Encapsulation of doxorubicin in magnesium acetate liposomes as a pH-sensitive drug carrier for tumor therapy

Summary

The pH-sensitive drug release is an appealing strategy to enhance the therapy efficacy of anti-tumor liposomal drugs. In this study, we constructed the pH-sensitive magnesium-doxorubicin (Mg-DOX) liposomes by remote-loading of DOX into MgAc₂ gradient liposomes. The prepared 100 nm Mg-DOX liposomes (Mg-DOX-Lip100) and folate receptor targeting Mg-DOX liposomes (FA-Mg-DOX-Lip100) were stable at physiological pH condition but gradually released DOX at acidic media. The FA-Mg-DOX-Lip100 exhibited much higher uptake by tumor cells and cytotoxicity in vitro than the Mg-DOX-Lip100. The Mg-DOX liposomes also showed stability in circulation in mice and delayed the growth of orthotopic EO771 tumor in C57BL/6 mice similar to Doxil-like liposomes at DOX dose of 5 mg/kg body weight every 4 days for 4 times, without observable morphologic change of the dissected organs at experiment endpoints. Thus, the pH-sensitive Mg-DOX liposomes have been successfully constructed and approved to be a potential antitumor delivery system to treat acidic tumors.

Graphical abstract

Highlights

• •Magnesium-doxorubicin (Mg-DOX) liposomes have been prepared by remote loading method
• •Mg-DOX liposomes are stable in physiological pH condition but release drug in acidic medium
• •FR targeted Mg-DOX liposomes show enhanced DOX uptake in tumor cells and cytotoxicity in vitro
• •Mg-DOX liposomes show anti-tumor effects similar to Doxil-like liposomes in tumor-bearing mice

Therapy; Biotechnology; Cancer

Introduction

Liposomes are promising drug carriers which can enhance drug stability and optimize the pharmacokinetics with effective delivery of drugs to pathologic site and thus decrease the systemic toxicity. Active or remote loading methods have been extensively studied for efficient encapsulation of weakly basic or acidic drugs into liposomes driven by transmembrane pH or non-metal ion gradient.^1,2,3,4,5,6 The liposomes with metal ion gradient have also been used for drug encapsulation.⁷ The weakly acidic molecules can be encapsulated into magnesium or calcium ion containing liposomes,^5,8 whereas some weakly basic molecules can be encapsulated into liposomes containing transition metal ions, like manganese, copper, and nickel.^{9,10,11,12,13,14,15,16,17,18} Several liposomal drugs, including the liposomal doxorubicin (DOX) prepared by remote loading methods have been clinically approved for the treatment of cancers.^19,20 Recently, many functionalization, including stimuli-response and specific targeting strategies have been investigated to improve the clinical efficacy of liposomal drugs.^21,22,23,24

The tumor extracellular microenvironment, endosomal and lysosomal compartments, are more acidic than blood and normal tissue adjacent to tumor,^{25,26,27,28,29} making pH-sensitive or pH-responsive release an appealing strategy to enhance therapy efficacy.^30,31,32 Various pH-sensitive liposomes have been constructed by using pH-sensitive lipids and amphiphilic derivatives inserted in the liposomal membrane or pH-sensitive polymer moieties on the liposome surface. These liposomes can be destabilized by structural conformation or chemical bond breakage in mild acidic environment.^30,31,32 However, the pH responsiveness of these liposomes is defined precisely by the lipid components. The pH-sensitive lipids for liposomes that show suitable in vivo drug delivery property are still restricted and no pH-sensitive liposomes have been developed into clinical use.

The complexation of metal ions with chemical drugs gives us another clue to develop liposomes as pH-sensitive drug delivery system to treat cancer. Magnesium is an abundantly nutritious mineral element in human body, mostly present inside cells or in the bone.³³ Mg²⁺ ions can form complexes with DOX^34,35 and daunomycin,³⁶ which are both anthracycline drugs. In this study, we investigated the pH-dependent complexation of Mg²⁺ with DOX and developed a remote loading strategy of DOX into liposomes with magnesium acetate gradient for a pH-sensitive liposomal drug delivery system (Scheme 1). The folate receptor (FR) specific targeting Mg-DOX liposomes were also prepared by the surface modification with folic acid (FA)^37,38 for a goal to enhance the intracellular drug delivery to various FR overexpressing tumors.^39,40 We conducted in vitro experiments to investigate the tumor cell uptake and cytotoxicity of the Mg-DOX liposomes and in vivo experiments to determine the pharmacokinetics of the liposomes in normal mice and the tumor treatment efficacy with breast tumor bearing mice.

Scheme 1.

Schematic representation of remote-loading of DOX into MgAc₂ gradient liposomes and pH-sensitive release of DOX for therapy of acidic tumor

For DOX remote-loading, MgAc₂ gradient liposomes are mixed with DOX in NaHCO₃ medium, acetic acid molecules in the liposome inner aqueous space will be released and reacted with OH⁻ and HCO₃⁻ ions, resulting in the increase of inner pH of liposomes. Meanwhile, DOX in the medium can cross the lipid bilayer and react with Mg²⁺ inside the liposomes to form hydrophilic complex ions, which was represented by [Mg(DOX-H)]⁺. While the Mg-DOX liposomes are exposed to tumor acidic microenvironment, the pH in the liposome inner space will be decreased by the transmembrane transport of weak acid molecule, represented by HL, causing the decomposition of [Mg(DOX-H)]⁺ ion, and free DOX will be released from liposomes.

Results and discussion

Complexation of DOX with Mg²⁺

The complexation of Mg²⁺ with DOX was investigated, with emphasis on the effects of Mg²⁺ concentrations and pH values. The UV-visible absorption spectra and fluorescence spectra of 0.144 mM DOX in the presence of different concentrations of Mg²⁺ in HEPES, pH 7.8 buffer were determined. The results showed that the visible absorption spectra of DOX in the weakly alkaline condition changed significantly by the adding of Mg²⁺ (Figure 1A). While Mg²⁺ concentration increased from 0 to 200 mM, the broad absorption band at 480–500 nm red-shifted and gradually weakened except a small leap appeared at 34.5 mM Mg²⁺, whereas a shoulder band with maximum absorption at about 540 nm gradually turned into the strongest one, another new absorption band with maximum absorption at 580 nm appeared and gradually strengthened, both of which may be attributed to the complexes of DOX with Mg²⁺. Further increase of Mg²⁺ concentration from 200 mM to 500 mM resulted in almost the same absorption spectrum, indicating that DOX was completely complexed in the presence of 200 mM Mg²⁺ (Figure 1B).

Figure 1.

Effect of Mg²⁺ concentration on the complexation of 0.144 mM DOX with different concentrations of MgSO₄ in 100 mM HEPES buffer, pH 7.8

(A) UV-visible absorption spectra.

(B) Variation of absorbances at specific 480 nm of DOX, 540 nm and 580 nm of the DOX complexes with Mg²⁺.

(C) Fluorescent emission spectra of the DOX complexes with Mg²⁺ excited by 480 nm light.

(D) Fluorescence emission intensities of DOX complexes with Mg²⁺.

