Multifunctional metal-organic framework-based nanoreactor for starvation/oxidation improved indoleamine 2,3-dioxygenase-blockade tumor immunotherapy

Inhibited immune response and low levels of delivery restrict starvation cancer therapy efficacy. Here, we report on the co-delivery of glucose oxidase (GOx) and indoleamine 2,3-dioxygenase (IDO) inhibitor 1-methyltryptophan using a metal-organic framework (MOF)-based nanoreactor, showing an amplified release for tumor starvation/oxidation immunotherapy. The nanosystem significantly overcomes the biobarriers associated with tumor penetration and improves the cargo bioavailability owing to the weakly acidic tumor microenvironment-activated charge reversal and size reduction strategy. The nanosystem rapidly disassembles and releases cargoes in response to the intracellular reactive oxygen species (ROS). GOx competitively consumes glucose and generates ROS, further inducing the self-amplifiable MOF disassembly and drug release. The starvation/oxidation combined IDO-blockade immunotherapy not only strengthens the immune response and stimulates the immune memory through the GOx-activated tumor starvation and recruitment of effector T cells, but also effectively relieves the immune tolerance by IDO blocking, remarkably inhibiting the tumor growth and metastasis in vivo.


Responses to Reviewers' Comments
Major changes made based on reviewers' comments: (1) We preformed the animal studies again and added MOF and GOx treatments as controls based on the suggestions from the reviewer #1. Please see Figure 8-10 in the revised manuscript, and Supplementary Fig. 13-18 and Supplementary Fig. 22 in the supporting information.
(2) We provided the SEM image of MOF based on the suggestions from the reviewer #1. Please see Supplementary Fig. 1a in the supporting information.   (11) We cited some recent papers in GOx-instructed synergistic cancer therapy based on the suggestions from the reviewer #3. Please see references 2-4 in the revised manuscript. (12) We further highlighted the novelty by comparing with previous works and cited the related references based on the suggestions from the reviewer #1 in the revised manuscript. Please see page 4 and page 5, as marked with YELLOW background. 3

C: Glucose oxidase (GOx) and indoleamine 2,3-dioxygenase (IDO) inhibitor 1-methyltryptophan (1-MT) are co-loaded with metal-organic framework (MOF)-based multifunctional nanoreactor.
The authors studied this nanoreactor for starvation/oxidation improved indoleamine 2,3dioxygenase-blockade tumor immunotherapy. This manuscript does not show enough advantage over previous Gox and IDO works. A: Thanks for your criticism with instructive questions. Based on your questions, we revised the manuscript one by one as follows.

Q1. Please highlight the novelty by comparing with previous works.
A: Thanks for your good suggestions. We highlighted the novelty by comparing with previous works and cited the related references in the revised manuscript according to your suggestion. Please see page 4 and page 5 in the revised manuscript. All changes were marked with YELLOW backgrounds.

Q2. MOF is inorganic materials and it might bring long-term toxicity over other nanoreactors. What is the advantage of MOF over other nanoreactors?
A: Thanks very much for your kind reminding. On the one hand, compared to other nanoreactors, MOFs have the advantages of facile preparation, high loading efficacy of enzymes, excellent fidelity characteristics of enzyme activity, multifunctionality and good biocompatibility, making them promising systems for developing the nanoreactors [1][2][3][4]. On the other hand, tumor-activated degradable MOF nanoreactor was synthesized for the first time in this work via covalent crosslinking with ROS-susceptible agents and Mn 2+ , which can be rapidly disassembled and excreted from body in response to the rich intracellular ROS of tumor cells over other nanoreactors, minimizing the potential long-term retention toxicity of conventional nanoreactors.
More importantly, we further investigated the long-term toxicity of MOF in this work. there was no significant long-term toxicity during the evaluation period. Furthermore, there was also no evident difference on the body weight and histological analysis of the major organs of two tumor models (B16F10 tumor-bearing C57BL/6 mice and 4T1 tumor-bearing Balb/c mice) after the i.v. injection with MOF nanoreactor-based nanosystem over 18 days (Supplementary Fig. 18 and Supplementary Fig. 22), confirming again the good biocompatibility. Above primary and comprehensive in vitro and in vivo results indicated that MOF nanoreactor-based nanosystem showed no obvious toxicity, which provided a chance for in vivo therapeutic applications. The related descriptions have been added in the revised manuscript. Please see Supplementary Fig. 8, Supplementary Fig. 18, Supplementary Fig. 21 and Supplementary Fig. 22 in the supporting information, page 4 and page 20 in the revised manuscript, as marked with YELLOW backgrounds.

