Xyloglucan as a mitomycin C carrier to reverse multidrug resistance

Abstract

Hepatocellular carcinoma (HCC) is still considered as the third highest cause of cancer death in developing countries. In our previous study, we confirmed that the galactose residues of natural polysaccharide xyloglucan (XG) turns out to be selected uptake by asialoglycoprotein receptor (ASGPR) and the xyloglucan can be used as drug carriers and achieve some enhancement in the therapeutic effect of HCC via ASGPR mediated endocytosis. In this study, we selected mitomycin C (MMC) as the anticancer model drug and a macromolecular drug delivery system (DDS) with different degrees of substitution that was synthesized by conjugating mitomycin C with xyloglucan. The degree of substitution of xyloglucan–mitomycin C conjugates (XG–MMC) was about 9.7%. Another sample obtained a higher degree of substitution (p-MMC) of approximately 31%. The content of MMC in XG–MMC and p-MMC was determined and calculated to be about 3.1% and 6.4%, respectively. In the in vitro cytotoxicity experiment, the prodrugs presented higher cytotoxicity than free MMC against the drug resistant HepG2 cells. Moreover, the IC₅₀ of XG–MMC (0.997 μg mL⁻¹) was much lower than that of p-MMC (2.56 μg mL⁻¹) and XG–MMC exhibited higher drug uptake amounts (187.9 ng) than that of p-MMC (63.5 ng). The in vivo cytotoxicity was also evaluated, and the result was in good agreement with the in vitro cytotoxicity experiment.

---

1 Introduction

Unmet medical needs in cancer diagnosis and therapy remain substantial despite the emergence of some powerful anticancer drugs.¹ Basically, the conventional therapeutic agents for cancer are usually low-molecular weight compounds and exhibit non-specific bio-distribution profile, short plasma circulation time and rapid systemic elimination. Consequently, a relatively low amount of drug would reach the target sites, indicating that the therapy is usually associated with low efficacy and some undesired side effects.^2,3 Moreover, the most devastating challenge in chemotherapy is the emergence of multidrug resistance (MDR). Although the biological background of multidrug resistance (MDR) is complex and generally contains numerous mechanisms, the most extensively studied and major mechanism of MDR is the active transport of anticancer drugs out of tumor cells and the increased drug efflux is mediated predominantly by ATP-driven extrusion pumps.^4–8 Particularly, in the liver, the over-expressed plasma membrane P-glycoprotein (P-gp), which is a member of the ATP-binding cassette (ABC) superfamily of multidrug transporters, is capable of purging various generally positively charged xenobiotics, including some anticancer drugs, out of the tumor cells, thus decreasing the drug intracellular accumulation.^7–12

Fortunately, several macromolecular drug delivery systems (DDS) have successfully improved the therapeutic potential of respective conventional drugs, particularly for cancer treatment. These macromolecular drug delivery systems can be classified as nano-particulate drug-delivery systems or as drug–polymer conjugates.^3,13 In drug–polymer conjugates, the macromolecular carriers chosen for the preparation of polymer therapeutics should ideally be water-soluble, nontoxic, non-immunogenic, and biodegradable and/or able to be eliminated from the organism. Moreover, the macromolecular carrier should exhibit suitable functional groups for attaching the respective drug or spacer. In addition, the drugs can be conjugated directly or via a degradable or non-degradable linker onto the polymeric carrier, which should allow the release of the active drug from the conjugate at the desired target site.^2,3,13–15

Natural polysaccharides are abundant and inexpensive macromolecules with diverse properties, including sustainability, biocompatibility and biodegradability, as well as non-toxic, which make them suitable to be used widely in pharmaceutical applications and drug delivery systems as promising conjugates and ideal polymeric carriers for drug delivery, especially in chemotherapy.^16–18

Xyloglucan, a high molecular weight (720–880 kDa) galacto-xyloglucan, is regarded as one of the most potential polysaccharides for its excellent biocompatibility, wide pH range tolerance, and heat and shear stability, as well as its various biological activities.¹⁹ This biopolymer, which is extracted from tamarind seed xyloglucan, has a (1–4)-β-D-glucan backbone chain with (1–6)-α-D-xylose branches that are partially substituted by (1–2)-β-D-galacto-xylose. Xyloglucan polysaccharide is composed of three xyloglucan oligomer units: hepta-saccharide (Glu₄Xyl₃), octa-saccharide (Glu₄Xyl₃Gal) and nona-saccharide (Glu₄Xyl₃Gal₂), which differ in the number of galactose side chains^19–25 and the molar ratio of glucose : xylose : galactose ranging from 4 : 3.4 : 1.5 to 2.8 : 2.25 : 1.^16,17,20,26 To date, xyloglucan has been applied widely in pharmaceutical technology to administer certain drugs, especially as a carrier for variety of drugs for controlled release.²⁷

This paper discusses the effects of xyloglucan as the macromolecular carrier of drug delivery system with different degrees of substitution to suppress the drug resistant hepatoma cells (HepG2/MMC). In this contribution, mitomycin C (MMC), which is a potent anticancer agent,^28–35 was used as the model drug of our designed xyloglucan macromolecular drug delivery systems. Xyloglucan anchored with MMC by the spacer of tri-peptide Gly-Leu-Gly is the same as in our previous reported study.^36,37 Gly-Leu-Gly used as the spacer provided more stability to the DDS in the bloodstream, and it can be degraded by lysosome enzymes and released drug after polymeric conjugates being transferred into the endosome.³⁸ The cytotoxicity effect of the polymeric conjugates and the therapeutic effect on HepG2/MMC tumor cells implanted in mice were also investigated.

