Microbe-Derived Extracellular Vesicles as Carriers for Doxorubicin Delivery to Colorectal Cancer Cells

Abstract

Background/Objectives: Microbe-derived extracellular vesicles (MEVs) provide a biocompatible, naturally derived platform for drug delivery. Methods: We encapsulated doxorubicin in Lactobacillus plantarum-derived EVs and evaluated their ability at delivering doxorubicin to colorectal cancer cells in vitro. Endocytosis inhibitors were used to investigate the mechanisms by which the MEVs entered the cells. Results: The MEVs maintained structural stability under physiological conditions. Cellular internalization of doxorubicin-loaded MEVs involve clathrin/caveolae-dependent endocytosis, and dynamin- and clathrin-mediated pathways. Conclusions: These findings highlight the role of the microbe–cancer cell biointerface in mediating drug uptake and enabling intracellular delivery. The study supports the potential of MEVs as nanocarriers for anticancer drugs and provides mechanistic insights into the intracellular trafficking pathways that influence drug activity.

1. Introduction

Extracellular vesicles (EVs) are nanoscale, lipid bilayer-enclosed particles actively secreted by living cells [1]. They carry a diverse repertoire of biomolecules, including proteins, lipids, nucleic acids, and metabolites, and serve as important mediators of intercellular communication [2,3]. By transferring functional cargoes between cells, EVs direct complex signaling networks and control a broad spectrum of physiological and pathological processes [4].

Microbe-derived EVs (MEVs) also play pivotal roles in intercellular communication but exert distinct biological effects [5]. In addition to their intrinsic bioactivities, EVs are increasingly recognized as potential cargos of therapeutic agents, offering advantages over synthetic nanocarriers [6,7]. MEVs are nanoscale, spherical structures that mediate both inter-microbial signaling and microbe-host interactions [8,9]. However, the understanding of their biological functions is limited [9]. By interacting with both bacterial and host cells, MEVs are not only key modulators of host physiology but also potential drug-delivery platforms [10]. MEVs dynamically interact with the host immune system regulating the systemic humoral and cellular immune response, and promote bacterial survival [11,12]. We previously demonstrated that Lactobacillus-derived EVs can modulate macrophage polarization [13]. MEVs can also transfer bacterial components to host cells, thereby influencing cellular, molecular, and immunological responses [14]. MEV subtypes have effects and roles within the tumor microenvironment [15,16]. Various strategies have been investigated for the use of EVs as drug-delivery platforms, including using different EV types [17,18], forming hybrid vesicles [19], and cell disruption-based biomimetic approaches [20]. MEVs, as with other EVs, possess the inherent capacity to encapsulate and deliver drugs, highlighting their potential as drug-delivery platforms.

In this study, we isolated MEVs derived from Lactobacillus plantarum and encapsulated doxorubicin (DOX) to evaluate their potential as nanocarriers for delivering doxorubicin to colorectal cancer cells. Furthermore, we first examined the underlying intracellular trafficking mechanism by which MEVs mediate drug delivery to colorectal cancer cells.

2. Results and Discussion

2.1. Physicochemical Characteristics of DOX-Loaded MEVs

Prior to DOX loading, the total amount of proteins in the MEVs was 1.62 ± 0.13 mg measured in 1 L of L. plantarum culture medium. The isolation and purification procedures yielded uniform MEVs (average hydrodynamic diameter 133.83 ± 0.48 nm) with a slightly negative surface charge (zeta potential −13.60 ± 0.20 mV) (

Table 1. Physicochemical properties of MEV formulations.

Formulation	Composition (MEV:DOX Ratio) a	Particle Size (nm)	PDI	Zeta Potential Value (mV)	Encapsulation Efficiency (%)
F1	5:0	133.83 ± 0.48	0.22 ± 0.007	−13.60 ± 0.20	—
F2	5:1	192.43 ± 1.68	0.24 ± 0.007	−15.23 ± 0.35	3.77 ± 0.099
F3	5:2	232.26 ± 5.03	0.18 ± 0.003	−16.80 ± 1.00	2.72 ± 0.105
F4	5:3	226.70 ± 1.67	0.17 ± 0.015	−16.30 ± 0.50	13.56 ± 0.309
F5	5:4	272.10 ± 1.43	0.17 ± 0.018	−18.20 ± 1.20	11.35 ± 0.201

DOX, doxorubicin; MEV, microbe-derive extracellular vesicle; PDI, polydispersity index. ^a The ratio of MEV to DOX was expressed as a weight ratio based on the amount of MEV quantified using a BCA protein assay and the amount of DOX.

).

