Cationic amphiphilic microfibrillated cellulose (MFC) for potential use for bile acid sorption

Carbohydrate Polymers 132 (2015) 598–605

Contents lists available at ScienceDirect

Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol

Cationic amphiphilic microfibrillated cellulose (MFC) for potential use for bile acid sorption Xuhai Zhu a , Yangbing Wen a,b,∗ , Dong Cheng a , Changmo Li c , Xingye An a , Yonghao Ni a,b a b c

Tianjin Key Laboratory of Pulp and Paper, Tianjin University of Science & Technology, Tianjin 300457, China Limerick Pulp & Paper Centre & Department of Chemical Engineering, University of New Brunswick, Fredericton, NB, Canada E3B 5A3 Key Laboratory of Food Nutrition and Safety, Ministry of Education of China, Tianjin University of Science and Technology, Tianjin 300457, China

a r t i c l e

i n f o

Article history: Received 2 March 2015 Received in revised form 29 May 2015 Accepted 15 June 2015 Available online 26 June 2015 Keywords: Microfibrillated cellulose Cationization Hydrophobization Bile acid Sorption

a b s t r a c t In this work, Micro-fibrillated Cellulose (MFC) was cationically modified by quaternary ammonium groups with different chemical structures aiming to improve the sorption capacity to bile acid. The invitro bile acid sorption was performed by investigating various factors, such as quaternary ammonium group content and length of its alkyl substituent of the modified cationic MFC (CMFC), ionic strength, initial concentration and hydrophobicity of bile acid. The results showed that the sorption behavior of the modified CMFC was strongly influenced by the quaternary ammonium group content and the lengths of its alkyl substituent, the sorption capacity for the modified CMFC with a C18 alkyl substituent, was approximately 50% of that of Cholestyramine. The experimental isotherm results were well fitted into the Temkin model. The effect of salts in the solution was smaller for the bile acid sorption onto the hydrophobic CMFC than the CMFC. It was also found that the binding capacity of CMFC was higher for more hydrophobic deoxycholate in comparison with cholate. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction It is well-known that cholesterol is critical in cell membrane formation and biosynthesis of bile acids, steroid hormones and vitamin D. On the other hand, too high level of cholesterol in blood would not be desirable, for example, causing atherosclerosis (Kazlauske, Ramanauskiene, & Liesiene, 2014), which is one of the main reasons for illnesses and deaths (Mason, 2011). Bile acid sorbents are polymeric compounds that can be used to control the cholesterol level. When these polymers are ingested, bile acids can be adsorbed onto them, which will lead to increased fecal bile acid excretion and, in turn, decreasing the levels of serum and/or tissue cholesterol (Kahlon & Smith, 2007; Einarsson et al., 1991). Bile acids consist of a curved steroidal skeleton with a hydrophobic face and a hydrophilic face, which includes a carboxylic acid group (Fig. 1). This provides the bile acids amphiphilic ˜ character and self-associative behavior (Yanez, Chianella, Piletsky, Concheiro, & Alvarez-Lorenzo, 2010). Bile acid sorbents, such as

∗ Corresponding author at: Tianjin University of Science & Technology, Tianjin Key Laboratory of Pulp and Paper, No. 29, 13th Avenue, Tianjin 300457, China. Tel.: +86 022 60602913. E-mail address: [email protected] (Y. Wen). http://dx.doi.org/10.1016/j.carbpol.2015.06.063 0144-8617/© 2015 Elsevier Ltd. All rights reserved.

Cholestyramine, Cholestipol and Cholestimide, have been extensively used in clinic for several decades (Ast & Frishman, 1990; Schulman et al., 1990; Honda & Nakano, 2000). However, the above sorbents are based on synthetic polymers. For example, Cholestyramine is quaternized styrene-divinilbenzene copolymer (Ast & Frishman, 1990); and Colestipol is synthesized from tetraethylenpentamine and epichlorohydrin (Schulman et al., 1990); while Cholestimide is an anion-exchange resin with an imidazolium salt on an epoxide polymer skeleton (Honda & Nakano, 2000). Therefore, in the clinical practice, these sorbents can cause some side effects such as constipation, nausea and meteorism (Scaldaferri, Pizzoferrato, Ponziani, Gasbarrini, & Gasbarrini, 2013). Cellulose is the most abundant natural polymer on earth. Cellulose-based materials, such as Cellulose Nanocrystal (CNC) and Microfibrillated cellulose (MFC) are emerging biomaterials having many unique properties, including excellent surface area and high aspect ratio (Moon, Martini, Nairn, Simonsen, & Youngblood, 2011). Previous studies have demonstrated that CNC, MFC or modified cellulose material can act as sorbents for various organic compounds, including dyes, aromatic compounds and herbicides (Pei, Butchosa, Berglund, & Zhou, 2013; Sun, Hou, He, Liu, & Ni, 2014; Sun et al., 2013; Maatar, Alila, & Boufi, 2013). Since glucosidic bonds in cellulose cannot be hydrolyzed by human digestive enzymes, once ingested, cellulose material maintains its binding capacity to bile

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

OH

OH

O

CH3

O Na

OH

H

CH3

CH3

OH OH

H

Sodium cholate

Sodium deoxycholate OH

O

CH3 CH3

CH3

OH

O

CH3

O Na

CH3

CH3

599

N H

O

Na

O

OH

H

Sodium glycocholate OH CH3

CH3

CH3

OH

H

O N H

O SO O

Na

OH

Sodium taurocholate Fig. 1. Examples of bile salts.

