Abstract

Micro-nano biorobots based on bacteria have demonstrated great potential for tumor diagnosis and treatment. The bacterial gene expression and drug release should be spatiotemporally controlled to avoid drug release in healthy tissues to induce toxicity. Herein, we describe an alternating magnetic field-manipulated tumor-homing bacteria developed by genetically modifying engineered Escherichia coli with Fe₃O₄@lipid nanocomposites. After accumulating in orthotopic colon tumors in female mice, the paramagnetic Fe₃O₄ nanoparticles enable the engineered bacteria to receive and convert magnetic signals into heat, thereby initiating expression of lysis proteins under the controlled of a heat-sensitive promoter. The engineered bacteria then lyse, releasing its anti-CD47 nanobody cargo, that is pre-expressed and within the bacteria. The robust immunogenicity of bacterial lysate cooperates with anti-CD47 nanobody to activate both innate and adaptive immune responses, generating robust antitumor effects against not only orthotopic colon tumors but also distal tumors in female mice. The magnetically engineered bacteria also enable the constant magnetic field-controlled motion for enhanced tumor targeting and significantly increase therapeutic efficacy. Thus, the gene expression and drug release behavior of tumor-homing bacteria can be spatiotemporally manipulated in vivo by a magnetic field, achieving tumor-specific CD47 blockage and precision tumor immunotherapy.

Figure 1. (A) Design of the alternating magnetic field-manipulated tumor-homing bacteria and its analogy with a machinery robot. (B) The assembly of five functional modules of a robotic system: “active navigation”, “signal decoding”, “signal feedback”, “signal process” and “signal output”.

The importance of spatiotemporal manipulation

There is a need to achieve the spatial or temporal control of bacterial gene expression. The in vivo manipulation of bacteria robots can be vitally beneficial in the treatment of disease. From the first generation of natural bacteria-based cancer therapy to the second generation of genetically engineered bacteria, and further to the third generation of customized nanomaterial-assisted bacteria robots, the functions of therapeutic bacteria are becoming increasingly sophisticated to afford promising therapeutic efficacy. However, the lack of effective strategies for in vivo manipulation is one of the major obstacles to the clinical use of bacteria robots. For the native oncolytic strains, Clostridium and Salmonella, that regress tumors by bacterial cytotoxicity, it is essential to limit these strains to specifically colonize and proliferate in tumors rather than normal tissues to avoid infection-associated toxicity. For genetically engineered Salmonella, Listeria and Clostridium, which can express antitumor cytokines or toxins in tumors, the spatiotemporal control of protein synthesis and secretion is essential to avoid side effects from toxic drugs. Therefore, despite their different anticancer mechanisms, precise control systems are crucial for bacteria to be effective therapeutic robots.

Specially for the CD47 blockade drug, it is also important to allow the bacteria to release the CD47 drug in a spatiotemporally controlled manner. The CD47 drug has high blood toxicity. If it is continuously secreted in the tumor, it may cause the drug to influx into the blood and cause side effects. During clinical use, for drugs with relatively high toxic and side effects, the dosage, frequency of administration, and interval of administration will be adjusted at any time according to the degree of side effects of the patient. If the CD47 drug was continuously released, it would be inconvenient to adjust the dosage and frequency of administration. Once the patient has side effects, it is difficult to quickly reduce the dosage. However, for our method, after the bacteria are injected once, the drug is released once, and the dosing interval can be adjusted very conveniently. In summary, the advantage of our bacterial drug delivery system is its convenience to adjust the dosage and frequency of administration, which is very important for therapeutic drugs with severe side effects.

Furthermore, we demonstrated the bacterial lysates could largely benefit the CD47 blockade drug. The bacterial lysates stimulate the type I IFN pathway in innate immune cells (e.g., DCs) through the activation of toll-like receptors (TLRs), while the CD47 drug block the “don’t eat me” signaling pathway and increase phagocytosis of tumor cells by macrophages and DCs; the activation of the type I IFN pathway and increased uptake of tumor antigen ultimately promotes antigen presentation and adaptive immune responses. This synergistic immune activation generated an excellent antitumor effect against not only orthotopic colon tumor but also abscopal subcutaneous tumor.

