Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Adsorption of protein onto double layer mixed matrix membranes Junfen Sun a,∗ , Lishun Wu b a State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, College of Material Science and Engineering, Donghua University, North People Road 2999, Shanghai 201620, PR China b Department of Chemistry and Chemical Engineering, Heze University, Daxue Road 2269, Heze, Shandong Province 274015, PR China
a r t i c l e
i n f o
Article history: Received 9 May 2014 Received in revised form 1 September 2014 Accepted 3 September 2014 Available online 16 September 2014 Keywords: Polyether sulfone Hydroxyapatite (HAP) Double layer mixed matrix membranes (MMMs) Bovine serum albumin (BSA) Adsorption
a b s t r a c t This work proposed a novel approach for protein purification by using double layer mixed matrix membranes (MMMs). The double layer MMMs consisting of an active support and separating layer were prepared by co-casting two polymer solutions onto a glass plate. The active support layer consisted of nano hydroxyapatite (HAP) particles embedded in macroporous polyether sulfone (PES) and the separating layer was particle free PES membrane. The influence of separating layer with different PES content on membrane morphology was studied. The double layer MMMs were further characterized concerning permeability and adsorption capacity. The double layer MMMs showed purification of protein via diffusion as well as adsorption. The bovine serum albumin (BSA) was used as a model protein. The properties and structures of double layer MMMs prepared by immersion phase separation process were characterized by pure water flux, BSA adsorption and scanning electron microscopy (SEM). © 2014 Elsevier B.V. All rights reserved.
1. Introduction In the recent years, there has been increasing interest in adsorptive membranes as carriers for macromolecules such as proteins [1], enzymes [2] and viruses [3]. Adsorptive membranes have shown great promise for bioseparations as an alternative to packed bed chromatography. Mixed matrix membranes (MMMs) as a kind of adsorptive membrane, a composition of polymers and active particles, combine the selectivity of the filler material with the low costs, manufacturing ease and flow behavior of polymer membranes [4,5]. The resultant MMMs have excellent separation performance and special properties from particles. To improve protein purification, some researchers prepared MMMs embedding adsorptive particles. Recently some researches combined ion exchange resins and polymer together to make MMMs which were applied in protein capturing, purifying and polishing steps. The MMMs combined the principles of chromatography and membrane filtration in a single separation device. Avramescu prepared ethylene-vinyl alcohol (EVAL)/Lewatit ionexchange resins MMMs to adsorb BSA [6] and separate BSA and bovine hemoglobin (Hb) [7], and prepared PES/Lewatit ionexchange resins adsorber for lysozyme (LZ) separation [8]. Saiful [9] prepared EVAL/ion-exchange resins MMMs to capture LZ. Kopec
∗ Corresponding author. Tel.: +86 18602105973; fax: +86 21 67792855. E-mail address: [email protected] (J. Sun). http://dx.doi.org/10.1016/j.colsurfb.2014.09.006 0927-7765/© 2014 Elsevier B.V. All rights reserved.
[10] prepared solvent-resistant P84-based MMMs to adsorb BSA and LZ. BSA adsorption capacity is 77 mg/g membrane and LZ adsorption capacity is 85.1 mg/g membrane. Saufi prepared some MMMs embedding anion-exchange resins and cation-exchange resins. EVAL/Lewatie anion-exchange resins MMMs [11], EVAL/SP SepharoseTM cation exchange resin MMMs [12] and a novel mixed mode interaction MMMs incorporating 42.5 wt% Lewatit MP500 anionic resin and 7.5 wt% SP Sepharose cationic resin into EVAL [13] were prepared for whey protein fractionation. The MMMs had good static binding capacities for -lactoglobulin, ␣-lactalbumin, BSA and lactoferin in individual protein solutions. Moreover different inorganic nanoparticles such as carbon nanotubes [14,15], titanium dioxide (TiO2 ) [16] and silicon dioxide (SiO2 ) [17] were dispersed in microporous and macroporous polymeric structures to prepare MMMs. Some inorganic nanoparticles such as copper sulfide nanoparticles [18], NiO nanoparticles [19], gold nanoparticles [20] and silver nanoparticles [21] process adsorption ability and were applied in the adsorptive field. Inorganic nanoparticles have some potential advantages over other polymeric nanoparticles because of their low susceptibility to immune response and low toxicity. Among the inorganic nanoparticles, hydroxyapatite (HAP, Ca10 (PO4 )6 ·(OH)2 ) has attracted much attention as a carrier for biomolecules because of its excellent biocompatibility and bioactivity. It has good adsorption to protein and is mainly used for repairing bone tissue and culturing sclerotin in the medical field because of its biocompatibility, bioactivity and osteoconductivity [22,23]. HAP has been used in adsorption
34
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
chromatography for many years, and widely applied for separating various proteins as a column in a high performance liquid chromatograph apparatus [24]. It was known that HAP has two different binding sites, the C and P sites on its surface respectively, which can provide proteins a multiple site binding opportunity [25]. The C sites are rich in calcium ions or positive charge and bind to acidic groups of proteins, and the P sites lack calcium ions or positive charge and attach to basic groups of proteins [26]. A lot of researches tried different ways to synthesis HAP with high affinity for protein. Mohandes synthesized nano HAP with controllable morphologies by the aid of various Schiff bases [27–29]. Liu [30] prepared calcium-deficient HAP for controlled drug delivery. Kandori [31] prepared positively charged calcium HAP which had high BSA adsorption. Dasgupta synthesized Zn and