acid in the physiological condition of gut, which accordingly makes it an attractive material for food supplement for people with high cholesterol. In a previous study (Zhu et al., 2014), we have carried out invitro bile acid sorption studies of cationic MFC (CMFC) that was cationized using trimethylammonium chloride. It was found that the primary interaction occurred between bile acid and CMFC was electrostatic in nature. However, the sorption capacity of CMFC was lower than that of Cholestyramine under the same condition. In such applications, one would expect that a higher hydrophobic substituent in the CMFC may enhance the interaction with bile acid. Modifications of polymers to increase the hydrophobic interaction have been reported in the literature. For example, when investigating the interaction between cationic dextran hydrogels with N-(2-hydroxypropyl)-N,N-dimethyl-N-alkyl ammonium pendant groups and bile acids, Nichifor, Cristea, and Carpov (2000) reported that the increase in the chain length of the alkyl substituent of the modified dextran hydrogel can strongly increase the rate of initial binding constant, K0 and the binding capacity, K. Also, the ionic strength has a small influence on the bile acid binding by hydrogels with alkyl substituent of C8 or C12 (Nichifor, Zhu, Cristea, & Carpov, 2001). It was suggested that when alkyl substituent is longer than C4 , the binding is governed by hydrophobic interaction between alkyl substituent and bile acid, and aggregation may occur via mixed micelle formation (Nichifor et al., 2001). The objective of this study was to enhance the interaction of the modified CMFC having cationic groups and long alkyl chains with bile acid. The influences of quaternary ammonium group content (cationic charge density) and length of its alkyl substituent of the modified CMFC, ionic strength, initial concentration and hydrophobicity of bile acid on the binding characteristics were examined.

2. Methods 2.1. Materials A sulfite- based dissolving pulp used to prepare the CMFC was from a mill in Shandong, China. A endoglucanase sample (Novozym 476, Novozym 435) was used without further purification. Tertiary amines, epichlorohydrin (ECH) and N-(2-3-epoxypropyl) trimethylammonium chloride (EPTMAC) were purchased from Sigma Aldrich and were used as received. The Total Bile Acids Test Kit for bile acid measurement was purchased from CY-BIO, Co., Shangyu, China. Other reagents used in this work were supplied by Tianjin Chemical Reagent Co. Ltd., China. Distilled and deionized water was used throughout experiments unless otherwise specified.

2.2. Preparation of sorbents The preparation of CMFC was as follows: (a) enzyme pretreatment of the sulfite-based dissolving pulp (details will be given in the next section); (b) chemical modification of enzyme-treated sulfite dissolving pulp for the attachment of the first pendant quaternary ammonium group with methyl substitutes (single-modified fibers); (c) chemical modification of single-modified fibers for the attachment of the second quaternary ammonium group with long alkyl chains (double-modified fibers); (d) the cationic pulp was diluted to 1% w/w of mixtures and then homogenized to produce the CMFC. The schematic representation of preparing the cationic amphiphilic MFC and chemical structure of the modified CMFC was illustrated in Fig. 2 and the chemical characteristics were listed in Table 1.

600

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

Fig. 2. Schematic representation of preparing the cationic amphiphilic MFC and the sorption of bile acid (Alkyl (R1 ) = methyl, (R2 ) = butyl, octyl, benzyl, dodecyl, hexadecyl, octadecyl).

Table 1 Physico-chemical characteristics of the samples. Samples

R1

R2

Dissolving pulp fibers Un-modified MFC Cholestyramine C1 (7.00) C1 (7.69) C1 (16.57) C1 (6.50)–C4 (0.41) C1 (6.50)–benzyl(0.42) C1 (6.50)–C8 (0.42) C1 (6.50)–C12 (0.43) C1 (6.50)–C16 (0.43) C1 (6.50)–C18 (0.42)

– – – CH3 CH3 CH3 CH3 CH3 CH3 CH3 CH3 CH3

– – – – – – C4 H9 Benzyl C8 H17 C12 H25 C16 H33 C18 H37

*

Content in amino groups DS1 (mol%)

DS2 (mol%)

Total charge density (meq/g)

– – – 7.00 7.69 16.57 6.50 6.50 6.50 6.50 6.50 6.50

– – – – – – 0.41 0.42 0.42 0.43 0.43 0.42

−0.06 −0.06 3.71* 0.32 0.40 0.99 0.30 0.30 0.32 0.30 0.32 0.30

The charge density of Cholestyramine was from the literature (Kazlauske et al., 2014).

(a) The enzymatic pretreatment followed a procedure reported elsewhere (Pääkkö et al., 2007). 100 g (calculated as dry fibers) of the pulp was dispersed in 2.0 l of phosphate buffer (pH 5.8, final pulp concentration 5% w/w) with 0.17 ␮l monocomponent endoglucanases per gram fiber (5 ECU/␮l) and was incubated at 58 ◦ C for 10 h with continuous mixing. Then, the samples were washed with deionized water followed by the denaturation of the mono-component endoglucanase at 80 ◦ C for 30 min. At the end, the pulp sample was washed with deionized water again. (b) The cationization (quaternization) reaction of enzyme-treated sulfite dissolving pulp was as follows (Zaman, Xiao, Chibante, & Ni, 2012; Olszewska et al., 2011): 7.5 g, or 15 g or 30 g EPTMAC was added to a 60 g of enzyme-treated hardwood sulfite dissolving pulp (25% w/w) dispersion, together with 2.25 g NaOH dissolved in 15 g distilled water to produce the cationic fibers with different charge densities. These dispersions were diluted with 180 g isopropanol and were allowed to react at 50 ◦ C for 3 h after which these cationic fibers were washed with distilled water three times. The resultant suspensions were then dialyzed (molecular weight cut-off 14,000) against distilled water for 3 days. At last, these cationic samples were dried under vacuum. (c) The second quaternary ammonium groups with long alkyl chains were attached to enzyme-treated sulfite dissolving pulp. This second chemical modification was done on the

