, not by the Cytcox peptide, I think the text should explicitly reflect that. Their data clearly show that the combination of Cytcox and KH9 is what gives the highest expression, making CytKH9 a *novel* and very versatile peptide for delivery, not restricted to mitochondria. Because authors already have plasmids with chloroplast promoters, I would suggest that they try to deliver them and see if they detect expression. • Regarding the claim of HDR in mitochondria, the PCR should be better explained and the sequencing data should be provided in the SI. Moreover, if there is HDR, the bottom panel of figure 6b should show a larger band corresponding to the locus with the introduced genes. Also, to sustain the claim that increased folate results in root growth, I would suggest that they show root folate quantification. Alternatively, a good negative control would be to deliver pDNA with only the GFP reporter lacking the sul1 gene. • Based on the provided FE-SEM micrographs, there are no SWNTs visible. The red arrows are pointing to features we see prolifically in plant leaf tissue EM of inclusion bodies. SWNT and peptides are carbon based and are therefore highly unlikely to be observable inside plant cells except in the vacuole. I suggest removing the SEM micrographs as they do not help support the claim of internalized SWNT. To this end, the resolution of confocal Raman (unspecified) is unlikely to resolve SWNT present in the mitochondria against those present in the cell or on the surface of the mitochondria or more broadly in plant tissue. Also please clarify why the baseline Raman intensity for the different SWNT treatments varies so greatly. • Figure 6a-b: It seems that given the design of the primers, a PCR of the delivered pDNA complex (not necessarily the mitochondrial genome) would give rise to amplification of the same region. It may be that the PCR amplicon has amplified to a greater extent with the SWNT sample because SWNTs are able to better protect the pDNA from degradation. It might be necessary to sequence regions that begin past the left and right homology arms to determine whether authors are amplifying a region from the pDNA or a mitochondrial genomic region.

Minor points:
• Root expression is particularly exciting. However I am unclear how root expression assays were done: were the roots vacuum infiltrated? Were these roots from plants that were leaf infiltrated? Hydroponic assays? Line 237 simply says "after infiltration" but it is unclear what was infiltrated and when. • Authors show that loading of the plasmid generates a closer to neutral SWNT-plasmid suspension. Authors should test whether a larger plasmid could be adsorbed to the SWNT (not suggesting all of the study be redone, just the gel shift assays and colloidal stability tests). in the manuscript). • Figure 5b: Please provide scale bars on your images. Also, it would be great to have some higher magnification images available. • Throughout the entire manuscript, authors should state prior to every result how many days-post-infiltration (dpi) their results were resultant from. • It is unclear from initial reading of the text and figure 1 whether the peptides are attached to the SWNT covalently or noncovalently. I suggest reworking the text and figure accordingly. I.e. line 74-27 refers to prior publications but this is critical information that needs to be included in this manuscript (it isn't until line 327 that authors mention that the polymer is noncovalently coated on SWNT). • In general, the authors suggest that a mitochondrial localization sequence tethered to these relatively large nanostructures enables trafficking into the mitochondria. Are the entire SWNT structures, which can apparently be on the same size magnitude as mitochondrion themselves be trafficked in their entirety into these organelles?Or are the SWNT constructs somehow shedding these peptides and pDNA cargo in the cell? It would be good for authors to speculate, in addition to providing information on the average length and polydispersity of the SWNT used. • Line 204. It is surprising that a human derived COX2 mitochondrial promoter is active in plant mitochondria. Are there literature to support this inter-kingdom adaptability? Are there any possible sources of non-specific transcription from this promoter? Additionally, the COX2 promoter elements are quite complex in their organization. How was the exact sequence to use determined? Additionally, could the authors explain the use of tNOS in their mitochondrial plasmid construct? It is surprising that a terminator used in nucleus expression is effective in this scenario as well. • Could authors cite literature for the statement that HDR occurs readily within plant mitochondria? • Line 286. In the literature cited for this selective marker, it appears that the sul1 protein was expressed by nucleus expression and then subsequently targeted to the mitochondria using a localization sequence. It was not developed for mitochondrial transformation. Is the transgene used in this experiment the same? If so, could there be errorneous nuclear expression and subsequent transport to the mitochondria that are driving the observed increase in folate? • Line 293. In the mt genotyping experiment, could the authors provide the primer sequences used as well as that of the donor plasmid? It is unclear if the authors used primers that would span the gaps of the HDR template and the mtgenome itself or if regions within the plasmid were used. If it is the latter case, could the amplified pcr product not originated from the delivered plasmid itself? Additionally, for plastid transformation, and HDR in mammalian cells, a linear donor template is preferred so it is surprising a circular construct is effective. Actual sequencing of an amplicon that spans the gaps of the HDR donor within the mtDNA would strongly supplement the claims made by the author here. Introduction Line 81 -Information about Cytcox are not exhaustive and the reference of its targeting properties in plant is not appropriate. Similarly, KH9 details and relative references are missing Results Lines 87-93 This sentence appears confused and should be rephrased Line 124. Please include a map of this plasmid Line 130. Looking at the graph in Fig. 2f (blue curve), the ratio SWNT-PM(KH9)/ pDNA (1 microgram/ 50 ng) seems not well estimated. Lines 134-137 What about the zeta potential of SWCNT-PM after plasmid binding? It should be reported. Lines 173-176. FE-SEM images of infiltrated plants must be integrated with micrographs of non infiltrated plants treated with nanotubes and not infiltrated. FE-SEM are not totally convincing and could not discriminate clearly SWCNT-PM from naturally occurring granules available in the organelles. Therefore, the author must prove with additional techniques the mitochondrial localization. For instance, TEM images reported in literature could be much more useful then SEM to prove mitochondria uptake of SWNTs as reported in: Ma X., et al 2021 ACS Nano. 2012;6(12): 10486-10496. doi:10.1021/nn302457v Battigelli et al., Nanoscale, 2013 -Alternatively , fluorescently -labelled nanotubes could be employed for mitochondrial colocalization analyses by confocal microscopy.
-Moreover, the uptake analysis must be strengthen by quantitative data, for instance by reporting the number of SWCNT positive mitochondria vs the total observed in the micrographs.
-Finally, the uncertainness to recognise SWNT from other organelle granules (as admitted by the authors at lines 178 -180) suggests to clearly state in the text that the SWCNT localization in other organelles cannot be excluded. The expression of nuclear promoters supports this.

Conclusion
Apart the valorisation of the results, the conclusion must mention also the limit of nuclear expression that could be a problem for those applications that may require specific mitochondrial expression. The author should include these considerations for proper information.
We thank the reviewers for their valuable comments to improve our manuscript and the opportunity to revise the manuscript. A summary of additional figures and changes to original figures precede our detailed reply to the reviewers' individual comments. Our replies are colored in red with direct changes to the main text highlighted in yellow. Relevant figures and supplementary information have also been reproduced.    with DyLight488 labelled SWNT-PM-CytKH9 to confirm the localization within mitochondria. Furthermore, additional experiments using Raman mapping of isolated mitochondria are now included to show the colocalization of the SWNT G-band peak with the mitochondria. The spectra in the Raman spectroscopy graph were originally shifted in the y-axis for clarity but have been reverted to the original positions.

