A bimetallic nanoplatform for STING activation and CRISPR/Cas mediated depletion of the methionine transporter in cancer cells restores anti-tumor immune responses

Lack of sufficient cytotoxic T lymphocytes (CD8+ T cells) infiltration and dysfunctional state of CD8+ T cells are considered enormous obstacles to antitumor immunity. Herein, we construct a synergistic nanoplatform to promote CD8+ T cell infiltration in tumors while restoring T cell function by regulating methionine metabolism and activating the STING innate immune pathway. The CRISPR/Cas9 system down-regulates the methionine transporter SLC43A2 and restricts the methionine uptake by tumor cells, thereby relieving the methionine competition pressure of T cells; simultaneously, the released nutrition metal ions activate the cGAS/STING pathway. In this work, the described nanoplatform can enhance the effect of immunotherapy in preclinical cancer models in female mice, enhancing STING pathway mediated immunity and facilitating the development of amino acid metabolic intervention-based cancer therapy.

could not only induce ROS storm to kill tumors but also stimulated the activation of the STING pathway. The design is interesting and each component plays a unique role, which could work in a cooperative manner. Overall, the findings are interesting, but extensive experiments are required to frame the results.
Major concerns: (1) In this study, the authors claimed that the nanoparticles could release ions in acid lysosome conditions and catalyze intracellular H2O2 to ROS to kill cancer cells. However, the tumor microenvironment (TME) is also acidic (pH 6.5-6.8) with a high concentration of H2O2. Are the nanoparticles stable in TME? Will they also catalyze TME H2O2 to ROS? Why does they affect T cell functions?
(3) Why the PMZH effectively knocked down SLC43A2 in tumors, but not in liver and kindy?
The biodistribution of this nanosystem in organs should be performed.
(4) The ROS increase upon nanoparticles accumulation in the tumor should be confirmed in vivo.
(5) It has been known that Mn2+ dramatically promotes pro-inflammation response and anti-tumor immunity dependent on macrophages. Would the nanoparticles affect other immune cells (such as macrophages) and remodel the tumor immune microenvironment?
How to demonstrate the nanoparticles specifically promotes CD8+ T cells antitumor function rather than the changes of the immune microenvironment?
(6) The authors claimed that PMZH remodels methionine metabolism in T cells. However, the authors don't provide any data to support that intervention of methionine metabolism in tumor cells could affect T cell immunity in this nanosystem.
(7) lipo has cytotoxicity in cells. However, the authors showed that the lipo treatment in Figure 3C decreased the apoptosis of T cells. Please explain.
Minor concerns: (1) The full name of ZIF should be spelled out in the manuscript.
(2) The quantitative colocalization analysis of plasmid with endo/lysosomes in 4T1 cells should be conducted. The color scatter plots and corresponding Pearson's correlation coefficient (PCC) values between the red and green fluorescence signals in the images should be provided.
(3) The title of the y-axis in Figure 4A should be labeled.
(4) The quality of WB results should be improved (such as fig2f, fig6g).
Reviewer #3 (Remarks to the Author): with expertise in cancer nano-therapy In this paper, Huang et.al constructed a synergistic nanoplatform by encapsulating CRISPR plasmids into Mn/Zn bimetallic MOF nanoparticles. The nanoplatform promoted CD8 + T cell infiltration in tumors while restoring T cell function by modulating methionine metabolism and activating the sting innate immune pathway. This study is based on previous work of Weiping Zou (nature volume 585, pages277 -282 (2020)) to exploit differences in methionine metabolism to design protocols to enhance T cell infiltration and restore T cell function . It is a topic of interest to researchers in cancer treatment, biomaterials, and other related fields. Some minor revisions should be considered before publication in Nature Communications.
1. The authors introduced a CMZH control group in their study in vivo to exclude the effect of a nonfunctional plasmid on methionine metabolism. However, they did not provide the effect of the CMZH control group on SLC43A2 gene expression. Please supplement. In addition, compared with CMZH, PMZH effectively reduced the level of apoptosis of T cells, but the effect of PMZH and CMZH on the secretion of functional cytokines was not evaluated. This needs further research and discussion.
2. Cytosolic DNA can be recognized by CGAs and other DNA sensors to produce type I interferons and antitumor innate immunity. Although the authors demonstrated the effects of methionine transport on the sting pathway in vivo, experiments at the cellular level are lacking. The CMZH group needs to be introduced in cytological experiments as well.