single-modified fibers. In an early literature (Nichifor, Stanciu, & Simionescu, 2010; Mocanu & Nichifor, 2014), the procedure for this modification and its reaction mechanism were elaborated for the quaternization of crosslinked dextran. The quaternization reactions were performed in isopropanol (single-modified fibers/isopropanol ratio = 1 g/10 ml), under constant stirring, with an equimolar mixture of epichlorohydrin and a tertiary amine for 6 h at 50 ◦ C. The reaction mixture was filtered and the cationic fibers left on the filter were sequentially rinsed with ethanol, 0.1 N HCl and water (until the absence of chloride ions in the filtrate). Therefore, the cationic charge and hydrophobic property were introduced to the enzyme-treated sulfite dissolving pulp in two steps, the first step renders the MFC with the cationic charge by the N-(2,3-epoxypropyl) trimethylammonium chloride, followed by the second step with N-(2-hydroxypropyl)-N,N-dimethyl-N-alkyl ammonium functional groups (Alkyl (R2 ) = butyl, octyl, benzyl, dodecyl, hexadecyl, octadecyl). (d) A 5 g of the dried cationic fibers were diluted to 1% w/w of 500 g mixtures. Afterward, an ultra-turrax was used to homogenize the fibers. After 15 min, the cationic fibers were homogenized in a high-pressure homogenizer (GYB-3004, Donghua Co. Ltd., Shanghai, China) for 20 passes at 100 MPa to produce the CMFC. For un-modified MFC, 5 g of enzyme-treated dissolving pulp was directly diluted to 1% w/w of 500 g mixtures and then homogenized to produce the un-modified MFC.

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

The degree of substitution of quaternary ammonium groups was calculated based on the nitrogen content of the cationically modified fibers as given in Eqs. (1) and (2) (Zaman et al., 2012; Mocanu & Nichifor, 2014). The nitrogen content was determined using a CHNS 2400 II Perkin Elmer analyzer. DS1 =

162 × N1 100 (mol/100 AGU) 100 × 14 − N1 × MS1

  DS2 =



X MS1 − MS2 + 162 × 100/DS1 100 − X × MS2



(1)

 − 1 × DS1

(mol/100 AGU)

(2)

where DS1 and DS2 represent the degree of the first and second group substitution, respectively. X = Nt /14; N1 and Nt are the nitrogen contents after the first and second quaternization step, respectively; MS1 and MS2 are the molecular weights (in g) of the first and the second pendent ammonium group, respectively. The DS is expressed as mol%. 2.4. Charge density measurements The charge densities of samples were determined using a Particle Charge Detector, Mütek PCD 04 titrator (Herrsching, Germany). Poly(diallyldimethylammonium chloride) (PDADMAC) solution (0.1 mN) or polyelectrolyte potassium polyvinyl sulfate (PVSK) solution (0.1 mN) was used as the titrant (Liu, Ni, Fatehi, & Saeed, 2011) 2.5. Fourier transform infrared spectroscopy (FTIR) For Fourier Transform Infrared (FTIR) analysis, the powder sample of Un-modified MFC, CMFC sample (C1 (16.57)) and CMFC sample (C1 (6.50)–C16 (0.43)) were used. The analysis was conducted by using a FTIR-650 spectrometer (China) over the range from 400 to 4000 cm−1 . The sample was prepared at approximately 2% by weight in KBr (potassium bromide). 2.6. Transmission electron microscopy (TEM) analysis Aqueous suspension (0.004 wt%) of CMFC sample (C1 (16.57)) and CMFC sample (C1 (6.50)–C16 (0.43)) were prepared. A drop of 20 ␮l suspension was transferred to a carbon-coated copper grid using a pipette. The grid was air-dried at room temperature overnight. The TEM observation was conducted by using a JEOL 2010 STEM instrument (Japan) operated at an accelerating voltage of 200 keV (Sun et al., 2014). 2.7. In vitro bile acid sorption In vitro bile acid binding of CMFC with different chemical structures was determined by following previously published procedures (Kim & White, 2009). The bile acid mixture was formulated with sodium cholate, sodium deoxycholate, sodium glycocholate, and sodium taurocholate with proportions as 35, 35, 15, and 15% (w/w) in a 50 mM phosphate buffer (pH 6.8) based on the composition of human bile. Cholestyramine was used as a positive control and sulfite-based dissolving pulp was used as a negative control. 50 mg of Cholestryramine, dissolving pulp fibers, and CMFC samples were treated with 20 ml of 0.01 N hydrochloric acid and then incubated in a shaking water bath at 37 ◦ C for 1 h, aiming to simulate gastric digestion. The pH of the materials was then adjusted to 6.8 with 0.1 N sodium hydroxide. 20 ml of bile acid mixture

