Introduction

Doxorubicin (DOX), a broad-spectrum anticancer drug, has extremely strong antitumor activity [1, 2]. However, as a cell cycle non-specific drug, it has a strong cytotoxic effect on both tumor and normal cells, which has also triggered its strong side effects. In addition to the common side effects such as bone marrow suppression, vomiting, nausea, and alopecia, it can damage myocardial cells through oxygen free radicals due to its significantly high affinity with the myocardium, thereby causing severe cardiac toxicity [3,4,5]. Liposomes have been well developed as drug delivery systems since their discovery in 1964 by Bangham et al. because of their unique advantages [6]. The encapsulation of DOX into liposomes not only increases drug accumulation in tumors but also effectively reduces its side effects, especially cardiotoxicity [7, 8].

Doxil® was the first liposomal DOX, with a lipid composition of hydrogenated soybean phosphatidylcholine (HSPC), cholesterol (CH), and mPEG2000-DSPE, approved in 1995 by Sequus Pharmaceuticals [9]. The coverage of a PEGylated layer on the surface of liposomes by “STEALTH” technology also avoids immune system recognition and phagocytosis and achieves long circulation in vivo [10, 11]. Although PEG-modified liposomes greatly reduce the cardiotoxicity of DOX, they cause skin toxicity, such as palmar-plantar erythrodysesthesia (PPE), with an incidence approaching 50% [12, 13]. Morever, Doxil® did not show significantly improved efficacy in clinical treatment [14, 15]. Therefore, DOX-loaded liposomes still need to be optimally formulated to increase stability, cycle time, and efficacy.

A previous study showed that SA-modified DOX-loaded liposomes prepared using HSPC/CH exhibited better antitumor effects in the S180 tumor-bearing mouse model than PEGylated liposomal DOX [16, 17]. Sialic acid (SA), an acidic amino sugar containing nine carbon atoms, plays an important role in many physiological and pathological processes [18]. Currently, SA binding proteins mainly include Selectin, Siglec, and factor H [19, 20]. Siglec-1 expressing tumor-associated macrophages (TAMs) affects tumor occurrence and development by promoting immunosuppression, angiogenesis, and tumor invasion and metastasis [21, 22]. Some TAMs tend to enter the hypoxic region of the tumor and deliver nano-drug carriers into the tumor tissue. Finally, the tumor is treated by specifically clearing TAMs and killing the tumor cells. In this study, we used a specific combination of SA and Siglec-1 to deliver drugs to TAMs.

Although SA-modified liposomes achieved a good antitumor effect [23, 24], the stability of this formulation needs to be further improved. Egg sphingomyelin (ESM) is a major lipid component of animal cell membranes, along with phosphatidylcholine (PC), and constitutes more than 50% of the membrane lipids. Although PC and SM have similar structures, their physicochemical properties are different [25,26,27]. Specifically, the fatty chains in the ESM and HSPC structures are linked by amide and ester bonds, respectively; therefore, ESM is less prone to acid hydration and enzymatic degradation than PC [28]. It was also pointed out that the nitrogen atom of the amide bond in the ESM structure could form an intermolecular hydrogen bond with the oxygen atom in CH, resulting in good retention characteristics and in vivo stability of water-soluble drugs [28, 29].

To the best of our knowledge, phospholipids play a crucial role in the stability of liposome preparations, and there has been no study reporting DOX-loaded liposomes prepared by ESM/CH. In this study, we prepared SA-modified DOX-loaded liposomes using HSPC/CH and ESM/CH as membrane materials to improve the stability and antitumor effect of DOX preparations.

