Magneto-aerotactic bacteria deliver drug-containing nanoliposomes to tumour hypoxic regions

Abstract

Oxygen-depleted hypoxic regions in the tumour are generally resistant to therapies. Although nanocarriers have been used to deliver drugs, the targeting ratios have been very low. Here, we show that the magneto-aerotactic migration behaviour of magnetotactic bacteria, Magnetococcus marinus strain MC-1 (ref. 4), can be used to transport drug-loaded nanoliposomes into hypoxic regions of the tumour. In their natural environment, MC-1 cells, each containing a chain of magnetic iron-oxide nanocrystals, tend to swim along local magnetic field lines and towards low oxygen concentrations based on a two-state aerotactic sensing system. We show that when MC-1 cells bearing covalently bound drug-containing nanoliposomes were injected near the tumour in severe combined immunodeficient beige mice and magnetically guided, up to 55% of MC-1 cells penetrated into hypoxic regions of HCT116 colorectal xenografts. Approximately 70 drug-loaded nanoliposomes were attached to each MC-1 cell. Our results suggest that harnessing swarms of microorganisms exhibiting magneto-aerotactic behaviour can significantly improve the therapeutic index of various nanocarriers in tumour hypoxic regions.

Figure 1. Assessment of the specificity of MC-1 antibody in HCT116 colorectal xenografts in SCID Beige mice

To ascertain specificity of MC-1 antibody, xenografts were injected with either MC-1 (a) or PBS (b). An unrelated IgG isotype (6.6 µg/ml) was used (c and d) along with the Polyclonal Rabbit MC-1 antibody (e and f) on two adjacent sections of the same xenograft. In both cases, the reaction was revealed using either FITC (green) or Texas red (red) secondary antibodies. Specific labelling could be observed only in the MC-1 injected xenografts incubated with MC-1 antibody. Control IgG was unable to identify MC-1. These results confirmed the specificity of the antibodies to detect MC-1 cells in the tumours.

Figure 2. MC-1 cells are preferentially located in the hypoxic regions of the xenografts

To determine the exact location of MC-1 with regards to the local oxygen tension in tissue, a hypoxyprobe-specific antibody labelled with biotin was used, as illustrated in panels a and b, where strands and islands of remaining hypoxic tumour tissue give a positive brownish reaction in the vicinity of necrotic areas, as illustrated in panel b. The MC-1 antibodies specifically detect MC-1 that is visualized by staining with anti-rabbit FITC-labelled secondary antibodies (panel c, 10× and 40×). In panel c, the adjacent sections of the same xenografts were incubated with MC-1 antibody and a FITC-conjugated specific secondary antibody to label MC-1, thus confirming the presence of abundant bacteria in areas of lowered oxygen concentration (10× and 40× images of MC-1 FITC fluorescently-labelled), high-resolution images were acquired manually using a DP71 digital camera mounted on a Olympus BX61 motorized upright microscope with fluorescence (FITC/TXRED/DAPI). In panel d, a representative portion of a paraffin-embedded section reprocessed for transmission electron microscopy (TEM) is illustrated, to identify the MC-1 bacteria according to their typical ultra-structural features including presence of magnetosomes. The representative pictures were selected from stained sections obtained using 10 different tumours isolated from the 10 tumour-bearing mice collected in three independent experiments.

Figure 3. Penetration of live MC-1 with and without magnetic field exposure in HCT116 xenografts following a peritumoral injection