(1.4 ␮mol/ml) and 20 ml of porcine pancreatin (6.25 mg/ml in a 50 mM phosphate buffer, pH 6.8) were added and incubated in a shaking water bath at 37 ◦ C for 1 h. The mixtures were transferred to 10 ml centrifuge tubes and centrifuged at 99,000 × g for 30 min at 6 ◦ C. Then clear filtrates were collected for analysis. To determine the sorption isotherms of sodium cholate onto CMFC sample (C1 (6.50)–C16 (0.43)), various sodium cholate solutions of known concentrations (10 ml) were added to 80 mg of the modified CMFC in a 80 ml Elenmeyer flask and shaken at 120 rpm and 37 ◦ C for 300 min. Water was used as the incubation media. Samples were transferred to 10 ml centrifuge tubes and centrifuged at 99,000 × g for 30 min at 6 ◦ C in an ultracentrifuge. Then the clear filtrate was collected for analysis. To investigate the selective sorption of CMFC on bile acids and the effect of the ionic strength on sorption, different concentrations of sodium chloride were introduced separately into the sorption system. The percent adsorbed was determined after 50 mg of CMFC samples were separately equilibrated with 50 ml of 3 mM of cholate and deoxycholate solution in water at 37 ◦ C for 3 h. Samples were transferred to 10 ml centrifuge tubes and centrifuged at 99,000 × g for 30 min at 6 ◦ C in an ultracentrifuge. Then the clear filtrates were collected for analysis. The concentrations of bile acid in the original solution and clear filtrate were analyzed by using a Total Bile Acids Test Kit (Kim & White, 2011). The samples were diluted to the desired range of the test kit. The concentration of bile acid was calculated on the basis of a calibration curve. Six replicates were carried out and the averages and standard error of the mean (SEM) were reported. 3. Results and discussion 3.1. Characterization of CMFC Results on chemical structure and DS of quaternary ammonium groups and charge density of samples are listed in Table 1. The sample codes of the modified CMFC, for example, C1 (6.50)–C12 (0.42) representing that the CMFC contains both of N-(2-hydroxypropyl)-N,N,N-trimethylammonium chloride groups with DS1 = 6.50 mol% and N-(2-hydroxypropyl)-N,N-dimethyl-Ndodecyl ammonium groups with DS2 = 0.42 mol%. Fig. 3 shows the FTIR spectrum of the adsorbent at the different steps of the modification. The un-modified MFC (Spectrum a) exhibits the characteristic bands of cellulose skeleton at 3417

un-modified MFC CMFC sample (C1(16.57))

1.0

CMFC sample (C1(6.50)-C16(0.43))

0.8

Transmittance (%)

2.3. Determination of degree of substitution (DS)

601

b

0.6 1644 901

0.4

2896 2925

0.2

1410 1644

2896 2896

c 1644

3417

1479

1112

a

1047 901

0.0 4000

3500

3000

2500

2000

1500

1000

500

-1 Wave number (cm ) Fig. 3. FTIR spectra of the un-modified MFC (Spectrum a), the CMFC sample (C1 (16.57)) (Spectrum b) and the CMFC sample (C1 (6.50)–C16 (0.43)) (Spectrum c).

602

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

and 2896 cm−1 , which were due to hydroxyl (-OH) stretching and symmetric C H vibrations, respectively (Zaman et al., 2012; Jahan, Saeed, He, & Ni, 2011; Sun, Sun, Liu, Fowler, & Tomkinson, 2002). An intense band at 1644 cm−1 was attributed to the absorbed moisture of the samples (Ren, Sun, Liu, Chao, & Luo, 2006). The absorbance at 901 cm−1 can be assigned to the C H deformation of the glycosidic linkage of the glucose units (Kacurakova, Ebringerova, Hirsch, & Hromadkova, 1994). Those between 1030 cm−1 and 1163 cm−1 were attributed to the C O stretching in MFC (Liu et al., 2011). Compared to Spectrum a, both Spectrum b (cationically modified MFC of sample C1 (16.57)) and Spectrum c (cationically modified MFC of sample C1 (6.50)–C16 (0.43)) in Fig. 3 shows characteristic bands at 1410 cm−1 (Ren, Sun, Liu, Lin, & He, 2007), which is clear evidence of quaternization of CMFC. An increase in intensity of the major ether bands between 1030 cm−1 and 1163 cm−1 provides evidence of both of N-(2-hydroxypropyl)-N,N,N-trimethylammonium chloride groups and N-(2-hydroxypropyl)-N,N-dimethyl-N-hexadecyl ammonium groups grafting onto the MFC surface (Zaman et al., 2012; Ren et al., 2006). Moreover, in Spectrum c of the CMFC sample C1 (6.50)–C16 (0.43), the emergence of a new peak at 1479 cm−1 was attributed to methyl groups of the cationic substituent (Zaman et al., 2012; Ren et al., 2006). An additional evidence of the success of the hydrophobic modification was at 2925 cm−1 , which is due to the presence of the aliphatic backbone (Alila & Boufi, 2009) that was attached. In the design of the present study, the modified CMFC properties, such as charge density, content of amino groups, and hydrophilic/lipophilic balance (HLB) were varied. The HLB of the modified CMFC was studied by: (a) increasing the DS1 in the single modified MFC: C1 (7.00), C1 (7.69), C1 (16.57); (b) changing the hydrophobicity of the alkyl substituent R2 in the double modified MFC (R2 = butyl, octyl, dodecyl, hexadecyl, octadecyl); A benzyl group as a substituent for R2 was also studied. 3.2. In-vitro bile acid sorption Shown in Table 2 are the sorption results of bile acid onto the modified CMFC with different chemical structures. For comparison, the dissolving pulp, un-modified MFC were also included. The sorption of Cholestyramine was also carried out under otherwise the same conditions. It can be seen that the bile acid binding was the highest (34.59 ␮mol/100 mg) for Cholestyramine based on the equal dry matter (DM) basis, while the dissolving pulp exhibited the lowest binding capacity (0.92 ␮mol/100 mg). Un-modified MFC, although having same charge density to the dissolving pulp, had a much higher binding capacity for cholate (3.23 ␮mol/100 mg) than

Table 2 In-vitro bile acid binding with CMFC having equal weight but different chemical structures (expressed as dry matter (DM) basis, reported as an average of 6 replicates). Sorbents