Materials and Methods

Materials

Doxorubicin hydrochloride (DOX, purity ≥ 99%) was obtained from Beijing Huafeng United Technology Co., Ltd. (Beijing, China); hydrogenated soybean phosphatidylcholine (HSPC) was obtained from Lucas Meyer (Germany); egg sphingomyelin (ESM) was obtained from Lipoid GmbH (Ludwigshafen, Germany); cholesterol (CH) was obtained from Nanjing Xinbai Pharmaceutical Co., Ltd. (Nanjing, China); a sialic acid-cholesterol conjugate (SA-CH) was self-synthesized in the laboratory; Sephadex G50 (G50) was obtained from Shanghai YSRIBIO Industrial Co., Ltd. (Shanghai, China); 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was obtained from Sigma-Aldrich (Saint Louis, USA); 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (DiR iodide), RPMI-1640 cell culture, and Trypsin were obtained from Dalian Meilun Biotechnology Co., Ltd. (Dalian, China); FBS was obtained from Zhejiang Tianhang Biotechnology Co., Ltd. (Hangzhou, China).

Cells and Animals

Mouse S180 sarcoma cells and RAW264.7 monocyte-macrophage leukemia cells were supplied by the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Male Wistar rats (weighing 180–220 g) and male Kunming mice (weighing 18–22 g) were purchased from the Central Animal Laboratory of Shenyang Pharmaceutical University (Liaoning, China). All animal experiments were conducted in accordance with the guidelines of the Animal Welfare Committee of Shenyang Pharmaceutical University.

Preparation of DOX-loaded Liposome

Ethanol was added to a prescribed amount of liposome membrane material (HSPC/ESM, CH, a sialic acid-cholesterol conjugate (SA-CH)) and dissolved in a certain temperature water bath (HSPC/65℃, ESM/70℃). After the complete dissolution of membrane materials, ethanol was volatilized, and citrate buffer (200 mM, pH 4.0) was injected into the membrane materials at a speed of 4 mL/min. The initial product was obtained by stirring in a water bath at a certain temperature (HSPC/65℃, ESM/70℃) for 1 h. The product was treated by extrusion through polycarbonate membranes of 400 nm (5 times), 200 nm (10 times), 100 nm (10 times), and 80 nm (5 times) sequentially, thereby yielding blank liposomes with different prescriptions (phospholipid concentration of 50 mg/mL).

A phosphate solution (500 mM) was used to adjust the outer aqueous phase pH value of blank liposomes to 7.0, thus establishing a gradient of the inner and outer aqueous phase pH of blank liposomes. Then, the above pH gradient blank liposomes and DOX solution (5 mg/mL) were mixed and incubated at a certain temperature (HSPC/65℃, ESM/70℃) in a water bath with stirring for 20 min (EPI/phospholipid ratio 1/10, w/w). DOX-loaded liposomes were obtained after incubation in an ice-water bath for 5 min.

Characterization of DOX-loaded Liposome

Morphology

A slightly diluted DOX-loaded liposome preparation was dropped onto a copper grid covered with carbon film and negatively stained with 1.0% (w/v) phosphotungstic acid for 2 min. After natural drying at 23 ± 2℃, the morphology of each DOX-loaded liposome was observed and photographed using a transmission electron microscope (TEM) (JEOL JEM-2100, Japan).

Particle size, Polydispersity Index, Zeta Potential, and Entrapment Efficiency (EE%)

Particle size, polydispersity index (PDI), and zeta potential were determined at 23 ± 2℃ using a particle sizer (Nicomp-380, Particle Sizing Systems, USA). Sephadex G-50 was used to remove free DOX, and methanol was used to extract DOX from liposomes. Then, the content of DOX was measured at 480 nm using a multifunctional plate reader (Thermo Scientific, USA). We used the following formula to calculate the EE%:

$$EE\%={\mathrm{A}}_{\mathrm{after}}/{\mathrm{A}}_{\mathrm{before}}\times 100\%$$

where Abefore and Aafter represent the absorbance of liposomes with and without the free drug, respectively.