Fig. 4a:
The NOS terminator sequence has been removed from the diagram for the Cox2 driven RLuc reporter construct. Plasmid sequencing showed that the NOS terminator was absent.

Fig. 5a:
The NOS terminator sequence has been removed from the diagram for the Cox2 driven GFP reporter construct. Plasmid sequencing showed that the NOS terminator was absent.   Supplementary Data 1 and 2 files containing the raw sequence reads from genotyped PCR products from the 5′ and 3′ arms of pAtMTTF1 transformants 7 days after infiltration with SWNT-PM-CytKH9/pDNA ( Fig. 6 and Supplementary Fig. 15) have also been added.

Reviewer #1:
The manuscript describes the transformation of mitochondria (with protein and nucleic acid) using a carbon nanotube approach. This is a useful and novel approach, and with some significance as plant mitochondria remain one of the challenges to transform. While overall I am convinced by the material presented to transform mitochondria, a significant increase in efficiency on previous reports I do have some clarifications.

Comment #1
While it is great to show mitochondria are transformed and that the DNA was integrated, is it inherited into the next generation -this would really show transformation compared to transient transformation. Moreover given the almost universal maternal inheritance of mitochondria, reciprocal crosses would be needed to carry out to show that it is a maternally inherited trait.

Response
Stable integration of DNA in the mitochondria that can be inherited into the next generation remains challenging due to low delivery efficiencies into plant mitochondria as well as the instability of the plant mitochondrial genomes and the lack of strong selection markers that can be maintained during the development of the adult plant after zygote formation. In this study, we have successfully tackled the first problem by the use of peptide conjugated SWNT NCs that has sufficient delivery and expression efficiency for genome integration and expression with an observable phenotype. Furthermore, we have augmented the characterization of this phenotype with additional experiments including mitochondrial and photosynthetic measurements with folate quantification to demonstrate the potential of this system in metabolic engineering ( Supplementary Fig. 21). Nevertheless, without the development of a strong selection system that can be used to tackle the instability of the mitochondrial genome, stable and inheritable transformation that can be demonstrated through maternal inheritance of the next generation remains challenging to study. This manuscript is focused on the development of a new and flexible method of DNA delivery using SWNT and peptides that would be an important first step with other applications including delivery to other organelles.
We have added this information into the conclusion section (pg. 18) of the manuscript to highlight the need of such a selection system to enable follow up studies in mitochondrial transformation and complementation.
(pg. 18) However, organelle-specific delivery methods within plants are still lacking due to uptake limitations and relatively low delivery efficiency, and is further compounded for mitochondria targeted transformation due to the instability of the plant mitochondrial genomes and the lack of strong selection markers that can be maintained during the development of the adult plant after zygote formation. In this study, we have successfully tackled the first of these problems by the use of peptide conjugated SWNT NCs with sufficient delivery and expression efficiency for genome integration and expression with an observable root phenotype. Further development of such a selection system would enable follow-up studies of maternal inheritance leading to genomic engineering of mitochondrial agronomic traits.

Comment #2
I was interested in the fact that the transformed plants grow better -this is an important finding, and while it cannot be fully explained in this manuscript, cannot be left without more explanation. I am presuming the plants transformed are not limited by nutrient etc, especially folate. So why are they growing better? Some basic measurements of mitochondrial activities (and photosynthetic parameters) are warranted.

Response
In order to evaluate the physiological effects of the mitochondrial integration in more detail, we evaluated the mitochondrial respiration rates via ATP generation and photosynthetic parameters by chlorophyll fluorescence (Fv/Fm) but did not observe any significant differences between the control (pDNA and SWNT-PM-CytKH9) treated samples and the samples infiltrated with the SWNT-PM-CytKH9/pDNA ( Supplementary Fig. 21). Furthermore, transformation with a negative control pDNA with the SUL1 removed did not show any enhanced root growth, providing evidence that the phenotype is due to SUL1 integration alone ( Supplementary Fig. 20).
These results suggest that overall cellular metabolism and stress were not affected by the integrating gene. Since the gene encodes for an enzyme directly used in folate synthesis, we next evaluated folate levels from plant lysates infiltrated with the complex and found that pDNA containing the SUL1 integration and expression had a significant effect on folate levels from samples extracted 1 day and 3 days after infiltration ( Supplementary Fig. 21). This suggests that the increased root growth that we observed may be a downstream pathway effect due to the increased folate concentrations as a result from the SUL1 expression.
The results of these experiments are summarized within the Results on pg. 17 as follows and the corresponding figures have been added to Supplementary Information.

(pg. 17)
A. thaliana seedlings treated with the SWNT-PM-CytKH9/pAtMTTF1 complex also showed greater root growth  than seedlings treated with only pAtMTTF1 or SWNT-PM-CytKH9, as quantified by the relative root growth in area over the course of 1 week after transformation ( Fig. 6d and e). No significant difference was observed between the DNA-only or SWNT-PM-CytKH9/pDNA complex treated seedlings when a negative control construct with the SUL1 and its promoter and terminator removed ( Supplementary Fig. 20).
Chlorophyll fluorescence and mitochondrial respiration measurements did not show any significant differences upon SWNT NCs infiltration, suggesting that the overall increase in root area was not due to overall metabolic changes and that the plants were not stressed by the SWNT NCs introduction (Supplementary Fig. 21a-b). We then quantified intracellular folate concentrations that showed significantly increased folate levels relative to the pDNA only and SWNT-PM-CytKH9/pDNA (-SUL1) infiltrated samples at days 1 and 3 postinfiltration ( Supplementary Fig. 21c). We hypothesize that the expression of SUL1 from the transformed construct led to the observed increased folate levels, thus increasing the growth rate observed within the roots.

Supplementary Fig. 20. Effect on A. thaliana root growth by infiltration of SWNT-PM-CytKH9/pDNA with and without the SUL1 insert.
A. thaliana root growth after infiltration with SWNT-PM-CytKH9 complexed with pAtMTTF1 pDNA containing the SUL1 insert (+SUL1) and the negative control with theinsert removed (-SUL1) relative to their respective pDNA only control. Significant differences were observed between the relative root areas (n=18 seedlings) on Day 3, 5, and 7. Arabidopsis. The complementation of at least one of these by expressing in mitochondria would show the utility of this system.

Response
We agree that complementation of mitochondrial mutants is the next important step in demonstrating the utility of this system for mitochondrial transformation. However, to achieve stable and homologous expression of transgenes in mitochondria, a well-developed selection marker system in mitochondria would be required to tackle problems associated with the instability of mitochondrial genome. We have begun efforts into developing a selection system by expressing the SUL1 sulfadiazine resistance gene using genetic integration in this work, but sufficient stable integration of transgenes has still not yet been achieved. Thus, we believe these complementation experiments would be beyond the scope of the current work. We hope the reviewer understands this technical issue and our current focus in this work.

