Tumor microenvironment-labile polymer–doxorubicin conjugate thermogel combined with docetaxel for in situ synergistic chemotherapy of hepatoma

Accepted Manuscript Full length article Tumor microenvironment-labile polymer−doxorubicin conjugate thermogel combined with docetaxel for in situ synergistic chemotherapy of hepatoma Yanbo Zhang, Jin Zhang, Weiguo Xu, Gao Xiao, Jianxun Ding, Xuesi Chen PII: DOI: Reference:

S1742-7061(18)30415-X https://doi.org/10.1016/j.actbio.2018.07.021 ACTBIO 5570

To appear in:

Acta Biomaterialia

Received Date: Revised Date: Accepted Date:

25 April 2018 1 July 2018 9 July 2018

Please cite this article as: Zhang, Y., Zhang, J., Xu, W., Xiao, G., Ding, J., Chen, X., Tumor microenvironmentlabile polymer−doxorubicin conjugate thermogel combined with docetaxel for in situ synergistic chemotherapy of hepatoma, Acta Biomaterialia (2018), doi: https://doi.org/10.1016/j.actbio.2018.07.021

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Tumor

microenvironment-labile

polymer−doxorubicin

conjugate

thermogel combined with docetaxel for in situ synergistic chemotherapy of hepatoma Yanbo Zhang

a,b,1

, Jin Zhang

c,1

, Weiguo Xu

a,1

, Gao Xiao d, Jianxun Ding a,*, Xuesi

Chen a

a

Key Laboratory of Polymer Ecomaterials, Changchun Institute of Applied Chemistry,

Chinese Academy of Sciences, Changchun 130022, P. R. China b

Department of Orthopedics, China-Japan Union Hospital of Jilin University,

Changchun 130033, P. R. China c

College of Chemical Engineering, Fuzhou University, Fuzhou 350108, P. R. China

d

Wyss Institute for Biologically Inspired Engineering, Harvard University, Cambridge,

MA 02115, USA

* Corresponding author. E-mail address: [email protected] (J. Ding). 1

Y. Zhang, J. Zhang, and W. Xu contributed equally to this work.

1

ABSTRACT Topical chemotherapy with complementary drugs is one of the most promising strategies to achieve an effective antitumor activity. Herein, a synergistic strategy for hepatoma therapy by the combination of tumor microenvironment-sensitive polymer−doxorubicin (DOX) conjugate thermogel, containing a DNA intercalator DOX, and docetaxel (DTX), a microtubule-interfering agent, was proposed. First, cis-aconitic anhydride-functionalized DOX (CAD) and succinic anhydride-modified DOX

(SAD)

were

conjugated

onto

the

terminal

hydroxyl

groups

of

poly(lactide-co-glycolide)-block-poly(ethylene glycol)-block-poly(lactide-co-glycolide) (PLGA−PEG−PLGA), yielding the acid-sensitive CAD-PLGA−PEG−PLGA-CAD and the insensitive SAD-PLGA−PEG−PLGA-SAD conjugates, respectively. The prodrug aqueous solution exhibited a thermoreversible sol–gel transition between room and physiological

temperature.

Meantime,

appropriate

mechanical

property,

biodegradability, as well as a sustained release profile were revealed in such prodrug thermogels. More importantly, the addition of DTX to the DOX-conjugated thermogels (i.e., Gel-CAD and Gel-SAD) was verified with enhanced curative effect against tumor, where the antitumor efficacy of Gel-CAD+DTX was obviously higher than the other groups. A reliable security in vivo was also showed in the Gel-CAD+DTX group. Taken together, such combination of tumor microenvironment-labile prodrug thermogel and a complementary drug exhibited fascinating prospect for local synergistic antineoplastic therapy. Keywords: Tumor microenvironment; Prodrug; Acid-sensitivity; Controlled drug

2

release; Local synergistic hepatoma chemotherapy

Statement of Significance Multidrug chemotherapy with synergistic effect has been proposed recently for hepatoma treatment in the clinic. However, the quick release, fast elimination, and unselectivity of multidrugs in vivo always limit their further application. To solve this problem, a synergistic combination of tumor microenvironment-sensitive polymeric doxorubicin

(DOX)

prodrug

thermogel

for

DNA

intercalation

and

a

microtubule-interfering agent docetaxel (DTX) is developed in the present study for the local chemotherapy of hepatoma. Interestingly, a pH-triggered sustained release behavior, an enhanced antitumor efficacy, and a favorable security in vivo is observed in the combined dual-drug delivery platform. Therefore, effectively combining tumor microenvironment-labile polymeric prodrug thermogel with a complementary drug provides an advanced system and a promising prospect for local synergistic hepatoma chemotherapy.

3

1. Introduction Hepatoma is one kind of major malignant tumors that seriously threatens human health with a high incidence and mortality. According to the statistics, the incidence of liver cancer in China is more than 0.02% [1], and the corresponding death toll caused by hepatoma accounts for almost half of the world [2]. Surgical intervention and chemotherapy are the most commonly-used methods for hepatoma therapy. However, due to the early asymptomatic disease and rapid proliferation of tumor cells, the majority of patients are diagnosed as the advanced hepatoma. At this stage, the surgical intervention is seriously limited by the high cost and shortage of liver donors. Systemic or local chemotherapy plays an increasingly significant role in treating hepatoma. In detail, systemic chemotherapy is carried out by oral or intravenous administration, which

makes

no

significant difference in

concentration

of

chemotherapeutic drugs between tumor and other tissues. Therefore, systemic chemotherapy always brings toxic damages to all organs, and thus results in gastrointestinal symptoms or bone marrow suppressions. By contrast, local chemotherapy can improve the chemotherapeutic efficacy and reduce the drug toxicity through upregulating the concentration of local drugs. To achieve a desired drug release profile and meantime reduce the negative effects, a local drug delivery system dependent on polymer–drug conjugates has attracted great attention nowadays, which is superior to the simple drug encapsulation due to the effect of chemical

bonding

[3,

4].