In Vitro Release Assay

Ammonium chloride (80 mM) and histidine salt buffer (10 mM) were selected as release media. Sodium hydroxide was also used to adjust the pH to 7.4, and 5% Glu was added to adjust for isotonicity. A dialysis bag (MW: 10 kDa) containing the DOX formulation was placed in a dissolution cup at 37℃ and stirred at a constant speed of 100 rpm/min. The dialysate was aspirated at 1 h, 2 h, 4 h, 8 h, 12 h, 24 h, and 48 h, and an equal amount of release medium was placed at the same temperature. The samples were spotted on a 96-well blackboard, and the concentration of DOX was determined using a multifunctional plate reader at λex = 482 nm and λem = 590 nm.

Long-term Storage Stability

Each DOX-loaded liposome was stored at 4 ± 2℃ with light-free. Particle size and EE% of each sample were determined at 0 month, 1 month, 2 months, and 3 months.

RAW264.7 Cell Uptake Assay

RAW264.7 cells were inoculated at a density of 6 × 105 cells/well into 6-well plates and incubated at 37℃ for 24 h. Next, the cell supernatants were replaced with DOX-S, DOX-HCL, DOX-HSL, DOX-ECL, and DOX-ESL (all at a DOX concentration of 50 μg/mL) diluted in a cell culture medium and incubated with the cells for 2 h. Meanwhile, SA pre-saturation groups, named DOX-HSL (SA) and DOX-ESL (SA), were set up. Briefly, excess SA solution was added to RAW264.7 cells to saturate the cell surface with SA receptors prior to incubation with the formulation. Finally, the cells were washed with PBS and quantified using flow cytometry (FCM) and confocal laser microscopy (CLSM).

MTT Assay

RAW264.7 and S180 cells were seeded in 96-well plates at a density of 5 × 103 cells/well. After 12 h in a CO2 incubator, the cells were incubated with different concentrations of DOX formulations (0.04 μM, 0.2 μM, 1 μM, 5 μM, 25 μM, and 125 μM) for 48 h. Subsequently, the MTT solution and the triple solution (10% SDS/5% isobutanol/12 mM HCl, w/v/v) were sequentially added to a 96-well plate. After the completion of the MTT assay, the optical density (OD) of each well was measured at a wavelength of 570 nm using a multifunctional plate reader.

In Vivo Pharmacokinetic Analysis

Male Wistar rats were randomly divided into five groups (n = 3) and then treated with DOX-S, DOX-HCL, DOX-HSL, DOX-ECL, or DOX-ESL at a dose of 5 mg/kg DOX. Subsequently, blood was sampled from the orbit at 0.0163 h, 0.087 h, 0.25 h, 0.5 h, 1 h, 2 h, 4 h, 8 h, 12 h, 24 h, and 48 h after injection, respectively, and the plasma was harvested after centrifugation at 5000 rpm for 10 min. The drug-containing supernatant was harvested after centrifugation at 3000 rpm for 5 min from a mixture consisting of 100 μL plasma and 900 μL methanol. The supernatants were transferred to 96-well plates, and the fluorescence intensity of the drug was measured at λex = 482 nm and λem = 590 nm.

Tumor Targetability

Tumor-bearing mice were randomly divided into four groups (n = 3) and injected with DiR-HCL, DiR-HSL, DiR-ECL, or DiR-ESL (DiR 1 mg/kg) via the tail vein. Mice were anesthetized at 1 h, 4 h, 8 h, 24 h, and 48 h after injection, placed under an in vivo FX Pro imaging system (Bruker, USA), and photographed (ex/em 750/790 nm). After taking photographs, the mice in each group were sacrificed. In addition, the heart, liver, spleen, lung, kidney, thymus, and tumor tissues of mice in each group were removed and similarly placed under the in vivo FX Pro imaging system for photography. The ROI live imaging software was used for the semiquantitative analysis of each tissue.