2 Materials and methods

2.1 Materials

Mitomycin C (MMC) was purchased from Chuangcheng Pharmaceutical (Wuhan) Ltd., China. N-t-Boc-glycyl-L-leucyl-glycine N-hydroxysuccinimide ester (Boc-Gly-Leu-Gly-OSu) was obtained from GL Biochem (Shanghai) Ltd., China. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and collagenase IV were purchased from Sigma-Aldrich Co, USA. All other chemicals used were of analytical grade unless stated otherwise. Organic solvents were purified and dried using common standard methods.

2.2 Isolation of xyloglucan

Xyloglucan was prepared in the following manner. In 250 mL distilled water, 5 g of tamarind kernel powder was added in batches with fierce stirring (THZ-82A, Guohua Instruments Co. China). This suspension was heated for 2 h with stirring in a boiling water bath, and then after cooled at room temperature, the mixture was centrifuged at 9000 rpm for 10 min to remove the water-insoluble protein fraction. The resulting thin clear solution was poured into twice the volume of anhydrous ethanol by continuous stirring to precipitate the polysaccharide. Suction filtered with a sintered glass funnel (G4, pore size, 5–15 μm), the product was lyophilized to obtain 3.5 g xyloglucan for further experiments. The molecular weight of xyloglucan was about 700 kDa according to GPC (Prominence GPC, Shimadzu Co. Japan).

2.3 Preparation of peptide-MMC derivatives

1.1 g of boc-Gly-Leu-Gly-OSu (2.5 mmol) and 0.87 g of MMC (2.6 mmol) were dissolved into 50 mL hydrous DMF. 0.55 g of diethylphosphoryl cyanide (DEPC) was added into the solution, and 0.4 mL triethylamine (TEA) was then added in 30 min at 0 °C with moderate agitation. After an overnight reaction in the dark at room temperature, the solvent was evaporated under vacuum to give the crude product (2.5 g). The residue was dissolved in 10 mL ethyl acetate and washed with a 10% citric acid solution (2 × 5 mL) and saturated sodium bicarbonate (2 × 5 mL). The organic layer was separated and the aqueous phase extracted with ethyl acetate (3 × 5 mL) and washed with saturated brine. The extracts were evaporated in vacuum, and the residue was purified by column chromatography on silica gel (CHCl₃/MeOH: 9/1). Dried over anhydrous sodium sulfate, the selected fraction was evaporated to dryness and the boc-Gly-Leu-Gly–MMC derivative was obtained as a blue solid.

0.3 g of boc-Gly-Leu-Gly–MMC was dissolved in 5 mL DMF; 0.5 mL of trifluoroacetic acid (TFA) was added into the solution and the reaction was conducted with stirring at room temperature for 1 h. After removing most of the solvent, the residue was diluted with 10 mL anhydrous methanol and filtered; furthermore, the solvent was evaporated under vacuum to dryness to afford the Gly-Leu-Gly–MMC conjugate (0.25 g) without purification.

2.4 Activation of xyloglucan

Activated xyloglucan with different degrees of substitution was prepared and the example of synthesis is described as follows.³⁹ Xyloglucan (3.1 g, 19 mmol unites) and 4-dimethylaminopyridine (DMAP) (0.23 g, 1.9 mmol, 0.1 eq.) were dissolved in 180 mL of dry DMSO/pyridine solution (vol. ratio 1/1). 4-Nitrophenyl chloroformate (1.15 g, 5.7 mmol, 0.3 eq.) was dissolved in 10 mL anhydrous DMSO and added dropwise with moderate agitation under an ice water bath. After 4 h reaction at 0 °C, the reaction mixture was precipitated in anhydrous ethanol (200 mL) and collected and washed several times with the same solvent. The white precipitate was dried in vacuo to give xyloglucan–COO(C₆H₄)NO₂ (XG–NPC) (3.0 g) (yield: 71.43%) (Scheme 1).

Scheme 1 Synthetic route of the XG–MMC conjugate (Boc-Gly-Leu-Gly-OSu: N-t-Boc-glycyl-L-leucyl-glycine N-hydroxysuccinimide ester).

2.4.1 Determination of the degree of substitution. The content of 4-nitrophenylcarbonate in activated xyloglucan was determined by UV analysis after being hydrolyzed in NaOH. 2.0 mg of XG–NPC was dissolved in 10 mL of 0.1 M sodium hydroxide solution; the absorption of sodium 4-nitrophenolate liberated was monitored by UV-visible spectroscopy at 402 nm (λ_max) (ε_M = 18 400 L mol⁻¹ cm⁻¹).

The percentage degree of substitution (DS) is expressed as the number of 4-nitrophenyl carbonate moieties per 100 anhydroglucoside units. The percentage degree of substitution in the sample described above (XG–NPC), calculated with Beer's law, was approximately 9.7%. Another sample (p-XG–NPC) was prepared using the same synthesis route and the percentage degree of substitution was calculated to be about 31%.

The percentage degree of substitution was calculated as follow:

2.5 Preparation of the xyloglucan–MMC conjugates

0.1 g xyloglucan–COO(C₆H₄)NO₂ (XG–NPC) and 0.1 g Gly-Leu-Gly–MMC were dissolved into 5 mL anhydrous DMSO, and a catalytic amount of TEA was added to the solution. After 48 h of reaction in the dark, the conjugate was precipitated by adding 5 mL anhydrous ethanol into the mixture. The product was collected and washed with the same solvent, and then purified by preparative GPC (Sephadex G25) with water as the eluent and lyophilized to obtain the xyloglucan–Gly-Leu-Gly–MMC conjugates (XG–MMC). The sample of p-MMC was synthesized by p-XG–NPC and Gly-Leu-Gly–MMC using the same procedure.