3. Please supplement the fluorescence intensity scale bar of biodistribution (supplement,Figure10b). 4. Flow gate strategies for DC cells, Figure7f-g should be provided in SI.
5. The targeting effect of HA needs to be characterized.
6. The authors should clarify why they choose plasmid DNA but not mRNA to encode the CRISPR/Cas9. Usually the immunogenicity of mRNA is much lower than plasmid DNA and can achieve higher protein expression in vivo.
7. Please check for formatting issues throughout this manuscript.
Reviewer #4 (Remarks to the Author): with expertise in cancer immunology, metabolism Here the authors generate nanoparticle formulations that allow for encapsulating a plasmid encoding for sgRNA targeting SLC43A2 and CRISPR/Cas9. Using this platform, the authors show that reduction of SLC43A2 in 4T1 tumors results in increased T cell function in vitro and in vivo (e.g. tumor killing/rejection) and this is due to activation of the STING/cGAS pathway.
My major comment is related to an unspecified mechanism of action off this nanoparticle.
On the one hand, there is reduced tumor methionine consumption and higher T cell function. On the other hand, the enhanced T cell function could be due a general enhancement in STING/cGAS activation or both. Some genetic loss of function experiments are essential to clarify how these nanoparticles are acting.
A second comment is the use of PBS as a control in all the in vitro experiments. At a minimum there should be a non-targeting sgRNA as well as testing some individual components of the nano-particle itself. Ideally, the authors should do this on the tumor cells alone, T cells alone, and then both in a coculture experiment.
Third, most, if not all, of the micrographs need to be carefully quantified and statistical analysis performed.
Forth, while the authors do examine DC markers and frequency of DCs, careful assessment of STING activation in these DC populations needs to be performed. In other words, does the nanoparticle activate STING in DCs or tumors or another immune population (e.g. myeloid). The authors suggest this happens in DCs but do not show these data.
While I am fine with some methods being deposited in the supplemental information, this is not sufficient for the reader to understand how certain experiments were performed. Therefore, the key methods of the main figures need to be placed in the manuscript. In particular, there needs to more details on the replicates and statistical test for each figure panel embedded directly in the captions.
Next is the summarization of our point-by-point response to the comments raised by the editor and reviewers.
Reviewer #1 (Remarks to the Author): with expertise in cancer nano-therapy 1. There are mainly two mechanisms of the nanoplatform constructed by the article. On one hand, modulation of methionine metabolism by CRISPR plasmids restores T cell function. On the other hand, methionine metabolism cooperates with nutrient metal ions to activate the sting immune pathway. However, the current description of the effects of methionine metabolism on the sting pathway is confusion. Please adjust the sequence of picture combinations to a more systematic description of the activation of the sting pathway. Moreover, it is not sufficient to introduce a CMZH control group in vivo only.
Complementing relevant experiments at the cellular level is necessary.
Our response: We appreciate very much for your constructive comments and kind recommendations. We introduced the CMZH group at the cellular level and adjusted the sequence of picture combinations"Figure 5, Supplement Figure 20#. As shown in Figure 5, compared to CMZH, the PMZH group significantly enhanced the generation of reactive oxygen species and the release of mtDNA, thereby enhancing the expression of STING signals. Methionine restriction can enhance the sensitivity of tumor cells to oxidative stress. This phenomenon indicated that the CRISPR plasmid regulated methionine metabolism, thereby enhancing ROS production and promoting the activation of the STING signaling pathway. This is consistent with the conclusion of the previous in vivo experiment.
At the same time, we further explored the relationship of methionine metabolism and the activation of STING"Supplement Figure 23#. In STING-KO 4T1 tumor bearing mice, PMZH could still restore T cell function, but its effect is weaker than in wild-type tumor bearing mice. This further elucidated the mechanism of PMZH: activation of the STING pathway increased T cell infiltration, while regulation of methionine metabolism restored T cell function. The synergistic effect of both effectively enhanced T cell immunity. 2. The authors did not adequately discuss the superiority for using a gene editing approach to inhibit SLC43A2; siRNA for example may work better considering the reduced delivery barriers.
Our response: We sincerely appreciate the insightful comments of the reviewer. We have added a discussion on this issue in the manuscript. CRISPR/Cas9 has the advantages of permanently modifying target genes, being exceedingly selective, and having low off-target likelihood. It is one of the most efficient, simple and low-cost gene editing technologies available, and is a very popular gene editing system present.