Mg doped HAP nanoparticles for controlled release of protein [32]. Moreover HAP has selective adsorption property to some protein by controlling preparation methods. Ozeki [33] prepared rod-shaped and plate-shaped HAP which selectively adsorbed BSA and lysozyme (LSZ), and BSA/LSZ adsorption ratio depended on the crystal shape. Kandori [34] found the sheet-like HAP particles could be applied to separate completely BSA from BSA/LSZ mixed solution. Fujii [35] prepared nano-crystalline Zn-containing HAP (ZnHAP) which had selective adsorption to pathogenic protein such as 2 -microglobulin (2 -MG) in the BSA/2 -MG mixed solution. Takemoto [36] synthesized hydroxyl-carbonate apatite which had higher selectivity for 2 -MG adsorption in the BSA/2 MG mixed solution. Some US patens reported HAP was used in polymer matrix. HAP was embedded in polytetrafluoroethylene (PTFE) matrix for absorption [37]. HAP/poly(etheretherketone) nanocomposites was prepared for a wide variety of applications, such as biological, medical, biochemical, biosensor, fuel cell, and aerospace applications [38]. HAP-targeting poly(ethylene glycol) having biologically active conjugates was provided [39]. The chitosan/graphene oxide/HAP nanocomposite with high bioactivity was prepared via a simple precipitation method with the aid of a new capping agent based on Schiff base compounds [40]. Previously, we prepared single layer PES/HAP MMMs which showed high BSA adsorption capacity and desorption rate [41]. In this study, the novel double layer MMMs combining diffusion and adsorption of BSA retention solutes in one step were prepared. The double layer MMMs consisting of an active support and separating layer were prepared by co-casting two polymer solutions onto a glass plate. The active support layer consists of HAP particles which are embedded in macroporous polyether sulfone (PES), and the separating layer is particle free PES membrane. The co-casting process opens the possibility to improve the mechanical stability and the biocompatibility of double layer MMMs while prevent particle loss during preparation and processing. Tijink prepared dual-layer MMMs embedded with activated carbon (AC) particles in flat membranes [42] and hollow fiber membranes [43]. The duallayer MMMs had a higher clean water flux (350 l m−2 h−1 bar−1 ) and higher creatinine adsorption (29 mg g−1 AC). Compared to single layer MMMs, the double layer MMMs prevent HAP particles from releasing into the circulation. The top particle free layer is important for the application of MMMs, especially for the application in the field of blood purification [42,43]. In this work, double layer MMMs were prepared by an immersion phase separation process. PES was used for the preparation of porous membrane matrix and top particle free layer for double layer MMMs. HAP particles were used as adsorptive particles and were incorporated into a porous PES matrix with a high particle loading. Bovine serum albumin (BSA, molecular weight 67, 000 Da, size 4 nm × 14 nm) was used as a model protein. This study investigated the combination of diffusion and adsorption in a single step, which probably leads to more efficient protein purification devices.
Fig. 1. (A) Schematic representation of the co-casting process. (B) Resulting membrane structure comprising a porous surface layer and a layer containing HAP particles.
2. Experiment 2.1. Materials Polyether sulfone (PES) (Mw = 58,000) and polyvinylpyrrolidone (PVP, K90) were produced by BASF Company (Germany). Hydroxyapatite (HAP, d = 40 nm) was purchased from Nanjing Emperor Nano Material Company (China). Dimethyl acetamide (DMAc) and bovine serum albumin (BSA, Mw = 67,000) were supplied by China Medicine Chemical Reagent Company (China).
2.2. Membrane preparation The membranes used in this study were prepared by the phase inversion method. PES, PVP and DMAc were mixed and heated until homogeneous mixed solutions with various compositions were obtained. The concentration of PES of PVP in DMAc was 14% and 5%, respectively. The HAP particles were added into the casting solution and were dispersed in the casting solution for 24 h in order to make HAP particles have good dispersion in the mixed solution and improve membrane performance. The amount of HAP was 60 wt% in dry PES/HAP MMMs. A slit of 300 m of casting knife for single layer MMMs was used. PES concentration for top layer of double layer MMMs was 5%, 10% and 15%, respectively. Fig. 1 shows schematic representation of the co-casting process. The heights of the slits of the first and second knife were 300 and 400 m, respectively. The polymeric mixture incorporating HAP were cast on a glass plate and immersed into 60% DMAc aqueous solution. The single layer PES/HAP MMMs formed a few moments after immersion. The pure polymer dope and polymeric mixtures incorporating HAP were co-cast on a glass plate and immersed into 60% DMAc aqueous solution and double layer MMMs formed. The MMMs were washed with tap water at room temperature to remove residual solvent.
2.3. Membrane characterization 2.3.1. Scanning electron microscopy For scanning electron microscopy (SEM), membranes were dried in air at room temperature and cryogenically broken in liquid nitrogen. The obtained cross-sections were dried overnight under vacuum at 30 ◦ C and gold coated. The cross sections, as well as the top and bottom surfaces of the membrane were characterized by scanning electron microscopy (SEM, JSM-5600LV, JEOL, Japan).
2.3.2. Pure water flux The membranes were subjected to pure water flux estimation at a trans-membrane pressure of 0.1 MPa under cross-flow
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
35
Fig. 2. Cross section of double layer PES/HAP MMMs with different PES top layers. PES content of top layer A, B, C is 5%, 10%, 15%. Cross section of top layer: A1, B1, C1; whole cross section: A2, B2, C2; cross section of bottom layer: A3, B3, C3.