Dissolving pulp fibers Un-modified MFC Cholestyramine C1 (7.00) C1 (7.69) C1 (16.57) C1 (6.50)–C4 (0.41) C1 (6.50)–Benzyl(0.42) C1 (6.50)–C8 (0.42) C1 (6.50)–C12 (0.43) C1 (6.50)–C16 (0.43) C1 (6.50)–C18 (0.42)

Bile acid binding (␮mol/100 mg DM) 0.92 ± 0.64 3.23 34.59 6.88 8.42 13.13 7.16 13.54 12.00 14.34 18.07 18.45

± ± ± ± ± ± ± ± ± ± ±

0.53 0.62 0.40 0.34 0.37 0.63 0.23 0.55 0.48 0.23 0.50

Bile acid binding relative to Cholestyramine (%) 2.65 9.34 100.00 19.89 24.34 37.97 20.70 39.13 34.68 41.46 52.27 53.34

Fig. 4. TEM images of the CMFC sample (C1 (16.57)) (a) and the CMFC sample (C1 (6.50)–C16 (0.43)) (b).

dissolving pulp, which can be attributed to the increased fiber surface area (Pei et al., 2013; Morandi, Heath, & Thielemans, 2009; Alila, Aloulou, Thielemans, & Boufi, 2011). The modified CMFC with a charge density of 0.32 meq/g and alkyl substituents of C18 , the binding capacity for bile acid was significantly increased to 18.45 ␮mol/100 mg under the studied conditions. The increased binding capacity of the modified CMFC in comparison with that of dissolving pulp can be explained as follows: (a) the increased fiber surface area, as shown in Fig. 4, the modified CMFC were well dispersed, with excellent physical characteristics for being good sorption materials (Pei et al., 2013), (b) the cationic nature due to the presence of quaternary amine groups, can facilitate the interaction of CMFC with bile acid through electrostatic force, (c) the presence of hydrophobic substituent at the quaternary nitrogen of the sorbent functional groups, can improve the hydrophobic interaction between the modified CMFC with bile acid.

20

700

18

600

Adsorbed sodium cholate (qe), mg/g

Bile acid binding (μmol/100 mg DM)

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

16 14 12 10 8

603

500 400 300 200 100

6

0 0

2

4

6

8

10

12

14

16

18

0.0

20

length of the alkyl chain

0.2

0.4

0.6

0.8

1.0

Sodium cholate remaining in the solution (Ce), g/l

Fig. 5. Influence of the alkyl chain length of R2 substituent on binding of bile acid by the modified CMFC with N-(2-hydroxypropyl)-N,N-dimethyl-N-alkyl ammonium functional groups.

Fig. 6. Sorption isotherm of sodium cholate (C1 (6.50)–C16 (0.43)) (at 37 ◦ C, 120 rpm for 300 min).

As shown in Table 2, at the same amount of sorbent, the CMFC sample (C1 (6.50)–C18 (0.42)) can have about 50% capacity in binding bile acid relative to Cholestyramine. The bile acid binding capacities were calculated on the basis of 100 mg of single cationically modified MFC samples (Table 2) to evaluate the effect of charge density on bile acid binding to CMFC. As shown in Table 2, the adsorbed amount of bile acid increased with increasing cationic charge density from 0.32 to 0.99 meq/g on the CMFC. The CMFC sample (C1 (16.57)) with a cationic charge density of 0.99 meq/g was able to adsorb 13.13 ␮mol/100 mg of bile acid. This can be compared to the sorption capacity of cationized cotton cellulose and microcrystalline cellulose of 11.3 and 6.2 ␮mol/100 mg cholic acid, respectively (Kazlauske et al., 2014). Chitosan can adsorb 1.5 ␮mol/100 mg cholic acid (Kazlauske et al., 2014). The sorption capacity of the CMFC sample (C1 (16.57)) is comparable to that of the oat ␤-Glucan, which can have about 38% capacity in binding bile acid relative to Cholestyramine (Kim & White, 2011), but lower than that of the quaternized crosslinkeddextran hydrogel (Nichifor et al., 2000). This is because the cationic dextran has a higher degree of substitution (about 20–25 mol% N(2-hydroxypropyl)-N, N-dimethyl-N-alkyl ammonium groups). The relative bile acid binding results of different alkyl chains were shown in Fig. 5 to demonstrate the optimum chemical structure of hydrophobic CMFC for bile acid binding. It can be found that the binding capacity of CMFC to bile acid increased significantly as the chain length of the alkyl substituent (R2 ) was higher than C4 , and this was true till the alkyl chain length was C16 . It was reported that modified cationic dextran samples showed a strong increase toward cholate sorption with the increase in the alkyl chain length of the substituent (Nichifor et al., 2000, 2001). The above results indicated that the optimum chemical structure of the modified CMFC providing the greatest bile acid binding, under the conditions studied, should contain a desired cationic charge density and a C16 –C18 alkyl chain length of the substituent. Therefore, it can be concluded that the hydrophilic/lipophilic balance is critical to develop MFC-based bile acid sorbent to obtain a desired physiological effect.