SiRNA suppresses the expression of its target gene at the post-transcriptional level by mRNA degradation. Because of the capability of siRNA in selective targeting, much attention has been directed toward using siRNA in treatment of different cancers.
However, this conventional method cannot remove the original copy of the oncogenes and the proteins are translated again with the next generation of genetic replication.
Although both strategies of siRNA and CRISPR-Cas9 could down regulate gene editing, knockdown strategy based on CRISPR-Cas9 has the advantages of permanently silencing the target gene, high effectiveness for accurate gene editing, and low off-target likelihood.
That's the reason we chose the CRISPR-Cas9 system.
3. Because PMZH is pH sensitive, was the dye released from PMZH and escaped from endo/lysosomes? The author needs a further explanation for this.
Our response: Thanks for your question. PMZH can effectively deliver pDNA to organelles through endocytosis, and further release it into the nucleus. This good endosome/lysosomal escape ability stems from the proton sponge effect of the imidazole ring. PMZH contains imidazole groups, which successfully introduces intracellular pH buffer element, thus improving transfection efficiency. The "proton sponge" hypothesis proposed a mechanism for this behavior, indicating that the buffer activity generates osmotic swelling and lysis of endocytic compartments. Reduced lysosomal degradation and greater cytoplasmic availability of delivered DNA would then give improvements in transfection efficiency.
4. The numerical changes in ROS detection in figure4a are not given, please supplement.
Our response: Thanks for your careful reading. We had supplemented its numerical changes ( Figure 5a). (1) In this study, the authors claimed that the nanoparticles could release ions in acid lysosome conditions and catalyze intracellular H2O2 to ROS to kill cancer cells. However, the tumor microenvironment (TME) is also acidic (pH 6.5-6.8) with a high concentration of H2O2. Are the nanoparticles stable in TME? Will they also catalyze TME H2O2 to ROS? Why does they affect T cell functions?
Our response: We sincerely appreciate the insightful comments of the reviewer. First, we evaluated the degradation of PMZH in a buffer mimicking the pH of the tumor microenvironment (pH = 6.5), and it can be seen that some degradation occurs in a buffer mimicking the TME ( Figure R1a). Compared to the pH=6.5 group, PMZH undergoes faster and more complete degradation in the pH=5.5 group. The ability of PMZH to release pDNA in a pH-responsive manner is one of the most desirable properties for gene The Mn 2+ from PMZH degradation could catalyze H2O2 into hydroxyl radicals in the TME ( Figure R1b). Both intracellular and extracellular ROS produced by metal ions promotes STING pathway activation, thereby increasing T cell infiltration. As a result,   (2) Why the PMZH effectively knocked down SLC43A2 in tumors, but not in liver and kindy? The biodistribution of this nanosystem in organs should be performed.
Our response: Thanks for your question. We evaluated the biodistribution of PMZH and could see that PMZH was effectively enriched in the tumor (Supplementary Figure 17c). (4) It has been known that Mn 2+ dramatically promotes pro-inflammation response and anti-tumor immunity dependent on macrophages. Would the nanoparticles affect other immune cells (such as macrophages) and remodel the tumor immune microenvironment?
How to demonstrate the nanoparticles specifically promotes CD8+ T cells antitumor function rather than the changes of the immune microenvironment?
Our response: Thanks for your valuable comments. Your insightful suggestion was very helpful for improving our study. We evaluated the effect of PMZH on macrophage polarization ( Figure R2). PMZH could promote macrophage polarization to the anti-tumor M1 phenotype, means that PMZH could remodel the tumor immune microeniroment.
PMZH is not only promote CD8+T cells antitumor function, it had affected other immune cells. The immune system is a complex network, not separate parts, in addition, immune effector cells are interrelated. Our PMZH promoted macrophage polarization ( Figure R2), activated the STING pathway of antigen-presenting cells (Supplementary Figure 14), and promoted the anti-tumor function of T cells. These synergistic effects make our immunotherapy more optimized.