filtration. The permeability was measured under steady-state flow. Pure water flux was calculated as follows: Jw =
Q At
where Q is the quantity of permeate collected (in l), and A is membrane area (m2 ), t is the sampling time (h), Jw is pure water flux (l m−2 h−1 ). 2.3.3. BSA adsorption capacity The bovine serum albumin (BSA) protein with molecular weight of 67 kDa and isoelectric point ∼4.74 was used for the protein adsorption study. BSA adsorption was investigated within a range of pH from 5 (acetate buffer) and 6–9 (phosphate buffer). The static protein adsorption capacity of membranes was determined with bovine serum albumin (BSA). The membranes were dried at 30 ◦ C in a vacuum oven before examination. The samples containing 2 g/l BSA were incubated with an exact amount of membranes in sealed containers under continuous shaking at 25 ◦ C. The PES/HAP MMM adsorbed the BSA thereby reducing the BSA concentration in the bulk. The equilibrium BSA concentration after 24 h was monitored in time with a UV-1800 spectrophotometer which was produced by SHIMADZU Company. The BSA depletion was measured at 280 nm with 5 mm quartz cuvettes. 2.3.4. BSA desorption capacity The MMMs were transferred into the desorption buffer after an adsorption and washing step. Desorption was accomplished in static by using a phosphate buffer of pH = 7, I = 0.5 M NaCl for static experiment. The desorption was carried out for 24 h in a shaking
bath at 25 ◦ C. The BSA desorption capacity was defined as the amount desorbed BSA from MMMs which had adsorbed BSA. 2.3.5. Two-compartment diffusion test A two-compartment diffusion device [41] was used to measure diffusion and adsorption of BSA onto MMMs at room temperature. The first compartment was filled with BSA buffer solution (2 mg/ml) and the second compartment was filled with water. The compartments were separated by one piece of MMM or two pieces of MMMs. The volume of each compartment was 100 ml and the active membrane area was 7.11 cm2 . Both solutions were stirred by magnetic rotors under 200 rpm. During the experiment, the samples were taken every 1 h to determine the BSA concentration in both compartments by UV-VIS spectrophotometer (UV-1800, SHIMADZU, Japan). The diffusion experiments were evaluated for 24 h before the process were stopped. 3. Results and discussion 3.1. Characterization of double layer mixed matrix membrane Fig. 2 shows the cross section of double layer PES/HAP MMMs with different PES top layers. The PES content of top layer is 5%, 10% and 15%, respectively. The left side of SEM picture is top layer, as shown in Fig. 2. The top PES layer is hard to form when PES content of top layer is 5% because of less PES in top layer. The top PES layer is inclined finger-like pores when PES content of top layer is 10%, and the top PES layer is straight finger-like pores when PES content of top layer is 15%. The formation of finger-like
36
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
Fig. 3. SEM pictures of top surface of single layer MMMs (A) and double layer PES/HAP MMMs with 10% PES as top layer (B).
3.2. Pure water flux and protein adsorption of double layer mixed matrix membrane 3.2.1. PES/HAP double layer MMMs Table 1 shows pure water fluxes and BSA adsorption capacities of single layer MMMs and double layer MMMs with different PES top layer. It can be seen from Table 1 that pure water flux of double layer MMMs decreases gradually with increasing PES solution concentration from 5 to 15% because top PES layer of double layer MMMs gradually become denser. The pure water flux of double layer MMMs with top layer with 15% PES (116.7 ± 3.0 l m−2 h−1 ) decreases a lot in comparison to that of single layer MMMs (217.7 ± 7.2 l m−2 h−1 ). As shown in Table 1, single layer MMMs keeps good BSA adsorption capacity (26.0 ± 2.1 mg/g membrane and 43.3 ± 2.1 mg/g HAP) compared to pure HAP particles (48.2 ± 3.2 mg/g HAP). BSA adsorption capacity of double layer MMMs decreases gradually with increasing PES concentration. BSA adsorption capacity appears a little decrease when PES concentration of top layer is 5% and 10%, which means that double layer MMMs keep good BSA adsorption capacity when top PES layer is prepared with low concentration of PES dope. BSA adsorption capacity appears abrupt decrease when PES concentration of top layer is 15% and BSA adsorption rate of HAP in double layer MMMs decreases sharply to 59.1%, which means
denser top PES layer inhibits adsorption efficiency of HAP in double layer MMMs. Fig. 4 shows BSA static adsorption capacity of HAP in double layer PES/HAP MMMs and double layer PES/HAP MMMs at different equilibrium time. The top PES layer of double layer MMMs was prepared with 10% PES. The BSA adsorption capacities of HAP in double layer PES/HAP MMMs and double layer PES/HAP MMMs increase in time during 10 h then equilibrium is reached, as shown in Fig. 4. The obtained equilibrium capacity after 24 h is 21.8 mg BSA/g membrane which is equal to 39.5 mg BSA/g HAP in the double layer MMMs. The BSA adsorption lines of HAP and double layer MMMs have similar tendency, which means HAP keeps good adsorption performance in the double layer MMMs and the PES top layer does not inhibit BSA from freely transporting in the matrix of double layer MMMs. 3.2.2. Effect of pH value on BSA adsorption The adsorption of BSA on the pure HAP particles, HAP in double layer MMMs and double layer PES/HAP MMMs as a function of pH (pH = 5–9) are shown in Fig. 5. The top PES layer of double layer MMMs was prepared with 10% PES. The BSA adsorption capacities of pure HAP particles, HAP in double layer MMMs and double layer MMMs increase with increasing pH value of solutions and reach the maximum at pH 7. Compared to the BSA adsorption capacity of pure HAP particles, the BSA adsorption capacity of HAP in double layer MMMs decreases a little and the maximum value decreases from 54.7 mg/g HAP to 41.3 mg/g HAP at pH 7 after HAP is embedded in double layer MMMs. BSA adsorption capacities of HAP in double layer MMMs (41.3 mg/g HAP) decrease slightly compared to that 50
Double layer MMMs HAP in MMMs Adsorbed BSA (mg/g)
pores means that the top PES solution starts to demix instantaneously [44]. With increasing PES content, the demixing delays and a thicker sublayer forms, which affect the formation and size of pores inside membrane. It also means that the phase separation speed decreases and PES molecular chains have more time to relax with increasing PES content of top layer, which results in the formation of different finger-like pores when different PES content is used. Moreover the thickness of top PES layer of double layer MMMs increases with increasing PES concentration for top PES layer. The filtration resistance of double layer MMMs increases with increasing PES concentration for top PES layer. So it is better to choose low PES concentration for top PES layer. The thickness of top layer prepared with 5% PES is very thin because of low concentration of casting solution, which can lead to more macropores in PES base and easily results in the leakage of HAP particles from MMMs. The thickness of top layer prepared with 10% PES concentration is better for double layer MMMs. Fig. 3 shows the surface structure of single layer PES/HAP MMMs and double layer PES/HAP MMMs. The top PES layer of double layer MMMs was prepared with 10% PES. Fig. 3A shows that HAP particles uniformly distribute at the surface of single layer MMMs and the surface of single layer MMMs is rough. Whereas the surface of double layer MMMs (Fig. 3B) is smooth. It is suggested that the top PES layer covers double layer MMMs completely.