The fluorescence spectra of DOX in HEPES buffer, pH 7.8 also changed significantly with the addition of Mg²⁺. The Ex480nm/Em595nm fluorescence band of free DOX gradually red-shifted with increasing Mg²⁺ concentration (Figure 1C). Two new Ex540nm/Em605nm and Ex580nm/Em640nm fluorescence bands attributed to the Mg-DOX complexes were discovered, and their fluorescent intensities gradually increased with increasing Mg²⁺ concentration, reaching maximum values at 200 mM Mg²⁺ and then decreased slowly (Figure 1D), in coincidence with the condition approaching maximum absorbance intensities of the 540 nm and 580 nm wavelengths. These results showed clearly the complexation degree of DOX with Mg²⁺ was dependent on the Mg²⁺ concentration, reflecting the modest complexation strength of the complexes. The relatively high Mg²⁺ concentration was required for the complete transformation of DOX into Mg-DOX complexes, supposed to be mainly DOX/Mg 1:1 type.

The complexation of DOX with Mg²⁺ was also severely affected by the pH values of medium. The color of 44.6 μM DOX·HCl and 500 mM MgAc₂·4H₂O mixtures showed a shift from orange-yellow to purplish-red with increasing pH from 5.25 to 8.24 (Figure S1). Simultaneously, the visible absorption spectra of the mixtures varied much with increasing pH values (Figure 2A). The visible absorption spectrum of the mixture at pH 5.25 was similar to that of free DOX solution at pH 5–7.8 without the presence of Mg²⁺. When the pH of DOX and MgAc₂ mixture increased to 6.0, the shoulder peak at 540 nm wavelength was more pronounced, and a new absorption peak at 580 nm appeared, indicating the complexation of Mg²⁺ with DOX. The intensities of absorption peaks at 540 nm and 580 nm both increased gradually with the increasing pH values, whereas the absorption at 480 nm decreased gradually, reflecting the enhanced complexation of DOX with Mg²⁺ at higher pH values. The curve of absorbance at 580 nm versus pH value was well fitted with the non-linear Slogistic1 model using OriginPro 8.0 software (OriginLab Corporation, Northampton, MA, USA), and the titration jump at pH 6.84 was derived from the fitted curve (Figures 2B and 2C).

Figure 2.

The complexation of DOX with Mg²⁺ at different pH conditions

(A) The absorption spectra of DOX and MgAc₂ mixed solutions.

(B) The absorbance at 580 nm versus pH value curve of DOX and MgAc₂ mixed solutions by non-linear fitting.

(C) One order differentiation result of the fitting curve of absorbance at 580 nm versus pH value.

(D) Fluorescence emission spectra of DOX ang MgAc₂ mixed solutions excited at 480 nm.

(E) Fluorescence emission spectra of DOX ang MgAc₂ mixed solutions excited at 540 nm.

(F) Fluorescence emission spectra of DOX ang MgAc₂ mixed solutions excited at 580 nm.

The fluorescence spectra of the DOX and MgAc₂ mixtures also changed significantly at different pH conditions (Figures 2D–2F). While the pH value of the mixture increased from 5.25 to 8.24, the fluorescence band with maximum emission at 590 nm by excitation at 480 nm red-shifted and the intensity decreased, whereas the fluorescence intensity of band peaked at 640 nm had an increase tendency (Figure 2D). The fluorescence intensities of emission bands peaked at 605 nm and 640 nm increased by excitation at 540 nm and 580 nm, respectively (Figures 2E and 2F). The changes of both absorption spectra and fluorescence spectra indicated that the pH values severely affect the complexation of DOX with Mg²⁺, and pH increase from acidic medium to basic medium can significantly promote the Mg-DOX complexation. These results suggested that Mg²⁺ gradient liposomes may be of potential for efficient DOX loading.

Remote loading of DOX into MgAc₂ gradient liposomes

The remote loading of doxorubicin into liposomes was investigated by mixing the blank MgAc₂ gradient liposomes with DOX in basic media. The mixtures in NaHCO₃ medium gradually changed from orange-yellow to purplish-red by heating. Notably, this color change was achieved by heating at 55°C for 5 min. The visible spectra of the heated mixtures showed that the DOX absorption peak at 480 nm red-shifted gradually to a peak at 510 nm with a shoulder peak at 540 nm (Figure 3A), concomitantly, the Ex480nm/Em595nm fluorescence band also declined significantly, almost quenched after incubation at the gel-liquid transition temperature of DSPC, 55°C for 5 min (Figure 3B). Moreover, the heated mixture had almost no specific Ex540nm/Em605nm and Ex580nm/Em640nm fluorescence emission spectra of Mg-DOX complexes. These results indicated the efficient heating promoted remote loading and complexation of DOX with Mg²⁺ inside the liposomes. The fluorescence quenching phenomenon may be contributed to the formation of high concentration of Mg-DOX inside the liposomes.

Figure 3.

UV-visible absorption spectra and fluorescence spectra of heated MgAc₂ gradient DSPC/Chol/DSPE-PEG2000 liposomes and DOX mixtures at drug/TLs molar ratio of 0.33 in medium containing 125 mM NaHCO₃

(A) UV-visible absorption spectra.

(B) Fluorescence emission spectra by excitation at 480 nm.

While HEPES buffers (pH 7.4 and pH 7.8) were used instead of NaHCO₃ solution as the media in the remote loading of DOX, the resulting visible absorption spectra of the mixed solutions by heating also showed red-shift of the absorption peaks, and the faster red-shift was achieved by higher pH (Figure S2A), furthermore, the florescence emission intensities by excitation at 480 nm and at 540 nm both decreased more quickly under higher pH condition following heating (Figures S2B and S2C), confirming that higher basic condition facilitated the Mg-DOX complexation inside the liposomes. It could be concluded that the remote loading of DOX into liposomes was faster in NaHCO₃ medium than in HEPES buffer, by comparison the fluorescence signals (Figure 3B; Figures S2B and S2C).

Furthermore, we prepared the 300 mM MgSO₄ gradient liposomes and 300 mM MgCl₂ gradient liposomes for the DOX remote loading in a similar way as the 300 mM MgAc₂ gradient liposomes. Though the mixture of MgAc₂ gradient liposomes and DOX in 125 mM HEPES, pH 7.8 changed color and the visible absorption spectrum gradually red-shifted by heating at 55°C, the substitution of MgAc₂ gradient liposomes with MgSO₄ gradient liposomes or MgCl₂ gradient liposomes caused no significant color change and red-shift of the absorption spectrum under same heating condition (Figure S3), indicating both Mg²⁺ and acetate gradients were crucial factors for the efficient remote loading of DOX into liposomes. The weak acid molecule, acetic acid, but not SO₄²⁻ or Cl⁻ inside the liposome vesicle could pass through the lipid bilayer driven by the pH gradient between the inner space and outside medium of liposomes, react with alkaline in the medium, and thereby improve the pH value of liposome inner space. The reaction of basic NaHCO₃ with acetic acid to produce volatile CO₂ under heating condition may also favor the transmembrane transport of acetic acid and thus enable the fast remote loading of DOX into liposomes (Scheme 1).

The quantitative remote loading of DOX into 100 nm MgAc₂ gradient liposomes to form the Mg-DOX liposomes (Mg-DOX-Lip100) was achieved by heating the liposomes and DOX mixtures in the NaHCO₃ medium, even faster than the DOX loading into the 100 nm ammonium sulfate gradient liposomes in saline medium to form DOX liposomes (DOX-Lip100) (Figure 4). The measured encapsulation efficiencies (EEs) of DOX into liposomes were generally more than 95% for the DSPC/Chol/DSPE-PEG2000 liposomes and DSPC/Chol/DSPE-PEG2000/DSPE-PEG2000-FA liposomes by heating at 50°C for 10 min or 55°C–60°C for 5 min. It was also observed that other anthraquinone chemotherapeutic drugs, epirubicin and daunomycin, could be quantitatively encapsulated into MgAc₂ gradient liposomes, in a similar way as DOX.