Cryo-TEM imaging revealed a predominantly spherical vesicular structure (Figure 1a). The physicochemical characteristics and DOX encapsulation efficiency of each DOX formulation are shown in Table 1. DOX loading resulted in a slight, dose-dependent increase in particle size compared with unloaded MEVs (F1), whereas the zeta potential remained slightly negative (−15 to −18 mV). The PDI indicated that particle uniformity was preserved for all formulations. The encapsulation efficiency of F4 and F5 was approximately three- to four-fold higher than that of F2 and F3, but the increase was not proportional to the DOX content, and drug loading was enhanced at higher DOX concentrations. F4 and F5 yielded uniform DOX-loaded MEVs with a predominantly spherical morphology, as confirmed by cryo-TEM imaging (Figure 1b,c). Based on these findings subsequent studies focused on F4 and F5.

2.2. Physical Stability of DOX-Loaded MEVs

Before evaluating the stability of DOX-loaded MEVs, unloaded MEVs were first examined under a various range of environmental conditions. The isolated MEVs remained stable in storage for more than 1 week and maintained colloidal stability despite variations in pH and osmolarity (Supplementary Figure S1), resulting from no significant changes of size and PDI values (Supplementary Figure S2) DOX-loaded MEVs similarly showed high physicochemical stability (Figure 2). DOX-loaded MEVs (F4 and F5) maintained colloidal stability over 1 week (Figure 2a), with <10% variation in particle size, consistent with their low PDI values (Supplementary Figure S3a). DOX-loaded MEVs retained their integrity in 1.6 mM Triton X-100 (Figure 2b), and under both hypertonic, hypotonic conditions (Figure 2c), and various pH conditions mimicking the colonic environment (Supplementary Figure S4), with no significant changes in the PDI values (Supplementary Figure S3b–e). The structural stability of F4 and F5 across diverse physiological environments suggests that these formulations exhibit physicochemical robustness relevant to drug delivery applications. Their stability under conditions relevant to the colonic milieu further supports their potential as nanocarriers for delivering doxorubicin to colorectal cancer cells.

2.3. In Vitro Cytotoxicity and Cellular Uptake of DOX-Loaded MEVs

Based on their stability profiles, DOX-loaded MEVs F4 and F5 were applied to two colorectal cancer cell lines (HT-29 and HCT 116) to assess their cytotoxicity. Both cell lines exhibited a dose-dependent decrease in cell viability (Figure 3a,c). Also, no statistically significant difference in cytotoxicity was observed between DOX-loaded MEVs (F4 and F5) and free DOX at the equivalent concentration (Supplementary Figure S5). Furthermore, LysoTracker staining of late endosomes and early lysosomes showed substantial colocalization with DOX, suggesting involvement of endocytic pathways in MEV internalization (Figure 3b,d). Confocal microscopy assessment of endocytic uptake was performed using a low concentration of DOX incubated with cells for 1 h to minimize cellular perturbation, based on previous reports [19,21]. Under these optimized settings, microbe-derived MEVs were observed to undergo cellular internalization in cancer cells, predominantly through endocytosis. Based on the additional experimental results, DOX-loaded MEVs are expected to be internalized via endocytosis, followed by DOX release within mildly acidic endosomal compartments, with subsequent translocation of the released drug into the cytosol (Supplementary Figure S6).

2.4. Endocytosis Mechanisms of DOX-Loaded MEVs

To further investigate the mechanisms governing the heterologous biointerface between MEVs and colorectal cancer cells, inhibitor-based assays were performed that targeted specific endocytic pathways. To assess the inhibition of ATP-dependent endocytosis, each cell line was pretreated with the corresponding inhibitors or incubated at 4 °C. More than 85% of cellular uptake was suppressed at 4 °C (Figure 4), indicating that internalization of the MEVs was driven predominantly by an ATP-dependent endocytic process. In HCT 116 cells F4 uptake was inhibited by CytD (15.94 ± 0.65%) and dynasore (14.89 ± 1.14%) (Figure 4a), indicating dominant involvement of clathrin/caveolae-dependent endocytosis, and dynamin-mediated pathways. F5 uptake (Figure 4b), was inhibited by CytD (18.36 ± 0.10%), dynasore (15.72 ± 0.06%), and CPM (11.44 ± 0.18%), indicating clathrin/caveolae-dependent endocytosis, dynamin-dependent, and clathrin-mediated endocytosis pathways. In HT-29 cells, F4 uptake was inhibited by CPM (19.14 ± 1.03%), WMN (14.87 ± 0.93%), CytD (14.15 ± 0.89%), and dynasore (13.88 ± 0.64%) (Figure 4c), reflecting combined contributions from clathrin-mediated endocytosis, macropinocytosis, clathrin/caveolae-dependent endocytosis and dynamin-dependent pathways. F5 uptake was inhibited by CytD (26.32 ± 0.86%), CPM (23.33 ± 0.44%), Dynasore (20.09 ± 0.36%), and WMN (11.83 ± 0.71%) (Figure 4d), indicating reliance on clathrin/caveolae-dependent endocytosis, clathrin-mediated endocytosis, dynamin-dependent mechanisms, and macropinocytosis pathways. Although studies using endocytosis inhibitors to investigate internalization mechanisms are inherently limited by the in vitro setting, as well as the potential off-target effects and limited specificity of the inhibitors, making definitive conclusions remains challenging. Nevertheless, based on the findings of our study, our findings suggest that DOX-loaded MEV internalization (F4 and F5) occurs via clathrin/caveolae-dependent endocytosis, dynamin- and clathrin-mediated endocytosis, with relative pathway contributions differing according to the formulation and colorectal cell type.