40
30
20
10
0
0
5
10
15
20
25
Time (hour) Fig. 4. BSA static adsorption of double layer PES/HAP MMMs (n = 3) at different equilibrium time. Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%).
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
37
Table 1 Pure water flux, adsorption to BSA of single layer MMMs and double layer PES/HAP MMM (n = 3) (PES content of top layer for double layer MMMs is 10%). PES/%
Pure water flux (l m−2 h−1 )
0 5 10 15
217.7 192.4 151.4 116.7
± ± ± ±
Adsorbed BSA (mg/g HAP)
Adsorbed BSA (mg/g membrane)
7.2 5.4 4.6 3.0
26.0 24.5 21.8 15.1
± ± ± ±
2.1 2.8 2.4 1.4
43.3 42.4 39.5 28.5
± ± ± ±
BSA adsorption rate/%
2.1 2.8 2.4 1.4
89.8 88.0 82.0 59.1
*Adsorbed BSA of pure HAP particles is 48.2 ± 3.2 mg/g HAP [41]. SD indicates standard deviations.
in single layer MMMs (49.8 mg/g HAP), as shown in previous work [41]. It is suggested that HAP particles keep good BSA adsorption efficiency in PES/HAP double layer MMMs at different pH value. BSA adsorption capacity of double layer MMMs reaches the maximum (22.8 mg/g membrane) at pH 7. It was reported that BSA as a “soft protein” has lower structural stability and is adsorbed even under unfavorable conditions of a hydrophilic and electrostatically repelling surface by a structural rearrangement with a higher adsorption affinity [45,46]. The isoelectric point of BSA is around 4.8 [47]. When pH value is lower than 4.8, BSA carries positive charges and BSA adsorption is enhanced due to the electrostatic affinity interaction between HAP with negative charges and BSA with positive charges. When pH value is higher than 4.8, BSA carries negative charges. The pH used in this study is from 5 to 9, higher than the isoelectric point of BSA. The electrostatic interactions are dependent on the surface charges between the negatively charged COOH group of BSA and positively charged calcium ion of HAP. The electrostatic interactions generating between BSA and HAP is the maximum at pH 7, which makes the adsorbed amount of BSA reach the maximum. 3.2.3. BSA desorption of HAP in double layer MMMs Fig. 6 shows BSA desorption capacity of HAP in double layer PES/HAP MMMs at different time. The top PES layer of double layer MMMs was prepared with 10% PES. The BSA desorption capacity increase quickly within the first 7 h and BSA desorption capacity reaches 19.9 mg/g HAP (50.4% BSA release form HAP), which may attribute to the BSA weakly adsorbed on the outer surface of HAP. The BSA desorption capacity increases slowly with extending time after 7 h and BSA desorption capacity reaches 34.5 mg/g HAP (87.3% BSA release form HAP) after 24 h, as shown in Fig. 6. The BSA desorption capacity reached 41.5 mg/g HAP and 94.4% BSA released from HAP for single layer MMMs, as shown in previous work [41]. It means that PES top layer of double layer MMMs prevents adsorbed BSA from releasing from HAP quickly. Not 100% BSA release from HAP in MMMs is maybe because some BSA is adsorbed on the inner surface of pores in HAP. It takes more time for BSA adsorbed on the
inner surface of pores in HAP to release. In Boonsongrit’s study [48], approximately 80% BSA released from HAP after 24 h in physiological buffer solution and 10 mM phosphate buffer solution. In Wu’s study [49], 75–98.6% BSA released from HAP with different pore volume and pore size. In Sarkar’s group, BSA adsorption/release on HAP nanoparticles [50] and HAP-gelatin nanobiocomposite [51] was studied. Rod HAP nanoparticles exhibited relatively higher BSA adsorption capacity (28 mg/g) compared to the counterpart spherical and fibroid nanoparticles and around 75% BSA released within 96 h. The adsorption capacity of HAP-gelatin nanobiocomposite was 51.3 mg/g and 60% BSA released within 24 h. The specificity of HAP–protein interactions in MMMs provides a potential application in the field of protein delivery due to adsorption–desorption activity of HAP and proteins. 3.2.4. BSA diffusion through double layer MMMs For estimating the transport of BSA retention solutes through MMMs, we tested diffusion through and adsorption onto double layer MMMs using a two-compartment diffusion device with a 2 mg/ml BSA feed solution. Fig. 7 shows the relative contribution of adsorption and diffusion to the total removal of BSA for double layer PES/HAP MMMs. The top PES layer of double layer MMMs is prepared with 10% PES. The BSA removal by adsorption contributes to far more than that by diffusion and contributes 62.6% of the total removal after 24 h when double layer MMMs were used during BSA diffusion test, as shown in Fig. 7. Double layer MMMs have more BSA adsorption rate in the total BSA removal compared to single layer MMMs (BSA adsorption rate in the total BSA removal is 38.4% when one piece of single layer MMMs was used), as shown in previous work [41]. It is suggested that the resistance of BSA diffusing across double layer MMMs increases when double layer MMMs are used. More BSA is absorbed by double layer MMMs, not by diffusion because top PES layer makes MMMs surface denser and limit more BSA to diffuse through double layer MMMs. BSA stays in MMMs for longer time due to the resistance of BSA diffusing across MMMs increasing, which makes HAP particles in MMM have enough time to capture BSA. The MMMs combine BSA retention solutes removal
60
40
Desorption of BSA (mg/g HAP)
Adsorbed BSA (mg/g)
50 40 30 20
Pure HAP particles HAP in double layer MMMs PES/HAP double layer MMMs
10 0
4
5
6
7
8
9
10
pH value Fig. 5. BSA adsorption of pure HAP particles, HAP in double layer MMMs, double layer PES/HAP double layer MMMs (n = 3) at different pH value. Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%).