The distribution of DOX in the tumor tissues of each group was further determined. Briefly, sixty tumor-bearing mice were randomly divided into four groups with 15 mice in each group, namely DOX-HCL, DOX-HSL, DOX-ECL, and DOX-ESL groups. The mice were killed at 1 h, 4 h, 8 h, 24 h, and 48 h after administration, respectively, and their tumor tissues were dissected. Tumor tissue was homogenized, and 100 μL of the homogenate and 900 μL of the hydrochloric acid–methanol-in-water mixture were centrifuged at 12,000 rpm for 10 min to extract the drug. The supernatants were then transferred to 96-well plates, and the fluorescence intensity of the drug was measured using a multifunctional plate reader (ex/em 482/590 nm).

Biocompatibility of Drug-free Liposomes

Fifteen healthy Kunming mice were randomly divided into four groups, three in each group, namely control, HCL, HSL, ECL, and ESL. The dosage was 100 mg·lipid·kg−1. The drug was administered continuously for 7 days, and the weight changes of mice were detected.

Antitumor Assay

S180 cells were inoculated into the axilla of male Kunming mice at a density of 1 × 107 cells/mL to establish a mouse xenograft model. When the tumor volume reached approximately 100 mm3, the mice were randomly divided into six groups (n = 6). The control group was administered 10 mL/kg of 5% Glu by tail vein injection. DOX-S, DOX-HCL, DOX-HSL, DOX-ECL, and DOX-ESL were administered by tail vein injection of 5 mg/kg DOX in the other groups. Mice in each group were dosed once every three days for a total of five times during the pharmacodynamic period, and tumor volume and body weight were measured every two days. Tumor volume (V, mm3) was calculated as follows: V = 1/2 × a × b2, where a and b represent the major and minor diameters of the tumor, respectively. The tumor volume inhibition rate (TIRv) and net body weight (g) were calculated as follows: TIRV = (VControl − VTreated)/VControl × 100% and net body weight = body weight − tumor weight. In addition, in order to simultaneously consider the inhibitory effect of the preparation on tumor cells and the non-specific damage to the body, the tumor inhibition index (TIindex) was proposed by our laboratory [30]. The TIindex was calculated as follows: TIindex = body weight/tumor weight.

Histopathological Examination

At the end of the pharmacodynamic experiment, the heart, liver, spleen, lungs, kidneys, and tumor tissues of each mouse were dissected and washed with PBS. The surface water of each tissue was sucked out of the filter paper and fixed in 4% paraformaldehyde for 48 h. After a series of treatments, such as embedding, slicing, and staining, each tissue was observed and photographed under a microscope (Nikon Eclipse Ti-SR, Japan).

Immunohistochemical Staining Examination

F4/80 (a mature mouse macrophage marker) was used to mark the tumor tissue in each group as a test index for TAMs. Briefly, sections were incubated with CD68 monoclonal antibody (Santa Cruz, USA) for 12 h at 4℃ and then with HRP-labeled goat anti-rabbit antibody (Zhongshan Jinqiao, China) for 4 h at 23 ± 2℃. Hematoxylin was used to stain nuclei. Finally, the sections were observed under a microscope (Nikon Eclipse Ti-SR, Japan), and the proportion of positive cells (TAMs) was calculated using Image-Pro Plus 6.0.

Results

The characterization of liposomes composed of two different membrane materials was conducted in this study. As shown in Table I, the particle sizes of the liposomes were 100–105 nm, the PDI was < 0.1, the zeta potential was in the range of − 20 to − 15 mV, and EE% was > 95%. Liposome morphology was further characterized using TEM. As shown in Fig. 1a, each DOX-loaded liposome was round or quasi-round with a relatively uniform size. In addition, the liposome particle size observed under TEM was in accordance with the particle size measurement data from the Nicomp-380 particle size meter (approximately 100 nm).