Comment #4
In summary as presented the manuscript reports aa substantial increase in efficiency for plant mitochondrial transformation. However, as this is not the first time plant mitochondria have been transformed, some useful utilisations of this system are necessary to show it can be inherited and address some of the issues that the authors raised in the introductionotherwise it would be a rather specialised system with limited uses.

Response
Although this method is not the first time plant mitochondria have been transformed, this is the first time that a SWNT-based delivery system has been used to deliver pDNA into plant mitochondria. Importantly, this system shows a 30-fold increase in expression efficiency compared with previous peptide/pDNA based systems and we show for the sufficient integration of the transgene with an observable root growth phenotype and elevated folic acid levels as a result of this integration, suggesting there is strong potential for metabolic engineering in mitochondrial transformation. Furthermore, the flexibility of the SWNT NCs developed in this manuscript allows facile customization of the conjugated peptides to develop delivery systems to other organelles carrying different cargoes.
We have highlighted these advances within the introduction (pg. 4) as follows.

(pg. 4)
Our results also showed that the expression of a folate metabolism-related gene conferred increased root growth and folate levels with potential for metabolic engineering, including the development of a selection marker system for stable transformation of plant mitochondria.
Results summary: The work by Lau et al uses polymer coated carbon nanotubes modified with peptides to enable plasmid delivery into the mitochondria of plants. The team uses different reporters to validate their platform and are able to quantify protein expression with western blots and fluorescence microscopy. The concept and advance is very high in its significance for mitochondria transformation in plants, which is likely unprecedented. However, given its potential very high impact, there are several points that should be addressed by the authors prior to publication as noted below.
Main points: Comment #1 Colorimetric assays such as evans blue relies on absorbance and emission of dyes. It is known that SWNT have very broad absorption and emission spectra that overlap with those of dyes. qPCR of plant stress genes should be performed for molecular-level validation of biocompatibility for the SWNT-PM-Peptide/pDNA complexes.

Response
We further confirmed the stress upon infiltration with the SWNT complexes by measurement of mitochondrial respiration and photosynthetic efficiency by ATP production and chlorophyll fluorescence, respectively (Supplementary Fig. 21). Both of these parameters did not show any significant decrease in efficiency, suggesting that toxicity and stress effects by the SWNT were relatively low. Furthermore, root growth was relatively unaffected by infiltration with the SWNT-PM-CytKH9 compared with those infiltrated with the pDNA only, also indicating that there are relatively low toxicity and stress effects by infiltration of the SWNT. The results of these experiments are summarized within the Results (pg. 17) as follows within the manuscript and the corresponding figures have been added to the Supplementary Information.

(pg. 17)
Chlorophyll fluorescence and mitochondrial respiration measurements did not show any significant differences upon SWNT NCs infiltration, suggesting that the overall increase in root area was not due to overall metabolic changes and that the plants were not stressed by the SWNT NCs introduction (Supplementary Fig. 21a-b). We then quantified intracellular folate concentrations that showed significantly increased folate levels relative to the pDNA only and SWNT-PM-CytKH9/pDNA (-SUL1) infiltrated samples at days 1 and 3 postinfiltration ( Supplementary Fig. 21c). We hypothesize that the expression of SUL1 from the transformed construct led to observed increased folate levels, thus increasing the growth rate observed within the roots.

Comment #2
Authors claim that the Cytcox peptide confers specific mitochondrial delivery while the data suggest that SWNT location is unspecific. SWNT-PM-KH9 in SFig6 are shown to be localized to mitochondria while there is no targeting peptide in this material (so localization should not be specific). Authors show that SWNT-PM-CytKH9 are able to deliver pDNA to both mitochondria and nucleus for expression. From the data they present, organelle-specific expression is given by the promoter in the pDNA, not by the Cytcox peptide, I think the text should explicitly reflect that. Their data clearly show that the combination of Cytcox and KH9 is what gives the highest expression, making CytKH9 a *novel* and very versatile peptide for delivery, not restricted to mitochondria. Because authors already have plasmids with chloroplast promoters, I would suggest that they try to deliver them and see if they detect expression.

Response
We agree that the delivery of the pDNA by SWNT-PM-CytKH9 were not organelle specific and the organelle-specific expression was conferred by the promoter in the pDNA. We have clarified this within the main text (pg. 14-15) as follows: (pg. 14-15) We note that the SWNT NCs likely had a degree on non-specificity and could localize within the nucleus in addition to the mitochondria as evidenced by the expression with the 35S promoter pDNA and that the observed organelle specific expression was from the promoters used in the respective constructs.
We also attempted to deliver the pDNA using SWNT-PM-CytKH9 for chloroplast expression of the same Renilla Luciferase expression construct with the psbA promoter but did not observe significant expression. Please see the luciferase expression graph below.

Comment #3
Regarding the claim of HDR in mitochondria, the PCR should be better explained and the sequencing data should be provided in the SI. Moreover, if there is HDR, the bottom panel of figure 6b should show a larger band corresponding to the locus with the introduced genes. Also, to sustain the claim that increased folate results in root growth, I would suggest that they show root folate quantification. Alternatively, a good negative control would be to deliver pDNA with only the GFP reporter lacking the sul1 gene.

Response
The description of the PCR has been expanded in the methods section (pg. 26) as follows: Primers were designed to amplify the wild-type mitochondrial genomic locus containing the integration sites for pAtMTTF1, outside of the homology arms. Primers were also designed for detecting the presence of the recombined construct, such that one primer binds outside of the homology arm and the other primer binds within the coding sequence of a gene in the construct, spanning either the left or right junction. Primers were also designed for amplifying the integrated construct, spanning either the left or right homology arm of pAtMTTF1. PCR was performed using PrimeSTAR GXL polymerase (Takara), with an annealing temperature of 64°C and 35 amplification cycles for the left and right arm reactions, and 60 °C and 30 cycles for the wild-type reaction. DNA sequencing of the corresponding extracted bands was performed using the respective primers in Fig. 6a and included as Supplementary Data 1-2.
We believe that HDR is occurring, but at a frequency too low to be detected by PCR using the primer pair in the bottom panel of Figure 6b. Thus, the only PCR product visible in that reaction would be amplified from the wild-type locus in the mitochondrial genome.
Sequencing data has been added to Supporting Fig. 15, and the raw sequencing data has also been included as Supplementary Data 1 and 2. The sequencing reads span the junction between the mitochondrial genome and the homologous sequence from the HDR donor plasmid, showing that recombination occurred at the expected location.
Supplementary Fig. 15. Sequencing of the genotyping PCR products from the 5` and 3` arms of the integrated pAtMTTF1 construct. Sequencing reads spanning the junction between the mitochondrial genome and the homologous sequence from the donor plasmid for recombination on both arms confirmed integration at the expected positions within the mitochondrial genome. Primers 1 and 4 refer to the primers labelled in Fig. 6a. Raw sequence reads can be found in the Supplementary Data 1 and 2.
We cloned the pDNA with the SUL1 gene and associated promoter and terminator deleted and repeated the root growth experiment with this negative control as mentioned ( Supplementary Fig. 20). There was increased root growth only when pDNA containing the SUL1 gene was delivered with SWNT, but not when the negative control was delivered with SWNT as compared to the respective pDNA only controls. We also evaluated folate levels from plant lysates and found that pDNA containing the SUL1 integration and expression had a significant effect on folate levels from samples extracted 1 day and 3 days after infiltration ( Supplementary Fig. 21). This suggests that the increased root growth that we observed may be a downstream pathway effect due to the increased folate concentrations as a result from the SUL1 expression.
These results are now included within the main text (pg. 17).
(pg. 17) No significant difference was observed between the DNA-only or SWNT-PM-CytKH9/pDNA complex treated seedlings when a negative control construct with the SUL1 and its promoter and terminator removed ( Supplementary Fig. 20).