30
20
10
0
0
5
10
15
20
25
Time (h) Fig. 6. BSA desorption of HAP in double layer PES/HAP MMMs (n = 3) at different time. Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%).
38
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
Removel (mg/g membrane)
40
Total removel Diffusion Adsorption
30
20
10
0
0
5
10
15
20
25
Time (hour) Fig. 7. BSA total removal, diffusion and adsorption plotted vs. time (n = 3). Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%; one piece of double layer PES/HAP MMM was used).
via both diffusion as well as adsorption in one single step. The top PES layer of double layer MMMs works as an isolation layer to prevent HAP from contacting protein solution directly and at the same time allows HAP in the matrix to adsorb protein. 4. Conclusions In this study, the double layer PES/HAP mixed matrix membranes (MMMs) were prepared by an immersion phase separation process. A particle free PES membrane layer was co-cast with the single layer MMMs which led to formation of double layer MMMs with good transport capacity and BSA adsorption capacity. The BSA adsorption capacity of double layer MMMs reaches the maximum at pH 7. 87.3% BSA releases form HAP after 24 h during desorption process. Double layer MMMs have high BSA removal rates and high BSA adsorption rate during the process of BSA diffusion through double layer MMMs. The double layer MMMs with top PES layer keep good filtration, adsorption and desorption properties. Acknowledgements The authors thank National Natural Science Foundation of China (51203020); Scientific Research Starting Foundation for Returned Overseas Chinese Scholars, Ministry of Education of China; Donghua University central university scientific research special fund (2232012D3-32). References [1] X. Su, Y. Tian, W. Zuo, J. Zhang, H. Li, X. Pan, Water Res. 50 (2014) 267. [2] K. Plate, S. Beutel, H. Buchholz, W. Demmer, S. Fischer-Frühholz, O. Reif, R. Ulber, T. Scheper, J. Chromatogr. A 1117 (2006) 81. [3] T. Muschin, S. Han, T. Kanamoto, H. Nakashima, T. Yoshida, J. Polym. Sci. Polym. Chem. 49 (2011) 3241.
[4] S.A. Hashemifard, A.F. Ismail, T. Matsuura, J. Colloids Interface Sci. 359 (2011) 359. [5] D.Q. Vu, W.J. Koros, S.J. Miller, J. Membr. Sci. 211 (2003) 311. [6] M.E. Avramescu, Z. Borneman, M. Wessling, J. Membr. Sci. 218 (2003) 219. [7] M.E. Avramescu, Z. Borneman, M. Wessling, J. Chromatogr. A 1006 (2003) 171. [8] M.E. Avramescu, Z. Borneman, M. Wessling, J. Membr. Sci. 322 (2008) 306. [9] S. Saiful, Z. Borneman, M. Wessling, J. Membr. Sci. 280 (2006) 406. [10] K.K. Kopec, S.M. Dutczak, L. Bolhuis-Versteeg, M. Wessling, D.F. Stamatialis, Sep. Purif. Technol. 80 (2011) 306. [11] S.M. Saufi, C.J. Fee, J. Membr. Sci. 444 (2013) 157. [12] S.M. Saufi, C.J. Fee, Sep. Purif. Technol. 77 (2011) 68. [13] S.M. Saufi, C.J. Fee, J. Chromatogr. A 1218 (2011) 9003. [14] M.M. Khan, V. Filiz, G. Bengtson, S. Shishatskiy, M.M. Rahman, J. Lillepaerg, V. Abetz, J. Membr. Sci. 436 (2013) 109. [15] J. Yin, G. Zhu, B. Deng, J. Membr. Sci. 437 (2013) 237. [16] J.H. Li, Y.Y. Xu, L.P. Zhu, J.H. Wang, C.H. Du, J. Membr. Sci. 326 (2009) 659. [17] S. Yu, X. Zuo, R. Bao, X. Xu, J. Wang, J. Xu, Polymer 50 (2009) 553. [18] S. Khodadoust, M. Ghaedi, R. Sahraei, A. Daneshfar, J. Ind. Eng. Chem. 20 (2014) 2663. [19] F. Motahari, M.R. Mozdianfard, M. Salavati-Niasari, Process Saf. Environ. (2014), http://dx.doi.org/10.1016/j.psep.2014.06.006. [20] A.T.N. Lam, J. Yoon, E.-O. Ganbold, D.K. Singh, D. Kim, K.-H. Cho, S.J. Son, J. Choo, S.Y. Lee, S. Kim, S.-W. Joo, J. Colloids Interface Sci. 425 (2014) 96. [21] A. Jaworska, K. Malek, J. Colloids Interface Sci. 431 (2014) 117. [22] F. Vázquez-Hernández, C. Mendoza-Barrer, V. Altuzar, Mater. Sci. Eng. B 174 (2010) 290. [23] M. Espanol, I. Casals, S. Lamtahri, M.T. Valderas, M.P. Ginebra, Biointerphases 7 (2012) 37. [24] J.M. Thomann, M.J. Mura, M.S. Behr, J.D. Aptel, A. Schmitt, E.F. Bres, J.C. Voegel, Colloids Surf. 40 (1989) 293. [25] T. Kawasaki, S. Takahashi, K. Ideda, Eur. J. Biochem. 