Figure 2. Illustration of potential immune responses induced by bacterial lysates. (A) The alternative magnetic field (AMF) induces the lysis of bacteria in colon tumors. (B) Bacterial lysates recruit and/or activate macrophages, neutrophils, natural killer (NK) cells and/or dendritic cell (DCs). The release of anti-CD47 nanobody (CD47nb) blocks the “Don’t eat me” pathway and enhances the phagocytosis of tumor cells by macrophages. (C) Bacterial lysates also activate the type I IFN pathway and adaptive immunity. TLRs, Toll-like receptors.

The advantage of spatiotemporal manipulation by magnetic field

Controlling bacterial gene expression in vivo by magnetic field exhibited significant advantages compared with the previous existing approaches. Inducible promoters are one of the common strategies to control bacterial gene expression. Hypoxia- and low pH-inducible promoters (e.g., the fumarate and nitrate reduction [FNR] system) have been introduced into bacteria, which can then respond to the hypoxic or acidic microenvironment of solid tumors to achieve spatial control of bacterial activity; however, it is still impossible to achieve temporal control with these systems. On the other hand, exogenously applied transcriptional inducers (L-arabinose, acetyl salicylic acid or tetracyclines) can tightly regulate the corresponding inducible promoters for controlling bacterial colonization or gene expression to achieve temporal control of bacterial activity, while unable to achieve spatial control. Radiation-inducible promoters (e.g. RecA) can simultaneously achieve temporal and spatial control, but lead to radiation damage of normal tissues and the bacteria (Nature Review Cancer 2018, 18, 727-743; Nature Reviews Cancer 2010, 10, 784-793). Blue light-inducible promoters (e.g. pDawn) can also affect temporal and spatial control, but the tissue penetration of blue light is shorter than one millimeter (Cell Reports 2021, 36, 109690). Magnetic field has superior tissue-penetration compared with light, and is virtually harmless compared with radiation (Advanced Drug Delivery Review 2019, 138, 326-343). Magnetic field also has the advantages of achieving non-invasive, real-time, spatiotemporal and exogenous manipulation, in contrast to hypoxia-/pH-inducible promoters. Therefore, Magnetic field has significant advantages compared with light, radiation and other inducible promoters.

Our method of spatial-temporally controlling gene express in vivo possibly has promising application in various scenarios. Optogenetics achieves the light control of neurons via transfecting cells with light-gated ion channels, which has advanced the entire field of neurobiology. However, the current methods around optogenetics are invasive, required fiber optics implantation into the brain. Magnetogenetics can achieve the non-invasive modulation of neurons using magnetic fields, but the method of fusing the iron-binding protein, ferritin, to transient receptor potential (TRP) channels in the neuronal membrane has been called into question for being contrary to physics principles (Nature Methods 2016, 13, 900-901). The method we described in the manuscript provided a possible mechanism to achieve the modulation of neurons using magnetic fields. Just in a few months ago, Jacob T. Robinson et al. reported a similar study using magnetic nanoparticle heating of temperature-sensitive ion channels enables remote, non-invasive activation of neurons for deep-tissue applications (Nature Materials 2022, 21, 951–958), demonstrating the great utility of our reported method. We also described its potential application for controlling the activity of chimeric antigen receptor T cells (CAR-T cells) in the section of “Discussion”. Magnetic field-manipulation of the activation of engineered T cells possibly facilitate the design of safer cell therapies with improved on-target off-tumor activity and mitigated side effects. The application for CAR-T cells has not been reported until now.

Another important advantage of our method is that alternating magnetic field (AMF)-manipulated bacteria (AMF-Bac) can also achieve constant magnetic field (CMF)-controlled motion. The ferromagnetism of Fe₃O₄ nanoparticles enables directional movement of AMF-Bac in CMF. With the guidance of CMF, more AMF-Bac can target to tumors. The sequential combination of CMF-controlled motion and AMF-controlled drug release can significantly enhance the therapeutic efficacy of AMF-Bac. Specifically, AMF-Bac can exhibit biased and directional motion towards tumors upon the guidance of CMF after administration; once reaching tumors, AMF-Bac can initiate the gene expression of BLPs (bacterial lysis proteins) for bacterial lysis to release the therapeutic cargo (CD47 blockade drug).

For more details of our work, please see the original article: "Modular-designed engineered bacteria for precision tumor immunotherapy via spatiotemporal manipulation by magnetic field" in Nature Communications.