Supplementary Fig. 20. Effect on A. thaliana root growth by infiltration of SWNT-PM-CytKH9/pDNA with and without the SUL1 insert.
A. thaliana root growth after infiltration with SWNT-PM-CytKH9 complexed with pAtMTTF1 pDNA containing the SUL1 insert (+SUL1) and the negative control with theinsert removed (-SUL1) relative to their respective pDNA only control. Significant differences were observed between the relative root areas (n=18 seedlings) on Day 3, 5, and 7.

Comment #4
Based on the provided FE-SEM micrographs, there are no SWNTs visible. The red arrows are pointing to features we see prolifically in plant leaf tissue EM of inclusion bodies. SWNT and peptides are carbon based and are therefore highly unlikely to be observable inside plant cells except in the vacuole. I suggest removing the SEM micrographs as they do not help support the claim of internalized SWNT. To this end, the resolution of confocal Raman (unspecified) is unlikely to resolve SWNT present in the mitochondria against those present in the cell or on the surface of the mitochondria or more broadly in plant tissue. Also please clarify why the baseline Raman intensity for the different SWNT treatments varies so greatly.

Response
Additional experiments using fluorescently labelled SWNT-PM-CytKH9 and wide area Raman mapping of the relevant SWNT G and G′ peaks have been added to support the localization of SWNT within the isolated mitochondria upon infiltration ( Fig. 3a-b). Raman measurements were shifted in the y-axis for clarity, but have been replotted to the original positions in the revised figure (Fig. 3c). We apologize for any confusion in the previous graphs. The results have been incorporated into the Results of the main text (pg. 9-11) as follows: (pg. 9-11) To investigate the delivery capabilities of the SWNT NCs to plant mitochondria, we prepared fluorescently labelled SWNT-PM-CytKH9 using DyLight 488-conjugated KH9Cys and examined its localization within the root cells of A. thaliana upon vacuum/pressure infiltration at 0.08 MPa for 60 seconds. In the cells infiltrated with the labelled SWNT-PM-CytKH9, clear colocalization between DyLight 488-labelled SWNT-PM-CytKH9 and MitoTracker-stained mitochondria could be observed in the samples infiltrated with the labelled SWNT-PM-CytKH9 as well as localization of the labelled SWNT-PM-CytKH9 within the cytosol of the root cells (Fig. 3a, upper panels). Conversely, most of the labelled SWNT-PM-CytKH9 samples remained on the surface of the root without vacuum infiltration (Fig. 3a, lower panels). Next, we analyzed isolated mitochondria from A. thaliana infiltrated with SWNT NCs by confocal Raman microscopy ( Fig. 3b-d). Raman mapping of the G band peak at 1590 cm -1 showed clear colocalization of SWNT signal with the isolated mitochondria from seedlings infiltrated with SWNT-PM-KH9 and SWNT-PM-CytKH9 (Fig. 3b) that is not present in the samples infiltrated with SWNT-PM. Similar levels of colocalization was also observed in samples infiltrated with SWNT-PM-CytKH9 but those infiltrated with SWNT-PM-KH9 and SWNT-PM exhibited considerably less G-band signal, suggesting that the Cytcox peptide played a role in directing the SWNT NCs into the mitochondria. To further quantify the effect of the peptide conjugation, we collected Raman spectra over an area of isolated mitochondria treated with the SWNT NCs and the overlay of the averaged Raman spectra (Fig. 3c) showed that similar characteristic G and G′ bands at 1590 cm -1 and 2640 cm -1 (Fig. 3c and d). Quantification of the normalized 1590 cm -1 (Fig. 3d) and 2640 cm -1 (Supplementary Fig. 8) peak heights showed that all three SWNT NCs were detected within the isolated mitochondria. In particular, the strongest Raman signals were detected from the samples infiltrated with SWNT-PM-CytKH9 with an approximately 10-fold increase in normalized G-band intensity relative to the samples infiltrated with SWNT-PM and SWNT-PM-KH9, suggesting that the dually functionalized SWNT was delivered most efficiently into the mitochondria (Fig. 3d and Supplementary Fig.  8). Taken together with the fluorescently-labelled SWNT results, these findings show that the SWNT NCs can be localized within plant mitochondria, with the Cytcox peptide conferring increased mitochondrial targeting. Representative Raman spectra taken from mitochondria isolated from A. thaliana 18 h after infiltration show characteristic G (1590 cm -1 ) and G` (2640 cm -1 ) bands from the SWNTs. (d) Quantified normalized G-band peak (1590 cm -1 ) intensities for isolated mitochondrial samples shown in (b). Relative values from 64 individual spectra (n=64) are shown for each individual sample. ns -not statistically significant, ***P<0.001, ****P<0.0001. Figure 6a-b: It seems that given the design of the primers, a PCR of the delivered pDNA complex (not necessarily the mitochondrial genome) would give rise to amplification of the same region. It may be that the PCR amplicon has amplified to a greater extent with the SWNT sample because SWNTs are able to better protect the pDNA from degradation. It might be necessary to sequence regions that begin past the left and right homology arms to determine whether authors are amplifying a region from the pDNA or a mitochondrial genomic region.

Response
We are sorry for this confusion. The original figure and description of the PCR showed the outer primers in the wrong position; they have been revised to show that the primer pairs do span past the left and right homology arms, so they show that the pDNA sequence has undergone recombination with the mitochondrial genome. The primer sequences are also included in Supplementary Table 1.

Fig. 6. Genotyping and phenotypic effects in A. thaliana upon infiltration with SWNT-PM-CytKH9/pAtMT pDNA complexes. (a)
Design of a plasmid DNA construct (pAtMTTF1) for integration into the mitochondrial genome of A. thaliana. SUL1 encodes dihydropteroate synthase type-2, which was previously used as a selection marker for mitochondrial transformation 36 .
Sequencing data of the genotyping PCR products has also been added (Supplementary Fig.  15 and Supplementary Data 1 and 2). The sequencing reads span the junction between the mitochondrial genome and the homologous sequence from the HDR donor plasmid, showing that recombination occurred at the expected location.
This information is clarified within the main text as well (pg. 16).