152 (1985) 361. [26] K. Kandori, A. Masunari, T. Ishikawa, Calcified Tissue Int. 76 (2005) 194. [27] F. Mohandes, M. Salavati-Niasari, Chem. Eng. J. 252 (2014) 173. [28] F. Mohandes, M. Salavati-Niasari, Mater. Sci. Eng. C 40 (2014) 288. [29] F. Mohandes, M. Salavati-Niasari, Z. Fereshteh, M. Fathi, Ceram. Int. 40 (2014) 1222–1227. [30] T.Y. Liu, S.Y. Chen, D.M. Liu, S.C. Liou, J. Control. Release 107 (2005) 112. [31] K. Kandori, S. Oda, M. Fukusumi, Y. Morisada, Colloids Surf. B: Biointerfaces 73 (2009) 140. [32] S. Dasgupta, A. Bandyopadhyay, S. Bose, Langmuir 26 (2010) 4958. [33] K. Ozeki, T. Hoshino, H. Aoki, T. Masuzawa, Biomed. Mater. Eng. 23 (2013) 239. [34] K. Kandori, H. Hamazaki, M. Matsuzawa, S. Togashi, Adv. Powder Technol. 25 (2014) 354. [35] E. Fujii, M. Ohkubo, K. Tsuru, S. Hayakawa, A. Osaka, K. Kawabata, C. Bonhomme, F. Babonneau, Acta Biomater. 2 (2006) 69. [36] S. Takemoto, Y. Kusudo, K. Tsuru, S. Hayakawa, A. Osaka, S. Takashima, J. Biomed. Mater. Res. A 69 (2004) 544. [37] D.F. Hagen, S.J. St. Mary, L.A. Errede, US patent 4,906,378. [38] K.K. Kar, S. Pramanik, US patent 8,652,373. [39] A. Kozlowski, M.J. Roberts, US patent 8,372,422. [40] F. Mohandes, M. Salavati-Niasari, RSC Adv. 4 (2014) 25993. [41] J. Sun, L. Wu, Appl. Surf. Sci. 308 (2014) 155. [42] M.S.L. Tijink, M. Wester, J. Sun, A. Saris, L.A.M. Bolhuis-Versteeg, S. Saiful, J.A. Joles, Z. Borneman, M. Wessling, D.F. Stamatialis, Acta Biomater. 8 (2012) 2279. [43] M.S.L. Tijink, M. Wester, G. Glorieux, K.G.F. Gerritsen, J. Sun, P.C. Swart, Z. Borneman, M. Wessling, R. Vanholder, J.A. Joles, D.F. Stamatialis, Biomaterials 34 (2013) 7819. [44] C.A. Smolders, A.J. Reuvers, R.M. Boom, I.M. Wienk, J. Membr. Sci. 73 (1992) 259. [45] W. Norde, C.I. Anuisem, Colloids Surf. 66 (1992) 73. [46] W. Norde, T. Arai, H. Shirahama, Biofouling 4 (1991) 37. [47] T.P.B. Nguyen, J.W. Lee, W.G. Shim, H. Moon, Microporous Mesoporous Mater. 110 (2008) 560. [48] Y. Boonsongrit, H. Abe, K. Sato, M. Naito, M. Yoshimura, H. Ichikawa, Y. Fukumori, Mater. Sci. Eng. Solid-State Mater. Adv. Tech. 148 (2008) 162. [49] X. Wu, X. Song, D. Li, J. Liu, P. Zhang, X. Chen, J. Bionic Eng. 9 (2012) 224. [50] S. Swain, D. Sarkar, Appl. Surf. Sci. 286 (2013) 99. [51] B. Basu, S.K. Swain, D. Sarkar, RSC Adv. 3 (2013) 14622.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Adsorption of protein onto double layer mixed matrix membranes Junfen Sun a,∗ , Lishun Wu b a State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, College of Material Science and Engineering, Donghua University, North People Road 2999, Shanghai 201620, PR China b Department of Chemistry and Chemical Engineering, Heze University, Daxue Road 2269, Heze, Shandong Province 274015, PR China
a r t i c l e
i n f o
Article history: Received 9 May 2014 Received in revised form 1 September 2014 Accepted 3 September 2014 Available online 16 September 2014 Keywords: Polyether sulfone Hydroxyapatite (HAP) Double layer mixed matrix membranes (MMMs) Bovine serum albumin (BSA) Adsorption
a b s t r a c t This work proposed a novel approach for protein purification by using double layer mixed matrix membranes (MMMs). The double layer MMMs consisting of an active support and separating layer were prepared by co-casting two polymer solutions onto a glass plate. The active support layer consisted of nano hydroxyapatite (HAP) particles embedded in macroporous polyether sulfone (PES) and the separating layer was particle free PES membrane. The influence of separating layer with different PES content on membrane morphology was studied. The double layer MMMs were further characterized concerning permeability and adsorption capacity. The double layer MMMs showed purification of protein via diffusion as well as adsorption. The bovine serum albumin (BSA) was used as a model protein. The properties and structures of double layer MMMs prepared by immersion phase separation process were characterized by pure water flux, BSA adsorption and scanning electron microscopy (SEM). © 2014 Elsevier B.V. All rights reserved.