In

particular,

in

situ

forming

poly(lactide-co-glycolide)-block-poly(ethylene glycol)-block-poly(lactide-co-glycolide)

4

(PLGA−PEG−PLGA) triblock copolymer thermogel-based drug delivery system has been demonstrated with unique injectable property, sustained release behavior, as well as high concentration of drugs at the tumor site [5, 6]. Stimuli-responsive drug delivery platform on the basis of photo [7], enzyme [8], temperature [9], and pH [10] has been widely developed so far to enhance the selectivity. For instance, highly-stable supramolecular vesicles composed by photo-responsive supra-amphiphiles were able to significantly alter the therapeutic effect against tumor cells upon UV light irradiation [11]. Temperature-sensitive liposomes have been reported with a rapid drug release at hyperthermic temperature, and the absorption amount of drugs within the tissue was directly relevant with heating duration [12]. Among them, pH-responsiveness is one of the vital signs as there exists pH difference between physiological state (~7.4) and acidic extracellular microenvironment (~6.8) [13, 14] or intracellular microenvironment of tumors (~5.5) [15, 16]. In our previous study, a pH-responsive reversible PEGylation was proposed to endow antineoplastic agent with enhanced intracellular drug release and upregulated security in vivo [17]. Naturally, a controlled drug delivery system, including dual advantages of PLGA−PEG−PLGA thermogel and pH sensitivity, is of great expectation for improving the antitumor efficacy and reducing the in vivo toxicity simultaneously [18]. Because of the heterogeneity and drug resistance of tumors, it is usually difficult to achieve good therapeutic effect by using one kind of drug alone. According to the synergistic mechanism, multidrug chemotherapy was put up recently for the

5

treatment of malignant tumors in the clinical application [19], which had great significance in minimizing drug dose, maximizing therapeutic efficacy, and reducing side effects. Compared to single or continuous administration, the use of complementary drugs made malignant tumor cells less likely to develop the compensatory resistance [20]. In the current study, a polymeric doxorubicin (DOX) prodrug thermogel encapsulated with docetaxel (DTX) was fabricated with the aim of improving antitumor efficacy. In detail, cis-aconitic anhydride (CA) functionalized DOX (CAD) and succinic anhydride (SA) modified DOX (SAD) were conjugated onto the terminal hydroxyl groups of PLGA−PEG−PLGA triblock copolymer, yielding the insensitive the acid-sensitive CAD-PLGA−PEG−PLGA-CAD (Gel-CAD) and the insensitive SAD-PLGA−PEG−PLGA-SAD (Gel-SAD) conjugates, respectively [17]. Compared with the previous studies in the field, novelty of the current work could be summarized

into

the

following

two

aspects:

1)

the

proposed

tumor

microenvironment-sensitive polymeric prodrug thermogel exhibited an efficient pH-triggered drug release behavior; 2) a synergistic combination of DOX for DNA intercalation and a microtubule-interfering agent DTX has been well-developed for the local multidrug chemotherapy of hepatoma. The complete process for synthesis, sol–gel phase transition, intratumoral injection, long-term degradation, and pH-triggered sustained drug release of Gel-CAD+DTX was shown in Scheme 1. First, chemical structural characterizations of above-mentioned two prodrug thermogels were identified by gel permeation chromatography (GPC) and proton nuclear magnetic resonance spectroscopy (1H NMR). Then, sol−gel phase transition,

6

rheology property, aggregation, and biodegradability of the prodrug thermogels were studied, which revealed an ideal gelation temperature, sufficient mechanical strength, and long-term biodegradation behavior. Afterwards, the prodrug thermogels encapsulated with DTX were localized administration into hepatoma-bearing mice subcutaneously, and the Gel-CAD+DTX group revealed the highest antitumor efficacy in all groups. More importantly, a favorable in vivo security based on tissue fluorescence and histopathological evaluations was presented in the Gel-CAD+DTX group. On the whole, a combination of acid-sensitive polymeric DOX prodrug thermogel and DTX as a controlled drug delivery system showed a promising potential for local synergistic chemotherapy of hepatoma.

Scheme 1. Schematic illustration for synthesis, sol–gel transition, intratumoral injection, long-term degradation, and pH-triggered sustained drug release of Gel-CAD+DTX.

7

2. Materials and methods 2.1. Materials SA and CA were obtained from Tokyo Chemical Industry (Tokyo, Japan) and Alfa Aesar (Shanghai, P. R. China), respectively. Doxorubicin hydrochloride (DOX·HCl) was obtained from Zhejiang Hisun Pharmaceutical Co., Ltd (Zhejiang, P. R. China). Elastase

was

supplied

from

4-N,N-dimethylaminopyridine

Merck

(DMAP)

Company and

(Darmstadt,

Germany).

1-ethyl-(3-(dimethylamino)propyl)

carbodiimide hydrochloride (EDC·HCl) were purchased from GL Biochem Co., Ltd. (Shanghai, P. R. China). Other reagents were supplied from Sigma-Aldrich (Shanghai, P. R. China). 2.2. Synthesis of PLGA−PEG−PLGA The triblock PLGA−PEG−PLGA copolymer (number-average molecular weight (Mn) = 6,900 g mol−1, LA/GA = 75:25, mol/mol) was prepared via the procedures according

to

our

preceding

publications

[21,

22].

In

brief,

ring-opening

copolymerization of LA and GA was performed to synthesize PLGA−PEG−PLGA with an initiator of PEG (Mn = 1,500 g mol−1) and a catalyst of stannous octoate (Sn(Oct)2). 2.3. Syntheses of CAD and SAD With triethylamine (TEA) as a catalyst, CAD or SAD was synthesized via the sequential ring opening and condensation reactions between DOX and CA or SA, respectively [23]. In detail, with regard to the synthesis of CAD, the dissolved CA (68.6 mg, 0.44 mmol) and DOX·HCl (232.0 mg, 0.40 mmol) in anhydrous dimethylformamide (DMF) were put in a dried flask, afterwards, 67 µL of TEA was