(pg. 16)
PCR genotyping of tissue samples 7 days after infiltration showed that the exogenous sequence was successfully integrated into the mitochondrial genome (Fig. 6b). DNA integration events were observed in all samples treated with SWNT-PM-CytKH9/pAtMTTF1 complexes, as indicated by positive genotyping PCR products for both the left and right homology arms (Fig. 6b). These PCR products span the junctions between the mitochondrial genome and the homologous sequence from the donor plasmid, and thus would only be amplified in the case of successful recombination. (Fig. 6b). In contrast, positive PCR products were detected faintly in only one sample out of six treated with pAtMTTF1 alone. Sequencing of the PCR products from the 5′ and 3′ arms of the construct confirmed integration at the expected positions within the mitochondrial genome ( Supplementary Fig.  15). Supplementary Fig. 15. Sequencing of the genotyping PCR products from the 5` and 3` arms of the integrated pAtMTTF1 construct. Sequencing reads spanning the junction between the mitochondrial genome and the homologous sequence from the donor plasmid for recombination on both arms confirmed integration at the expected positions within the mitochondrial genome. Primers 1 and 4 refer to the primers labelled in Fig. 6a. Raw sequence reads can be found in the Supplementary Data 1 and 2.
Minor points:

Comment #6
Root expression is particularly exciting. However I am unclear how root expression assays were done: were the roots vacuum infiltrated? Were these roots from plants that were leaf infiltrated? Hydroponic assays? Line 237 simply says "after infiltration" but it is unclear what was infiltrated and when.

Response
We apologize for this unclear description. Whole plant seedlings (7 day old) were infiltrated with the SWNT-PM-Peptide/pDNA solutions and were plated on MS media plates after infiltration. Root lengths and all other characterization were determined as days after initial infiltration. We have clarified this within the main text as follows: (pg. 11) "To investigate the delivery efficiency of the pDNA within plants, we evaluated the expression of a pDNA encoding a Renilla luciferase reporter construct (RLuc) upon the same vacuum/pressure infiltration conditions (0.08 MPa for 60 s) in the previous experiment within whole A. thaliana seedlings." (pg. 24) "Seven-day-old A. thaliana seedlings were used to assess the delivery of pDNA and its expression and integration in A. thaliana. Vacuum/pressure infiltration was performed by incubation of whole seedlings (10 μL solution per seedling) in the respective solution and subjected to 0.08 MPa vacuum followed by pressure for 60 seconds each. The seedlings were allowed to recover in the solution for 1 h at room temperature before being plated on 0.5x MS medium plates. For root growth measurements, the plates were placed at an orientation of approximately 75°. Seedlings were allowed to grow under 16-/8-h light/dark periods at 22°C for 1-7 days."

Comment #7
Authors show that loading of the plasmid generates a closer to neutral SWNT-plasmid suspension. Authors should test whether a larger plasmid could be adsorbed to the SWNT (not suggesting all of the study be redone, just the gel shift assays and colloidal stability tests).

Response
The integrating plasmid DNA containing the SUL1 was relatively large (10 kb) compared to the transiently expressed plasmid (5 kb). Gel shift assays of this plasmid with the respective SWNT samples are included in Supplementary Fig. 14.   Supplementary Fig. 14. pAtMTTF pDNA complexation with SWNT-PM-CytKH9. Quantification of gel shift electrophoresis mobility assays showing the amount of residual uncomplexed pAtMTTF pDNA (10 kb) at different SWNT-PM-Peptide NPs/pDNA ratios.
This information was also made more explicit within the main text (pg. 15) as: (pg. 15) SWNT-PM-CytKH9 complexed similarly with the pAtMTTF1 pDNA (10 kb) compared to the previously used pDONR pDNA (5 kb), suggesting that the SWNT NCs are able to complex with larger plasmids (Supplementary Fig. 14).  Figure 4

figure nor caption mentions what the biological substrate is, please mention it is leaves (I am assuming?) Response
The Renilla luciferase assays were evaluated using whole seedling soluble lysate fractions as performed in previous studies. The Evans blue test for toxicity was evaluated using intact seedlings. This information is now clarified within the Fig. 4 caption. "(b) Transfection efficiency of the SWNT-PM-Peptide and pDNA was evaluated 18 h after infiltration using a Renilla luciferase reporter construct under the control of mitochondriaspecific (Cox2) and nuclear (35S) promoters using whole seedling soluble lysate fractions. Data from 6 biologically independent samples (n = 6) are shown. (c) Evans blue assays for intact A. thaliana seedlings 18 h post infiltration with each respective NC, normalized to the absorbance of a boiled sample. A. thaliana infiltrated with DNA alone was used as a control."

Comment #9
How come 200 ng of pDNA was loaded on the SWNT-PM-peptide complex? Please include reasoning in line 203. Response 200 ng pDNA was loaded based on the gel-shift assays and the binding capacity of the SWNT-PM-CytKH9. This information has been added into the text (pg. 11) as "Based on DNA binding capacity of SWNT-PM-CytKH9, SWNT-PM-Peptide (400 µg) was complexed with 200 ng of pDNA for infiltration experiments (Fig. 2e and f)".

Comment #10
Line 48: Carbon nanotubes can be made from graphite but are not made of graphite. Response This is now corrected within the text (pg. 3) as "Carbon nanotubes (CNTs) are cylindricaltubule structures made from graphite with exceptional physical properties that have been harnessed in biological applications such as drug delivery and biosensors 14,15 ."  Based on the SEM images and AFM results, we expect the coating on the surface of the SWNT to be relatively uniform, so we expect the increases in the observed AFM heights (~2 nm) to be also reflective of the different in widths observed between the unmodified and modified SWNTs. This is clarified within the main text (pg. 8) as follows:

Comment #11
(pg. 8) The profiled AFM heights were relatively uniform for the measured samples (Fig. 2j) and increased from 0.81 ± 0.26 nm to 2.14 ± 0.24 nm upon polymer coating, giving a polymer layer thickness of approximately 1.3 nm which agrees with our previous results 24 . Figure 2g: Please include zeta potentials of the different SWNT samples after DNA has been loaded, I imagine the zeta potential would be highly dependent on amount of DNA. Response Zeta potentials in the presence of pDNA are now included in Fig. 2g. We have also added the following information into the main text (pg. 8):   Supplementary Fig. 2: From what I understand, SWNT-PM-CytKH9 is derived from SWNT-PM, so it would be logical that mass % remaining at 500C (that is, what we attribute to pure SWNT) would be lower for the SWNT-PM-CytKH9 sample than SWNT-PM. Why is this not so? Response Changes in solubility of the peptide conjugated sample (SWNT-PM-CytKH9) as well as ashing of the peptides themselves may have resulted in a higher residual weight at higher temperatures during TGA analysis. To confirm this, we performed TGA analysis on peptide samples (CytcoxCys and KH9Cys) and both peptides showed similar residual weight at the highest temperature, with KH9 having higher residual weight than the SWNT-PM sample. These results have been included in Supplementary Fig. 2 and the main text (pg. 6) as follows.

