Close this search box.

Bacterial protoplast-derived nanovesicles carrying CRISPR-Cas9 tools re-educate tumor-associated macrophages for enhanced cancer immunotherapy – Nature Communications

Ethical statement

This research complies with all relevant ethical regulations. All animal experimental procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Ethical Board of Nanjing University (IACUC-2005005-2).

Chemical and biological reagents

1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-Succinimidyl Ester (DSPE), mPEG2000-OH and Boc-NH-PEG2000-NH2 were purchased from Ruixi Biological Technology CO., LTD (Xi’an, China). Isopropyl β-D-1-thiogalactopyranoside (IPTG), the BCA protein assay kit, PBS, and the E. coli BL21 (DE3) strain were obtained from Sangon Biotech (Shanghai, China). IPI549, purchased from MedChemExpress LLC (Shanghai, China), was dissolved in a 5% 1-methyl-2-pyrrolidinone solution in polyethylene glycol 400 for in vivo use. Other reagents were sourced from Sigma-Aldrich (St. Louis, MO, USA). 4T1 cells (catalog number (cat. no.): SCSP-5056), Raw 264.7 cells (cat. no. SCSP-5036), 293 T cells (cat. no. SCSP-502) were obtained by Cell Bank, Chinese Academy of Sciences (Shanghai, China). MC38 cells (cat. no.: 1101MOU-PUMC000523) was obtained from the Cell Resource Center, Peking Union Medical College (Beijing, China).

Mouse tumor model

Female BALB/c mice and C57BL/6 J mice were procured from Beijing Vital River Laboratory Animal Technology Co. Ltd (Beijing, China). C57BL/6Smoc-Tlr9em1Smoc (TLR9 KO) mice were obtained from Shanghai Model Organisms Center Inc. (Shanghai, China). All animals were housed in a specific pathogen-free (SPF) environment with 21 ± 2 °C and a relative humidity of 55 ± 10%, with free access to standard food and water. For maximal tumor burden, we complied with the guideline of Animal Ethical and Welfare Committee of Nanjing University and the maximal tumor size did not exceed 2000 mm3. To establish the mouse breast cancer model, 100 μL PBS containing 1 × 106 4T1 cells were subcutaneously injected into the right back of 6-week-old female BALB/c mice. For the colorectal cancer model, 1 × 106 MC38 cells in 100 μL PBS were subcutaneously injected into the right back of 6-week-old female C57BL/6 J mice or 6-week-old female TLR9 KO mice.

Preparation and characterization of sgPik3cg-DHP/DGA-NV

E. coli harboring CRISPR plasmids (BPK764) were cultured in LB medium at 37 °C. When the OD 600 reached 0.5, isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to induce the expression of the geneome editing system. The bacteria were then cultured overnight at 25 °C with shaking at 200 rpm and finally resuspended in 50 mM Tris-HCl buffer (pH 8.0) containing 20% (w/v) sucrose. Lysozyme (final concentration:1 mg/mL) and EDTA (final concentration: 50 mM) were added to the cell suspension and it was further incubated at 37 °C for 35 min to obtain protoplasts. The protoplasts were washed three times with the fresh Tris-HCl buffer to remove the remaining outer membrane components49.

The synthesis protocols for a pH-responsive phospholipid derivative (DHP) and a TAM-targeted phospholipid derivative (DGA) were provided in the supplementary information. To create protoplast-derived NVs, protoplasts and DHP/DGA were transferred into a mini-extruder (Avanti Polar Lipids, Birmingham, AL, USA) and passed through 5, 1, and 0.4 μm polycarbonate membrane filters (Whatman, Maidstone, UK) sequentially. The crude extrudate was further purified using iodixanol density-gradient ultracentrifugation (100,000 × g for 2.5 h at 4 °C), and NV fractions were collected from the interface between 50 % and 10 % iodixanol (Axis Shield Diagnostics Ltd, Dundee, Scotland). Finally, the NVs were washed twice, resuspended in PBS, and stored at −80 °C until use. To determine the decoration efficiency of DHP/DGA in NVs, fluorescence-labeled DHP/DGA were added during the extrusion process and the collected NVs were used for flow cytometer and fluorescence resonance energy transfer (FRET) analysis. OMVs were prepared following a previous report50.

The diameters of NVs after undergoing multiple freeze-thaw cycles, NVs pre-incubated with fetal bovine serum for 24 h, or NVs pre-incubated in PBS for 8 days were examined by Nanosight NS300 (Malvern, United Kingdom). The morphology, Zeta potential and polydispersity index (PDI) of NVs were examined by TEM (JEM-2100, JEOL, Tokyo, Japan), Zetasizer Nano-Z (Malvern) and NanoBrook 90Plus Zeta (Brookhaven, Holtsville, NY, USA), respectively. The yield of NVs was calculated by quantifying the protein content of NVs produced from a known amount of E. coli. To assess the pH-responsive ability of DHP, NVs incorporated with RhB-DHP were incubated with PBS (pH 7.4 or pH 6.5) for flow cytometry and FRET analysis. Additionally, 50 μL PBS containing 5 × 109 CFSE-labeled sgPik3cg-RhB-DHP/DGA-NVs were directly injected into 4T1 tumor tissues and tumor-adjacent tissues, respectively. The tissues were harvested for fluorescence photography, H&E staining and ELISA assay (IL-1β and IL-6) 24 h post-injection.