Figure 4.

Encapsulation efficiencies of DOX in 100 nm ammonium sulfate gradient liposomes and in 100 nm MgAc₂ gradient liposomes by heating promoted remote loading at initial drug/TL molar ratio of 0.16

(A) heating at 50°C.

(B) heating at 55°C.

(C) heating at 60°C.

The data are represented as mean +SD (n = 3). Independent-samples t test for two groups, ∗p < 0.05.

The hydrodynamic particle sizes and Zeta potentials of the 100 nm liposomes determined with the Zetasizer Nano ZS90 analyzer were listed in Table 1. The blank liposomes and DOX loaded liposomes were all around 100 nm of diameters and had narrow size distribution with polydispersity index (PDI) values ≤0.23. All the liposome surfaces were slightly negatively charged, shown by the negative Zeta potential values.

Table 1.

The hydrodynamic particle sizes, PDI, and Zeta potentials of the liposomes

liposomes	Diameter (nm)	PDI	Zeta potential
Blank ammonium sulfate liposomes	123.2 ± 2.6	0.085 ± 0.026	−6.25 ± 1.17
Blank MgAc2 liposomes	123.8 ± 4.5	0.164 ± 0.030	−8.36 ± 0.65
DOX-Lip100	104.4 ± 0.9	0.149 ± 0.034	−6.30 ± 0.16
Mg-DOX-Lip100	97.1 ± 4.3	0.229 ± 0.061	−11.67 ± 0.15
FA-Mg-DOX-Lip100	128.7 ± 2.3	0.116 ± 0.051	−13.33 ± 1.01

n = 3 for each sample.

Micromorphology of Mg-DOX liposomes

The cryo-transmission electron microscopy (cryo-TEM) observation showed that the Mg-DOX-Lip100 particles were mostly unilamellar spherical vesicles and bilamellar spherical vesicles with some degree of distortion at approximately 1:1 ratio (Figure 5). The irregular dense aggregation was found inside the vesicles, including the intercalation of bilamellar vesicles, which may represent the nanocrystals of high concentration of Mg-DOX complexes.

Figure 5.

Cryo-TEM micrograph of the Mg-DOX-Lip100 vesicles

Storage stability of Mg-DOX liposomes

The hydrodynamic diameters of Mg-DOX liposomes in 10 mM NaHCO₃/saline stored for 1 month and 3 months (Table 2) were close to that of the freshly prepared liposomes (Table 1). The percentages of DOX retention inside Mg-DOX-Lip100 and FA-Mg-DOX-Lip100 were 98.0 ± 0.01% and 99.0 ± 0.28% after 1 month of storage and 93.7 ± 0.04% and 96.8 ± 0.09% after 3 months of storage, respectively, suggesting the low temperature of 4°C is an appropriate condition for the stable storage of Mg-DOX liposomes.

Table 2.

The hydrodynamic particle sizes, PDI, and DOX retention of the stored Mg-DOX liposomes

Liposomes	Storage at 4°C	Diameter ±SD (nm)	PDI	Zeta potential	DOX retention (%)
Mg-DOX-Lip100	1 month	125.9 ± 5.6	0.302 ± 0.019	−7.4 ± 0.47	98.0 ± 0.01%
Mg-DOX-Lip100	3 months	100.9 ± 14.6	0.303 ± 0.099	−10.9 ± 0.96	93.7 ± 0.04%
FA-Mg-DOX-Lip100	1 month	122.1 ± 2.2	0.206 ± 0.155	−8.4 ± 0.21	99.0 ± 0.28%
FA-Mg-DOX-Lip100	3 months	118.4 ± 3.1	0.298 ± 0.085	−8.91 ± 0.84	96.8 ± 0.09%

n = 3 for each sample.

Release of DOX from Mg-DOX liposomes under different pH conditions

The stability of DOX-Lip100 and Mg-DOX-Lip100 under different pH conditions was evaluated using the dialysis method. The percentages of release of DOX from liposomes into the dialysis media measured by the fluorescence assay for a period of 48 h were shown in Figure 6. The release rates of the DOX-Lip100 in all the 3 media increased slowly with time but were lower than 10% after 48 h of incubation, indicating the DOX liposomes were very stable and almost not affected by the various pH conditions. On the contrary, the DOX release of Mg-DOX liposomes was severely affected by the pH values of media. Only a few percentages of DOX released from the Mg-DOX-Lip100 during 48 h in 5 mM NaHCO3/saline, pH 7.4, indicating that the liposomes were both stable in the weakly basic medium. However, the Mg-DOX-Lip100 released significantly DOX in the neutral and weakly acidic media. The accumulated DOX release increased gradually in 10% FBS in saline, pH 7.0, and the release rate approached 34 ± 1.5% at 48 h. The Mg-DOX-Lip100 released DOX rapidly in 10 mM NaAc/HAc buffer in saline, pH 5.5, with about 60% of DOX released in 12 h, afterward the dialysis system seemed to reach an equilibrium state and little more DOX released from the Mg-DOX liposomes with time, the accumulated DOX release at 48 h was 64.3 ± 1.1%. These results indicated that the Mg-DOX liposomes have pH-sensitive release property, supporting the proposed mechanism (Scheme 1) that in relative acidic medium, weak acid molecules will enter the inner space of liposomes driven by pH gradient, decrease the inner pH, and induce the decomposition of Mg-DOX complexes, then free DOX molecule can be readily released from the liposomes.

Figure 6.

The release rate of DOX from DOX-Lip100 and Mg-DOX-Lip100 at media of different pH values

(A) 10 mM NaAc/HAc buffer in saline, pH 5.5.

(B) 10% FBS in saline, pH 7.0.

(C) 5 mM NaHCO₃ in saline, pH 7.4.

The data are represented as mean +SD (n = 3).

Tumor cell uptake of Mg-DOX liposomes in vitro

The tumor cell uptake of DOX·HCl, DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 at equivalent DOX concentration of 10 μg/mL by MCF-7 cells was observed by the confocal microscopy imaging. As the remote loading of DOX into liposomes caused the quenching of DOX fluorescence, the intracellular release of DOX from liposomes can be monitored by the microscopic fluorescent imaging of free DOX. The microscopic images showed that free DOX could be rapidly captured by the tumor cells and accumulated in the nuclei in 3 h of incubation in vitro, whereas the cellular DOX fluorescence signal was relatively weak for all the 3 liposomes groups (Figure 7A), implying the shortage of liposomes intake by the tumor cells or the lack of DOX release from liposomes inside the cells. The cellular DOX fluorescence intensities of all the 3 liposomes groups increased with increasing incubation time, moreover, they showed a tendency of FA-Mg-DOX-Lip100 > Mg-DOX-Lip100 > DOX-Lip100 at 6 h and 12 h (Figures 7B and 7C).

Figure 7.

The confocal microscopic images of MCF-7 cells in upper row and MDA-MB-231 cells in lower row incubated with DOX·HCl, DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100

A scale bar representing 30 μm is shown in the lower right corner.

(A) and (D) Incubation for 3 h.

(B) and (E) Incubation for 6 h.

(C) and (F) Incubation for 12 h.