(pg. 6)
Thermogravimetric analysis (TGA) of the SWNT NCs was used to compare the mass loss profiles from 30 to 500°C (Supplementary Fig. 2) and evaluate their composition. Pristine SWNTs had minimal weight loss during pyrolysis, as expected for CNTs. SWNT-PM and SWNT-PM-CytKH9 showed a large mass loss at approximately 300°C, which was completed at approximately 420°C. This agrees well with residual amounts after pyrolysis of the maleimide and peptide samples and suggests that it corresponds to the adsorbed polymer layer and conjugated peptides. The residual weights at 500°C for SWNT-PM (22%) and SWNT-PM-CytKH9 (33%) suggested that the weight ratio of polymer to SWNT to peptide is approximately 10:3:2 in SWNT-PM-CytKH9.

Comment #17
Throughout the entire manuscript, authors should state prior to every result how many days-post-infiltration (dpi) their results were resultant from.

Response
The number of days-post-infiltration (dpi) has been explicitly added to the results without this information in the main text.

Comment #18
It is unclear from initial reading of the text and figure 1 whether the peptides are attached to the SWNT covalently or noncovalently. I suggest reworking the text and figure accordingly. I.e. line 74-27 refers to prior publications but this is critical information that needs to be included in this manuscript (it isn't until line 327 that authors mention that the polymer is noncovalently coated on SWNT).

Response
Information regarding the noncovalent nature of the maleimide methacrylate layer adsorbed on the surface of the SWNT and the covalently linked peptides have been added to the main text. Additional information regarding the peptides have also been added to the introduction and results sections as follows: (pg. 4) The maleimide methacrylate layer is non-covalently adsorbed on the surface of the SWNT using micelle-polymerization and can be covalently conjugated with thiol-containing moieties. The Cytcox peptide contains the mitochondrial targeting presequence that has been previously characterized and used for mitochondrial pDNA delivery and expression within plant and animal cells 26 . KH9 is a cationic peptide that facilitates electrostatic interactions with anionic pDNA for binding.

(pg. 5)
Furan-protected maleimide was polymerized with polyethylene glycol (PEG) methacrylate as a non-covalently bound polymer layer on dispersed SWNTs (Supplementary Fig. 1a). Upon deprotection, the maleimide group can be covalently modified with thiol-containing molecules via Michael addition ( Supplementary Fig. 1b and c). To target SWNT-PM for DNA delivery into mitochondria, we covalently conjugated two distinct Cys-containing functional peptides to yield SWNT-PM-CytKH9 (Fig. 1a).

Comment #19
In general, the authors suggest that a mitochondrial localization sequence tethered to these relatively large nanostructures enables trafficking into the mitochondria. Are the entire SWNT structures, which can apparently be on the same size magnitude as mitochondrion themselves be trafficked in their entirety into these organelles?Or are the SWNT constructs somehow shedding these peptides and pDNA cargo in the cell? It would be good for authors to speculate, in addition to providing information on the average length and polydispersity of the SWNT used.

Response
The lengths of the conjugated SWNT NCs introduced into the plants averaged approximately 200-300 nm based on DLS measurements with PDI around 0.4-0.5. The overall images taken using AFM also support these size and distribution. Relative to the size of the plant mitochondria (1-3 µm), it would be feasible that the entire SWNT complex is able to enter, and our experimental data using fluorescently labelled SWNT and confocal Raman microscopy support this. However, based on the resolution achievable with these techniques we cannot rule out the fact that the SWNT NCs could be embedded on the surface or only partially translocated into the mitochondria with the cargo released. This explanation has been added to the main text (pg. 8) and Supplementary Fig. 6 as follows: (pg. 8) Long strands of SWNT-PM could be observed on the graphite substrate and on average ranged from 200-1000 nm that corresponded well with the average lengths (200-400 nm and PDI: 0.4-0.55) observed by dynamic light scattering (Supplementary Fig. 6). Supplementary Fig. 6. Dynamic light scattering measurements of SWNT-PM-Peptide NCs. DLS measurements of particle size showed similar results with those observed using AFM.

Comment #20
Line 204. It is surprising that a human derived COX2 mitochondrial promoter is active in plant mitochondria. Are there literature to support this inter-kingdom adaptability? Are there any possible sources of non-specific transcription from this promoter? Additionally, the COX2 promoter elements are quite complex in their organization. How was the exact sequence to use determined? Additionally, could the authors explain the use of tNOS in their mitochondrial plasmid construct? It is surprising that a terminator used in nucleus expression is effective in this scenario as well.

Response
In the pAtMTTF1 construct, GFP is under the control of an Arabidopsis thaliana COX2 promoter, and SUL1 is under the control of a Saccharomyces cerevisiae COX2 promoter. In both of these cases, COX2 refers to the cytochrome c oxidase subunit 2 in the mitochondrial genome. These promoter sequences were determined by taking approximately 1 kilobase upstream from the respective translation start sites in the mitochondrial genomes. In the pDONR-cox2p-RLuc and GFP constructs, the promoter is from S. cerevisiae COX2. There is no terminator sequence according to the sequence map, and this has been removed from the construct diagrams in Figs. 4 and 5. We apologize for the initial error in the diagram label.
Previous studies have demonstrated that deletion of the untranslated terminator sequence did not affect expression of pDNA containing reporter constructs introduced within plant mitochondria, possibly due to the complex mRNA maturation and editing processes in Information regarding the constructs is now included within the main text (pg. 23) as follows: (pg. 23) Plasmids containing the GFP and luciferase reporter constructs were previously prepared and used for expression within A. thaliana in previous studies (Plasmid maps are shown in Supplementary Fig. 22) 12,13,46, . The pDONR cox2p-RLuc and GFP constructs contain the promoter from S. cerevisiae COX2 and the reporter construct, without a terminator sequence. Previous studies have demonstrated that deletion of the untranslated terminator sequence did not affect expression of pDNA containing reporter constructs introduced within plant mitochondria 47,48 .
The respective plasmid maps have also been added into Supplementary Fig. 22.   Fig. 15. Sequencing of the genotyping PCR products from the 5` and 3` arms of the integrated pAtMTTF1 construct. Sequencing reads spanning the junction between the mitochondrial genome and the homologous sequence from the donor plasmid for recombination on both arms confirmed integration at the expected positions within the mitochondrial genome. Primers 1 and 4 refer to the primers labelled in Fig. 6a. Raw sequence reads can be found in the Extended Data files. Supplementary Fig. 22. Plasmid maps of the reporter constructs pAtMTTF1, pDONR-Cox2Rluc, and pDONR-Cox2GFP.

Comment #21
Could authors cite literature for the statement that HDR occurs readily within plant mitochondria?

Response
The references for HDR occurring readily within plants is now included in the text.
"This construct is expected to be integrated into the mitochondrial genome via homologous recombination, which occurs readily within plant mitochondria 49,50 ."

Comment #22
Line 286. In the literature cited for this selective marker, it appears that the sul1 protein was expressed by nucleus expression and then subsequently targeted to the mitochondria using a localization sequence. It was not developed for mitochondrial transformation. Is the transgene used in this experiment the same? If so, could there be errorneous nuclear expression and subsequent transport to the mitochondria that are driving the observed increase in folate?