Table I Composition, Particle Size, Polydispersity Index, Zeta Potential, and EE% of Different DOX Liposome Preparations (Data Represented as mean ± S.D.) (n = 3)
Full size table
Fig. 1
figure 1

(a) The transmission electron microscopy images of a DOX-HCL, b DOX-HSL, c DOX-ECL, and d DOX-ESL. (b) In vitro cumulative release of DOX from different DOX formulations (n = 3). (c) Change of particle size a and EE% b of different DOX liposomes during a storage period of three months. Stability data were represented for initial, 1 month (1 M), 2 months (2 M), and 3 months (3 M) (n = 3) (*p < 0.05, **p < 0.01)

Full size image

The release behavior of each DOX prescription is shown in Fig. 1b. DOX-S release was the fastest, with almost complete release at approximately 8 h. The four DOX-loaded liposomes exhibited a significantly lower release rate compared with DOX-S (P < 0.05), indicating that the preparation had a drug-sustained-release effect. After 48 h, the release rates of DOX-loaded liposomes were in the order: DOX-HCL ≈ DOX-HSL > DOX-ESL > DOX-ECL.

Figure 1c illustrates the changes in particle size and EE% of each liposome after three months of storage at 4 ± 2℃. The particle size of DOX-HCL and DOX-HSL increased significantly (P < 0.05), while those of DOX-ECL and DOX-ESL showed no significant change (P > 0.05). Moreover, the EE% of DOX-HCL, DOX-HSL, and DOX-ESL decreased significantly after three months (P < 0.05), whereas no significant change was observed in the EE% of DOX-ECL (P > 0.05).

As shown in Fig. 2a and Table II, each DOX preparation had a dose-dependent inhibitory effect on the growth of S180 and RAW264.7 cells. The inhibitory effect of each liposome on the cells was lower than that of DOX-S. Regardless of whether the DOX-loaded liposomes were prepared using HSPC/CH or ESM/CH, the SA modification significantly increased the inhibitory effect of drugs on the cells (P < 0.05). In addition, liposomes prepared with ESM/CH exhibited significantly weaker cell inhibition than DOX-loaded liposomes prepared with HSPC/CH (P < 0.05).

Fig. 2
figure 2

(a) Cell viability of the RAW264.7 cells a and S180 cells b incubated with different DOX formulations (n = 3). (b) The cellular uptake of different DOX formulations in RAW264.7 cells. a Flow cytometry analysis of the RAW264.7 cells treated with different DOX formulations at 37℃. b Quantitative results of flow cytometry analysis (n = 3). (c) Confocal images of the RAW264.7 cells incubated with different DOX formulations at 37℃. Note: scale bar = 50 μm (**p < 0.01)

Full size image
Table II IC50 Values of Different DOX Formulations in S180 and RAW264.7 Cell Lines (Data are Presented as mean ± S.D.) (n = 3)
Full size table

The results of FCM (Fig. 2b) and CLSM (Fig. 2c) collectively revealed that the fluorescence intensity of each DOX prescription taken up by RAW264.7 cells was in the following order: DOX-S > DOX-HSL > DOX-HCL ≈ DOX-HSL (SA) > DOX-ESL > DOX-ECL ≈ DOX-ESL (SA). The fluorescence intensity of DOX-HSL was significantly higher than that of DOX-HCL and DOX-HSL (SA) (P < 0.01), and the fluorescence intensity of DOX-ESL was significantly higher than that of DOX-ECL and DOX-ESL (SA) (P < 0.01).

As shown in Fig. 3 and Table III, DOX-S was completely eliminated from blood circulation within 15 min, whereas DOX-loaded liposomes significantly delayed drug elimination (P < 0.05). The AUC0–48 h values of the DOX-HCL, DOX-HSL, DOX-ECL, and DOX-ESL groups were 1168-, 1232-, 2117-, and 1783-fold higher than that of the DOX-S group, respectively.