The cell uptake of DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 at 10 μg DOX/mL by MDA-MB-231 cells was also observed by the confocal microscopy imaging. The cellular DOX fluorescence intensities of MDA-MB-231 cells for the 4 groups were DOX·HCl > FA-Mg-DOX-Lip100 > Mg-DOX-Lip100 > DOX-Lip100, showing same tendency as that of MCF-7 cells (Figures 7D–7F). The quantitative assay of DOX uptake by the tumor cells incubated with DOX containing liposomes for 3 h showed that both MCF-7 cells and MDA-MB-231 cells took more DOX amount from FA-Mg-DOX-Lip100 than from non-specific targeting DOX-Lip100 and Mg-DOX-Lip100, which had no difference in cellular DOX uptake (Figure S4).

The higher cellular DOX accumulation of FA-Mg-Dox-Lip100 than the other liposomes under same incubation conditions for either MCF-7 cells or MDA-MB-231 cells reflected that the folic acid modification of the Mg-DOX liposomes enables the active intake of the liposomes by folic acid receptor positive tumor cells and enhances the intracellular DOX release in vitro.

In vitro cytotoxicity of Mg-DOX liposomes to tumor cells

The in vitro cell viabilities of MCF-7 cells and MDA-MB-231 cells incubated with DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 for 24 h were measured by the MTT assay. The cell viability decreased with increasing liposomal DOX concentration, and three types of liposomes exhibited different tumor cell growth inhibition abilities (Figure 8). The cytotoxicity of FA-Mg-DOX-Lip100 was significantly stronger than both Mg-DOX-Lip100 and DOX-Lip100 to either MCF-7 cells or MDA-MB-231 cells, while Mg-DOX-Lip100 had stronger cytotoxicity than DOX-Lip100 to MCF-7 cells, and comparable cytotoxicity as DOX-Lip100 to MDA-MB-231 cells. The strong cytotoxicity of FA-Mg-DOX-Lip100 may be related to the folate receptor targeting mediated active tumor cell intake of liposomes and enhanced intracellular DOX release, as observed by confocal microscopic imaging (Figure 7). The IC50 values of FA-DOX-Lip to MCF-7 cells and MDA-MB-231 cells were equivalent DOX concentrations of 6.73 ± 2.35 μg/mL and 4.66 ± 2.58 μg/mL, respectively. The in vitro cell viabilities of tumor cells incubated with DOX·HCl, DOX-Lip100, and Mg-DOX-Lip100 for 48 h were measured by the CCK8 assay. The results showed that Mg-DOX-Lip100 produced stronger cytotoxicity than DOX-Lip100 to both MCF-7 and MDA-MB-231 cells, whereas free DOX produced much stronger cytotoxicity than the DOX liposomes to the tumor cells (Figure S5). This result was consistent with the much faster uptake of lipophilic DOX by tumor cells and approaching the nuclei than the DOX liposomes (Figure 7). The blank 100 nm MgAc₂ gradient liposomes were also tested and showed no any cell inhibition effect to MCF-7 cells and MDA-MB-231 cells for 48 h of incubation, indicating that the blank liposomes are non-cytotoxic drug carriers.

Figure 8.

Cell viability of tumor cells incubated with liposomes for 24 h measured by MTT assay

(A) MCF-7 cells.

(B) MDA-MB-231 cells.

The data are represented as mean +SD.

Pharmacokinetics and biodistribution of liposomes in C57BL/6 mice

The DOX concentrations in plasma samples post intravenous (IV) injection of DOX-Lip100, Mg-DOX-Lip100, and FA-MG-DOX-Lip100 were measured. The plasma DOX concentration versus time curves (Figure 9A) showed that the blood clearance speed of DOX was FA-Mg-DOX-Lip100 > Mg-DOX-Lip100 > DOX-Lip100. The non-compartmental analysis results of the blood DOX clearance curves showed the blood half-lives (T_1/2) of DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 were 15.4 ± 3.7 h, 13.7 ± 3.8 h and 6.6 ± 1.7 h, respectively. The area under curves (AUCs) were 725.0 ± 136.7, 390.6 ± 171.1, and 198.5 ± 40.0 μg·h/kg, respectively. These results indicated that Mg-DOX-Lip100 liposomes were stable in circulation, but was less stable than DOX-Lip100. FA-Mg-DOX-Lip was least stable in circulation among the DOX containing liposomes. Mg-DOX-Lip100 liposomes were stable in pH 7.4 medium as shown by the drug release curve in Figure 6. The faster clearance of Mg-DOX-Lip100 than the stable DOX-Lip100 in vivo implied that the interaction of bilayer of the Mg²⁺ containing liposomes with protein components in blood probably change the bilayer conformation and then lead to the DOX release from the vesicles. The interaction of liposomes with blood cells, liver cells, and other tissue cells may also contribute to the relatively faster clearance of Mg-DOX liposomes. The detailed mechanism on the pharmaceutical behavior of Mg-DOX liposomes requires further investigation.

Figure 9.

Pharmacokinetics and biodistribution of liposomal DOX

(A) Pharmacokinetic curves of DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 in C57BL/6 mice.

(B) DOX concentrations of dissected organs/tissue 48 h post IV injection of DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 into C57BL/6 mice.

The data are represented as mean +SD. Independent samples t test for two groups, ∗p < 0.05 and ∗∗∗p < 0.001.

The biodistribution of liposomal DOX in the dissected heart, liver, spleen, lung, kidneys, and muscle of C47BL/6 mice 48 h post IV injection of the liposomes were determined. The liver, spleen, and kidneys of all the three groups had higher DOX concentrations than muscle. FA-Mg-DOX-Lip100 group had significant higher concentration of DOX in liver than the other two groups, whereas the DOX concentrations of each other organ/tissue had no statistical difference among DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 groups, except the FA-Mg-DOX-Lip had a little bit higher concentration of DOX in heart than the DOX-Lip100 (Figure 9B). These biodistribution and the pharmacokinetic data suggest that Mg-DOX liposomes accumulated in the organs/tissue were also stable in comparison to the DOX-Lip100. The worse pharmaceutic property and increased liver accumulation of the folate receptor targeting liposomes were also demonstrated in various studies.^38,41 However, the mechanism of this in vivo characteristic of FR targeted liposomes has not been clarified until recent studies discovered that IgM mediated opsonization plays important role in the clearance of such liposomes.^42,43 It was found that folic acid modification of liposomes surface at a molar ratio of 2% to lipids can significantly register the deposition of natural IgM on surface of liposome vesicles in blood circulation. Such protein corona of liposomes with high absorption of IgM is ready to induce rapid blood clearance and decrease the tumor targeting yields of liposomes.⁴²