The degree of MMC substitution in the conjugates was determined by UV analysis in water (λ = 364 nm). The content of MMC in XG–MMC and p-MMC was determined and found to be about 3.1% and 6.4%, respectively.

2.6 In vitro studies

2.6.1 In vitro release of MMC from the conjugates. The release study was conducted in serum, phosphate buffered saline (PBS) at pH 7.0 and pH 5.0 (pH of endosomes or lysosomes), and incubation buffer with collagenase IV (0.3 mg mL⁻¹) at 37 °C with moderate stirring. The XG–MMC conjugate (10 mg) and p-MMC conjugate (10 mg) were dissolved in PBS (10 mL) respectively, then transferred into the dialysis tube (8000–14000). At selected time intervals, the medium was removed and then replaced with fresh PBS; the content of MMC was detected by UV at 360 nm. A calibration curve was made by detecting different concentrations of MMC solution (0.5, 1, 2, 4, 8, 16 mg mL⁻¹, respectively) at 360 nm.

2.6.2 In vitro cytotoxicity assay.
2.6.2.1 Cell culture. HepG2 cells (Yanyu Biotech Co., LTD, Shanghai) were cultured and maintained in an RPMI-1640 medium (GIBCO), supplemented with 10% fetal bovine serum (FCS, Sigma). The drug resistant HepG2 cell line (HepG2/MMC) was developed from HepG2 cells and incubated with MMC by increasing the concentration (from 0.01 to 2 μg mL⁻¹) stepwise during a few months. The drug resistance cells were obtained after removing the dead cells and preserved in 1 μg mL⁻¹ MMC.
2.6.2.2 Cytotoxicity assay in vitro. The cytotoxicity of these samples (XG–MMC, p-MMC, and free MMC) were evaluated by a 3-(4,5-diemethylthiazol-2-yl)-2,5-diphenyl-tetrazolium (MTT) assay (Sigma Co. USA) against human hepatoma cell line (HepG2). HepG2/MMC cells were treated with three formulations at a dose equivalent free MMC for 48 h at 37 °C.

The statistics of the cell number were determined by the MTT cell viability assay and the reversal of MDR was evaluated by the IC₅₀ value. Control groups were only treated with physiological saline solution. The cell viability was calculated as follows:

where A is the absorbance of the control group and B is the absorbance of treated group.

2.6.2.3 Cellular uptake of drug. HepG2/MMC cells were pre-incubated with the three samples (XG–MMC, p-MMC, and free MMC) for 2 h at a dose equivalent free MMC (100 μg mL⁻¹); 300 mL TM-2 buffer solution (10 mmol L⁻¹ Tris–HCl, pH 7.4, 2 mmol L⁻¹ MgCl₂, 0.5 mmol L⁻¹ PMSF) was added to a 1 mL cell suspension (10⁷ HepG2/MMC cells) in an ice bath and maintained for 5 min. After being combined with 300 μL 1.0% Triton X-100, the mixture was maintained in an ice-bath for 5 min and then filtrated through the membrane (0.22 μm) several times. Furthermore, 1 mL suspension of the cell (HepG2/MMC 10⁷ cells per mL) was mixed with MMC standard solution at different concentrations and the following steps were similar, as described previously.^36,40

These samples were analyzed by HPLC at different points of time with a Shimadzu HPLC system, equipped with a SPD-10Avp ultraviolet detector (Shimadzu Corporation, Japan) and two pumps (LC-10Avp and LC-10AS) in the reverse phase mode. Extend-C18 column (4.6 × 250 mm I.D., 5 μm) was the stationary phase and methanol–acetonitrile–phosphate buffer (50 : 30 : 20) for analysis was used as the mobile phase with a flow rate of 1 mL min⁻¹. The drug content was detected by UV at 360 nm.

2.7 In vivo studies

All study performed on animals was in accordance with the “Guidelines for the Care and Use of Laboratory Animals” published by the National Institute of Health (NIH Publication No. 85-23, revised 1985). This study was approved by the Ethics Committee of Central China Normal University and written informed consents of prior subject study were obtained.

2.7.1 Safety assessment of the conjugates in normal mice. Male BALB/c nude mice aged 4 weeks were injected with different doses of 8.0, 16.0 and 24.0 mg (MMC eq.) kg⁻¹ every week for four doses (days 0, 7, 14, and 21) (n = 10 mice per group). The toxicity and mortality data were recorded afterward for the next 2 weeks. At the end of the 5^th week, blood samples were collected from each mouse, and the serum was separated and examined using aspartate aminotransferase (AST) and alanine aminotransferase (ALT) by enzymatic reagent kits.

2.7.2 Drug bio-distribution. Every specific pathogen-free grade male BALB/c nude mouse (provided by the Experimental Animal Center of Wuhan Institute of Biological Products) was inoculated with HepG2/MMC cells (10⁷ cells per animal) in the right axillary region and solid tumor growth was established in most mice after three weeks. A total of 20 tumor-bearing mice were randomly divided into 2 groups, each group received XG–MMC conjugate or free MMC at a single dose (equivalent dose of MMC = 8 mg kg⁻¹) as a single dose on 10^th day from tumor inoculation.

At various time points after the injection (n = 10 at each time point), 0.5 mL blood sample was collected by retro-orbital venous plexus puncture from the tail vein of the tumor-bearing mice and plasma was prepared by centrifugation at 1200 rpm for 5 min. At the 6^th h after the injections, the mice were sacrificed by cervical dislocation. The livers, hearts, tumors, spleens and kidneys of the mice were immediately separated and washed with 10 mM Na₂HPO₄ buffer, followed by homogenization with triple volume of ethyl acetate as the solvent.