a, Magnetotaxis directional control system used to generate the magnetic field necessary to guide the MC-1 cells towards the xenograft. b, MC-1 peritumoral injection in HCT116 tumour xenograft in mice and representation of the applied directional magnetic field used in this study to direct the bacteria towards the xenograft. The directional magnetic field, B⃗ was aligned towards the centre of the tumoral volume. c and d, MC-1 average distribution in transverse tumour (n=10) sections visualized by staining with anti-rabbit FITC-labelled secondary antibodies at 2, 4, 6 and 8 mm from the peritumoral injection site for (c) group I and (d) group II. The results show not only a significant increase of the targeting ratios of MC-1 cells that was magnetically-guided compared to non magnetically-guided MC-1, but also led to a good distribution in the tumoral volume and more specifically in the tumor hypoxic regions. The higher populations of MC-1 cells of group I contribute in targeting all hypoxic regions in the tumour. e, Standard deviation and average number of MC-1 for the transverse tumor sections in groups I and II. Statistical analysis was performed using unpaired t-student. Significant difference was considered for **P < 0.01. The results clearly show a significant increase of MC-1 cells for group I in the tumoral volume where magnetotactic directional guidance was used prior to remove the magnetic field to enable aerotactic displacements of the MC-1 cells towards the hypoxic areas once inside the xenograft. Without magnetotactic directional guidance, only a smaller proportion of MC-1 cells that randomly swam in the direction of the xenograft could potentially be influenced by oxygen gradients in the tumoral tissues. f, Summary of the total average number of injected and detected MC-1 for all tumours in groups I and II. Data represent as Mean + S.D. (n = 10 for each of c, d, e, and f). Transverse tumor sections were scanned at 40× magnification and the numbers of bacteria in the tumours were estimated by image processing techniques (see Methods: MC-1 count). The results show that a significant number of the peritumorally injected MC-1 being magnetically-guided using a relatively simple static directional magnetic field was able to penetrate the xenograft, unlike for the bacteria which were not guided into the tumour. These results likely represent a simplified case scenario as many other more sophisticated modulated directional magnetic fields could be investigated to increase further the targeting ratios achieved in this study.

Figure 4. Superior penetration of MC-1 cells over passive diffusion in HCT116 xenografts demonstrated by two methods

(1) peritumoral injection of a mixture of MC-1 cells and similar size polymer fluorescent microspheres (a,b,c,d) and (2) peritumoral injection of a mixture of dead and live MC-1 (e). For both methods, the injection was followed by 30 min of magnetic guidance directed towards the centre of the tumoral volume. a, 20× images of MC-1 Texas red fluorescent and FITC fluorescent microspheres at various tumoral depths in a longitudinal tumour section showing the superior penetration of MC-1 cells inside the tumour. The small wide strip marks the site where the tumour slice was cut into two parts – two rectangles: one larger and one smaller. b, Quantity of fluorescent microspheres vs. live MC-1 at various tumoral depths related to the experiment in (a) and depicting the superior penetration of live MC-1 well passed the diffusion limits of the fluorescent microspheres. c, 20× images of MC-1 Texas red fluorescent and FITC fluorescent beads at 1, 3, and 6 mm inside the HCT116 xenograft, again showing the much higher penetration depths of the live MC-1. d, Co-localization of MC-1 bead complexes at a tumoral depth of 4 mm suggesting the capability of MC-1 cells to transport large therapeutic payloads deep in tumoral tissues. Dead MC-1 are unable to move in the magnetic field, and are located in the far left section, whereas the live MC-1 which typically migrate along the magnetic field are in the right section confirming that they indeed moved along the direction of the magnetic field. e, 20× images of live and dead MC-1 at different tumoral depths showing the superior penetration inside the tumor of live cells.

Figure 5. Targeting ratios of MC-1-LP in HCT116 xenografts

a, Scanning electron microscopy images of unloaded MC-1 vs. MC-1-LP (right) with ~70 SN-38 loaded liposomes (diam. ~170 nm) attached to the surface of each cell. b, Estimated mean ratios of MC-1-LP found in the tumour after targeting with a directional magnetic field. The total numbers of bacteria in the tumours were estimated by direct count of the MC-1 cells obtained from homogenization of the tumours or image processing techniques of transverse tumor sections. The data represents the mean ± S.D. (n = 10 tumours). Number of bacteria was calculated using 10 different tumours isolated from the 10 tumour-bearing mice collected in three independent experiments. Five tumours were homogenized to obtain exact count and the other five were cut into numerous sections, stained and counted. The preliminary results already show a mean targeting ratio of 55% similar to the targeting results achieved with the unloaded MC-1 under the same experimental and physiological conditions. These results suggest that the MC-1 can be loaded without significantly affecting the targeting ratios in tumours. c, Transverse tumour sections of MC-1-LP after targeting, images of each section were acquired using a fluorescence optical microscope equipped with a 40× magnification objective lens. The images show a good distribution of the loaded MC-1 cells throughout the tumour. d, Example of MC-1-LP distribution inside four different tumors. The differences in the locations of the hypoxic regions among the various xenografts lead to variations of the distributions of the loaded MC-1 between the targeted xenografts.