In vivo therapeutic efficacy of Mg-DOX liposomes

The orthotropic breast cancer tumor model in C57BL/6 mice was set up with EO771 cells, which had higher FOLR1 mRNA expression than 4T1 cells by qPCR assay (Figure S6). The mice were IV injected with liposomes at equivalent DOX dose of 5 mg/kg body weight every 4 days for consecutive 4 times. The tumor of control group had a continuous growth tendency (Figure 10A) and the first mouse reached the endpoint criteria on day 16. Comparably, the tumor growth of the DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 groups was delayed following treatment and the tumor sizes were significantly smaller than the control group on days 8, 10, 14, and 16 by ANOVA analysis. The three experimental groups showed no significant difference of tumor sizes at same time (Figure 10A). The treatment with DOX containing liposomes caused the mean body weight measured on same day was smaller than that of control group, with statistical differences on day 8 to day 14, however, the 3 liposomes groups had no significant body weight difference following the treatment (Figure 10B). The mice of control, DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 groups reached the endpoint criteria on day 16–22, day 20–26, day 20–24, and day 20–24, respectively. The estimated average survival times were 19 days, 23 days, and 22 days, respectively. These results indicated that Mg-DOX liposomes had comparable tumor inhibition effect as the ammonium sulfate gradient DOX liposomes in vivo. Though FA-Mg-DOX-Lip100 had stronger tumor cell cytotoxicity compared to non-specific DOX liposomes in vitro, no enhanced therapeutic effect was observed in vivo with the tumor bearing mice. Both the stability in blood circulation and efficient release inside tumor are important for the therapeutic efficacy of drug delivery system. The relatively faster blood clearance of Mg-DOX-Lip100 and FA-Mg-DOX-Lip100 than DOX-Lip100 found in the pharmaceutical study may suggest that these Mg-DOX liposomes have no advantage over DOX-Lip100 in the delivery of DOX to tumor. However, the pH-sensitive intratumoral release of Mg-DOX liposomes, especially the FA-Mg-DOX-Lip100 may produce more efficient tumor growth inhibitory effect than the DOX-Lip100 at same DOX amount delivered inside tumor.

Figure 10.

Therapeutic efficacy of the DOX liposomes on the orthotropic EO771 cell breast cancer mice

(A) Tumor sizes, (B) Body weight, (C) Representative microscopic images of H&E stained histochemical slides of dissected organs.

The data are represented as mean +SD. One-way ANOVA for multiple groups and independent-samples t test for two groups, ∗p < 0.05 and ∗∗p < 0.01.

The investigation of toxic side effect in animal model is important for the safety evaluation of new drug delivery system. The treatment of EO771 tumor bearing mice by DOX-Lip100, Mg-DOX-Lip100, and FA-Mg-DOX-Lip100 at same DOX doses led to similar side effect in body weight loss (Figure 10B). The hematoxylin and eosin (H&E) staining slides of the dissected heat, liver, spleen, lung, and kidneys of the mice of control and DOX containing liposomes groups were observed by microscopic imaging. The results showed that all the vital organs had normal histopathological characterization and no observable microstructural damage caused by either the conventional DOX-Lip100 liposomes or Mg-DOX liposomes (Figure 10C), suggesting their safety with neglectable side effect in the study.

Though acidic tumor microenvironment has been recognized in many solid tumors, the actual tumor acidity of the breast cancer model and the intra-tumoral DOX release, which may be crucial for the therapeutic efficacy of liposomes were not measured in our study. Monitoring both the acidity of tumor microenvironment and DOX release inside tumor are necessary in further study to more comprehensively evaluating the therapeutic potential of the Mg-DOX liposomes. Both the pharmacokinetics and biodistribution are predominant features in determining the therapeutic performance of each type of drug carriers, the in vivo dosimetry of Mg-DOX liposomes needs to be investigated further to achieve the optimal efficacy to treat tumor.

The FA-Mg-DOX-Lip100 liposomes was modified with folic acid at the tail of DSPE-PEG2000 component of liposomes, and were mostly spherical vesicles, thus could hardly avoid the deposition of IgM on liposomes surface depicted in recent studies.^42,43 Interestingly, earlier studies of liposomes with long polyethylene glycol (PEG) chain (PEG5000 and PEG10000) conjugated folic acid could improve the pharmacokinetic characteristic compared to FA targeting liposomes coated using relatively short chain of PEG.^44,45 We will further investigate the new formulations of FR targeted Mg-DOX liposomes modified with long chain PEG to fully elucidate the antitumor potential of the Mg-DOX liposomes drug delivery system.

The liposomes with pH-sensitive bilayers have been studied extensively.^30,31,32 The pH-sensitive lipids used in these liposomes are usually neutral molecules but have ionizable moieties or acid liable linkers, which can be positively charged or cleaved by protonation resulting in the destabilization of liposomal bilayers in acidic condition, and subsequent release the drug payload by the conformation change of liposome bilayer or by the increased interactions of liposomes with the negatively charged endosome membrane. The most common pH-sensitive liposomes containing DOPE/CHEMS lipids are easily recognized by opsonins in blood circulation and can be cleared quickly.³² The DOX encapsulated liposomes with DOPE/CHEMS showed significant in vitro release of DOX in 80% human plasma at 37°C.⁴⁶ The surface PEGylation of such pH-sensitive liposomes could improve the liposomes stability in blood at some degree by reducing the opsonization.^47,48 The estrogen-anchored pH-sensitive DOX liposomes with DOPE/CHEMS showed pH-sensitive DOX release in vitro and similar pharmacokinetic curve in mice as the non pH-sensitive, long circulating liposomes in mice.⁴⁷ The folate targeting liposomes with DOPE/CHEMS showed good in vitro stability in plasma at 37°C, and the biphasic clearance profile of the non FA targeting, long circulating liposomes and FA targeting liposomes radiolabeled by ^99mTc in mice had long half-lives of 7.9 h and 3.8 h, respectively.⁴⁸ The DOX liposomes with pH sensitive molecule, malachite green carbinol base (MG) inserted in the bilayer showed pH sensitive DOX release and stronger anti-tumoral effect than the non pH-sensitive liposomes in tumor bearing mice. However, the pharmaceutic data were not provided in the study and the toxic MG component limits the clinical possibility of the liposomes.⁴⁹ The DOX liposomes containing alpha-tocopheryl succinate as a pH-sensitive lipid showed shorter half-life in blood circulation in mice than the Doxil-like liposomes but higher DOX concentration in xenografted 4T1 tumor and tumor inhibitory effect.⁵⁰

Various new pH-sensitive lipids and polymers have been synthesized for the construction of pH-sensitive liposomes, providing flexible formulation and great potential for improved drug delivery efficiency to tumor cells in acidic microenvironment.^30,31,32 However, the pharmaceutical development of such type of pH-sensitive liposomes with new lipid substance meets many challenges, including the both requirement of stability in blood circulation and efficient delivery to tumor cells, in addition to complicated production and strict evaluation. Comparably, the pH-sensitivity of Mg-DOX liposomes relies on the complexation of Mg²⁺ with DOX, producing the unique drug loading and release mechanism different from that of the conventional pH-sensitive liposomes. The Mg-DOX liposomes require no pH-sensitive lipids and can be easily formulated with pharmaceutical acceptable components, which provide better clinical translation potential than the liposomes with engineered pH-sensitive lipids.³² Further studies are warranted to fully elucidate the antitumor potential of this Mg-DOX liposomes drug delivery system.

In summary, we successfully constructed pH-sensitive DOX carrying liposomes based on the pH dependent complexation of DOX with Mg²⁺. The 100 nm Mg-DOX liposomes had good pH-dependent DOX release feature, could efficiently deliver DOX inside folate receptor positive tumor cells by folic acid modification and thus enhanced the tumor cell killing ability in vitro. The pharmacokinetics in mice showed Mg-DOX-Lip100 were stable in circulation, but FA-Mg-DOX-Lip100 were cleared faster. IV injection of Mg-DOX-Lip100 and FA-Mg-DOX-Lip100 liposomes significantly delayed the tumor growth of orthotropic breast cancer mouse model. Therefore, the pH-sensitive Mg-DOX liposomes are a promising drug delivery system, which merits further optimization to explore the potential therapy of acidic tumors.