We also supplemented experiments to further confirm the effect of PMZH on T cell immunity. PMZH restored T cell immunity through the regulation of methionine metabolism. Methionine deficiency decreased H3K79me2 in T cells (Nature 2020,585, 277-282), thereby impairing T cell immunity, and we assessed the effect of different treatments on H3K79me2 expression in T cells (Supplementary Figure 10). It can be seen that PMZH could restore H3K79me2 in T cells, which helps to restore T cell immunity. (5) The authors claimed that PMZH remodels methionine metabolism in T cells. However, the authors don't provide any data to support that intervention of methionine metabolism in tumor cells could affect T cell immunity in this nanosystem. (6) lipo has cytotoxicity in cells. However, the authors showed that the lipo treatment in Figure 3C decreased the apoptosis of T cells. Please explain.
Our response: Thanks for your question. Firstly, lipo has cytotoxicity, and direct incubation with T cells using lipo can indeed lead to an increase in T cell apoptosis (Supplementary Figure 9). However, the specific operation of Figure 4c   Minor concerns: (1) The full name of ZIF should be spelled out in the manuscript.
Our response: Thanks for your careful comments. We had supplement the full names of ZIF (Mn/Zn-ZIF-8) in the manuscript.  (3) The title of the y-axis in Figure 4A should be labeled.
Our response: Thanks for your comments. We had labeled the y-axis in Figure 5a (original Figure 4a). (4) The quality of WB results should be improved (such as fig2f, fig6g).
Our response: Thanks for your comments. We had re-conducted the WB experiment to obtain higher quality images (Figure 3h, Figure 4g, Figure 5d, and Figure 7g).    Reviewer #3 (Remarks to the Author): with expertise in cancer nano-therapy 1. The authors introduced a CMZH control group in their study in vivo to exclude the effect of a nonfunctional plasmid on methionine metabolism. However, they did not provide the effect of the CMZH control group on SLC43A2 gene expression. Please supplement. In addition, compared with CMZH, PMZH effectively reduced the level of apoptosis of T cells, but the effect of PMZH and CMZH on the secretion of functional cytokines was not evaluated. This needs further research and discussion.
Our response: Thanks for your professional comments. Your insightful suggestion was very helpful for improving our study. We had supplemented the related experiments on SLC43A2 gene expression by CMZH control group, and it could be seen that CMZH did not affect the expression of SLC43A2 gene (Figure 3g-h, Figure 4f-h). We further evaluated the effect of PMZH and CMZH on the secretion of functional cytokines (Figure 4e), and PMZH promoted the secretion of TNF-P more effectively than CMZH, which was consistent with our previous conclusion that PMZH effectively restored T cell immunity through methionine metabolism regulation.  Our response: Thanks for your professional comments. We introduced a CMZH control group at the cellular level ( Figure 5), and the experimental results were consistent with those in vivo (Supplementary Figure 20). Compared with CMZH, PMZH promoted ROS generation, mtDNA release, and STING signaling activation more efficiently. Methionine restriction can enhance the sensitivity of tumor cells to oxidative stress. This phenomenon indicated that the CRISPR plasmid regulated methionine metabolism, thereby enhancing ROS production and promoting the activation of the STING signaling pathway. This is consistent with the conclusion of the previous in vivo experiment. 6. The authors should clarify why they choose plasmid DNA but not mRNA to encode the CRISPR/Cas9. Usually the immunogenicity of mRNA is much lower than plasmid DNA and can achieve higher protein expression in vivo.
Our response: Thank you very much for your valuable comments. Firstly, mRNA does not need to enter the nucleus for transcription. In addition, recent studies have also found that the addition of modified nucleotides can reduce the immunogenicity of mRNA and improve the translation efficiency of proteins. These advantages all indicated that mRNA has priority in clinical applications.
In our study, we chose plasmid instead of mRNA, mainly considering the efficiency of protein self-supply. The CRISPR / cas9 system mainly consists of cas9 components and sgRNA components. The two components can be co constructed in a single plasmid system, but using mRNA requires two separate RNAs with a large difference in length (cas9 mRNA: ~ 5000 NT; sgRNA: ~ 300 NT). Determining the ratio of cas9 mRNA and sgRNA for optimal protein self-supply efficiency and regulating the efficiency of MOF for simultaneous loading of different lengths mRNA poses a challenge for the conduct of experiments. In addition, DNA is more stable than mRNA, which may make the protein self-supply of CRISPR/Cas9 more durable.
Therefore, we chose CRISPR/Cas9 plasmid for this study.

Please check for formatting issues throughout this manuscript.
Our response: Thanks for your careful reading. We have tried our best to correct the formatting issues in the revised manuscript and re-examine the whole article to revise similar errors.