Mitomycin C was extracted after being incubated with acidic isopropanol (81 mM HCl in isopropanol) for 4 hours at 4 °C. The mixture was treated with a vortex mixer for 1 min and then centrifuged at 1200 rpm for 15 min. The concentration of MMC in the supernatant was determined by HPLC, the released MMC or free MMC were extracted and determined in the same way.³⁷

2.7.3 In vivo cytotoxicity of XG–MMC conjugates against drug resistant HepG2 cells in mice. Specific pathogen-free grade male BALB/c naked mice (provided by the Experimental Animal Center of Wuhan Institute of Biological Products, 4 weeks old, 20–35 g) were inoculated subcutaneously with HepG2/MMC cells (1 × 10⁷ cells per animal). After 3 weeks, solid tumor growth was noticeably established in most mice; XG–MMC conjugate, p-MMC conjugate or free MMC (equivalent dose of MMC = 8 mg kg⁻¹) suspended in PBS were injected to the tail veins of animals every week for four doses (days 0, 7, 14, and 21). The major and minor axes of the tumors were measured by calipers and the tumor volume was then determined. The survival time and the number of long-term survivors (LTS) until day 50 were also monitored.

2.8 Statistical analysis

The data are expressed as means ± standard deviations (SD) of multi-replicated determinations. The results were analyzed by one-way analysis of variance (ANOVA) with Student–Newman–Keuls multiple comparisons or t-test when comparing the differences with the means of two groups at the same time point. Differences were considered to be statistically significant if P < 0.05.

3 Results

3.1 Preparation and characterization of the XG–MMC conjugate

To introduce the MMC into the macromolecular drug carrier (XG), 4-nitrophenyl chloroformate was selected for the activation of the hydroxyl groups of xyloglucan and then, attachment of MMC to xyloglucan was accomplished using glycyl-L-leucyl-glycine (Gly-Leu-Gly) as the spacer. The conjugate was identified by FITR and ¹H-NMR and the spectra are shown in Fig. 1.

Fig. 1 FTIR spectrum (up) of mitomycin C (MMC), xyloglucan (XG), xyloglucan–MMC (XG–MMC) conjugate and ¹H-NMR (down) spectrum of xyloglucan (XG) and xyloglucan–MMC (XG–MMC) conjugate.

In the FITR spectrum, the strong absorptions between 2966 and 2860 cm⁻¹ might be assigned to the methyl and methylene groups of the Gly-Leu-Gly conjugate, which are different from the absorption peaks groups of xyloglucan due to the tripeptide spacer residue. The absorption bands at about 1730 and 1550 cm⁻¹ suggested that the MMC had been successfully conjugated with xyloglucan.

In the ¹H-NMR spectrum, the new peaks at 8.020 ppm and 7.953 ppm compared with xyloglucan were attributed to the protons in –CO–NH– of tripeptide and the peaks at 0.8 ppm were attributed to –CH₃ of tripeptide. The signals at 7.210 ppm might be due to the two protons in –CO–NH₂ of MMC and signals at 1.085, 1.340, 1.383 might be –CH₃ of MMC.

The particle size of XG–MMC conjugate and distribution were measured by dynamic light scattering (DLS; Zetasizer3000, Malvern Instruments LTD, UK). According to the measurements of the DLS (Fig. 2), the mean diameter of the macromolecular pro-drug is about 24.4 nm with a PDI of 0.384.

Fig. 2 Dynamic laser scattering (DLS) results of xyloglucan (XG) and xyloglucan–MMC (XG–MMC) conjugate.

3.2 Release of MMC from the XG–MMC conjugate

The in vitro release behaviors of MMC from XG–MMC conjugate were tested by incubating the conjugate with collagenase IV (or PBS) at 37 °C, and the results are shown in Fig. 3.

Fig. 3 Release profiles of xyloglucan–MMC (XG–MMC) conjugate [incubating with collagenase IV (◆), pH 5.0 buffer (▼), serum (▲), pH 7.0 buffer (■)]. Data were given as mean + SD (n = 6) (P < 0.05).

In both PBS at pH 7.0 and serum, the amount of MMC released was almost negligible, indicating that the conjugates are stable during plasma circulation. However, we also noticed that the released rate of MMC in native serum was slightly higher than in buffer pH 7.0. This may indicate that the presence of some types of enzymes in serum would affect the MMC release.

On treating with collagenase IV, the drug was released with a noticeable accumulation as time proceeded. The total release was approximately 60% over a period of 12 h and then the released rate of MMC slowed down gradually. Half-maximal release time (t_1/2) that freed 50% of loaded MMC from conjugates was 8 h in buffer with collagenase IV. In the buffer at pH 5.0, the release from the conjugate was much lower when compared with the release from the conjugate incubated with collagenase IV, and apparently the release did not exceed 35% after 48 h. On the other hand, because the amide bonds were hydrolyzed easier at pH 5.0 than at pH 7.0, the released rate of MMC, dissociated from XG–MMC conjugate, at pH 5.0 is much more rapid than at pH 7.0. Therefore, we believe that the release of MMC can be controlled by the amide group in the spacer under intracellular lysosomal condition.

3.3 In vitro cytotoxicity of XG–MMC conjugate against tumor cells

The cytotoxicity of the XG–MMC conjugate compared with that of free MMC and p-MMC conjugate was determined by the cell growth inhibition assay of the drug resistant HepG2/MMC cells. The results for the MTT assay are shown in Fig. 4. In comparison with free MMC, the MMC conjugate achieved better therapeutic effects. Furthermore, the XG–MMC conjugate showed lower cytotoxicity against the drug resistant HepG2/MMC cells at the experimental concentration. The calculated IC₅₀ was 0.997 μg·mL⁻¹, 2.56 μg mL⁻¹ and 187 μg mL⁻¹ for XG–MMC conjugate, p-MMC conjugate and free MMC, respectively, which indicated that xyloglucan loaded with MMC had better inhibition effects than that of free MMC.