adsorbed onto the modified CMFC as a function of the amount remaining in the solution. The sorption of sodium cholate depends on a number of factors, e.g., the charge interaction; hydrophobicity interaction; in addition, the presence of a large amount of surface area of the modified CMFC had a positive effect on the cholate sorption. In an early study (Maatar et al., 2013), it was found that the sorption of dissolved organics, including aromatic organic solutes or herbicides, onto cellulose-based aerogel was also affected by its molecular weight and surface characteristics. Cholate structure has both hydrophobic property (due to its steroid nucleus) and charge interaction potential (due to its ionic characteristics) (Rub et al., 2013), thus can have both hydrophobic interactions and electrostatic interactions with potential absorbents. Therefore, the sorption of cholate onto the modified CMFC can occur via the hydrophobic/hydrophobic interactions and electrostatic force, similar mechanism was proposed in the literature to account for the results of surface hydrophobized cellulose fibers as adsorbents for dissolved organic pollutants (Maatar et al., 2013), and those of ˜ acrylic polymers as adsorbents for bile acids (Yanez et al., 2010). Due to the self-assembly of sodium cholate (Rub et al., 2013), the diffusion of sodium cholate into the pores of cellulose was probably limited. This implies that the sorption of sodium cholate onto the surface of the modified CMFC would mainly occur on the surface. The experimental sorption data plotted in Fig. 6 was employed to better understand the sorption isotherm of sodium cholate onto the CMFC. Three appropriate sorption models, i.e., Langmuir, Freundlich and Temkin isotherms, were adapted. Results demonstrated that neither Langmuir model nor Freundlich model can adequately fit the data (results not shown). On the other hand, the Temkin model was in a good agreement with the experimental data, as show below.

3.3. Sorption isotherms In this set of experiments, the CMFC sample C1 (6.50)–C16 (0.43), which exhibited higher sorption capacities (Table 2), was selected for further analysis. Fig. 6 shows the amounts of sodium cholate

=

qe RT In K0 Ce = qmax Q

onto

the

CMFC

sample

(3)

where  is fractional coverage, R is universal gas constant (J/mol K), T is temperature (K), Q = (−H) is variation of the sorption energy (kJ/mol), and K0 is Temkin equilibrium constant (l/mg) (Temkin, 1941). The parameters determined by fitting the data of Fig. 6 in the Temkin model are listed in Table 3. As shown in Fig. 7, the Temkin model can fit into the experimental data. It was proposed that the electrostatic interaction and the hydrophobic interaction would be responsible for the two regions observed. In an early study (Maatar et al., 2013), similar result was

604

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

Table 3 Constant parameters derived from fitting the experimental data of Fig. 6 into isotherm models presented in Eq. (3). CMFC sample (C1 (6.50)–C16 (0.43)) Temkin model H1 (kJ/mol) H2 (kJ/mol) K0 (ml/mg) R2

19.46 3.95 75.70/5.31 0.923/0.953

salt concentration from zero to 0.15 mM. This is because an increase in the ionic strength increases the compression of the electrostatic double layer (Cole, 1932; Zakrajˇsek, 2014; Bobacka & Eklund, 1999), causing a decrease in CMFC’s binding with oppositely charged cholate (Cadotte, Tellier, Blanco, van de Ven, & Paris, 2007). It is noted that the more hydrophobic CMFC sample (C1 (6.50)–C16 (0.43)) is less sensitive to the presence of salt (Fig. 8) than the less hydrophobic CMFC sample (C1 (7.00)). This implies that the sorption due to the hydrophobic interaction between long alkyl R2 substituent and bile acid hydrophobic skeleton becomes the dominant mechanism.

3.5. Comparison of deoxycholate and cholate

Fig. 7. Fitting of the isotherm data of sodium cholate adsorbed onto the CMFC sample (C1 (6.50)–C16 (0.43)) by using the Temkin model.

CMFC sample (C1(6.50)-C16(0.43))

Fraction of bile acids adsorbed (%)

25

CMFC sample (C1(7.00))

20

15

The results on two types of bile acid: deoxycholate and cholate, are given in Fig. 8. For the same CMFC, the binding capacity was higher for deoxycholate than that for cholate. The above may be explained by the increased hydrophobic interaction between the modified CMFC and bile acid, because deoxycholate is significantly more hydrophobic than cholate (Honda & Nakano, 2000). Fig. 8 further shows that CMFC sample (C1 (6.50)–C16 (0.43)) exhibited better sorption for the same bile acid than CMFC sample (C1 (0.70)). This is because CMFC sample (C1 (6.50)–C16 (0.43)) has both the desired charge density and the hydrophobic moieties, both of which have positive contributions toward binding bile acid. In the literature, it was also found that the binding performance of sorbents was higher for more hydrophobic deoxycholate than those for cholate (Honda & Nakano, 2000; Nichifor et al., 2001). For example, the sorption capacities of both Cholestimide and Cholestyramine were lower for cholate than that for deoxycholate (Honda & Nakano, 2000). The above results suggested that the structure of bile acid had direct influence on their sorption onto the modified CMFC. Evidently, deoxycholate, which is more hydrophobic, had a higher sorption onto the modified CMFC samples. In the literature, it was proposed that deoxycholate is the main factor leading to colon cancer (Ota, Tanaka, Bamba, Kato, & Matsuzaki, 1999; McMillan, Butcher, Wallis, Neoptolemos, & Lord, 2000).