Fig. 3
figure 3

Plasma concentrations of DOX in Wistar rats injected with different DOX formulations (n = 3)

Full size image
Table III The Main Pharmacokinetic Parameters of Different DOX Formulations in Wistar Rats (Data are Presented as mean ± S.D.) (n = 3)
Full size table

To further evaluate the biodistribution of each liposome in vivo, DiR dye was chosen as a fluorescent probe in live imaging experiments in mice (Fig. 4a,b). The fluorescence intensities of the tumors in the DiR-HCL and DiR-HSL groups reached a maximum at 8 h and then gradually decreased, showing only a weak fluorescence at 48 h. The fluorescence intensities of the tumors in the DiR-ECL and DiR-ESL groups reached a maximum of 24 h. On the other hand, the results of the DOX distribution experiment in tumor tissue further proved the different distribution behaviors of different liposomes in vivo (Fig. 4c). In addition, although the fluorescence intensity of DOX-ECL decreased at 48 h, it was significantly higher than that of DOX-loaded liposomes prepared with HSPC/CH (P < 0.05).

Fig. 4
figure 4

(a) In vivo fluorescence images of S180 tumor-bearing mice at different time points following injection of DiR-labeled liposomes. (b) Ex vivo DiR fluorescence images of tumor and major organs at 48 h. (c) Quantification of the DOX distribution in the tumor after injection of different DOX formulations (n = 3)

Full size image

As shown in Fig. 5, the body weight of mice in each group showed a steady rising trend within 7 days after the administration of drug-free liposomes, and there was no significant difference with the control group, which indicated the low toxicity of drug-free liposomes.

Fig. 5
figure 5

Body weight of mice administrated with different drug-free liposomes (n = 3)

Full size image

As shown in Fig. 6a, the tumors grew rapidly in the control group mice, and each DOX formulation inhibited the tumor growth rate to some extent. The tumor volume of DOX-HSL was smaller than that of DOX-HCL in liposomes prepared with HSPC/CH (P < 0.01), and the tumor volume of mice in the DOX-ECL group was smaller than that of the DOX-ESL group (P < 0.01). To better compare the antitumor efficacy of different DOX formulations, TIRV after 30 days was calculated separately for each group of mice to quantify the antitumor efficacy of each DOX formulation. Figure 6b indicates that the inhibitory effect of each DOX formulation on S180 tumors in mice occurred in the order of DOX-ECL > DOX-ESL > DOX-HSL > DOX-HCL > DOX-S, in accordance with the results shown in Fig. 6a,c. As shown in Fig. 6d,e, the percent change in body weight and net percent change in body weight before day 24 did not significantly differ between DOX-loaded liposomes during the pharmacodynamic experiment. However, the body weight gain after 24 days was significantly higher in the ESM/CH-prepared liposomes than in the HSPC/CH-prepared liposomes. As shown in Fig. 6f, all the rats in the untreated group and the DOX-S group died on days 32 and 31 after modeling, respectively, and all the rats in the DOX-HCL group died on day 60. On day 75, the survival rates of DOX-HSL, DOX-ECL, and DOX-ESL groups were 37.5%, 75%, and 50%, respectively.

Fig. 6
figure 6

(a) Tumor growth curves of tumor-bearing mice receiving diverse DOX formulations (n = 1 ~ 6). (b) Tumor inhibition index change in tumor-bearing mice receiving diverse formulations (n = 1 ~ 6). (c) Tumor volume inhibition rate on day 30 in tumor-bearing mice receiving diverse formulations (n = 1 ~ 6). Percentage of body weight change (d) and net weight change (e) of tumor-bearing mice treated with different DOX formulations (n = 1 ~ 6) (**p < 0.01). (f) Percent survival of the tumor-bearing mice following administration of DOX formulations (n = 8)

Full size image

Figure 7a depicts the pathological examination of each tissue. The heart, liver and spleen were all damaged to different extents in the DOX-S group, while DOX-HCL showed slight heart and liver injuries. No abnormalities were found in the heart, liver, spleen, lungs, or kidneys in the other preparation groups. Notably, a large number of necrotic areas were observed in the tumor tissues of the DOX-HSL, DOX-ECL, and DOX-ESL cohorts, and almost all of the tumor tissues in the DOX-ECL group were necrotic.