Reviewer #4 (Remarks to the Author): with expertise in cancer immunology, metabolism My major comment is related to an unspecified mechanism of action off this nanoparticle. On the one hand, there is reduced tumor methionine consumption and higher T cell function. On the other hand, the enhanced T cell function could be due a general enhancement in STING/cGAS activation or both. Some genetic loss of function experiments are essential to clarify how these nanoparticles are acting.
Our response: We sincerely appreciate the insightful comments of the reviewer. We performed further evaluations of the mechanism of PMZH affecting T cells. We assessed T cell viability and function after different treatments in wild-type 4T1 tumor-bearing mice and in STING-KO 4T1 tumor-bearing mice (Supplementary Figure 23). The results showed that the enhancement of T cell function by PMZH was more pronounced in wild-type 4T1 tumor-bearing mice. At the same time, we also supplemented the effects of CMZH and PMZH groups on T cells. We used pDNA inserting a non-targeting sgRNA as the control plasmid to construct CMZH. PMZH restored T cell immunity through the regulation of methionine metabolism.
Methionine deficiency decreased H3K79me2 in T cells (Nature 2020,585, 277-282), thereby impairing T cell immunity, and we assessed the effect of different treatments on H3K79me2 expression in T cells (Supplementary Figure 10). It can be seen that PMZH could restore H3K79me2 in T cells, which helps to restore T cell immunity.
These phenomena all indicated that the activation of the STING pathway increased T cell infiltration, while the regulation of methionine metabolism restored T cell function. The synergistic effect of the both effectively enhanced T cell immunity. A second comment is the use of PBS as a control in all the in vitro experiments. At a minimum there should be a non-targeting sgRNA as well as testing some individual components of the nano-particle itself. Ideally, the authors should do this on the tumor cells alone, T cells alone, and then both in a co-culture experiment.
Our response: Thanks to the reviewer for the professional comment. We used pDNA inserting a non-targeting sgRNA as the control plasmid to construct CMZH. Neither gene expression nor protein expression of SLC43A2 was affected by CMZH ( Figure   3g-h). In contrast to PMZH, CMZH does not have the ability to regulate methionine metabolism but the other components are all the same. We added the pDNA, Mn 2+ , Zn 2+ , and CMZH control groups in experiments in vitro as required (Figure 4a, c-e).
As can be seen, in tumor cells alone and co-cultured tumor cells, pDNA alone did not affect cell viability; The Mn 2+ /Zn 2+ mixture group and CMZH groups showed similar degrees of cell killing, while PMZH showed the strongest ability to kill tumor cells.
This illustrated that the antitumor ability of PMZH not only stems from nutrient metal ions, but is also closely related to the gene editing effect of CRISPR. In the T cell alone group, there was no significant difference between CMZH and PMZH groups in T cell killing because T cells did not rely on SLC43A2 protein for methionine transport, unlike tumor cells. While in the co-culture group, PMZH limited tumor cell methionine uptake, relieved T cell metabolic stress, and thus reduced T cell apoptosis.  Third, most, if not all, of the micrographs need to be carefully quantified and statistical analysis performed.
Our response: Thanks for your valuable comments. We had quantified and statistically analyzed fluorescence microscopy photographs (Figure 3a     Forth, while the authors do examine DC markers and frequency of DCs, careful assessment of STING activation in these DC populations needs to be performed. In other words, does the nanoparticle activate STING in DCs or tumors or another immune population (e.g. myeloid). The authors suggest this happens in DCs but do not show these data.
Our response: Thanks for your valuable comments. Your insightful suggestion was very helpful for improving our study. The activation of STING signaling in bone marrow-derived dendritic cells (BMDCs) was further evaluated by the method reported in the literature (ACS Nano 2023, 17, 5, 4495-4506). The results Figure 14) showed that PMZH could effectively enhance sting signaling in BMDCs, and promote BMDC cell maturation and secretion of IFN-Q$ thereby enhancing systemic antitumor immunity. While I am fine with some methods being deposited in the supplemental information, this is not sufficient for the reader to understand how certain experiments were performed. Therefore, the key methods of the main figures need to be placed in the manuscript. In particular, there needs to more details on the replicates and statistical test for each figure panel embedded directly in the captions.

(Supplementary
Our response: Thanks for your valuable comments. We have supplemented important experimental methods of the figures in the Methods section of the manuscript. In addition, all bar graphs had replaced with plots that feature information about the