8

added and stirred overnight under a pure nitrogen condition at 25 °C protected from light [24]. Afterwards, cold acetic ether (100.0 mL) was mixed with the solution, and flushed with saturated sodium chloride liquor of cold acid under pH 2−3 and ultimately washed with the saturated liquor in normal pH. By using anhydrous sodium sulfate, the organic layer was gathered for drying overnight. To acquire the red powder, the filtrate was dried in a vacuum environment at 25 °C after desiccant filtration. SAD was synthesized using the similar approaches as CAD. 2.4. Preparation of Gel-CAD+DTX and Gel-SAD+DTX The acid-sensitive Gel-CAD or insensitive Gel-SAD was acquired via the condensation between CAD or SAD and PLGA−PEG−PLGA, respectively, where EDC·HCl was served as a condensing reagent and DMAP was employed as a catalyst. Specifically, PLGA−PEG−PLGA (75.0 mg, 0.1 mmol) was dissolved in dimethyl sulfoxide (DMSO), afterwards, CAD (81.8 mg, 0.24 mmol), DMAP (1.2 mg, 0.01 mmol), and EDC·HCl (57.5 mg, 0.3 mmol) were added into the solution. Under the room temperature, the reaction solution was blended for 72 h away from light. Afterwards, Gel-CAD was prepared via a dialysis procedure with molecular weight cutoff values of 3500 Da for 48 h. Using the similar procedures as Gel-CAD, Gel-SAD was synthesized. The aqueous solution of Gel-CAD or Gel-SAD (75.0 mg, 20 wt %) was then mixed with DTX (25.0 mg) at 4 °C for 12 h, and the prodrug thermogel was finally obtained at 37 °C that can be used for the following experiments of in vitro drug release, in vivo tissue distribution, and tumor suppression. 2.5. Structural and physical characterizations

9

1

H NMR spectra were used to determine the composition of prodrug thermogels

(400 MHz NMR spectrometer, Billerica, MA, USA). Following the potassium bromide procedures, Fourier transform infrared (FTIR) spectra were obtained through a FTIR spectroscopy (Cambridge, MA, USA). GPC consisted of a pump (515 HPLC) with an interferometric refractometer detector (OPTILAB DSP, Wyatt Technology) and linear Styragel columns (HT3 and HT4) was used to acquire the molecular weight distribution. Dynamic light scattering (DLS) was performed on an He–Ne laser spectrometer (Santa Barbara, CA, USA) in a range of 10 to 50 °C. 2.6. Biodegradation Tris hydrochloric acid (HCl) buffer solution (0.05 M, pH 7.4) consisted of elastase (0.2 g L−1) or pure phosphate-buffered saline (PBS), calcium chloride (CaCl2, 10.0 mM), and sodium azide (NaN3, 0.2 wt.%) was employed as the degradation medium. The solutions (3.0 mL) were added onto top of the prodrug thermogels. Every two days, the medium was replaced and the remaining gel was accurately weighed for measuring the biodegradation rate. 2.7. In vitro DOX release In vitro drug release behaviors of prodrug thermogels were studied at different pH values of 7.4 (a simulant biological condition), 6.8 (an acidic extracellular microenvironment of tumors), and 5.5 (an intracellular microenvironment of tumors). At 37 °C, the prodrug solutions (20 wt.%) in glass vials (diameter = 16 mm) were put into a water bath for 30 min. Then, PBS (3.0 mL) was poured onto top of prodrug thermogels with shaking at 70 rpm in different pH values. At predetermined time

10

intervals, the top buffer was removed and the same PBS volume was refilled. The release amount of DOX was quantitively assessed via an ultraviolet–visible spectrophotometer (Lawrenceville, NJ, USA) and the detection wavelength was determined as 480 nm. 2.8. Tissue distribution Kunming mice were maintained in line with the protocol approved by the Use Committee and School of Life Sciences Animal Care of Jilin University. For the tissue distribution assay, Kunming mice were subcutaneously inoculating with 3.0 × 106 hepatocellular carcinoma cells (H22) to obtain the tumor-bearing models. Mice were randomly divided into 9 groups (six mice in each group) and weighed as the tumor volume reached to ~200 mm3. Then, different formulations of normal saline (NS) as a control, DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX (5.0 mg kg−1, DOX) were administered intratumorally. At 1 week or 2 weeks post-injection, major organs including liver, heart, lung, kidney, and spleen and tumors were obtained. The DOX fluorescence images were recorded by the Maestro in vivo imaging system (Cambridge Research & Instrumentation Inc., Woburn, MA). 2.9. In vivo assessment of antitumor efficacy The tumor model was established by inoculating 3.0 × 106 H22 cells subcutaneously. As the tumor volume was approximately 200 mm3, the mice were treated with different formulations of NS, DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX (DOX concentration, 5.0 mg kg−1) by intratumoral injection on 0, 6, 12, 18, and 24 days. Antitumor efficacy was evaluated

11

via measuring the tumor volume every other day and then was calculated according to the following Equation (1):

V (mm) =

L × S2 2

(1)

Where, L and S (mm) represent the longest and the shortest diameters of tumors, respectively. 2.10. Histopathological analyses At 4 days after the end of treatment, all mice were sacrificed. Major organs and tumors were gathered and fixed in paraformaldehyde (4%). The histopathological analyses were carried out by staining the sections of tumors and organs with hematoxylin and eosin (H&E). A microscope (Nikon Eclipse Ti, Ardmore, PA) was used to detect the histological alterations. The necrotic area was defined as the dissolution or even the disappearance of nuclei and organelles in the cytoplasma. The relative necrotic area (%) was acquired on the basis of Equation (2): Relative necrotic area (%) =

Necrotic area in tumor section × 100% Total area in tumor section

(2)

2.11. Statistical analyses All data were shown as mean ± standard deviation (SD). According to Student’s t-tests, the differences between groups were carefully analyzed. *P < 0.05 was regarded as statistically significant, **P < 0.01 and ***P < 0.001 were thought as highly statistically significant.