Response
The transgene used in this work is the same as the previous studies. It is under the transcriptional control of a S. cerevisiae COX2 mitochondrial promoter previously demonstrated to be specifically expressed within the plant mitochondria, so we do not believe erroneous nuclear expression would be an issue. This is clarified within the main text (pg. 23) as follows: (pg. 23) Plasmids containing the GFP and luciferase reporter constructs were previously prepared and were previously shown to express exclusively within the mitochondria of A. thaliana (Plasmid maps are included in Supplementary Fig. 22) 12,13,46, . The pDONR cox2p-RLuc and GFP constructs contain the promoter from S. cerevisiae COX2 and the reporter construct, without a terminator sequence. Previous studies have demonstrated that deletion of the untranslated terminator sequence did not affect expression of pDNA containing reporter constructs introduced within plant mitochondria 47,48 .

Comment #23
Line 293. In the mt genotyping experiment, could the authors provide the primer sequences used as well as that of the donor plasmid? It is unclear if the authors used primers that would span the gaps of the HDR template and the mtgenome itself or if regions within the plasmid were used. If it is the latter case, could the amplified pcr product not originated from the delivered plasmid itself? Additionally, for plastid transformation, and HDR in mammalian cells, a linear donor template is preferred so it is surprising a circular construct is effective. Actual sequencing of an amplicon that spans the gaps of the HDR donor within the mtDNA would strongly supplement the claims made by the author here.

Response
The primers used and donor plasmid maps are now included in Supplementary Table 1 and Supplementary Fig. 22, respectively.
The primers used amplifies a PCR product spanning the junction between the HDR template and the mitochondrial genome and would only be amplified upon successful integration in the mitochondrial genome. The original figure in Fig. 6a showing primer binding sites was incorrect and has been amended.
Sequencing data has been added to Supporting Fig. 15, and the raw sequencing data has also been included as Supplementary Data 1 and 2.
The main text (pg. 16) has been amended as follows (pg. 16) DNA integration events were observed in all samples treated with SWNT-PM-CytKH9/pAtMTTF1 complexes, as indicated by positive genotyping PCR products for both the left and right homology arms (Fig. 6b). These PCR products span the junctions between the mitochondrial genome and the homologous sequence from the donor plasmid, and thus would only be amplified in the case of successful recombination. (Fig. 6b). In contrast, positive PCR products were detected faintly in only one sample out of six treated with pAtMTTF1 alone.  Supplementary Fig. 15. Sequencing of the genotyping PCR products from the 5` and 3` arms of the integrated pAtMTTF1 construct. Sequencing reads spanning the junction between the mitochondrial genome and the homologous sequence from the donor plasmid for recombination on both arms confirmed integration at the expected positions within the mitochondrial genome. Primers 1 and 4 refer to the primers labelled in Fig. 6a. Raw sequence reads can be found in the Supplementary Data 1 and 2.  Supplementary Fig. 22. Plasmid maps of the reporter constructs pAtMTTF1, pDONR-Cox2Rluc, and pDONR-Cox2GFP.

Name
Sequence Primer Number (Fig. 6a Fig. 6 and Supplementary Fig. 15. In the present paper the authors designed single-walled carbon nanotubes (SWCNT) for gene delivery in mitochondria. In particular they reported a detailed chemico-physical characterization of SWCNTs functionalised with peptides and plasmids. They employed two peptides to confer mithocondria targeting properties to SWCNT and increase DNA binding on the nanotubes. Then they checked the pDNA binding properties of this material before infiltrating them in plantlet of A. thaliana. They also proved that SWCNT-CA-pDNA enable expression of reporter genes in Arabidopsis and drive genetic integration through the homologous recombination of sul1 gene involved in the folate pathway Nanomaterials for genetic engineering are highly appealing although a technological gap still limits their use in plant. The paper has the merit to provide a new strategy for SWCNTmediated transformation in plant mitochondria. The chemical-physical characterization of the nanomaterials is solid while some flaws are in the description of the uptake mechanisms. The specificity of mitochondrial transformation is not ensured but the approach sounds interesting. The significance and the novelty of the manuscript is suitable for publication on Nature communications. However, before publishing the following points need to be revised:

Comment #2
Line 81 -Information about Cytcox are not exhaustive and the reference of its targeting properties in plant is not appropriate. Similarly, KH9 details and relative references are missing Response Additional information regarding Cytcox and KH9 peptides have been added to both the introduction and results sections.

(pg. 4)
The Cytcox peptide contains the mitochondrial targeting presequence that has been previously characterized and used for mitochondrial pDNA delivery and expression within plant and animal cells 26 . KH9 is a cationic peptide that facilitates electrostatic interactions with anionic pDNA for binding." (pg. 5) CytcoxCys (MLSLRQSIRFFKC), abbreviated as Cyt in SWNT-PM-CytKH9, has been previously shown to direct the yeast cytochrome c oxidase subunit IV into the mitochondrial matrix and deliver pDNA for transient expression into mitochondria within plant and animal cells 26 . KH9 (KHKHKHKHKHKHKHKHKHC) is a cationic peptide consisting of alternating lysine and histidine residues that facilitates electrostatic complexation with anionic pDNA to increase loading efficiency. Our previous studies using these peptides have shown that they are able to direct nucleic acid and protein cargoes to the mitochondria within intact plants 12,13 .

Comment #3
Lines 87-93 This sentence appears confused and should be rephrased Response This sentence has been rewritten and simplified as follows: (pg. 5-6) "The SWNT nanocarriers (NCs) were quantitatively and qualitatively evaluated for their physical and chemical properties. Field-emission scanning electron microscopy (FE-SEM) of SWNT-PM and SWNT-PM-CytKH9 (6 kV on Si) ( Fig. 2a and b) show the typical bundled morphology of SWNTs approximately 200 nm to 2 µm in length, suggesting that the adsorption of the polymer layer and conjugation of the peptides did not significantly alter the physical dimensions of the SWNTs."

Comment #4
Line 124. Please include a map of this plasmid Response A map of this plasmid has been added to Supplementary Fig. 22. Supplementary Fig. 22. Plasmid maps of the reporter constructs pAtMTTF1, pDONR-Cox2Rluc, and pDONR-Cox2GFP.

Response
The blue curve may have looked to be not well estimated due to the smearing of the DNA band in the gel that was not well depicted originally in Fig. 2e. Fig. 2e has been adjusted to show the band clearer. Furthermore, the uncropped versions of the gel are also now included in Supplementary Fig. 5. The revised versions are as follows: Supplementary Fig. 5. Binding of pDNA to SWNT-PM-Peptide NCs.
Representative gel shift electrophoresis mobility assay used for quantification in Fig. 2e. The main unbound band at approximately 4kb was used for comparison in the binding abilities of the SWNT-PM-Peptide NCs.

Comment #6
Lines 134-137 What about the zeta potential of SWCNT-PM after plasmid binding? It should be reported.