Fig. 4 In vitro antitumor activity of free MMC (■), XG–MMC conjugate (▲), and p-MMC conjugate (●) against the drug resistant HepG2 cells. Data were given as mean + SD (n = 6) (P < 0.05).

3.4 Cellular uptake of drugs

Drug resistant HepG2/MMC cells were incubated in a free MMC solution, XG–MMC conjugate and p-MMC conjugate with equivalent doses of MMC for 2 h. As shown in Fig. 5, XG–MMC conjugate displayed higher uptake amounts of 187.9 ng after 2 h. In contrast, the free MMC and p-MMC internalized by tumor cells was relatively lower; the cellular uptake was 46.3 ng and 63.5 ng, respectively.

Fig. 5 Uptake of the drug by MDR cells after incubation with free MMC (■), XG–MMC conjugate (▲), and p-MMC conjugate (●). Values are means + SD (n = 3) (P < 0.05).

3.5 In vivo distribution studies of XG–MMC conjugate in tumor-bearing mice

In plasma, the concentration of free MMC was 1.62 mg L⁻¹ in 0.5 h, about 11-fold lower than XG–MMC conjugate, which was attributed to the rapid elimination from the circulation system by passive convection (Fig. 6). The formulating XG–MMC conjugate exhibited excellent characteristic of prolonged retention time in circulation before arriving at the tumor site. In Fig. 7, as it was expected, compared with free MMC, XG–MMC conjugate reduced the amount of MMC dramatically in the heart, lung, kidney, and spleen, and no drug was detected in the brain tissue of mice because of the blood–brain barrier. In contrast, the concentration of XG–MMC conjugate in the liver was much higher than that of the other tissues though the released MMC was only 33% of the free drug in the liver. Moreover, for XG–MMC conjugate, the drug uptake into tumor cell was 12-fold higher than that of free MMC, and the released MMC was 2.5-fold higher than the free MMC.

Fig. 6 Concentrations of MMC in plasma of free MMC (■), XG–MMC conjugate (▲), administered at [8 (MMC eq.) mg kg⁻¹] in tumor-bearing mice. Data are reported as mean + SD (n = 6) (P < 0.05).

Fig. 7 Drug bio-distribution in tumor-bearing mice (n = 10 per group, equivalent dose of MMC was 58 mg kg⁻¹). Data are reported as mean + SD (n = 6) (P < 0.05).

3.6 Safety assessment of XG–MMC conjugate

In comparison with free MMC, as described in Fig. 7, the amount of XG–MMC was much lower than free MMC in the remaining organism tissue (spleen, kidney, lung, heart); moreover, the XG–MMC accumulated in the tumor was higher than that in the liver. This reflected the safety of XG–MMC from different angles.

The administration of XG–MMC conjugate caused no mortality at all doses, and the median lethal dose (LD₅₀) of XG–MMC was up to 63.3 mg (MMC eq.) kg⁻¹. Compared to MMC (LD₅₀ of MMC: 13.4 mg kg⁻¹), the security of the XG–MMC conjugate was improved. We also monitored the mice body weights, following the treatments with the conjugate. After being treated with the conjugate, the mice did not show any observable side-effects and gained weight.

The hepatotoxicity of the conjugates was evaluated by the serum biochemical parameters and displayed in Table 1. The conjugate XG–MMC did not produce significant change in ALT or AST, even at doses of up to 24.0 mg (MMC eq.) kg⁻¹ for four times (Table 2).

Table 1 ¹H-NMR values of XG–MMC

Chemical shift	Assignment	Multiplicity
0.802–0.795	CH3 groups of tripeptide	Doublet
1.085, 1.340, 1.383	CH3 groups of MMC	Singlet
1.472–1.534	CH2 groups of tripeptide	Multiplet
3.603–4.113	CH groups of sugar moiety	Broad multiplet
4.739–4.842	Hydroxyl groups of sugar moiety	Broad multiplet
7.210	NH2 groups of MCC	Broad singlet
7.690, 7.953, 8.020	NH groups of tripeptide	Broad singlet

---

Table 2 Serum biochemical parameters and relative liver weight at 2 week after administration of different doses of the conjugates to mice^a

Dose (mg kg^−1)	ALT (U L^−1)	AST (U L^−1)
a ALT, alanine aminotransferase; AST, aspartate aminotransferase; (a) compared with the control group, p < 0.05; (b) compared with the MMC group, p < 0.05.
Control	53.56 ± 5.36	198.29 ± 17.35
MMC	42.32 ± 3.79	134.96 ± 15.29
8	52.78 ± 6.92(b)	193.14 ± 13.92(b)
16	50.93 ± 8.73(b)	184.97 ± 14.16(b)
24	47.31 ± 6.96(a, b)	174.49 ± 11.91(b)

---

3.7 In vivo anti-tumor activity of XG–MMC conjugate

The in vivo cytotoxicity effect of free MMC, p-MMC conjugate and XG–MMC conjugate was investigated to compare their ability to suppress the growth of drug resistant HepG2/MMC cells in BALB-C nude mice. Consequently, XG–MMC conjugate demonstrated improved therapeutic efficacy in suppressing tumor growth as compared to free MMC in animal models. The tumor volumes increased rapidly when mice were treated with PBS buffer alone or free MMC, and there was no significant difference between the two treatment groups in terms of inhibiting the tumor volume (Fig. 8).