1. Introduction In the recent years, there has been increasing interest in adsorptive membranes as carriers for macromolecules such as proteins [1], enzymes [2] and viruses [3]. Adsorptive membranes have shown great promise for bioseparations as an alternative to packed bed chromatography. Mixed matrix membranes (MMMs) as a kind of adsorptive membrane, a composition of polymers and active particles, combine the selectivity of the filler material with the low costs, manufacturing ease and flow behavior of polymer membranes [4,5]. The resultant MMMs have excellent separation performance and special properties from particles. To improve protein purification, some researchers prepared MMMs embedding adsorptive particles. Recently some researches combined ion exchange resins and polymer together to make MMMs which were applied in protein capturing, purifying and polishing steps. The MMMs combined the principles of chromatography and membrane filtration in a single separation device. Avramescu prepared ethylene-vinyl alcohol (EVAL)/Lewatit ionexchange resins MMMs to adsorb BSA [6] and separate BSA and bovine hemoglobin (Hb) [7], and prepared PES/Lewatit ionexchange resins adsorber for lysozyme (LZ) separation [8]. Saiful [9] prepared EVAL/ion-exchange resins MMMs to capture LZ. Kopec
∗ Corresponding author. Tel.: +86 18602105973; fax: +86 21 67792855. E-mail address: [email protected] (J. Sun). http://dx.doi.org/10.1016/j.colsurfb.2014.09.006 0927-7765/© 2014 Elsevier B.V. All rights reserved.
[10] prepared solvent-resistant P84-based MMMs to adsorb BSA and LZ. BSA adsorption capacity is 77 mg/g membrane and LZ adsorption capacity is 85.1 mg/g membrane. Saufi prepared some MMMs embedding anion-exchange resins and cation-exchange resins. EVAL/Lewatie anion-exchange resins MMMs [11], EVAL/SP SepharoseTM cation exchange resin MMMs [12] and a novel mixed mode interaction MMMs incorporating 42.5 wt% Lewatit MP500 anionic resin and 7.5 wt% SP Sepharose cationic resin into EVAL [13] were prepared for whey protein fractionation. The MMMs had good static binding capacities for -lactoglobulin, ␣-lactalbumin, BSA and lactoferin in individual protein solutions. Moreover different inorganic nanoparticles such as carbon nanotubes [14,15], titanium dioxide (TiO2 ) [16] and silicon dioxide (SiO2 ) [17] were dispersed in microporous and macroporous polymeric structures to prepare MMMs. Some inorganic nanoparticles such as copper sulfide nanoparticles [18], NiO nanoparticles [19], gold nanoparticles [20] and silver nanoparticles [21] process adsorption ability and were applied in the adsorptive field. Inorganic nanoparticles have some potential advantages over other polymeric nanoparticles because of their low susceptibility to immune response and low toxicity. Among the inorganic nanoparticles, hydroxyapatite (HAP, Ca10 (PO4 )6 ·(OH)2 ) has attracted much attention as a carrier for biomolecules because of its excellent biocompatibility and bioactivity. It has good adsorption to protein and is mainly used for repairing bone tissue and culturing sclerotin in the medical field because of its biocompatibility, bioactivity and osteoconductivity [22,23]. HAP has been used in adsorption
34
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
chromatography for many years, and widely applied for separating various proteins as a column in a high performance liquid chromatograph apparatus [24]. It was known that HAP has two different binding sites, the C and P sites on its surface respectively, which can provide proteins a multiple site binding opportunity [25]. The C sites are rich in calcium ions or positive charge and bind to acidic groups of proteins, and the P sites lack calcium ions or positive charge and attach to basic groups of proteins [26]. A lot of researches tried different ways to synthesis HAP with high affinity for protein. Mohandes synthesized nano HAP with controllable morphologies by the aid of various Schiff bases [27–29]. Liu [30] prepared calcium-deficient HAP for controlled drug delivery. Kandori [31] prepared positively charged calcium HAP which had high BSA adsorption. Dasgupta synthesized Zn and Mg doped HAP nanoparticles for controlled release of protein [32]. Moreover HAP has selective adsorption property to some protein by controlling preparation methods. Ozeki [33] prepared rod-shaped and plate-shaped HAP which selectively adsorbed BSA and lysozyme (LSZ), and BSA/LSZ adsorption ratio depended on the crystal shape. Kandori [34] found the sheet-like HAP particles could be applied to separate completely BSA from BSA/LSZ mixed solution. Fujii [35] prepared nano-crystalline Zn-containing HAP (ZnHAP) which had selective adsorption to pathogenic protein such as 2 -microglobulin (2 -MG) in the BSA/2 -MG mixed solution. Takemoto [36] synthesized hydroxyl-carbonate apatite which had higher selectivity for 2 -MG adsorption in the BSA/2 MG mixed solution. Some US patens reported HAP was used in polymer matrix. HAP was embedded in polytetrafluoroethylene (PTFE) matrix for absorption [37]. HAP/poly(etheretherketone) nanocomposites was prepared for a wide variety of applications, such as biological, medical, biochemical, biosensor, fuel cell, and aerospace applications [38]. HAP-targeting poly(ethylene glycol) having biologically active conjugates was provided [39]. The chitosan/graphene oxide/HAP nanocomposite with high bioactivity was prepared via a simple precipitation method with the aid of a new capping agent based on Schiff base compounds [40]. Previously, we prepared single layer PES/HAP MMMs which showed high BSA adsorption capacity and desorption rate [41]. In this study, the novel double layer MMMs combining diffusion and adsorption of BSA retention solutes in one step were prepared. The double layer MMMs consisting of an active support and separating layer were prepared by co-casting two polymer solutions onto a glass plate. The active support layer consists of HAP particles which are embedded in macroporous polyether sulfone (PES), and the separating layer is particle free PES membrane. The co-casting process opens the possibility to improve the mechanical stability and the biocompatibility of double layer MMMs while prevent particle loss during preparation and processing. Tijink prepared dual-layer MMMs embedded with activated carbon (AC) particles in flat membranes [42] and hollow fiber membranes [43]. The duallayer MMMs had a higher clean water flux (350 l m−2 h−1 bar−1 ) and higher creatinine adsorption (29 mg g−1 AC). Compared to single layer MMMs, the double layer MMMs prevent HAP particles from releasing into the circulation. The top particle free layer is important for the application of MMMs, especially for the application in the field of blood purification [42,43]. In this work, double layer MMMs were prepared by an immersion phase separation process. PES was used for the preparation of porous membrane matrix and top particle free layer for double layer MMMs. HAP particles were used as adsorptive particles and were incorporated into a porous PES matrix with a high particle loading. Bovine serum albumin (BSA, molecular weight 67, 000 Da, size 4 nm × 14 nm) was used as a model protein. This study investigated the combination of diffusion and adsorption in a single step, which probably leads to more efficient protein purification devices.