Response
Zeta potentials of the respective SWNT/pDNA complexes have been added to the graph. We have also added the following information into the main text:

Response
Fluorescently-labelled SWNT-PM-CytKH9 using DyLight 488 were prepared and repeated for the infiltration experiments. Labelled SWNT-PM-CytKH9 were detected within the mitochondria, providing evidence for the uptake of the SWNT complexes into the mitochondria within the plants (Fig. 3a). Comparison between the non-infiltrated samples and infiltrated samples show that the SWNT complex does not readily enter the cell without vacuum infiltration (Fig. 3a). infiltrated with SWNT-PM. This suggests that the Cytcox peptides are able to confer the ability of the SWNT to enter the mitochondria. The results of these experiments have been summarized within the main text and Fig. 3 as follows: To investigate the delivery capabilities of the SWNT NCs to plant mitochondria, we prepared fluorescently labelled SWNT-PM-CytKH9 using DyLight 488-conjugated KH9Cys and examined its localization within the root cells of A. thaliana upon vacuum/pressure infiltration at 0.08 MPa for 60 seconds. In the cells infiltrated with the labelled SWNT-PM-CytKH9, clear colocalization between DyLight 488-labelled SWNT-PM-CytKH9 and MitoTracker-stained mitochondria could be observed in the samples infiltrated with the labelled SWNT-PM-CytKH9 as well as localization of the labelled SWNT-PM-CytKH9 within the cytosol of the root cells (Fig. 3a, upper panels). Conversely, most of the labelled SWNT-PM-CytKH9 samples remained on the surface of the root without vacuum infiltration (Fig. 3a, lower panels). Next, we analyzed isolated mitochondria from A. thaliana infiltrated with SWNT NCs by confocal Raman microscopy ( Fig. 3b-d). Raman mapping of the G band peak at 1590 cm -1 showed clear colocalization of SWNT signal with the isolated mitochondria from seedlings infiltrated with SWNT-PM-KH9 and SWNT-PM-CytKH9 (Fig. 3b) that is not present in the samples infiltrated with SWNT-PM. Similar levels of colocalization was also observed in samples infiltrated with SWNT-PM-CytKH9 but those infiltrated with SWNT-PM-KH9 and SWNT-PM exhibited considerably less G-band signal, suggesting that the Cytcox peptide played a role in directing the SWNT NCs into the mitochondria. To further quantify the effect of the peptide conjugation, we collected Raman spectra over an area of isolated mitochondria treated with the SWNT NCs and the overlay of the averaged Raman spectra (Fig. 3c) showed that similar characteristic G and G′ bands at 1590 cm -1 and 2640 cm -1 (Fig.  3c and d). Quantification of the normalized 1590 cm -1 (Fig. 3d) and 2640 cm -1 (Supplementary Fig. 8) peak heights showed that all three SWNT NCs were detected within the isolated mitochondria. In particular, the strongest Raman signals were detected from the samples infiltrated with SWNT-PM-CytKH9 with an approximately 10-fold increase in normalized G-band intensity relative to the samples infiltrated with SWNT-PM and SWNT-PM-KH9, suggesting that the dually functionalized SWNT was delivered most efficiently into the mitochondria ( Fig. 3d and Supplementary Fig. 8). Taken together with the fluorescently-labelled SWNT results, these findings show that the SWNT NCs can be localized within plant mitochondria, with the Cytcox peptide conferring increased mitochondrial targeting. Representative Raman spectra taken from mitochondria isolated from A. thaliana 18 h after infiltration show characteristic G (1590 cm -1 ) and G` (2640 cm -1 ) bands from the SWNTs. (d) Quantified normalized G-band peak (1590 cm -1 ) intensities for isolated mitochondrial samples shown in (b). Relative values from 64 individual spectra (n=64) are shown for each individual sample. ns -not statistically significant, ***P<0.001, ****P<0.0001.

Comment #8
Moreover, the uptake analysis must be strengthen by quantitative data, for instance by reporting the number of SWCNT positive mitochondria vs the total observed in the micrographs.

Response
Quantification and comparison of uptake between the different types of SWNT NCs was performed by analysis of the Raman spectra. Mitochondria infiltrated with SWNT-PM-CytKH9 showed approximately 10-fold increase in normalized G-band intensity (1590 cm -1 ) compared to the SWNT-PM and SWNT-PM-KH9 infiltrated samples. We have added this information within the main text (pg. 11) and Fig. 3d as follows:

(pg. 11)
To further quantify the effect of the peptide conjugation, we collected Raman spectra over an area of isolated mitochondria treated with the SWNT NCs and the overlay of the averaged Raman spectra (Fig. 3c) showed that similar characteristic G and G` bands at 1590 cm -1 and 2640 cm -1 (Fig. 3d). Quantification of the normalized 1590 cm -1 (Fig. 3d) and 2640 cm -1 (Supplementary Fig. 8) peak heights showed that all three SWNT NCs were detected within the isolated mitochondria. In particular, the strongest Raman signals were detected from the samples infiltrated with SWNT-PM-CytKH9 with an approximately 10-fold increase in normalized G-band intensity relative to the samples infiltrated with SWNT-PM and SWNT-PM-KH9, suggesting that the dually functionalized SWNT was delivered most efficiently into the mitochondria (Fig. 3c). Taken together with the fluorescently-labelled SWNT results, these findings show that the SWNT NCs can be localized within plant mitochondria, with the Cytcox peptide conferring increased mitochondrial targeting. Representative Raman spectra taken from mitochondria isolated from A. thaliana 18 h after infiltration show characteristic G (1590 cm -1 ) and G′ (2640 cm -1 ) bands from the SWNTs. (d) Quantified normalized G-band peak (1590 cm -1 ) intensities for isolated mitochondrial samples shown in (b). Relative values from 64 individual spectra (n=64) are shown for each individual sample. ns -not statistically significant, ***P<0.001, ****P<0.0001.

Comment #9
Finally, the uncertainness to recognise SWNT from other organelle granules (as admitted by the authors at lines 178 -180) suggests to clearly state in the text that the SWCNT localization in other organelles cannot be excluded. The expression of nuclear promoters supports this.

Response
We have added the information within the text to clearly state that SWNT localization cannot be excluded from other organelles including the nucleus as shown the expression observed with the nuclear promoters with both the RLuc and GFP reporter constructs.
(pg. 14-15) We note that the SWNT NCs likely had a degree of non-specificity and could localize within the nucleus in addition to the mitochondria as evidenced by the expression with the 35S promoter containing pDNA.

Comment #10
Apart the valorisation of the results, the conclusion must mention also the limit of nuclear expression that could be a problem for those applications that may require specific mitochondrial expression. The author should include these considerations for proper information.

Response
We have added statements into the conclusion regarding the considerations of the nonspecific localization of the SWNT NCs that could limit applications into the conclusion. In the case of specific organelle gene expression, organelle-specific promoters can be used to ensure that expression is only observed in the target organelle.

(pg. 18)
Although the SWNT NCs also localized within the nucleus of the plant cells and could potentially limit their applications where mitochondrial-specific localization of the SWNT NCs are required, we believe sufficient specificity can be achieved by the tailoring the cargo such as by using organelle-specific promoters or translocation sequences.