12

Fig. 1. Chemical structure characterizations. (a) 1H NMR and (b) FTIR spectra,

and (c) GPC chromatograms of PLGA−PEG−PLGA, CAD-PLGA−PEG−PLGA-CAD, and SAD-PLGA−PEG−PLGA-SAD. 3. Results and discussion

3.1. Characterizations of polymeric DOX prodrugs The composition and chemical structure of polymeric DOX prodrugs were evaluated by 1H NMR spectra, FTIR spectra, and GPC chromatograms. A successful synthesis of triblock copolymer was proved by the characteristic peaks that appeared at 3.6 (a), 5.2 (b), 4.8 (c), and 1.5 ppm (d) in the 1H NMR spectra [25] (Fig. 1a), which referred to the methylene proton of PEG, the methine proton of D,L-lactide unit, the

13

methylene proton of glycolide units, and the methyl proton of D,L-lactide units, respectively [26]. The peak appeared at 6.97 ppm (g) was assigned to the methylidyne proton in CAD, meanwhile the peaks at 2.33 (e) and 2.68 ppm (f) belonged to the methylene protons in SAD. The drug-binding efficiencies (DBEs) of CAD and SAD were calculated to be 75.6% and 83.9% based on the 1H NMR spectra, respectively. In FTIR spectra, signal at 1660 cm−1 marked by a dotted line belonged to the stretching vibration of amide bonds in both CAD-PLGA−PEG−PLGA-CAD and SAD-PLGA−PEG−PLGA-SAD (Fig. 1b) [17]. As shown in Fig. 1c, the same elution time of both CAD-PLGA−PEG−PLGA-CAD and SAD-PLGA−PEG−PLGA-SAD (15.2 min) in the single-peaked GPC curves was shorter than that of PLGA−PEG−PLGA (15.6 min), indicating the successful syntheses of DOX-conjugated prodrugs with large molecular weight and sufficient purity quotient [27, 28]. To assess the aggregation behavior of the polymeric DOX prodrug thermogels, hydrodynamic radii (Rhs) of the Gel-CAD and Gel-SAD prodrug thermogels and their distributions in water (0.5 wt.%) at different temperatures were shown in Figs. 2a and 2b, respectively. With an increase of temperature from 10 to 20 °C, Rhs of the prodrug thermogels increased slowly. While, the Rhs exhibited a drastic increase and a wide distribution as the temperature rose up to 30 °C. Along with the gradually elevated temperature, Rhs of the prodrug thermogels kept on increasing until the appearance of an aggregation at 40 °C, which was caused by the sol−gel transition of prodrug aqueous solution [29]. With the further increase of temperature until 50 °C, a higher value and a wider distribution of Rhs were revealed as a result of hydrophobic

14

aggregation [30].

Fig. 2. Gelation and rheological properties. (a, b) Rhs and distributions of Gel-CAD

(a) and Gel-SAD (b) in water (0.5 wt.%) at different temperatures. (c, d) G′ and G′′ of Gel-CAD (c) and Gel-SAD (d) in PBS (20 wt.%). Rheological tests were conducted to evaluate the changes in storage modulus (G′) and loss modulus (G′′) of Gel-CAD (Fig. 2c) and Gel-SAD (Fig. 2d) in PBS solution (20 wt.%) with the increasing temperature. Admittedly, gelling temperature is defined as the point when G′ is higher than G′′ [31]. It can be seen from Figs. 2c−d that the sol−gel transition temperatures of the two prodrug solutions were approximately 30 °C, where the value ranged from room temperature (~25 °C) to body temperature (~37 °C) was conducive to in situ injection [32, 33].

15

Fig. 3. Mass remaining and pH-responsive DOX release in vitro. (a) Degradation

properties of Gel, Gel-CAD, and Gel-SAD (20 wt.%) at 37 °C soaked in 3.0 mL of Tris-HCl buffer containing 0.2 g L−1 elastase with PBS as a control. (b−d) Cumulative release curves of DOX from Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX (20 wt.%) at different pH values of 7.4 (b), 6.8 (c), and 5.5 (d). All statistical data were presented as mean ± SD (n = 3; *P < 0.05, ***P < 0.001). In vitro degradation tests were carried out to evaluate the degradation properties of the thermogels. As shown in Fig. 3a, the degradation rate of the elastase group (> 60% at 45 days) was faster than that of the control group (< 40% at 45 days) due to the effect of enzyme catalysis [34]. While no matter in which kind of situation, the degradation rate of Gel-CAD or Gel-SAD was a little bit faster than that of the pure Gel. It was likely because that the introduction of CA or SA with amide bonds 16

promoted the hydrolysis [35]. Besides, the degradation rate of Gel-CAD was faster than that of Gel-SAD in the elastase group (P < 0.05), which could be ascribed to the synergistically-accelerated breaking of the amide bonds in the Gel-CAD group with the existence of elastase [36]. 3.2. In vitro DOX release The drug-binding contents (DBCs) of CAD-PLGA−PEG−PLGA-CAD and SAD-PLGA−PEG−PLGA-SAD were calculated to be 14.3% and 15.6% based on DBEs, respectively. As the weight ratio of DTX to the prodrug thermogel (Gel-CAD or Gel-SAD) was predetermined as 1:3, the drug-loading content (DLC) and the drug loading efficiency (DLE) of DTX were calculated to be around 25 wt.% and 100 wt.%, respectively. The in vitro release performances of DOX from Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX were investigated in PBS (pH 7.4, 6.8, and 5.5). The cumulative release profiles of DOX under different situations are shown in Figs. 3b−d. The DOX was delivered from the hydrogel in a sustained way without any initial burst release. At pH 7.4 (Fig. 3b), there was negligible difference in the release behavior of DOX between the Gel-CAD and Gel-SAD groups. While as the pH value was 6.8 (Fig. 3c), an enhanced DOX release from the Gel-CAD and Gel-CAD+DTX groups was observed compared to the Gel-SAD and Gel-SAD+DTX conditions. Quantitatively, the cumulative release amount of DOX from the Gel-CAD group and Gel-CAD+DTX groups was about 50% at day 30, which was significantly higher (P < 0.001) than that of the Gel-SAD and Gel-SAD+DTX groups (only around 37%). As the