4. Conclusions

10

5

(b) Cholate

(a) Deoxycholate

0 0.00

0.05

0.1

0.15

0.00

0.05

0.1

0.15

Concentration of electrolyte added (mM) Fig. 8. Effect of sodium chloride on the sorption of (a) deoxycholate and (b) cholate to different modified CMFC samples in Water at 37 ◦ C.

obtained from surface hydrophobized cellulose fibers as adsorbents for dissolved organic pollutants. Given the chemical structure of the modified CMFC surface, it is reasonable to assume that the former would be due to the quaternary ammonium groups, whereas the later would be due to the apolar hexadecyl chains attached. 3.4. Effect of ionic strength on sorption The binding of cholate with CMFC sample (C1 (7.00)) was primarily based on charge interaction, as shown in Fig. 8, its sorption of cholate was found to decrease somewhat when increasing the

Cationic amphiphilic MFC samples with different contents of quaternary ammonium group and lengths of its alkyl substituent were successfully synthesized and characterized. Subsequently, the in-vitro bile acid sorption onto the thus prepared CMFC was performed in order to systematically investigate the influence of the chemical structure of the modified CMFC for adsorbing bile acid. The sorption capacity was affected by both the charge density (amino content) and the alkyl length of the substituent: (1) increasing the content of quaternary ammonium groups can improve the sorption of bile acid, (2) with the increase in the alkyl group from C4 to C16 , the sorption capacity improved drastically, while a further increase to C18 , only had a minor increase. With a C18 alkyl substituent, the modified CMFC had a bile acid sorption equivalent to about 50% of that of Chelestyramine under the same conditions. A longer alkyl chain of the CMFC substituent will enhance the hydrophobic interaction between the modified CMFC and the steroid skeleton of the cholate anion. Additionally, the experimental isotherm results were well fitted into the Temkin model, which implies that two sorption sites within which the interaction is driven by two different interaction processes. The ionic strength of the bile acid solution exhibited a somewhat influence on bile acid sorption onto the CMFC, but the effect was smaller for the modified CMFC with more hydrophobic alkyl substituent. It was also found that the binding capacity was higher for deoxycholate than that for

X. Zhu et al. / Carbohydrate Polymers 132 (2015) 598–605

cholate, which can be explained by the fact that the former is more hydrophobic. All these results indicate the potential of the cationic amphiphilic MFC to be used as bile acid sorbent. Further in vivo studies are in progress to identify the biological potential of this new material. Acknowledgement The authors wish to acknowledge the financial support from the Tianjin Municipal Science and Technology Commission (Grant no. 12ZCZDGX01100). References Ast, M., & Frishman, W. H. (1990). Bile acid sequestrants. Journal of New Drugs, 30, 99–106. Alila, S., Aloulou, F., Thielemans, W., & Boufi, S. (2011). Sorption potential of modified nanocrystals for the removal of aromatic organic pollutant from aqueous solution. Industrial Crops and Products, 33, 350–357. Alila, S., & Boufi, S. (2009). Removal of organic pollutants from water by modified cellulose fibres. Industrial Crops and Products, 30(1), 93–104. Bobacka, V., & Eklund, D. (1999). The influence of charge density of cationic starch on dissolved and colloidal material from peroxide bleached thermomechanical pulp. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 152, 285–291. Cole, K. S. (1932). A theory of surface conductance at an electrolyte-solid interface. Applied Physics, 3, 114–118. Cadotte, M., Tellier, M. E., Blanco, A., van de Ven, T. G., & Paris, J. (2007). Flocculation, retention and drainage in papermaking: A comparative study of polymeric additives. Canadian Journal of Chemical Engineering, 85, 240–248. Einarsson, K., Ericsson, S., Ewerth, S., Reihner, E., Rudling, M., Ståhlberg, D., et al. (1991). Bile acid sequestrants: Mechanisms of action on bile acid and cholesterol metabolism. European Journal of Clinical Pharmacology, 40(1), S53–S58. Honda, Y., & Nakano, M. (2000). Studies on adsorption characteristics of bile acids and methotrexate to a new type of anion-exchange resin, colestimide. Chemical and Pharmaceutical Bulletin, 48, 978–981. Jahan, M. S., Saeed, A., He, Z. B., & Ni, Y. H. (2011). Jute as raw material for the preparation of microcrystalline cellulose. Cellulose, 18(2), 451–459. Kazlauske, J., Ramanauskiene, K., & Liesiene, J. (2014). Binding of bile acids by cellulose-based cationic adsorbents. Cellulose Chemistry and Technology, 48(1–2), 11–17. Kahlon, T. S., & Smith, G. E. (2007). In vitro binding of bile acids by bananas, peaches, pineapple, grapes, pears, apricots and nectarines. Food Chemistry, 101, 1046–1051. Kacurakova, M., Ebringerova, A., Hirsch, J., & Hromadkova, Z. (1994). Infrared study of arabinoxylans. Journal of the Science of Food and Agriculture, 66(3), 423–427. Kim, H. J., & White, P. J. (2009). In vitro bile-acid binding and fermentation of high, medium, and low molecular weight ␤-glucan. Food Chemistry, 58, 628–634. Kim, H. J., & White, P. J. (2011). Optimizing the molecular weight of oat ␤-glucan for in vitro bile acid binding and fermentation. Food Chemistry, 59, 10322–10328. Liu, Z., Ni, Y., Fatehi, P., & Saeed, A. (2011). Isolation and cationization of hemicelluloses from pre-hydrolysis liquor of kraft-based dissolving pulp production process. Biomass and Bioenergy, 5, 1789–1796. Mason, R. P. (2011). Optimal therapeutic strategy for treating patients with hypertension and atherosclerosis: Focus on olmesartan medoxomil. Vascular Health and Risk Management, 7, 405. Maatar, W., Alila, S., & Boufi, S. (2013). Cellulose based organogel as an adsorbent for dissolved organic compounds. Industrial Crops and Products, 49, 33–42. Mocanu, G., & Nichifor, M. (2014). Cationic amphiphilic dextran hydrogels with potential biomedical applications. Carbohydrate Polymers, 99, 235–241. Moon, R. J., Martini, A., Nairn, J., Simonsen, J., & Youngblood, J. (2011). Cellulose nanomaterials review: Structure, properties and nanocomposites. Chemical Society Reviews, 40, 3941–3994. Morandi, G., Heath, L., & Thielemans, W. (2009). Cellulose nanocrystals grafted with polystyrene chains through surface-initiated atom transfer radical polymerization (SI-ATRP). Langmuir, 25, 8280–8286.