Fig. 8 Tumor volume changes in vivo of the treated xenograft nude mice bearing the HepG2/MMC tumors. [PBS (◆) free MMC (■), XG–MMC conjugate (▲), and p-MMC conjugate (●)]. The tumor bearing mice were treated with equivalent drug (8 mg kg⁻¹ MMC) by tail injections every week for four doses (days 0, 7, 14, and 21).

At a similar dose, the XG–MMC conjugate was found to be more effective than free MMC in inhibiting the growth of the drug resistant HepG2 cells. The volume of tumors treated with the XG–MMC conjugate was about 55% less than those treated with free MMC after 21 days (Fig. 9). The results were also mirrored by the survival data. Treatment with the XG–MMC conjugate displayed longer life spans (42.9 days) of tumor-bearing mice than those treated with free MMC (20.7 days) (Fig. 9).

Fig. 9 Kaplan–Meier survival curve of the treated xenograft nude mice bearing the drug resistant HepG2 tumors. [Saline (◆), free MMC (■), XG–MMC conjugate (▲), and p-MMC conjugate (●)]. After injecting the HepG2/MMC cells for 3 weeks, these drug resistant HepG2 tumor-bearing mice were treated with drugs (8 mg MMC kg⁻¹) via a tail vein injection every week for four doses (days 0, 7, 14, and 21). The survival time and number of long-term survivors (LTS) until the 50^th day were monitored (p < 0.05).

4 Discussion

Hepatocellular carcinoma (HCC) is still considered as the third highest cause of cancer death in developing countries, and its limited optional treatment and exhibited multidrug resistance (MDR) may be responsible for the poor survival rate.^10,11,41 This study examined the effects of xyloglucan as the macromolecular carrier of drug delivery systems, with different degrees of substitution against the drug resistant hepatoma cells (HepG2/MMC). The therapeutic potential of MMC led us to choose it as the model drug to verify our hypothesis.^29–35

Xyloglucan as the macromolecular carrier of drug delivery systems to confront HCC has the following three main superiorities to relieve the urgent need for improving the treatment options.

First, xyloglucan is water-soluble, though the individual macromolecules tend not to fully hydrate, and consequently shows a balance between hydrophobic and hydrophilic character. The branches present on the backbone structure of xyloglucans may also contribute to the variation in viscosity as well as the solubility and conformational flexibility of the main chain; therefore, different branches may give different properties to the polysaccharide as a result of the presence of these hydrophilic and hydrophobic groups. Thus, the xyloglucan chain shows substantial stiffness. Among the drug delivery systems, conjugation to polymers is one of the main approaches for prolonging the circulation time of therapeutic agents, in particular due to the steric hindrance of the branches of the polymer, which dramatically decrease the drug uptake by the mononuclear phagocyte system (MPS), also known as reticuloendothelial system (RES).^2,16,18

Second, due to the high molecular weight of xyloglucan, polysaccharide based drug–polymer conjugates DDS would accumulate in hepatoma passively, known as the enhanced permeability and retention (EPR) effect. In contrast, low molecular weight substances should not be retained but returned to the circulating blood through a diffusion process. Enhanced permeability and retention (EPR) effect is a universal phenomenon in solid tumors (with the exception of some hypo-vascular tumors such as pancreatic cancer or prostate cancer) and the most widespread strategy used as a potential biological target for tumor-selective drug delivery.^5,8,42–44 A particle diameter ranging from 10 to 100 nm may be the optimal size for in vivo targeted delivery based on the EPR effects.⁸ According to the measurements of the DLS (Fig. 2), the mean diameter of xyloglucan is about 11.7 nm and the macromolecular pro-drug is about 24.4 nm, which is enough to accumulate in tumor tissue by the EPR effects.

Third, natural galactose rich polysaccharide xyloglucan, as the drug carrier, would construct a drug delivery system with enhanced potential and the ability to bypass multidrug resistance mechanisms to combat multidrug resistant HCC via receptor mediated endocytosis (RME). The asialoglycoprotein receptor (ASGPR) is a selectin E type receptor, not only present in high density (1–5 × 10⁵ receptors per cell) on the surface of hepatocytes (also hepatoma cells) but expressed minimally on extra-hepatic.⁴⁵ In addition, ASGPR turns out to be the selected uptake of moieties with terminal galactose residues of polysaccharide; it will render the drug delivery system hepato-specific if only the drug is conjugated with galactose rich polysaccharide (such as xyloglucan).⁴⁶ ASGPR mediated endocytosis would reverse multidrug resistance by enhancing the drug accumulation within the cell and also assure a minimum concentration at the off-target sites, thereby amalgamating high efficacy with low toxicity.^41,45,46

The XG–MMC conjugation was synthesized and its successful structure synthesis was confirmed by FTIR and ¹H-NMR (Fig. 1). We introduced a tri-peptide chain into the conjugate as the lysosomal enzyme degradable spacer. The tri-peptide spacer Gly-Leu-Gly has sufficient length and the sequence followed the rules of the oligopeptide design. The stability of the XG–MMC conjugate was relatively low in buffer at pH 5.0, a typical microenvironment in tumor cells. However, the conjugate released MMC by the specific hydrolysis of collagenase IV and hardly released MMC in pH 7.0 buffer or serum (Fig. 3). Therefore, the release of MMC from its carrier (xyloglucan) mainly dependent on the degradation rate of the tri-peptide spacers.