17

further decrease of pH to 5.5, mimicking the intracellular microenvironment (Fig. 3d), Gel-CAD and Gel-CAD+DTX showed a more obvious increase in DOX release (60%) on day 30 than the Gel-SAD and Gel-SAD+DTX groups (37%, P < 0.001). The result should be reasonably explained by the quick break of the acid-sensitive amide bonds [37] and the accelerated degradation of the prodrug thermogel [38]. The release of DOX from the DOX-conjugated thermogels is primarily dependent on the hydrogel degradation as well as the cleavage of covalent bonds between DOX-conjugated thermogels and DOX [39, 40]. Specifically, in an acidic environment, the pH-sensitive cis-aconitic linkage of CAD would be cleaved more quickly [41]. On the basis of the above results, the DOX release amount of Gel-CAD and Gel-CAD+DTX was relatively low at neutral pH but increased markedly at acid pH. Additionally, as shown in Fig. S1, the cumulative release of DTX from Gel+DTX showed a sustained release without initial burst release. Such stimuli-responsive performance in the Gel-CAD and Gel-CAD+DTX groups could minimize the drug loss within the circulatory system and meanwhile decrease the harmful side effects in vivo, revealing vast potential for cancer chemotherapy. 3.3. In vivo tissue distribution The tissue distributions of NS, DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX in vivo were examined by the fluorescence imaging of tumors and major organs, and the Maestro 2.4 software was used to do the semi-quantitative analyses. As shown in Fig. 4a, the Gel-CAD and Gel-CAD+DTX groups exhibited lower expressions in normal organs compared with

18

the DOX and DOX+DTX groups, indicating the lessened side effect and minor toxicity of such prodrugs [42]. The fluorescent intensities in the DOX and DOX+DTX groups were noticeably attenuated one week later due to the rapid excretion from the body [43]. At the meantime, a short circulating half-life and a quick elimination of DOX were also revealed [44]. Furthermore, the fluorescent intensity of tumors in the Gel+DOX group was stronger than that in the Gel-CAD and Gel-CAD+DTX groups at 1 week post-surgery, but a contrary result was exhibited at 2 weeks post-injection. The fluorescent intensity of the Gel+DOX group started to weaken significantly on account of blood circulation and metabolism. Compared with DOX encapsulated within the gel that was directly released to tumor tissue, a sustained release behavior caused by the cleavage of acid-response amide bonds was shown in the Gel-CAD and Gel-CAD+DTX groups [45]. Additionally, the Gel-CAD and Gel-CAD+DTX groups exhibited stronger fluorescent intensities of DOX accumulation than the Gel-SAD and Gel-SAD+DTX groups. The fluorescence quenching of DOX in polymer-grafted DOX conjugates has already been confirmed. Owing to such self-quenching effect of fluorescent molecules, the free DOX showed stronger fluorescence compared with the DOX in prodrug thermogels at the same concentration [46]. Therefore, the higher fluorescent expression in the Gel-CAD and Gel-CAD+DTX groups could be attributed to the pH-triggered DOX release within the tumor microenvironment [47]. Overall, the Gel-CAD+DTX group showed the highest fluorescent intensity of DOX at tumor site among all the groups due to the sustained release as well as the pH sensitivity of antineoplastic drugs [45, 47].

19

Fig. 4. Biodistribution in vivo. (a) DOX fluorescence images and (b)

semi-quantitatively analyzed average fluorescence intensities of tumors isolated at 1 or 2 weeks post-injection of NS, DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, 20

Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX in H22 hepatoma-bearing mouse models. All statistical data were presented as mean ± SD (n = 3; ***P < 0.001).

Fig. 5. Antitumor efficacy in vivo. (a, b) Tumor volume, (c) tumor inhibition ratio, (d)

percentage of

necrotic

area, and (e) histopathological analyses

of H22

hepatoma-bearing mice after treatment with DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX with NS as a control. All statistical data were presented as mean ± SD (n = 6; **P < 0.01, ***P < 0.001). Scale bar = 100 µm.

21

The semi-quantitative fluorescent intensities of DOX distributed in major organs and tumors were shown in Fig. 4b. No matter at 1 or 2 weeks post-injection, Gel-CAD seemed to have remarkably higher fluorescent intensity than Gel-SAD in terms of all tested organs and tumors because of the pH sensitivity (P < 0.001) [23]. It was also observed that the fluorescent intensity for the Gel-CAD+DTX group was the highest among all situations at 2 weeks post-injection, indicating that Gel-CAD especially Gel-CAD+DTX could achieve a more sustained release and a more effective accumulation at tumor site as compared to the DOX and DTX encapsulates. 3.4. In vivo tumor suppression In vivo antitumor efficacy towards female H22 hepatoma-bearing mouse model was investigated to explain the advantages of the as-manufactured prodrug thermogels. The treatment proposals of nine groups including NS, DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX were administrated by in situ injection. During the whole experimental process, the tumor volume changes were recorded. As shown in Figs. 5a−b, the tumor volume of the control group was rapidly growing with time, while all the rest groups exhibited good effects on tumor suppression. At approximately 24 days after the last treatment, the average tumor inhibition ratio in the DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and the Gel-SAD+DTX groups was 46.7%, 40.3%, 58.6%, 50.9%, 59.8%, 46.9%, 71.5%, and 61.1% (Fig. 5c), respectively (P < 0.001). Interestingly, the tumor inhibition efficacy of Gel-CAD was significantly higher than that of the Gel-SAD group (P < 0.001), which could be explained by the

22

acid-responsive DOX release of Gel-CAD at tumor site. In addition, the DOX+DTX group showed a strong inhibitory effect on tumor growth, because the drug concentration was relatively higher in the DOX+DTX group compared with the other conditions for maintaining the same initial dose of drugs. Beyond that, the tumor inhibition efficacy of the Gel-CAD+DTX group (P < 0.001) and the Gel-SAD+DTX group (P < 0.05) was higher than that of the DOX+DTX group, which could be ascribed to the sustained drug release profile of the DOX-conjugated thermogels. Taken together, the most efficient tumor inhibition among all the testing situations was observed in the Gel-CAD+DTX group due to the synergistic effect of a DNA intercalator DOX and a microtubule-interfering DTX [19, 48, 49]. To further determine the tumor inhibition efficacy of the prodrugs, tissue sections of the tumors were assessed using H&E staining analyses. As shown in Fig. 5e, normal cell pattern and large nuclei were found in the tumor cells of the control group. While with regard to the Gel-SAD group, a poor tumor inhibition effect was verified by the appearance of a great quantity of spindle-shaped nuclei. In clear contrast, a serious cell shrink, a decreased density, a condensed nucleoplasm, and even a disappearance of tumor cells were observed in the Gel-CAD group, especially in the Gel-CAD+DTX group a large area of tumor cell apoptosis and necrosis were also exhibited. Furthermore, the histopathological analyses were conducted to confirm the antitumor efficacy as well as the potential of Gel-CAD and Gel-CAD+DTX for in situ antineoplastic therapy. As shown in Fig. 5d, the percentage of necrotic area in the Gel-CAD group was significantly higher than that of the Gel-SAD group (P < 0.001).