2.5. Limitations of MEV Strategy

Microbe-derived extracellular vesicles (MEVs) provide a unique biointerface between microbes and cancer cells, retaining microbial membrane components and surface properties that can modulate cellular recognition and uptake via distinct endocytic pathways. This proof-of-concept study demonstrates the feasibility of MEVs for anticancer drug delivery and endocytic internalization; however, several limitations remain. Passive incubation preserved vesicle integrity but resulted in relatively low drug encapsulation, indicating a need for strategies that balance vesicle stability with efficient loading. Mechanistic insight was also limited, as omics-based analyses correlating MEV internalization with specific cellular components were not performed. Finally, the study was confined to in vitro models, and further in vivo investigations are required to evaluate the translational potential and therapeutic efficacy of MEV-mediated drug delivery in colorectal cancer. Despite these limitations, our findings highlight the potential of MEVs as a versatile platform for targeted drug delivery, motivating future studies to optimize loading strategies and explore in vivo applications.

3. Materials and Methods

3.1. In Vitro Cell Culture

Prior to separation and isolation of MEV from microorganisms, Lactobacillus plantarum (L. plantarum, KCTC3108), provided by Seong-Bo Kim (Yonsei University, Republic of Korea), was cultured in deMan, Rogosa and Sharpe (MRS) broth (BD Difco, Franklin Lakes, NJ, USA) at 37 °C in the ARA P150 chamber (Hanil Scientific, Gimpo, Republic of Korea). A 1:100 subculture was prepared and maintained under the same incubation conditions. Colorectal cancer cell lines (HT-29 and HCT 116) were obtained from the Korean Cell Bank for in vitro cancer cell studies such as cytotoxicity and cellular uptake mechanisms. Both cell lines were maintained in RPMI-1640 medium (Corning, New York, NY, USA) with 25 mM HEPES, and 25 mM NaHCO₃ supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (Corning, New York, NY, USA) at 37 °C in a 5% CO₂ incubator.

3.2. Preparation of MEVs and DOX Encapsulation

L. plantarum was cultured and harvested by sequential centrifugation at 1800× g for 20 min and 10,000× g for 20 min at 4 °C when the optical density at 600 nm reached 2.3–2.5. The resulting supernatants were passed through bottle-top filters with 0.45 µm pore size (Hyundai Micro, Seoul, Republic of Korea) and concentrated using 100 kDa MWCO Amicon ultrafiltration units (Merck Millipore, Burlington, MA, USA). Concentrated filtrates were then layered onto a sucrose gradient and ultracentrifuged at 150,000× g for 3 h using an Optima L-100K ultracentrifuge (Beckman Coulter, Brea, CA, USA).

The MEVs isolated from L. plantarum were characterized by dynamic light scattering and zeta potential measurements, and their protein contents in MEVs were quantified using the bicinchoninic acid (BCA, Thermo Fisher Scientific, Waltham, MA, USA) assay following ultracentrifugation-based purification. For DOX loading, in brief, MEVs were incubated with DOX hydrochloride (BLD Pharmatech, Shanghai, China) at 200 rpm and 37 °C for 4 h in a shaking incubator to minimize potential physical damage to the vesicles, as described previously [22]. Based on the protein content determined by BCA assay, 200 μL of MEV solution (1 mg/mL) was mixed with 0–80 μL of DOX solution (2 mg/mL). The amount of DOX incorporated into MEVs was quantified by high-performance liquid chromatography (Agilent 1100 series, Agilent Technologies, Santa Clara, CA, USA) using a C₁₈ reverse-phase column (Supersil^® 120 ODS II, 4.6 × 150 mm, 5 µm; LB Science, Gyeonggi, Republic of Korea). The mobile phase consisted of 25 mM monobasic potassium phosphate buffer and methanol (37:63, v/v), adjusted to pH 2.8, and delivered at 1 mL/min. The injection volume was 20 µL, and elution was monitored at 254 nm.