Component identification of sgPik3cg-DHP/DGA-NV

To analyze the nucleotide content in NVs, DNA and RNA isolated from E. coli and NVs were assessed using LabChip GX (PerkinElmer, Waltham, MA, USA) and Agilent 2100 Bioanalyzer (Santa Clara, CA, USA). The total protein from E. coli, E. coli protoplasts, OMVs and NVs was extracted and subjected to LC-MS analysis on a Shimadzu UFLC 20ADXR HPLC system connected to an AB Sciex 5600 Triple TOF mass spectrometer (AB SCIEX, Waltham, MA, USA). Subsequently, MaxQuant software was employed to analyze protein mass spectrometry data. The intensity-based absolute quantification (iBAQ) method was applied to rank the abundance of distinct proteins within each group. The protein levels across different samples were compared by label-free quantification (LFQ) intensity, represented by a normalized intensity profile generated using a specific algorithm51.

The encapsulation of sgPik3cg, Cas9 protein and CpG-rich DNA fragments by NVs was assessed through western blotting and PCR. Absolute quantification of Cas9 protein and sgRNA copies in NVs were conducted using ELISA (Cell Biolabs, Inc, San Diego, CA, USA) and qRT-qPCR, respectively. A luciferase reporter assay was applied to evaluate the stimulatory effect of NVs-encapsulated CpG-rich genomic DNA fragments on the TLR9 pathway. Furthermore, endotoxin content, cell toxicity, hemolysis assay and in vivo toxicity of NVs were performed to examine the biosafety of NVs. Detailed information on the above-mentioned methods is provided in the supplementary information.

Cellular uptake of NVs in M2-BMDMs and subcellular distribution

0.5 mL of medium containing 5 × 105 M2-BMDMs was seeded into 24-well plates overnight before the addition of NVs. Then, the medium was removed, and 6 × 108 CFSE stained sgPik3cg-NV, sgPik3cg-DGA-NV, or sgPik3cg-DHP/DGA-NV treated with either PBS at pH 7.4 or pH 6.5 in 0.5 mL DMEM medium, was introduced into the wells. After incubation at 37 °C for 3 h, the medium containing NVs was removed, and M2-BMDMs were washed twice with PBS before fresh medium was added. Uptake efficiency was examined using a ZEISS confocal microscope (ZEISS LSM 980, Oberkochen, Germany), a flow cytometer (Thermo Fisher Attune, Waltham, MA, USA), and the fluorescence intensity of the treated cell lysate was quantified using a microplate reader (Thermo Fisher, Em: 488 nm; Ex: 530 nm). To investigate the role of MGL-mediated endocytosis in the uptake of DGA-labeled NVs, potential competitors such as GalNAc or GlcNAc (100 mmol/L) were pre-incubated with BMDMs for 1 h prior to adding NVs. Moreover, BMDMs were transfected with siRNA targeting Mgl1/2 for 48 h before the addition of NVs.

To investigate the subcellular localization of NVs in M2-BMDMs, CFSE-labeled DHP/DGA-NVs were incubated with M2-BMDMs for 3 h. Subsequently, the cells were stained with LysoTracker Red (Beyotime Biotechnology, Shanghai, China) to visualize lysosomal escape, and observations were made using a ZEISS confocal microscope. The co-localization of NVs and lysosomes was analyzed using ImageJ Software. To assess the nuclear import of Cas9 protein, immunofluorescence staining and western blotting assays were conducted. Additionally, the presence of nuclear sgRNA, total sgRNA, and CpG-rich DNA sequences in M2-BMDMs, brought in by NVs, was characterized 12 h (except CpG-rich DNA sequences for 3 h), after the addition of NVs using PCR amplification and agarose gel electrophoresis. The information of primers and antibodies used was shown in Supplementary Data 1011.

Cell treatments

To examine the effects of NVs, 1.5 × 106 M2-BMDMs were seeded in 6-well plates and allowed to incubate for 24 h prior to the addition of sgPik3cg-DHP/DGA-NVs. These NVs were pre-treated with pH 6.5 PBS, and then were introduced to BMDMs for 6 h, followed by the incubation with fresh medium for 42 h. In certain experimental conditions, M2-BMDMs were treated with ODN1826 (a classical mouse TLR9 agonist, final concentration: 10 μmol/L) and IPI549 (a PI3Kγ inhibitor, final concentration: 1 μmol/L) for 48 h. To further investigate the role of NVs encapsulated CpG-rich DNA sequences in macrophage repolarization, 10 μmol/L ODN2088 (a TLR9 inhibitor) was added to BMDMs 30 min prior to NVs treatment and co-incubated for 48 h. A T7E1 assay was conducted to examine gene-editing efficiency, and next-generation sequencing (NGS) was performed to evaluate induced indel patterns. Western blotting was utilized to determine the protein levels of molecules related to the PI3Kγ and TLR9 pathways. The phenotype of BMDMs was characterized by quantifying mRNA levels of M1/M2 markers (iNOS and Arg1), performing ELISA assays for cytokines, and analyzing M1/M2 markers (CD86 and CD206) using flow cytometry. Detailed methods are provided in the supplementary information.