23

While the relative necrotic area in the Gel-CAD+DTX group was drastically higher than that of the Gel-CAD group, being consistent with the trend of the tumor inhibition rate. 3.5. In vivo security evaluation Many chemotherapy drugs have been reported with serious side effects in the clinical application, which always result in an acute cardiotoxicity and renal toxicity [50]. In this context, in vivo security of the chemotherapeutic agents becomes another crucial standard for the clinical chemotherapy, because it has a direct relationship with the survival rate of cancer patients. Herein, the safety of prodrugs was evaluated by monitoring of the body weight as well as the histopathological analyses of the major organs. All mice remained alive throughout the entire experiment. As shown in Figs. 6a−b, no significant difference was shown in the body weight of all groups, indicating the lower systemic toxicity and less injury to the body by the intratumor injection compared with the intravenous injection. Furthermore, the toxic side effects of DOX were examined by the histological analyses of the various organs, including heart, liver, spleen, lung, and kidney. As shown in Fig. 6c, there existed some myocardial inflammations and liver injuries in the DOX, DOX+DTX, Gel-SAD, and Gel-SAD+DTX groups. However, myocardial cells and nuclei of liver cells seemed regular in the DTX and Gel+DOX groups, especially in the control, Gel-CAD, and Gel-CAD+DTX groups the toxic side effects were obviously suppressed. In addition, varying degrees of congestion, necrosis of the spleen and granular degeneration of the kidney were also found in the DOX,

24

DOX+DTX, Gel-SAD, and Gel-SAD+DTX groups, whereas there were little spleen and kidney injuries in the other groups of control, Gel-CAD, and Gel-CAD+DTX. Such reduced cardiotoxicity, hepatotoxicity, and nephrotoxicity were benefit from the pH-sensitive amide bonds in the cis-aconitic anhydride that could be selectively reacted with the tumor cells [51]. 4. Conclusion

A synergistic combination of tumor microenvironment-sensitive polymeric DOX prodrug thermogel for DNA intercalation and a microtubule-interfering agent DTX was developed in the present study for local synergistic chemotherapy of hepatoma. The as-manufactured prodrug thermogel exhibited an ideal gelation temperature, sufficient mechanical strength, and long-term biodegradation behavior. Interestingly, the release rate of Gel-CAD in acidic extracellular microenvironment or intracellular endosome of tumors was significantly faster than that of Gel-SAD due to the existence of pH-sensitive amide bonds, which could lead to an effective DOX accumulation at tumor sites as well as an enhanced antitumor efficacy. More importantly, the addition of DTX to the DOX-conjugated thermogels (i.e., Gel-SAD and Gel-CAD) further upregulated the curative effect against tumors according to the synergistic mechanism, especially for the Gel-CAD+DTX group, the highest antitumor activity and the most reliable security in vivo were revealed. In summary, an effective combination of tumor microenvironment-labile polymeric prodrug thermogel with a complementary drug offered great prospect for in situ antineoplastic therapy.

25

Fig. 6. Biological safety in vivo. (a, b) Body weight changes and (c)

histopathological

analyses

of

the

visceral

organ

sections

of

the

H22

hepatoma-bearing mice after treatment with NS, DOX, DTX, DOX+DTX, Gel+DOX, Gel-CAD, Gel-SAD, Gel-CAD+DTX, and Gel-SAD+DTX. Scale bar = 100 µm.

26

Notes The authors declare no competing financial interest.

Acknowledgments

This work was financially supported by the National Natural Science Foundation of China (Grant Nos. 51673190, 51603204, 51673187, 81671804, 81171681, and 81701811) and the Science and Technology Development Program of Jilin Province (Grant Nos. 20160204015SF and 20160204018SF).

References

[1] K. Wei, X. Yu, R. Zheng, X. Peng, S. Zhang, M. Ji, Z. Liang, Z. Ou, W. Chen, Incidence and mortality of liver cancer in China, 2010, Chin. J. Cancer 33 (2014) 388–394. [2] Y. Liu, C. Chang, G.M. Marsh, F. Wu, Population attributable risk of aflatoxin-related liver cancer: systematic review and meta-analysis, Eur. J. Cancer 48 (2012) 2125–2136. [3] D.R. Vogus, V. Krishnan, S. Mitragotri, A review on engineering polymer drug conjugates to improve combination chemotherapy, Curr. Opin. Colloid Interface Sci. 31 (2017) 75–85. [4] E. Markovsky, H. Baabur, R. Satchi, Anticancer polymeric nanomedicine bearing synergistic drug combination is superior to a mixture of individually-conjugated drugs, J. Controlled Release 187 (2014) 145–157. [5] L. Yu, T. Ci, S. Zhou, W. Zeng, J. Ding, The thermogelling PLGA-PEG-PLGA block copolymer as a sustained release matrix of doxorubicin, Biomater. Sci. 1 (2013) 411–420. 27

[6] Y. Zhang, J. Ding, D. Sun, H. Sun, X. Zhuang, F. Chang, J. Wang, X. Chen, Thermogel-mediated sustained drug delivery for in situ malignancy chemotherapy, Mater. Sci. Eng. C Mater. Biol. Appl. 49 (2015) 262–268. [7] E.R. Ruskowitz, C.A. DeForest, Photoresponsive biomaterials for targeted drug delivery and 4D cell culture, Nat. Rev. Mater. 3 (2018) 17087. [8] C. Zhang, D. Pan, J. Li, J. Hu, A. Bains, N. Guys, H. Zhu, X. Li, K. Luo, Q. Gong, Z. Gu,