3.3. Particle Size and Zeta Potential

Particle size, zeta potential, and polydispersity index (PDI) were measured using a Zetasizer Nano ZS90 (Malvern Panalytical, Malvern, UK) at 25 °C. All measurements were performed in triplicate, and data were presented as the mean ± standard deviation.

3.4. Cryo-Transmission Electron Microscopy

The MEVs and DOX-loaded MEVs were visualized using a 200 kV Glacios2 cryo-transmission electron microscopy (cryo-TEM) instrument (Thermo Fisher Scientific, Waltham, MA, USA) [23]. In brief, 3 μL of each sample was applied to a glow-discharged holey carbon grid at 24 °C under saturated humidity. Excess liquid was blotted for 2 s, and the grids were rapidly plunge-frozen in liquid ethane cooled below −170 °C using a Vitrobot Mark IV vitrification system (FEI Company, Hillsboro, OR, USA). Images were acquired at an accelerating voltage of 200 kV.

3.5. Physical Stability of MEVs and DOX-Loaded MEVs

The storage stability of MEVs and DOX-loaded MEVs was evaluated at 4 °C in HBS buffer (25 mM HEPES, and 150 mM NaCl, pH 7.4) over 7 days. Colloidal stability under these conditions was assessed using a Zetasizer system, and particle size changes were monitored at predetermined time points. Physical stability was investigated by exposure to varying pH; Triton X-100 (0–1.6 mM), a vesicle destabilizer; and NaCl (0–200 mM). All measurements were performed in triplicate.

3.6. In Vitro Cell Viability and Cellular Uptake Mechanism

HT-29 (1.5 × 10⁴ cells/well) and HCT 116 (1 × 10⁴ cells/well) cells were seeded into 96-well plates and incubated at 37 °C for 24 h (Section 3.1). After adhesion, cells were treated with DOX or DOX-loaded MEVs. Cell viability was determined using a Cell Counting Kit-8 (CCK-8) assay (DoGenBio, Seoul, Republic of Korea) by measuring absorbance at 450 nm with a multimode microplate reader (BioTek Laboratories, Shoreline, WA, USA). Cellular uptake of DOX-loaded MEVs was visualized by confocal laser scanning microscopy (LSM800, Zeiss, Oberkochen, Germany). Cells were seeded onto cover glass-bottom dishes (SPL, Pocheon, Republic of Korea) and incubated at 37 °C for 24 h. DOX-loaded MEVs were then added and incubated for 1 h. Lysosomes were stained with LysoTracker™ Yellow HCK-123 (Invitrogen, Waltham, MA, USA) at a concentration of 75 nM for 5 min, and nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Fluoroshield, excitation 353 nm, emission 465 nm). DOX was detected at excitation 565 nm and emission 576 nm, and LysoTracker fluorescence was recorded using the appropriate excitation and emission settings. For uptake mechanism studies, cells were seeded into 6-well plates and incubated for 24 h. Before treatment with DOX-loaded MEVs, the cells were pre-incubated for 1 h at 4 °C or with the respective endocytosis inhibitors at 37 °C: dynasore (80 µM, Sigma-Aldrich, St. Louis, MO, USA), pertussis toxin (PTX, 100 ng/mL, Invitrogen, Waltham, MA, USA), wortmannin (WMN, 100 ng/mL, Invitrogen, Waltham, MA, USA), methyl-β-cyclodextrin (MβCD, 5 mg/mL, Sigma-Aldrich, St. Louis, MO, USA), cytochalasin D (CytD, 10 µg/mL, Sigma-Aldrich, St. Louis, MO, USA), and chlorpromazine (CPM, 10 µg/mL, Sigma-Aldrich, St. Louis, MO, USA) [24,25].

3.7. Statistical Analysis

Results were reported as the mean ± standard deviation. Comparisons between two groups were performed using t-tests, and comparisons among more than two groups were performed using one-way analysis of variance (ANOVA). Statistical analyses were performed using Social Science Statistics for Student’s t-test and StatsKingdom for one-way analysis ANOVA. Two-tailed p values less than 0.05 were considered statistically significant.

4. Conclusions

MEVs offer a stable, biocompatible platform for the delivery of DOX to colorectal cancer cells in vitro. Their internalization mechanisms are mainly mediated by ATP-dependent endocytosis, predominantly through clathrin/caveolae-dependent endocytosis, dynamin- and clathrin-mediated pathways. Our findings suggest that MEVs could be potential carriers for chemotherapeutic agents.