To assess the persistence of the genome editing system in macrophages, a conditioned medium (CM) was prepared using 4T1 cells. These cells were cultured in serum-free 1640 medium for 24 h, and the resulting supernatant was filtered through a 0.22 μm membrane filter to create the 4T1 cell-CM. Next, 1.5 × 106 BMDMs in 6-well plates were treated with 1.8 × 109 sgPik3cg-DHP/DGA-NVs for 6 h and further cultured in fresh DMEM medium for an additional 18 h. Subsequently, the BMDMs were exposed to 4T1 cell-CM for 24 h. Additionally, another group of BMDMs was treated with 1 μmol/L IPI549 for 24 h, followed by the addition of 4T1 cell-CM containing an equivalent concentration of IPI549 for another 24 h. Finally, the collected macrophages were analyzed for phenotype using RT-qPCR and flow cytometry.

Biodistribution assay of NVs in mouse tumor models

To establish the most effective dosing strategy for NVs in mouse tumor models, we initially conducted a series of preliminary tests with varied doses and schedules. The optimal regimen was determined to 1 × 1010 NVs per mouse, administered bi-daily. This specific dosage and frequency were found to maximize targeting efficiency and gene-editing capability in TAMs. In exploring the biodistribution of NVs within these models, we injected 4T1 tumor-bearing mice intravenously via the tail vein with 100 μL of PBS containing 1 × 1010 Cy5-labeled sgPik3cg-NVs, sgPik3cg-DGA-NVs, or sgPik3cg-DHP/DGA-NVs. For control purposes, a group of mice received only PBS. To monitor in vivo distribution following repeated doses, the aforementioned Cy5-labeled sgPik3cg-DHP/DGA-NVs were administered every two days. The IVIS Spectrum system (PerkinElmer) was then employed to capture images of mice and their organs at specified time points. Tumor tissues, other organs and blood were collected for immunofluorescence staining, quantification of fluorescence intensity, and flow cytometry. Additionally, we isolated TAMs to assess gene-editing efficiency at different time points. For detailed methods, please refer to the supplementary information.

Anti-tumor activity of sgPik3cg-DHP/DGA-NVs

Tumor-bearing mice were randomly assigned to receive 100 μL PBS containing 1 × 1010 different type of NVs (sgPik3cg-NVs, sgPik3cg-DHP/DGA-NVs, and sgControl-DHP/DGA-NVs) via tail vein injection every two days. Alternatively, they received IPI549 solution by oral gavage once a day at 15 mg/kg, starting on day 4 after tumor cell inoculation. In some cases, mice were co-treated with other reagents: 1) ODN2088 or its corresponding control (50 μg per mice every two days), 2) anti-TNF-α antibody (500 μg per mouse every two days), 3) anti-CD8α antibody (100 μg per mouse every 3 days), 4) anti-PD-1 antibody (250 μg per mouse every 3 days) via intraperitoneal injection. The relative tumor volume (RTV) = (tumor volume on measured day)/(tumor volume on day 0). TGI ratio was calculated as following described: TGI (%) = [1 − (RTV of the treated group)/(RTV of the control group)] × 100 (%)52. Tumor volume was calculated using the formula: (length × width2)/2. Tumor tissues were excised, weighed, and used for H&E staining, immunopathological staining, RT-qPCR assay, transcriptome sequencing, and ELISA assay on day 16 for the 4T1 animal model and day 19 for the MC38 animal model after tumor cell inoculation. Tumor and peripheral blood leukocytes were harvested for flow cytometry. TAMs were purified for ELISA assay and western blotting analysis. Body weight change during treatment, biochemical indicator assay (Jiancheng Bioengineering, Nanjing, China), and histological examination were performed to evaluate the in vivo safety of NVs. Detailed methods are provided in the supplementary information.

Statistical analysis

Results are expressed as the Mean ± SD. Data were processed in GraphPad Prism 8 software (GraphPad Software Inc. La Jolla, CA, USA) by two-tailed Student’s t-test, Mann-Whitney test, one-way ANOVA, two-way ANOVA or Kruskal-Wallis test. Survival rates were analyzed using a survival curve with the Log-rank (Mantel-Cox) test. The exact sample size and statistical test for each experiment are outlined in the corresponding figure legends. Statistical significance was considered when P < 0.05, and “ns” indicates no significance.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.