Enzyme-responsive

peptide

dendrimer-gemcitabine

conjugate

as

a

controlled-release drug delivery vehicle with enhanced antitumor efficacy, Acta Biomater. 55 (2017) 153–162. [9] J. Akimoto, M. Nakayama, T. Okano, Temperature-responsive polymeric micelles for optimizing drug targeting to solid tumors, J. Controlled Release 193 (2014) 2–8. [10] M. Karimi, A. Ghasemi, P. Sahandi Zangabad, R. Rahighi, S.M. Moosavi Basri, H. Mirshekari, M. Amiri, Z. Shafaei Pishbad, A. Aslani, M. Bozorgomid, D. Ghosh, A. Beyzavi, A.Vaseghi, A.R. Aref, L. Haghani, S. Bahrami, M. R. Hamblin, Smart micro/nanoparticles in stimulus-responsive drug/gene delivery systems, Chem. Soc. Rev. 45 (2016) 1457–1501. [11] C. Hu, N. Ma, F. Li, Y. Fang, Y. Liu, L. Zhao, S. Qiao, X. Li, X. Jiang, T. Li, F. Shen, Y. Huang, Q. Luo, J. Liu, Cucurbit[8]uril-based giant supramolecular vesicles: highly stable, versatile carriers for photoresponsive and targeted drug delivery, ACS Appl. Mater. Interfaces 10 (2018) 4603–4613. [12] C. Rossmann, M.A. McCrackin, K.E. Armeson, D. Haemmerich, Temperature sensitive liposomes combined with thermal ablation: effects of duration and timing of heating in mathematical models and in vivo, PLoS ONE 12 (2017) e0179131. [13] A. Kondo, S. Yamamoto, R. Nakaki, T. Shimamura, T. Hamakubo, J. Sakai, T. Kodama, T. Yoshida, H. Aburatani, T. Osawa, Extracellular acidic pH activates the sterol regulatory element-binding protein 2 to promote tumor progression, Cell Rep. 18 (2017) 2228–2242. [14] Z. Fan, S. Zhou, C. Garcia, L. Fan, J. Zhou, pH-Responsive fluorescent graphene quantum dots for fluorescence-guided cancer surgery and diagnosis, Nanoscale 9 (2017) 4928–4933. 28

[15] C. Lin, C. Kuan, L. Wang, H. Wu, Y. Chen, C. Chang, R. Huang, T. Wang, Integrated self-assembling drug delivery system possessing dual responsive and active targeting for orthotopic ovarian cancer theranostics, Biomaterials 90 (2016) 12–26. [16] B. He, T. Tan, H. Wang, H. Hu, Z. Wang, J. Wang, J. Li, K. Sun, Z. Zhang, Y. Li, Rational design of tumor microenvironment-activated micelles for programed targeting of breast cancer metastasis, Adv. Funct. Mater. 28 (2018) 1705622. [17] D. Sun, J. Ding, C. Xiao, J. Chen, X. Zhuang, X. Chen, pH-Responsive reversible PEGylation improves performance of antineoplastic agent, Adv. Healthcare Mater. 4 (2015) 844–855. [18] Y. Zhang, D. Sun, J. Ding, C. Li, F. Chang, X. Zhuang, X. Chen, J. Wang, Doxorubicin prodrug thermogel as sustained drug reservoir for in situ malignant therapy, J. Controlled Release 213 (2015) e126–127. [19] H. Wu, H. Jin, C. Wang, Z. Zhang, H. Ruan, L. Sun, C. Yang, Y. Li, W. Qin, C. Wang,

Synergistic

cisplatin/doxorubicin

combination

chemotherapy

for

multidrug-resistant cancer via polymeric nanogels targeting delivery. ACS Appl. Mater. Interfaces 9 (2017) 9426–9436. [20] X. Liang, C. Gao, L. Cui, S. Wang, J. Wang, Z. Dai, Self-assembly of an amphiphilic

janus

camptothecin−floxuridine

conjugate

into

liposome-like

nanocapsules for more efficacious combination chemotherapy in Cancer, Adv. Mater. 29 (2017) 1703135. [21] Y. Zhang, J. Ding, W. Xu, J. Wu, F. Chang, X. Zhuang, X. Chen, J. Wang, Biodegradable thermogel as culture matrix of bone marrow mesenchymal stem cells for potential cartilage tissue engineering, Chin. J. Polym. Sci. 32 (2014) 1590–1601. [22] Y. Zhang, J. Zhang, F. Chang, W. Xu, J. Ding, Repair of full-thickness articular cartilage defect using stem cell-encapsulated thermogel, Mater. Sci. Eng., C 88 (2018) 79–87. [23] D. Sun, J. Ding, C. Xiao, J. Chen, X. Zhuang, X. Chen, Preclinical evaluation of antitumor activity of acid-sensitive PEGylated doxorubicin, ACS Appl. Mater. Interfaces 6 (2014) 21202–21214. 29

[24] W. Hou, X. Zhao, X. Qian, F. Pan, C. Zhang, Y. Yang, J.M. de la Fuente, D. Cui, pH-Sensitive self-assembling nanoparticles for tumor near-infrared fluorescence imaging and chemo-photodynamic combination therapy, Nanoscale 8 (2016) 104–116. [25] K. Shi, Y. Wang, Y. Qu, J. Liao, B. Chu, H. Zhang, L. Feng, Z. Qian, Synthesis, characterization, and application of reversible PDLLA-PEG-PDLLA copolymer thermogels in vitro and in vivo, Sci. Rep. 6 (2016) 19077. [26] M. Kitagawa, T. Maeda, A. Hotta, PEG-based nanocomposite hydrogel: Thermo-responsive sol-gel transition and degradation behavior controlled by the LA/GA ratio of PLGA-PEG-PLGA, Polym. Degrad. Stab. 147 (2018) 222–228. [27] X. Zhu, C. Liu, J. Duan, X. Liang, X. Li, H. Sun, D. Kong, J. Yang, Synthesis of three-arm block copolymer poly(lactic-co-glycolic acid)–poly(ethylene glycol) with oxalyl chloride and its application in hydrophobic drug delivery, Int. J. Nanomed. 11 (2016) 6065–6077. [28] L. Yu, H. Hu, L. Chen, X. Bao, Y. Li, L. Chen, G. Xu, X. Ye, J. Ding, Comparative studies of thermogels in preventing post-operative adhesions and corresponding mechanisms, Biomater. Sci. 2 (2014) 1100–1109. [29] L. Chen, T. Ci, L. Yu, J. Ding, Effects of molecular weight and its distribution of PEG block on micellization and thermogellability of PLGA–PEG–PLGA copolymer aqueous solutions, Macromolecules 48 (2015) 3662–3671. [30] L. Yu, Z. Zhang, H. Zhang, J. Ding, Mixing a sol and a precipitate of block copolymers