Limitations of the study

The folate receptor targeted Mg-DOX liposomes showed pH dependent DOX release in vitro but reduced stability in blood circulation in vivo compared to the non-specific ones. The optimization of the folate receptor targeted Mg-DOX liposomes formulation will improve the pharmacokinetics and therapy outcomes. The experimental therapy of tumor model with Mg-DOX liposomes was performed without monitoring tumor microenvironment pH value and measurement of DOX release inside tumor.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Shihong Li (lishihong@suda.edu.cn).

Materials availability

The liposomes generated in this study are available from the lead contact upon request.

Data and code availability

• •Data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

This research was financially supported by the National Natural Science Foundation of China (grant no. 12175163) and partially supported by NHC Key Laboratory of Nuclear Technology Medical Transformation (MIANYANG CENTRAL HOSPITAL) (grant no. 2021HYX027). The authors gratefully acknowledge Tingyu Liu in the Center of Cryo-Electron Microscopy (CCEM), Zhejiang University for her technical assistance on cryo-transmission electron microscopy, and thank Professor Ande Bao at Case Western Reserve University for his helpful suggestions on the revision of this manuscript.

Author contributions

Conceptualization, S.L.; methodology, Y.Y., M.Z., N.L., and S.L.; investigation, Y.Y., M.Z., Z.H., and S.L.; formal analysis, Y.Y., M.Z., and H.M.; writing-original draft, Y.Y. and M.Z.; writing review and editing, S.L., G.Z., and L.M.; funding acquisition, S.L.; supervision, N.L., G.Z., and S.L.

Declaration of interests

The authors declare that a patent application in China related to this research was submitted by The First Affiliated Hospital of Soochow University.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
1,2-disteraroyl-sn-glycero-3-phosphocholine (DSPC)	Shanghai Ponsure Biotech, Inc.	CAS: 816-94-4
1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-methoxy polyetheneglycol-2000 (DSPE-PEG2000)	Shanghai Ponsure Biotech, Inc.	CAS: 147867-65-0 Cat No. PS1-DE-2k
1,2-distearoyl-sn-glycero-phosphoethanolamine-polyethyleneglycol 2000-folic acid (DSPE-PEG2000-FA)	Xi’an ruixi Biological Technology Co., Ltd.	CAS: 1236288-25-7 Cat No. R-0043-2
Cholesterol (Chol)	Shanghai Acmec Biochemical Technology Co., Ltd.	CAS: 57-88-5
Ethanol (EtOH)	Sinopharm Chemical Reagent Co. Ltd.	CAS: 64-17-5
Doxorubicin Hydrochloride (DOX·HCl)	Shanghai yuanye Bio-Technology Co., Ltd.	CAS: 25316-40-9
Magnesium acetate tetrahydrate (MgAc2·4H2O)	Sangon Biotech (Shanghai) Co., Ltd.	CAS: 16674-78-5
Magnesium sulfate heptahydrate (MgSO4·7H2O)	Sinopharm Chemical Reagent Co. Ltd.	CAS: 10034-99-8
Magnesium chloride hexahydrate (MgCl2·6H2O)	Sinopharm Chemical Reagent Co. Ltd.	CAS: 7791-18-6
Sodium bicarbonate (NaHCO3)	Sinopharm Chemical Reagent Co. Ltd.	CAS: 144-55-8
2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES)	Sinopharm Chemical Reagent Co. Ltd.	CAS: 7365-45-9
Isopropanol (IPA)	Sinopharm Chemical Reagent Co. Ltd.	CAS: 67-63-0
ProLong™ Gold Antifade Mountant with DAPI	Thermo Fisher Scientific (China) Co., Ltd.	Cat No. P36941
Fetal bovine serum (FBS)	ExCell Bio Inc.	Cat No. FSP500
Penicillin/Streptomycin solution	Beyotime Biotechnology Co., Ltd.	Cat No. C0222
0.25% Trypsin-EDTA Solution	Beyotime Biotechnology Co., Ltd.	Cat No. C0201
Critical commercial assays
MTT test kit	Hangzhou Fude Biological Technology Co., Ltd.	FD8326
CCK8 test kit	Hangzhou Fude Biological Technology Co., Ltd.	FD3788
RNA-Quick Purification Kit	Shanghai Yishan Biotechnology Co., Ltd.	ES-RN001
Experimental models: Cell lines
MCF-7 cell line	ATCC	HTB-22
MDA-MB-231 cell line	ATCC	CRM-HTB-26
EO771 cell line	CH3 Biosystems	94A001
Experimental models: Organisms/strains
Female C57BL/6 mice	Soochow University Laboratory Animal Center	A202312260498
Oligonucleotides
FOLR1 primers: forward 5 –‘ TGGCCGAAT GTGCTCAGTC –3‘ and reverse 5 –‘ GTCG TGTAAATTGTCCTCAGGG –3‘	Suzhou GENEWIZ Biological Technology Co., Ltd.	N/A
β-actin primers: forward 5 –‘ GGCTGTAT TCCCCTCCATCG –3‘ and reverse5 –‘ CCA GTTGGTAACAATGCCATGT –3‘	Suzhou GENEWIZ Biological Technology Co., Ltd.	N/A
Software and algorithms
GraphPad Prism 8.0	GraphPad Software	http://www.graphpad.com/
OriginPro8.0	OriginLab	https://www.originlab.com
SPSS22.0	IBM Corp.	https://www.ibm.com/

Experimental model and study participant details

Cell lines

The human breast carcinoma MCF-7 cell line and MDA-MB-231 cell line were obtained from American Type Culture Collection (ATCC). The cells were cultured in Gibco Dulbecco’s Modified Eagle medium (DMEM) (Thermo Fisher Scientific) supplemented with 10% FBS and 1% penicillin–streptomycin at 37°C in a humidified 5% CO₂ incubator. The mouse mammary carcinoma EO771 cell line was supplied by CH3 Biosystems (NY, USA). The EO771 cells were cultured in DMEM medium with 10% FBS and 1% penicillin–streptomycin at 37°C in a humidified 5% CO₂ incubator. The folate receptor α (FOLR1) gene mRNA expressions of EO771 cells and a known folate receptor-expressing mouse mammary carcinoma 4T1 cells supplied by ATCC were determined by the qPCR assay.

Mice

The female C57BL/6 mice were provided by Soochow University Laboratory Animal Center. Mice were housed in ventilated animal cabinets under controlled lighting conditions (12/12 h light-dark cycle) at environment temperature of 20-25°C. All the animal experiments were approved by the Animal Care and Use Committee of Soochow University (No. 202311A0044).

Method details

Complexation of DOX with Mg²⁺

To investigate the complexation of DOX with different concentrations of Mg²⁺, a series of mixed solutions of 0.144 mM DOX·HCl and different concentrations of MgSO₄ (0 – 500 mM) were prepared in 100 mM HEPES buffer, pH 7.8. 180 μl of each mixed solution was loaded into the well of 96 microplate, and the UV-visible absorption spectra and fluorescence spectra were measured with a Synergy NEO microplate reader (BioTek, VT, USA).

To further investigate the complexation of DOX with Mg²⁺ at different pH values, a series of mixed solutions containing 44.6 μM DOX·HCl and 500 mM MgAc₂·4H₂O with varying pH values adjusted by 1 M HCl for acidic solution or by 500 mM HEPES buffer for alkaline solution were prepared. The absorption spectra and fluorescence spectra were measured with the microplate reader.