As mentioned before, ASGPR turns out to be selected uptake of moieties with terminal galactose residues and also overexpressed on the surface of HepG2 hepatoma cells. We assumed that XG–MMC can be identified by ASGPR and internalized through receptor-mediated endocytosis (RME). The in vitro cytotoxicity study proved that the XG–MMC demonstrated a lower IC₅₀ value (Fig. 4) and exhibited higher drug uptake rate (Fig. 5) compared to free MMC in cell.

To our surprise, the cell viability and cellular uptake of sample p-MMC were not better than XG–MMC. p-MMC (higher MMC loaded) exhibited relatively higher cell viability (IC₅₀) and much lower cellular uptake than XG–MMC. We considered that two reasons should be responsible for this phenomenon. First, higher drug loaded not only meant that much more moieties with terminal galactose of xyloglucan might be conjugated with drug but those moieties cannot be recognized by ASGPR. Second, even if the drug was not conjugated with terminal galactose, the length of the spacer of tri-peptide (Gly-Leu-Gly), which was used to connect the drugs with the carrier, might be long enough to make the galactose “hide” and hinder the galactose recognized by ASGPR.

Similar to the in vitro results, much more relevant trends were found after the in vivo testing. Drug bio-distribution studies were carried out to test the ability of XG–MMC to truly mediate drug targeting to HepG2 tumors and tumor-selectivity is also seen from the distribution to other organs in the tumor-bearing mice. The XG–MMC uptake into kidney was not sensitive to the nature of drug formulation, whereas the spleen and heart uptake was reduced significantly (Fig. 7). The relatively higher MMC uptake by the spleen for free MMC may explain the rapid clearance of MMC from the circulation.

XG-MMC exhibited the excellent characteristics of prolonging the retention time in circulation when compared with free MMC, probably because of the hydroxyl groups, which were widely presented in the polysaccharide chain. Large number of hydroxyl groups provide the conjugate a “hydrophilic coat” which was needed for the long term circulation (Fig. 6). Although the XG–MMC pro-drug showed an improved potential to deliver the drug to the targeted sites, the side effects of treatment with XG–MMC remained obvious. XG–MMC uptake into liver and HepG2 tumor cell was similar because of the ASGPR also expressed on the normal hepatocytes. Thus, the cytotoxicity of intracellular MMC on these cells might be enhanced by conjugates. In Fig. 7, although the XG–MMC uptake into liver was much higher than free MMC, the released MMC was much lower. This suggests that the safety of XG–MMC due to the sustained release of XG–MMC with relatively low concentration in various tissues, lower cardiotoxicity and hepatotoxicity was achieved when MMC was given in the conjugate form. Taking together the measures of tumor response to treatment, the tumor size and survival rate (Fig. 6 and 7) reflected that the group treated with free MMC showed little differences compared to the group treated with PBS. The XG–MMC suppressed tumor growth with concomitant smaller tumors and highly prolonged mice life span, (Fig. 8 and 9) compared to the treatment with the free drug.

5 Conclusion

The objective of the current study was to develop a macromolecular pro-drug delivery system (DDS) for overcoming drug resistant cancers. The results of this study highlighted the great potential for MMC conjugated with xyloglucan by a lysosome triggered spacer. XG–MMC conjugate can passively accumulate into the tumor tissues by the effect of EPR and then become internalized through receptor-mediated endocytosis without introducing new target-ligand, thereby decreasing the carrier-derived risks of toxicity and enhancing intracellular drug, reduced cardiotoxicity and hepatotoxicity of free MMC. Overall, xyloglucan as the macromolecular carrier of the drug delivery system, which was obtained simultaneously with distinct advantages of intracellular drug delivery, long circulation, and high affinity to the target, would be a valid option for MDR tumor chemotherapy.

Acknowledgements

This research was supported by the grants from the self-determined research of the Central China Normal University (Fundamental Research Funds for the Central Universities CCNU15A02062) and the National Natural Science Foundation of China (51603081).