with

different

block

ratios

leads

to

an

injectable

hydrogel,

Biomacromolecules 10 (2009) 1547–1553. [31] C.D.F. Moreira, S.M. Carvalho, H.S. Mansur, M.M. Pereira. Thermogelling chitosan–collagen–bioactive glass nanoparticle hybrids as potential injectable systems for tissue engineering, Mater. Sci. Eng., C 58 (2016) 1207–1216. [32] A. Fakhari, J. Anand Subramony, Engineered in-situ depot-forming hydrogels for intratumoral drug delivery, J. Controlled Release 220 (2015) 465–475. [33] J. Luan, W. Shen, C. Chen, K. Lei, L. Yu, J. Ding, Selenium-containing thermogel for controlled drug delivery by coordination competition, RSC Adv. 5 (2015) 30

97975–97981. [34] H. Renata, Z. Wang, F.H. Arnold, Expanding the enzyme universe: accessing non-natural reactions by mechanism-guided directed evolution, Angew. Chem., Int. Ed. 54 (2015) 3351–3367. [35] D. Yuan, J. Shi, X. Du, N. Zhou, B. Xu, Supramolecular glycosylation accelerates proteolytic degradation of peptide nanofibrils, J. Am. Chem. Soc. 137 (2015) 10092–10095. [36] J. Wang, J. Bhattacharyya, E. Mastria, A. Chilkoti, A quantitative study of the intracellular fate

of pH-responsive

doxorubicin-polypeptide nanoparticles, J.

Controlled Release 260 (2017) 100–110. [37] Q. Zhou, Y. Hou, L. Zhang, J. Wang, Y. Qiao, S. Guo, Dual-pH sensitive charge-reversal nanocomplex for tumor-targeted drug delivery with enhanced anticancer activity, Theranostics 7 (2017) 1806–1819. [38] N. Murthy, Y.X. Thng, S. Schuck, M. Xu, J.M.J. Fréchet, A novel strategy for encapsulation and release of proteins: hydrogels and microgels with acid-labile acetal cross-linkers, J. Am. Chem. Soc. 124 (2002) 12398–12399. [39] Y. Liu, X. Chen, S. Li, Q. Guo, J. Xie, L. Yu, X. Xu, C. Ding, J. Li, J. Ding, Calcitonin-loaded thermosensitive hydrogel for long-term antiosteopenia therapy, ACS Appl. Mater. Interfaces 9 (2017) 23428–23440. [40] Y. Li, K. Xiao, W. Zhu, W. Deng, K.S. Lam, Stimuli-responsive cross-linked micelles for on-demand drug delivery against cancers, Adv. Drug Delivery Rev. 66 (2014) 58–73. [41] X. Guan, Y. Li, Z. Jiao, J. Chen, Z. Guo, H. Tian, X. Chen, A pH-sensitive charge-conversion system for doxorubicin delivery, Acta Biomater. 9 (2013) 7672–7678. [42] G. Yu, W. Yu, L. Shao, Z. Zhang, X. Chi, Z. Mao, C. Gao, F. Huang, Fabrication of a targeted drug delivery system from a pillar[5]arene-based supramolecular diblock copolymeric amphiphile for effective cancer therapy, Adv. Funct. Mater. 26 (2016) 8999–9008. [43] Y. Zhang, C. Yang, W. Wang, J. Liu, Q. Liu, F. Huang, L. Chu, H. Gao, C. Li, D. 31

Kong, Q Liu, J. Liu, Co-delivery of doxorubicin and curcumin by pH-sensitive prodrug nanoparticle for combination therapy of cancer, Sci. Rep. 6 (2016) 21225. [44] E.M. Mastria, M. Chen, J.R. McDaniel, X. Li, J. Hyun, M.W. Dewhirst, A. Chilkoti, Doxorubicin-conjugated polypeptide nanoparticles inhibit metastasis in two murine models of carcinoma, J. Controlled Release 208 (2015) 52–58. [45] H. Ma, C. He, Y. Cheng, Z. Yang, J. Zang, J. Liu, X. Chen, Localized co-delivery of doxorubicin, cisplatin, and methotrexate by thermosensitive hydrogels for enhanced osteosarcoma treatment, ACS Appl. Mater. Interfaces 7 (2015) 27040–27048. [46] E.S. Lee, K.T. Oh, D. Kim, Y.S. Youn, Y.H. Bae, Tumor pH-responsive flower-like micelles of poly(L-lactic acid)-b-poly(ethylene glycol)-b-poly(L-histidine), J. Controlled Release 123 (2007) 19–26. [47] J. Chen, J. Ding, Y. Zhang, C. Xiao, X. Zhuang, X.Chen, Polyion complex micelles with gradient pH-sensitivity for adjustable intracellular drug delivery, Polym. Chem. 6 (2015) 397–405. [48] G.M. Alushin, G.C. Lander, E.H. Kellogg, R. Zhang, D. Baker, E. Nogales, High-resolution microtubule structures reveal the structural transitions in αβ-tubulin upon GTP hydrolysis, Cell 157 (2014) 1117–1129. [49] S. Imstepf, V. Pierroz, R. Rubbiani, M. Felber, T. Fox, G. Gasser, R. Alberto, Organometallic rhenium complexes divert doxorubicin to the mitochondria, Angew. Chem., Int. Ed. 55 (2016) 2792–2795 [50] L. Jing, L. Li, J. Zhao, J. Zhao, Z. Sun, S. Peng, Zinc-induced metallothionein overexpression prevents doxorubicin toxicity in cardiomyocytes by regulating the peroxiredoxins, Xenobiotica 46 (2016) 715–725. [51] J. Wang, W. Xu, H. Guo, J. Ding, J. Chen, J. Guan, C. Wang, Selective intracellular drug delivery from pH-responsive polyion complex micelle for enhanced malignancy suppression in vivo, Colloids Surf., B 135 (2015) 283–290.

32

33