Fig. 7
figure 7

(a) Histopathology of heart, tumor, liver, lung, kidney, and spleen sections with hematoxylin and eosin (H&E) staining of different DOX groups. The sites where the solid arrows point represented the abnormal region. The scale bar represented 100 μm. (b) a Immunohistochemistry staining of F4/80 of different experimental groups within tumor stroma. The scale bar represents 100 μm. b The positive area of TAMs (%) in each group. The positive rate was calculated with the positive cell area/the total cell area × 100% (n = 1 ~ 6) (**p < 0.01, ***p < 0.001)

Full size image

As shown in Fig. 7b, the F4/80 positive expression rates decreased after treatment with DOX in all preparation groups when compared with the control group. Specifically, the order of positive F4/80 expression in each DOX preparation group was as follows: DOX-S > DOX-HCL > DOX-HSL > DOX-ESL > DOX-ECL.

Discussion

There existed obvious intermolecular hydrogen bonds in the ESM/CH system that resulted in the close accumulation of the bilayer, which played a good role in protecting the drugs in the internal water phase and the drugs were not easy to leak [29]. There was no statistically significant difference in the release rates of DOX-HCL and DOX-HSL (P > 0.05), which suggests that SA modification in liposomes prepared by HSPC/CH did not change the in vitro release behavior of drugs in the preparations [17, 31]. However, in the liposomes prepared using ESM/CH, the cumulative drug release in the DOX-ESL group was significantly higher than that in the DOX-ECL group (P < 0.05). This finding suggests that the SA modification accelerated the release of DOX in the liposomes prepared using this membrane material [32].

As mentioned previously, the internal aqueous phase of the liposomes is citrate buffer (pH = 4), which induces liposome hydrolysis to some extent. However, ESM is less sensitive to acid and enzymatic degradation than HSPC because it is linked only by amide and ether bonds. Thus, liposomes prepared from the ESM/CH system had the lowest sensitivity to acid, but the addition of SA reduced the stability of this system [28].

To evaluate whether SA-modified DOX-loaded liposomes could effectively increase the internalization effect of TAMs on drugs, RAW264.7 cells were subjected to an uptake profile analysis as an in vitro macrophage model [17, 33]. The fluorescence intensity of DOX-S was significantly higher than that of other DOX-loaded liposome groups, indicating that free DOX could more easily enter cells via passive diffusion. In each liposome group, the SA modification significantly increased the accumulation of formulations in the cells, regardless of whether the liposomes were prepared with HSPC/CH or ESM/CH (P < 0.01). However, this increased uptake phenomenon disappeared after the SA solution pre-saturated RAW264.7 cell surface receptors (P > 0.05), which illustrates that the cellular uptake was mediated by the specific binding of SA to Siglec-1 expressed by macrophages [34]. In addition, the results showed that the uptake of DOX-loaded liposomes prepared with ESM/CH by RAW264.7 cells was significantly lower than that of liposomes prepared with HSPC/CH. This result indicates that ESM/CH system had excellent drug retention characteristics in the inner aqueous phase [28, 35].

The AUC0–48 h of DOX-HCL and DOX-HSL were not significantly different (P > 0.05), which suggests that the SA-CH modification did not alter the pharmacokinetic characteristics of the formulation prepared by HSPC/CH [17, 31]. However, DOX-ECL significantly increased the AUC0–48 h of DOX in DOX-loaded liposomes prepared by ESM/CH (P < 0.05). In addition, DOX-loaded liposomes prepared with ESM/CH had a longer circulation effect than DOX-loaded liposomes prepared with HSPC/CH. A possible reason for this is that the liposomes prepared by ESM/CH had a significantly reduced interaction with plasma proteins, thus effectively avoiding the recognition and phagocytosis of monocyte macrophages [28, 29]. Moreover, the drug retention ability of liposomes is closely related to their antitumor activity [36], which was confirmed in the “antitumor assay.”