Preparation of blank liposomes

The blank DSPC/Chol/DSPE-PEG2000 liposomes were prepared via the typical thin-film hydration and extrusion methods. Briefly, a mixture of DSPC, Chol and DSPE-PEG2000 with predetermined molar ratio of 55:40:5 was dissolved in dehydrated ethanol, distilled by rotary evaporation under reduced pressure at 65°C and dried in vacuo overnight. The dried lipid film was hydrated to 60 mM total lipids with 300 mM of aqueous MgAc₂ solution or with 300 mM of aqueous (NH₄)₂SO₄ solution, heated at 60°C and vortexed. The dispersed suspension was subjected to 5 freeze-thaw cycles and then stepwise extruded at 55°C with Nuclepore polycarbonate membrane filters of 400 nm and 100 nm pore sizes (twenty passes for each pore size). The 100 nm blank folate receptor targeting liposomes composed of DSPC/Chol/DSPE-PEG2000/DSPE-PEG2000-FA with a predetermined molar ratio of 55:40:3:2 were prepared following the same procedure. The 100 nm DSPC/Chol/DSPE-PEG2000 liposomes using 300 mM MgSO₄ or MgCl₂ instead of 300 mM MgAc₂ were also prepared in a similar way. The hydrodynamic particle sizes and Zeta potentials of the liposomes were determined with a Zetasizer Nano ZS90 analyzer (Malvern Instruments, England). The liposomes were stored at 4°C for later use.

Remote loading of DOX into MgAc₂ gradient liposomes

For the remote loading of DOX into liposomes, blank MgAc₂ gradient liposomes with or without folate receptor targeting were prepared by loading 2 ml of the blank liposomes in MgAc₂ solution on PD-10 desalting columns filled with Sephadex G-25 gel and collecting the opaque liposomes fractions eluted with saline. An aliquot of MgAc₂ gradient liposomes were mixed with 5 mg/ml of DOX·HCl stock solution to drug/total lipids (drug/TLs) molar ratios of 0.125-0.5, and 125 mM of final concentration of NaHCO₃ solution or HEPES buffers (pH 7.4 and pH 7.8). The mixed solutions were placed at room temperature or heated at different temperatures (45-60°C) for different times. An aliquot of each mixture was diluted with the medium, and the UV-visible absorption spectrum and fluorescence spectrum were measured with the microplate reader.

The MgAc₂ gradient liposomes and DOX mixtures incubated at 50-60°C for 5-15 minutes were purified using Sephadex G-25 gel column and 10 mM NaHCO₃/saline eluent. The purplish-red DOX loaded liposomes fractions were collected. The 100 nm DOX liposomes (DOX-Lip100) were also prepared via the typical remote loading method with the ammonium sulfate gradient liposomes.³ An aliquot of each purified liposomes sample, together with the unpurified counterpart as standard were digested by 1:2 (V/V) dilution with 150 mM HCl in isopropanol (IPA) (1:39, V/V of 6 M HCl/IPA). The absorbance at 480 nm was measured for DOX quantification. The DOX encapsulation efficiency was calculated as the ratio of DOX in purified liposomes to the unpurified counterpart. The hydrodynamic sizes and Zeta potentials of the purified liposomes were determined with the Zetasizer Nano ZS90 analyzer.

Cryo-transmission electron microscopy of Mg-DOX liposomes

The micromorphology of the prepared 100 nm MgAc₂ gradient liposomes remote loaded with DOX (Mg-DOX-Lip100) was observed with a FEI Talos F200C cryogenic transmission electron microscope (Cryo-TEM) equipped with a tilt cryo-holder and Ceta 4KX4k detector and Xplore3D software. For the sample preparation, about 2.5 μl of liposomes was dropped onto the 300 mesh copper grid coated with a holey carbon film and blotted with a filter paper to form a thin liquid film, then quickly plunged into pre-cooled liquid ethane using a Vitrobot cryo-preparation station. The vitrified sample was stored in liquid nitrogen and then transferred to the microscope and the cryo-TEM images were acquired.

Storage stability of the Mg-DOX liposomes

The prepared 100 nm Mg-DOX-Lip100 containing 9 mM of total lipids and 1.44 mM of DOX in 10 mM NaHCO₃/saline were stored at 4°C. The particle sizes of the liposomes and DOX retention inside liposomes were monitored during 3 months of storage. For the measurement of DOX retention inside liposomes, the liposomes were separated with the G25 gel desalting column and DOX in the eluted liposome fractions was determined by the colorimetric method. The percentage of DOX inside liposomes was calculated by dividing the DOX amount in the collected liposomes fraction by DOX amount in the initial liposomes before separation.

DOX release from liposomes under different pH conditions

The DOX release from the Mg-DOX-Lip100 and DOX-Lip100 each containing 1.44 mM DOX was studied using the dialysis method. 10 mM NaAc/HAc buffer in saline, pH 5.5, 10% FBS in saline, pH 7.0, and 5 mM NaHCO₃ in saline, pH 7.4 were separately prepared as dialysis solution. Each 2 ml of the liposomes were loaded into a 50 kDa MWCO dialysis membrane tube (28 mm Width, Spectra/Por(R) 6 CR), sealed and immersed in 90 ml of the dialysis solution. The solution was kept under 37±0.5°C and gently stirred with a magnetic bar at a speed of 100 rpm. 1 ml of the dialysis solution was pipetted at 0, 1, 2, 3, 6, 12, 24, 36 and 48 h, and 1 ml stock dialysis solution was supplied at each time point to keep the dialysis volume constant. DOX concentrations of the pipetted dialysis solutions were determined by the Ex480nm/Em590nm fluorescence assay with microplate reader after 1:2 V/V dilution with 150 mM HCl/IPA, and used to calculate the percentage of DOX released from liposomes.

In vitro assay of tumor cells uptake of Mg-DOX liposomes

The MCF-7 cells and MDA-MB-231 cells were cultured in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin at 37 °C in a humidified 5% CO₂ incubator. The cells approaching 90% confluence were washed twice with sterile PBS and digested with 0.25% trypsin-EDTA solution for about 2 minutes. Then the cells were centrifuged and resuspended in the DMEM medium to 5 × 10⁵ cells/ml and each 1 ml of the suspension was pipetted into a well of 24-well microplates planted with a 13 mm diameter of round glass slide. After 24 h incubation, DOX-Lip100, Mg-DOX-Lip100 or the purified 100 nm folate receptor targeted Mg-DOX liposomes (FA-Mg-DOX-Lip100) were added into the wells to 108 μM of total lipids and 17.2 μM of DOX. After continuous incubation for 3 h, 6 h or 12 h, the wells were washed twice with PBS, and 4% paraformaldehyde was added to fix the cells (3 wells at each time point for each drug). Then the round glass slides were transferred onto clean glass slides and stained with DAPI solution. The cells were observed with a Confocal laser scanning microscope (Olympus FV1200, Japan) and the fluorescent microscopic images were acquired with Ex405nm/Em461nm for DAPI stained nuclei and Ex559nm/Em572nm for DOX.