Q4. SEM images are better to be supplied.
A: Thanks for your kind suggestions. We provided the SEM image of MOF in the revised supporting information. Please see Supplementary Fig. 1a in the supporting information.

Q5. Can the authors characterize the MOF with HRTEM to understand the detailed structure?
A: Thanks for your suggestion. We characterized the MOF with HRTEM, described its detailed 5 structure and added the related discussions, according to your suggestion. As shown in Supplementary Fig. 1b

Q3. Since GOx could consume glucose and inhibit ATP production by starvation treatment, the level of intracellular ATP upon above treatments thus should be investigated.
A: Thanks for your kind suggestion. We investigated the level of intracellular ATP upon above treatments according to your suggestion. In addition, we also measured the levels of H2O2 production and added the related discussions, since GOx and GOx-loaded nanosystem could consume glucose, produce H2O2 and inhibit ATP production by starvation treatment. Please see page 13 and page 14 in the revised manuscript, and Supplementary Fig. 10 in the 8 supporting information. All changes were marked out with YELLOW backgrounds.

Q4. Please added the live-dead assay of tumor cells in vitro upon different treatments to comprehensively evaluate tumor killing effect mediated by nanosystem.
A: Thanks for your kind suggestion. We added the live-dead assay of tumor cells in vitro upon different treatments and described the correspondingly results, as marked with YELLOW backgrounds. Please see page 13 and page 14 in the revised manuscript, and Supplementary  Fig. 11 in the supporting information.

Q5.
In the section of in vivo biosafety study, although the authors investigated the changes of mouse body weight, major organ pathological analysis, and liver function evaluations, more biosafety experiments need to be provided, for example, the biosafety of kidney, blood compatibility, etc. A: Thanks for your kind suggestion. We performed the biosafety of kidney and blood compatibility assays, including the blood biochemical levels and hematological parameters (hematocrit value (HCT), mean platelet volume (MPV), hemoglobin (HGB), platelets (PLT),

mean corpuscular hemoglobin (MCH), red blood cells (RBC), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), mean corpuscular volume (MCV) and
white blood cell (WBC)), and kidney function indexes (blood urea nitrogen (BUN) and creatinine (CREA)) of the mice after various administrations, and provided the corresponding discussions. Please see page 20 in the revised manuscript, and Supplementary Fig. 21 in the supporting information. All changes were marked with YELLOW backgrounds.

Q6. Please add the detailed procedures about tumor re-challenge assay.
A: Thanks for your kind suggestion. We added the detailed procedures and schematic illustration about tumor re-challenge assay according to your suggestion in the revised manuscript. Please see page 31 and Figure 10d in the revised manuscript. All changes were marked with YELLOW backgrounds.

Q7. Some recent works on GOx-instructed synergistic cancer therapy should be cited.
A: Thanks for your kind suggestion. We cited some recent papers regarding GOx-instructed synergistic cancer therapy. Please see references 2-4 in the revised manuscript, as marked with YELLOW backgrounds.

Reviewer #2 (Remarks to the Author):
The authors have addressed all my concerns in the revisions.It is now suitable for publication in Nature Communications.

Reviewer #3 (Remarks to the Author):
The authors supplied another relevant tumor model (4T1 tumor cell-bearing Balb/c mouse model) and studied the in vivo immune response level and in vivo antitumor effect of their nanosystem, they observed similar results to that on B16F10-bearing C57 mouse model. Both two tumor model results primarily confirmed that MOF nanoreactor-based nanosystem exhibited good antitumor efficiency with reinforced immune response activation and relieved immune resistance, which provided a common chance for in vivo therapeutic applications. Meanwhile, the authors added MOF and GOx treatments as controls in animal studies. In general, this manuscript has been revised mostly according to the comments and suggestions of the reviewers. I recommend the acceptance of the manuscript for publication in Nature Communications.

Reviewer #4 (Remarks to the Author):
The paper by Dai et al entitled "Multifunctional metal-organic framework-based nanoreactor for starvation/oxidation improved indoleamine 2,3-dioxygenase-blockade tumor immunotherapy" describes the generation of a novel metal organic framework based multifunctional nanoreactor co-loaded with glucose oxidase (GOx) and indoleamine 2,3-dioxygenase (IDO) inhibitor 1methyltryptophan (1-MT). They show that this new nano drug is able to penetrate tumor cells in a pH dependent manner and release it's cargo and affects tumor cells. In addition, they speculate that starvation/oxidation combined IDO-blockade strengthens antitumor immune response and stimulates immune memory. They also speculate that they observe an inhibition of tumor growth and metastasis in vivo. While the premise of this work is interesting and promising the results do not support the claims of the paper. In addition the tumor immunology studies shown in the paper are a bit limited and would benefit from tumor immunology expertise. To test for immune memory mice should be implanted at least 100 days after the last treatment to test immune memory. 5-It's unclear if the treatment is influencing tumor invasion or that the observed effect is simply due to an impairment of these cells to proliferate. 6-The starvation of the cells of the cells should also in theory affect immune cells. It's unclear how this treatment doesn't affect immune cells negatively. 7-The paper needs English proof reading as it was difficult to read.