605

McMillan, L., Butcher, S., Wallis, Y., Neoptolemos, J. P., & Lord, J. M. (2000). Bile acids reduce the apoptosis-inducing effects of sodium butyrate on human colon adenoma (AA/C1) cells: Implications for colon carcinogenesis. Biochemical and Biophysical Research Communications, 273(1), 45–49. Nichifor, M., Cristea, D., & Carpov, A. (2000). Sodium cholate sorption on cationic dextran hydrogel microspheres, 1. Influence of the chemical structure of functional groups. International Journal of Biological Macromolecules, 28, 15–21. Nichifor, M., Zhu, X. X., Cristea, D., & Carpov, A. (2001). Interaction of hydrophobically modified cationic dextran hydrogels with biological surfactants. Journal of Physical Chemistry B, 105, 2314–2321. Nichifor, M., Stanciu, M. C., & Simionescu, B. C. (2010). New cationic hydrophilic and amphiphilic polysaccharides synthesized by one pot procedure. Carbohydrate Polymers, 82, 965–975. Ota, S., Tanaka, Y., Bamba, H., Kato, A., & Matsuzaki, F. (1999). Nonsteroidal anti-inflammatory drugs may prevent colon cancer through suppression of hepatocyte growth factor expression. European Journal of Pharmacology, 367, 131–138. Olszewska, A., Eronen, P., Johansson, L. S., Malho, J. M., Ankerfors, M., Lindström, T., et al. (2011). The behaviour of cationic nanofibrillar cellulose in aqueous media. Cellulose, 18, 1213–1226. Pei, A., Butchosa, N., Berglund, L. A., & Zhou, Q. (2013). Surface quaternized cellulose nanofibrils with high water absorbency and adsorption capacity for anionic dyes. Soft Matter, 9, 2047–2055. Pääkkö, M., Ankerfors, M., Kosonen, H., Nykänen, A., Ahola, S., Österberg, M., et al. (2007). Enzymatic hydrolysis combined with mechanical shearing and high-pressure homogenization for nanoscale cellulose fibrils and strong gels. Biomacromolecules, 8, 1934–1941. Ren, J. L., Sun, R. C., Liu, C. F., Chao, Z. Y., & Luo, W. (2006). Two-step preparation and thermal characterization of cationic 2-hydroxypropyltrimethylammonium chloride hemicellulose polymers from sugarcane bagasse. Polymer Degradation and Stability, 91(11), 2579–2587. Ren, J. L., Sun, R. C., Liu, C. F., Lin, L., & He, B. H. (2007). Synthesis and characterization of novel cationic SCB hemicelluloses with a low degree of substitution. Carbohydrate Polymers, 67(3), 347–357. Rub, M. A., Sheikh, M. S., Asiri, A. M., Azum, N., Khan, A., Khan, A. A. P., et al. (2013). Aggregation behaviour of amphiphilic drug and bile salt mixtures at different compositions and temperatures. Journal of Chemical Thermodynamics, 64, 28–39. Schulman, K. A., Kinosian, B., Jacobson, T. A., Glick, H., Willian, M. K., Koffer, H., et al. (1990). Reducing high blood cholesterol level with drugs: Cost-effectiveness of pharmacologic management. Journal of the American Medical Association, 264, 3025–3033. Scaldaferri, F., Pizzoferrato, M., Ponziani, F. R., Gasbarrini, G., & Gasbarrini, A. (2013). Use and indications of cholestyramine and bile acid sequestrants. Internal and Emergency Medicine, 8, 205–210. Sun, B., Hou, Q., He, Z., Liu, Z., & Ni, Y. (2014). Cellulose nanocrystals (CNC) as carriers for a spirooxazine dye and its effect on photochromic efficiency. Carbohydrate Polymers, 111, 419–424. Sun, B., He, Z., Hou, Q., Liu, Z., Cha, R., & Ni, Y. (2013). Interaction of a spirooxazine dye with latex and its photochromic efficiency on cellulosic paper. Carbohydrate Polymers, 95, 598–605. Sun, R., Sun, X. F., Liu, G. Q., Fowler, P., & Tomkinson, J. (2002). Structural and physicochemical characterization of hemicelluloses isolated by alkaline peroxide from barley straw. Polymer International, 51(2), 117–124. Temkin, M. I. (1941). Adsorption equilibrium and the kinetics of processes on non-homogeneous surfaces and in the interaction between adsorbed molecules. Zhurnal Fizicheskoi Khimii, 15, 296–332. ˜ Yanez, F., Chianella, I., Piletsky, S. A., Concheiro, A., & Alvarez-Lorenzo, C. (2010). Computational modeling and molecular imprinting for the development of acrylic polymers with high affinity for bile salts. Analytica Chimica Acta, 659(1), 178–185. Zhu, X., Wen, Y., Wang, L., Li, C., Cheng, D., Zhang, H., et al. (2014). Binding of sodium cholate in vitro by cationic microfibrillated cellulose. Industrial and Engineering Chemistry Research, 53(48), 18508–18513. Zaman, M., Xiao, H., Chibante, F., & Ni, Y. (2012). Synthesis and characterization of cationically modified nanocrystalline cellulose. Carbohydrate Polymers, 89, 163–170. Zakrajˇsek, N. (2014). Influence of inorganic salts on the adsorption of cationically modified starch to fibers. Food and Nutrition Science, 5(3), 245–249.