In another experiment, each 1 ml of the tumor cell suspension containing 5 × 10⁵ cells was transplanted in a well of 24-well microplate and incubated for 24 h. Then DOX-Lip100, Mg-DOX-Lip100 or FA-Mg-DOX-Lip100 were added into the wells, respectively, to final DOX concentrations of 10, 20, 40 and 80 μg/ml. After 3 hours of incubation, the culture medium was aspirated, and the cells were washed twice with PBS. Then 1 ml of water and 150 mM HCl/IPA mixture (1:2 V/V) was added to each well and stored avoiding light in an incubator for 10 hours. The solution in the well was collected and centrifugated. The DOX concentration in the supernatant was determined by the fluorescence assay.

In vitro cytotoxicity assay of Mg-DOX liposomes

The MCF-7 and MDA-MB-231 cells cultured in the DMEM medium containing 10% FBS in the logarithmic growth phase were digested with 0.25% trypsin-EDTA solution, centrifuged and resuspended in the culture medium to 6 × 10⁴ cells/ml. Each 100 μl of the cell suspension was transferred into a well of a 96-well microplate and cultured for one day to allow the cells to attach on the well surface. Then 100 μl of DOX-Lip100, Mg-DOX-Lip100 or FA-Mg-DOX-Lip100 each diluted in the culture medium was added into the well to 0 - 64 μg/ml of final DOX concentration, respectively. After 24 h incubation at 37°C and 5% CO₂ atmosphere, the culture media were aspirated and the wells were washed twice with PBS. Thereafter, 100 μl of DMEM medium and 10 μl of 5 mg/ml Tetrazolium salt (MTT) from the MTT test kit was added to the well. After further 4 h incubation, the medium in the well was replaced by the solution used to dissolve the produced formazan. The OD value at 570 nm of the well was measured with the microplate reader according to the kit instruction. The cell viability was calculated according to the following equation.

$$Cellviability(\%)=\frac{{OD}_{test}-{OD}_{blank}}{{OD}_{control}-{OD}_{blank}}\times 100\%$$

OD_test, OD_control and OD_blank represent the OD values of experimental drug group, the DMEM medium as control group and the blank well group, respectively.

The cell viability versus liposomal DOX concentration curves were fitted with GraphPad Prism 8.0 (GraphPad Software, CA, USA) to calculate the IC50 value.

The in vitro cytotoxicity assay of MCF-7 and MDA-MB-231 cells incubated with DOX·HCl, DOX-Lip100 and Mg-DOX-Lip100 for 48 h at 0 - 32 μg/ml of final DOX concentration was performed by CCK assay. After 48 h incubation at 37°C and 5% CO₂ atmosphere, the culture media of the well culturing the tumor cells were aspirated and the wells were washed twice with PBS. Thereafter, 100 μl of 1:1 diluted CCK8 reagent in culture medium was added to the well. After further 2 h incubation, the OD value at 450 nm of the well was measured with the microplate reader according to the kit instruction. Then the cell viability was calculated according to the same equation as the MTT assay.

Pharmacokinetics and biodistribution of Mg-DOX liposomes

The 6-8 weeks old female C57BL/6 mice provided by Soochow University Laboratory Animal Center were used for the pharmacokinetics and biodistribution study. Totally 60 mice were used for the pharmacokinetic experiment of the DOX-Lip100, Mg-DOX-Lip100 and FA-Mg-DOX-Lip100 liposomes, and each liposomes group was assigned randomly with 20 mice. The liposomes were IV administered to the mice at an equivalent DOX dose of 5 mg/kg body weight by tail vein injection, respectively. Blood samples were taken in heparinized tubes by retro-orbital puncture at 2 min, 5 min, 15 min, 30 min, 1 h, 2 h, 6 h, 12 h, 24 h and 36 h (5 samples at each time point for each group). The collected blood samples were centrifuged at 5000 rpm for 15 min at 4°C, the supernatant was aspirated, and 1:10 V/V diluted with water, and then added 2X volume of 150 mM HCl/IPA. 1 h later, the mixture was centrifuged, and DOX concentration of the supernatant was measured by the fluorescence assay at Ex480nm/Em590nm. The DOX standard curve was established with a series of DOX·HCl solutions diluted in 10% FBS and using same pretreatment procedure. The plasma DOX concentration versus time curves were analyzed by Non-compartmental Analysis with Phoenix WinNonlin (version 6.4, Certara USA Inc., NJ, USA).

Fifteen mice were used for the biodistribution characterization of the DOX-Lip100, Mg-DOX-Lip100 and FA-Mg-DOX-Lip100 liposomes, and each liposomes group was assigned with 5 mice. Liposomes were IV administered to the mice (5 mice for each drug group) at an equivalent DOX dose of 5 mg/kg body weight by tail vein injection, respectively. The mice were euthanized 48 h post liposomes injection. The lung, heart, liver, spleen, kidneys and muscle samples were dissected. About 0.1 g portion of each tissue sample was weighed accurately and mixed with 1 ml of 150 mM HCl/IPA, then homogenized with Zirconia grinding beads using a HZ-II high speed tissue homogenizer (Wuhan Servicebio Technology, China) for 10.5 minutes and stored at 4°C overnight. The samples were centrifuged at 5000 rpm and 4°C for 15 minutes. The supernatant was aspirated and mixed with water (2:1, V/V). The DOX concentration of the mixture was measured by the fluorescence assay at Ex480nm/Em590nm and used to calculate the DOX concentration of the tissue.

Experimental therapy with breast tumor bearing mice

The therapeutic effect of the Mg-DOX liposomes was studied in vivo with an orthotropic EO771 cells breast tumor mouse model. To establish the orthotropic breast cancer tumor mouse model, in vitro cultured EO771 cells approaching 90% confluence were digested with the trypsin-EDTA solution, centrifuged and resuspended in PBS. An aliquot of 0.2 ml of 2 × 10⁵ EO771 cells in PBS was inoculated into the subcutaneous mammary gland near the left lower limb of the female C57BL/6 mouse (6-8 weeks, 17-20 g body weight). The palpable tumor size was measured with a digital caliper. The tumor volume (V) was calculated using the equation $V=L\times W\times W/2$, L and W represent the maximal length of the tumor and the width vertical to the maximal length, respectively. Twenty-four female C57BL/6 mice of average tumor volume of 134±54 mm³ were assigned randomly into 4 groups, including control group, DOX-Lip100 group, Mg-DOX-Lip100 group and FA-Mg-DOX-Lip100 group (6 mice for each group) with balanced body weights and tumor sizes. Each mouse of the experimental groups was IV injected by tail vein at equivalent DOX dose of 5 mg/kg body weight on day 0, 4, 8 and 12. The body weights and tumor sizes were measured every two days. The endpoints of the study were tumor size reaching volume of 2000 mm³. Average survival times of the groups were calculated by the Kaplan-Meier Estimation using SPSS22.0 software (IBM Corp., Armonk, NY). The heart, liver, spleen, lungs and kidneys were dissected from the mice reaching the in vivo endpoints and cut into 2-3 mm thick pieces. These tissue samples were separately fixed in 4% paraformaldehyde, dehydrated with ascending concentrations of ethanol solutions, cleaned in xylene, and embedded in paraffin. The 4 and 10 μm thick histological sections of the tissue samples were prepared and H&E stained for routine histopathological examination with microscopic imaging.

Quantification and statistical analysis

The experimental data were expressed as mean ± standard deviation (SD). The inter-group differences of data were assessed with independent-samples t test for two groups and one-way ANOVA for multiple group comparisons using SPSS10.0 (IBM Corp., Armonk, NY). The levels of statistical significance of difference were set at probabilities of ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

Published: August 9, 2025