References

1. A. D. Miller, J. Drug Delivery, 2013, 2013, 1–9.
2. E. Markovsky, H. Baabur-Cohen, A. Eldar-Boock, L. Omer, G. Tiram, S. Ferber, P. Ofek, D. Polyak, A. Scomparin and R. Satchi-Fainaro, J. Controlled Release, 2012, 161, 446–460.
3. R. Haag and F. Kratz, Angew. Chem., Int. Ed., 2006, 45, 1198–1215.
4. Q. Yin, J. Shen, Z. Zhang, H. Yu and Y. Li, Adv. Drug Delivery Rev., 2013, 65, 1699–1715.
5. C. Yi, X. Yang, N. Luo and C. Gong, Curr. Drug Metab., 2014, 15, 632–649.
6. M. M. Gottesman, Annu. Rev. Med., 2002, 53, 615–627.
7. Y. D. Livney and Y. G. Assaraf, Adv. Drug Delivery Rev., 2013, 65, 1716–1730.
8. Z. Gao, L. Zhang and Y. Sun, J. Controlled Release, 2012, 162, 45–55.
9. I. Brigger, C. Dubernet and P. Couvreur, Adv. Drug Delivery Rev., 2012, 64, 24–36.
10. S. W. Warmann and J. Fuchs, Curr. Pharm. Biotechnol., 2007, 8, 93–97.
11. M. Bebawy and M. Chetty, Curr. Drug Metab., 2009, 10, 322–328.
12. Y. A. Luqmani, Med. Princ. Pract., 2005, 14, 35–48.
13. F. Greco and M. J. Vicent, Adv. Drug Delivery Rev., 2009, 61, 1203–1213.
14. G. Pasut and F. M. Veronese, Prog. Polym. Sci., 2007, 32, 933–961.
15. R. Duncan, Curr. Opin. Biotechnol., 2011, 22, 492–501.
16. N. Dilbaghi, H. Kaur, M. Ahuja and S. Kumar, Carbohydr. Polym., 2013, 94, 286–291.
17. A. M. Avachat, K. N. Gujar and K. V. Wagh, Carbohydr. Polym., 2013, 91, 537–542.
18. T. A. Jó, D. F. S. Petri, L. M. Beltramini, N. Lucyszyn and M. R. Sierakowski, Carbohydr. Polym., 2010, 82, 355–362.
19. Y. Yan, M. Takemasa, C. Zhao, L. Yu and K. Nishinari, Food Hydrocolloids, 2016, 52, 447–459.
20. R. Freitas, S. Martin, G. Santos, F. Valenga, M. Buckeridge, F. Reicher and M. Sierakowski, Carbohydr. Polym., 2005, 60, 507–514.
21. S. Miyazaki, F. Suisha, N. Kawasaki, M. Shirakawa, K. Yamatoya and D. Attwood, J. Controlled Release, 1998, 56, 75–83.
22. C. Ribeiro, G. G. C. Arizaga, F. Wypych and M. R. Sierakowski, Int. J. Pharm., 2009, 367, 204–210.
23. S.-J. Seo, T. Akaike, Y.-J. Choi, M. Shirakawa, I.-K. Kang and C.-S. Cho, Biomaterials, 2005, 26, 3607–3615.
24. T. Coviello, P. Matricardi, C. Marianecci and F. Alhaique, J. Controlled Release, 2007, 119, 5–24.
25. N. Lucyszyn, A. F. Lubambo, K. F. Matos, I. Marvilla, C. F. Souza and M. R. Sierakowski, Mater. Sci. Eng., C, 2009, 29, 552–558.
26. M. J. Gidley, P. J. Lillford and D. W. Rowlands, Carbohydr. Res., 1991, 214, 299–314.
27. S. Gupta, S. Jain, G. K. Rao, V. Gupta and R. Puri, Syst. Rev. Pharm., 2010, 1, 50.
28. R. Cheung, A. Rauth, P. Ronaldson, R. Bendayan and X. Wu, Eur. J. Pharm. Biopharm., 2006, 62, 321–331.
29. Y. Patil, Y. Amitay, P. Ohana, H. Shmeeda and A. Gabizon, J. Controlled Release, 2016, 225, 87–95.
30. Y. Amitay, H. Shmeeda, Y. Patil, J. Gorin, D. Tzemach, L. Mak, P. Ohana and A. Gabizon, Pharm. Res., 2016, 33, 686–700.
31. Y. Li, J. Lin, X. Yang, Y. Li, S. Wu, Y. Huang, S. Ye, L. Xie, L. Dai and Z. Hou, ACS Appl. Mater. Interfaces, 2015, 7, 17573–17581.
32. T. Golan, T. Grenader, P. Ohana, Y. Amitay, H. Shmeeda, N. M. La-Beck, E. Tahover, R. Berger and A. A. Gabizon, Cancer Med., 2015, 4, 1472–1483.
33. Y. Li, H. Wu, X. Yang, M. Jia, Y. Li, Y. Huang, J. Lin, S. Wu and Z. Hou, Mol. Pharm., 2014, 11, 2915–2927.
34. Z. Hou, Y. Li, Y. Huang, C. Zhou, J. Lin, Y. Wang, F. Cui, S. Zhou, M. Jia, S. Ye and Q. Zhang, Mol. Pharm., 2013, 10, 90–101.
35. A. Volpe, M. Racioppi, D. D'Agostino, E. Cappa, A. Filianoti and P. F. Bassi, Minerva Urologica E Nefrologica, 2010, 62, 133–144.
36. S. Luo, Y. Gu, Y. Zhang, P. Guo, J. F. Mukerabigwi, M. Liu, S. Lei, Y. Cao, H. He and X. Huang, Mol. Pharm., 2015, 12, 2318–2327.
37. Y. Cao, D. Chen, P. Zhao, L. Liu, X. Huang, C. Qi, Y. Liu, H. He, Q. Wang, Y. Liu and S. Chen, Ann. Biomed. Eng., 2011, 39, 2456–2465.
38. A. K. Varkouhi, M. Scholte, G. Storm and H. J. Haisma, J. Controlled Release, 2011, 151, 220–228.
39. F. Vandoorne, R. Vercauteren, D. Permentier and E. Schacht, Makromol. Chem., 1985, 186, 2455–2460.
40. Y. Zhang, H. Wang, J. F. Mukerabigwi, M. Liu, S. Luo, S. Lei, Y. Cao, X. Huang and H. He, RSC Adv., 2015, 5, 71164–71173.
41. A. A. D'Souza and P. V. Devarajan, J. Controlled Release, 2015, 203, 126–139.
42. J. W. H. Maeda, T. Sawa, Y. Matsumura and K. Hori, J. Controlled Release, 2000, 65, 271–284.
43. J. W. Nichols and Y. H. Bae, J. Controlled Release, 2014, 190, 451–464.
44. H. Maeda, H. Nakamura and J. Fang, Adv. Drug Delivery Rev., 2013, 65, 71–79.
45. P. O. Pathak, M. S. Nagarsenker, C. R. Barhate, S. G. Padhye, V. V. Dhawan, D. Bhattacharyya, C. L. Viswanathan, F. Steiniger and A. Fahr, Carbohydr. Res., 2015, 408, 33–43.
46. X. Xie, S. Luo, J. F. Mukerabigwi, J. Mei, Y. Zhang, S. Wang, W. Xiao, X. Huang and Y. Cao, RSC Adv., 2016, 6, 26137–26146.

---

Footnote

† Equal contributors to the work.

---