Tumor tissues were homogenized to further verify the distribution of the formulation. Among them, the fluorescence intensity of DOX-HSL in liposomes prepared with HSPC/CH was higher than that of DOX-HCL, which indicated that DOX-HSL was able to enter tumor-recruited macrophages through Siglec receptor-mediated endocytosis and thus accumulate more in tumor tissues. The fluorescence intensity of DOX-ECL in liposomes prepared by ESM/CH was higher than that of DOX-ESL, and in general, DOX-ECL accumulated the most in tumor tissues and had the longest retention time. This can be attributed to the long cycle times of the ESM/CH system, which was disrupted by SA insertion (Fig. 4C) [37,38,39].

Next, the in vivo antitumor activity of the liposomes in each group was further determined. To take into account the level of tumor inhibition by the preparation and the non-specific damage to the body, this study combined the tumor volume and body weight. A new evaluation index “TIindex” (tumor inhibition index) was used, in which the larger the TIindex, the better the overall treatment effect [40]. The TIindex of DOX-S was similar to that of the control group, which indicated that DOX-S produce serious toxicity in the body and did not exert therapeutic effects. However, each DOX-loaded liposome presented reduced toxicity relative to free drugs and had a good tumor suppressive effect to some extent. The SA modification significantly improved the therapeutic efficacy of DOX in liposomes prepared from HSPC/CH (P < 0.01), whereas the SA modification also significantly decreased the therapeutic efficacy of DOX in liposomes prepared from ESM/CH (P < 0.01). This further verifies that the SA modification disrupted the excellent retention characteristics of the drugs in the ESM/CH system, reducing the antitumor effect. Similarly, in the experiment of mouse survival analysis, the DOX-loaded liposome group significantly prolonged the survival time of mice, among which DOX-ECL had the best effect. The modification of SA plays a positive role in HSPC/CH system and a negative role in ESM/CH system.

The ultimate goal of cancer therapy is to improve patient survival time and quality of life by reducing the systemic toxicity of chemotherapy [41]. The percent change in body weight and the percent change in net body weight of each DOX-loaded liposome were higher than those of DOX-S, indicating that each DOX-loaded liposome played a protective role in tumor-bearing mice to a certain extent, and the liposome group prepared by ESM/CH had the optimal effect. Histopathology analysis further illustrated that DOX-ECL was the most formulation for killing tumor tissues without causing damage to other tissues.

Macrophages are the most abundant inflammatory cells that infiltrate into tumors, and their content can reach more than 50% of the mass of solid tumors. Macrophages also play the most predominant role in the inflammatory response of the TME (tumor microenvironment) [42, 43]. To further test the effect of each DOX formulation on the colonization of TAMs in tumor tissue, F4/80 (a mature mouse macrophage marker) was selected as an indicator for TAMs detection. Then, immunohistochemical labeling of tumor tissues from each group was performed. Among them, in the liposome group prepared by HSPC/CH, the SA modification significantly reduced TAMs colonization in the TME (P < 0.01). This may be attributed to the fact that SA binds to Siglec-1 on the surface of TAMs to release drugs to kill TAMs. However, in the liposome group prepared with ESM/CH, the SA modification did not reduce TAMs colonization. In addition, the colonization of TAMs in the DOX-ECL group was significantly less than that in the other DOX-loaded liposome groups because of the long circulation function of the vehicle and good drug retention capacity.

Conclusions

In this study, SA-CH was modified on the DOX-loaded liposomes prepared using ESM/CH and HSPC/CH systems. Among them, DOX-HSL showed a stronger inhibitory effect on tumors than DOX-HCL, and SA enhanced tumor targeting based on TAMs. However, the modification of SA did not enhance the inhibitory effect of DOX-ESL on tumors but destroyed the stability of the ESM/CH system. Notably, DOX-ECL provides optimal tumor suppression with minimal toxicity, efficient depletion of TAMs, and inhibition of tumor growth. In conclusion, our study showed that DOX-loaded liposomes prepared using ESM/CH showed excellent stability and antitumor activity and provide a promising strategy for the development of DOX.