Major changes made based on reviewers' comments:
(1) We provided the cell gating strategy with assistance of a tumor immunologist, and also adjusted the presentation of immune characterization from "the flow plots" to "the flow zebra" or "the flow contour" based on the suggestions from reviewer #4. Please see Figure   8a,c and Figure 10d,d' in the revised manuscript, and Supplementary Fig. 14, Fig. 15, Fig.  17-19, and Fig. 26 in the Supplementary Information. A: Thanks for your kind suggestion. We corrected the immune characterization with assistance of a tumor immunologist according to your suggestion. First, the cell gating strategy was respectively provided and referred as follows: we firstly select cells by FSC vs SSC to remove the influence of cell debris, and then remove dead cells by live / dead staining, followed by analyzing the targeted proportion of immune cells via special antibody co-staining approach (Supplementary Fig. 14). Second, all immune characterization data shown in multiple figures using flow cytometry were supplemented with cell gating strategy respectively (i.e., Figure  8a,c vs. Supplementary Fig. 14a; Figure 10d,d' vs. Supplementary Fig. 14f; Supplementary Fig.   15 vs. Supplementary Fig. 14b; Supplementary Fig. 17 vs. Supplementary Fig. 14c;  Supplementary Fig. 18 vs. Supplementary Fig. 14d; Supplementary Fig. 19 vs. Supplementary   Fig. 14e; Supplementary Fig. 26a,c vs. Supplementary Fig. 14f) and provided with corresponding quantitative analysis chart based on flow data in the revised manuscript and Supplementary Information (Figure 8a,c vs. Figure 8b,d & Supplementary Fig. 16;  Supplementary Fig. 17-19 vs. Figure 8b,d; Figure 10d,d' vs. Figure 10e,e'; Supplementary Fig.   26a,c vs. Supplementary Fig. 26b,d). Furthermore, all the original data of the immune characterization shown in multiple or single figures using flow cytometry were provided as well. More importantly, we also adjusted the presentation of immune data from "the flow plots" to "the flow zebra" (Figure 8a,c; Figure 10d, d'; Supplementary Fig. 18, Fig. 19, Fig. 26) or "the flow contour" (Supplementary Fig. 15, Fig. 17) for the accuracy of the data, according to your suggestion. Please see Figure 8a,c and Figure 10d,d' in the revised manuscript, and Supplementary Fig. 14, Fig. 15, Fig. 17-19 and Fig. 26 in the Supplementary Information.

Q2. The tumor survival and progression are only showing a relative volume. The exact volume
should be shown, and the treatment should be started once the tumors are established. A: Thanks for your kind suggestion. The presentation of tumor volume was modified from the relative volume to the exact volume (Figure 9 b,b'). Please see Figure 9 b,b' in the revised manuscript.
Q3. The breast tumor model is grown on the flank while it should be implanted in the mammary fat pad. A: Thanks for your kind suggestion. We are very sorry for the misunderstanding caused by the missing description about the establishment of breast tumor model. Notably, 4T1 cells were inoculated subcutaneously into the right flank of female Balb/c mice rather than the mammary fat pad. This is another widely recognized and adopted breast tumor model establishment method proven by previous literature [1][2][3][4][5]. We added the related descriptions about the establishment of breast tumor model in the revised manuscript. Please see page 28.

Q4. The immune memory experiments are inconclusive. To test for immune memory mice should be implanted at least 100 days after the last treatment to test immune memory.
A: Thanks for your kind suggestion. Generally, the immune memory experiments are performed over 8-40 days after the mice are re-implanted with tumor cells, due to the limitation of the survive time of mice treated with various administrations having the tumor metastasis and recurrence [1][2][3][4][5][6]. Therefore, we additionally performed an immunological memory experiment for 50 days after the last treatment according to your suggestion (the mice treated with control/sample groups (20 per group) were generally died at 18 days after tumor resection, attributing to the inherent tumor metastasis and recurrence. About 6 mice survived at 50 days. Taking the validity and significance of the data (the number of parallel groups is at least more than 6) into account, we ended the experiments at 50 days and measured the corresponding immune memory). We discussed and explained the related results and inserted the corresponding data and figures in the revised manuscript and Supplementary Information. Please see page 20 in the revised manuscript, and Supplementary Fig. 26 in the Supplementary Information.