Fig. 1. (A) Schematic representation of the co-casting process. (B) Resulting membrane structure comprising a porous surface layer and a layer containing HAP particles.
2. Experiment 2.1. Materials Polyether sulfone (PES) (Mw = 58,000) and polyvinylpyrrolidone (PVP, K90) were produced by BASF Company (Germany). Hydroxyapatite (HAP, d = 40 nm) was purchased from Nanjing Emperor Nano Material Company (China). Dimethyl acetamide (DMAc) and bovine serum albumin (BSA, Mw = 67,000) were supplied by China Medicine Chemical Reagent Company (China).
2.2. Membrane preparation The membranes used in this study were prepared by the phase inversion method. PES, PVP and DMAc were mixed and heated until homogeneous mixed solutions with various compositions were obtained. The concentration of PES of PVP in DMAc was 14% and 5%, respectively. The HAP particles were added into the casting solution and were dispersed in the casting solution for 24 h in order to make HAP particles have good dispersion in the mixed solution and improve membrane performance. The amount of HAP was 60 wt% in dry PES/HAP MMMs. A slit of 300 m of casting knife for single layer MMMs was used. PES concentration for top layer of double layer MMMs was 5%, 10% and 15%, respectively. Fig. 1 shows schematic representation of the co-casting process. The heights of the slits of the first and second knife were 300 and 400 m, respectively. The polymeric mixture incorporating HAP were cast on a glass plate and immersed into 60% DMAc aqueous solution. The single layer PES/HAP MMMs formed a few moments after immersion. The pure polymer dope and polymeric mixtures incorporating HAP were co-cast on a glass plate and immersed into 60% DMAc aqueous solution and double layer MMMs formed. The MMMs were washed with tap water at room temperature to remove residual solvent.
2.3. Membrane characterization 2.3.1. Scanning electron microscopy For scanning electron microscopy (SEM), membranes were dried in air at room temperature and cryogenically broken in liquid nitrogen. The obtained cross-sections were dried overnight under vacuum at 30 ◦ C and gold coated. The cross sections, as well as the top and bottom surfaces of the membrane were characterized by scanning electron microscopy (SEM, JSM-5600LV, JEOL, Japan).
2.3.2. Pure water flux The membranes were subjected to pure water flux estimation at a trans-membrane pressure of 0.1 MPa under cross-flow
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
35
Fig. 2. Cross section of double layer PES/HAP MMMs with different PES top layers. PES content of top layer A, B, C is 5%, 10%, 15%. Cross section of top layer: A1, B1, C1; whole cross section: A2, B2, C2; cross section of bottom layer: A3, B3, C3.
filtration. The permeability was measured under steady-state flow. Pure water flux was calculated as follows: Jw =
Q At
where Q is the quantity of permeate collected (in l), and A is membrane area (m2 ), t is the sampling time (h), Jw is pure water flux (l m−2 h−1 ). 2.3.3. BSA adsorption capacity The bovine serum albumin (BSA) protein with molecular weight of 67 kDa and isoelectric point ∼4.74 was used for the protein adsorption study. BSA adsorption was investigated within a range of pH from 5 (acetate buffer) and 6–9 (phosphate buffer). The static protein adsorption capacity of membranes was determined with bovine serum albumin (BSA). The membranes were dried at 30 ◦ C in a vacuum oven before examination. The samples containing 2 g/l BSA were incubated with an exact amount of membranes in sealed containers under continuous shaking at 25 ◦ C. The PES/HAP MMM adsorbed the BSA thereby reducing the BSA concentration in the bulk. The equilibrium BSA concentration after 24 h was monitored in time with a UV-1800 spectrophotometer which was produced by SHIMADZU Company. The BSA depletion was measured at 280 nm with 5 mm quartz cuvettes. 2.3.4. BSA desorption capacity The MMMs were transferred into the desorption buffer after an adsorption and washing step. Desorption was accomplished in static by using a phosphate buffer of pH = 7, I = 0.5 M NaCl for static experiment. The desorption was carried out for 24 h in a shaking
bath at 25 ◦ C. The BSA desorption capacity was defined as the amount desorbed BSA from MMMs which had adsorbed BSA. 2.3.5. Two-compartment diffusion test A two-compartment diffusion device [41] was used to measure diffusion and adsorption of BSA onto MMMs at room temperature. The first compartment was filled with BSA buffer solution (2 mg/ml) and the second compartment was filled with water. The compartments were separated by one piece of MMM or two pieces of MMMs. The volume of each compartment was 100 ml and the active membrane area was 7.11 cm2 . Both solutions were stirred by magnetic rotors under 200 rpm. During the experiment, the samples were taken every 1 h to determine the BSA concentration in both compartments by UV-VIS spectrophotometer (UV-1800, SHIMADZU, Japan). The diffusion experiments were evaluated for 24 h before the process were stopped. 3. Results and discussion 3.1. Characterization of double layer mixed matrix membrane Fig. 2 shows the cross section of double layer PES/HAP MMMs with different PES top layers. The PES content of top layer is 5%, 10% and 15%, respectively. The left side of SEM picture is top layer, as shown in Fig. 2. The top PES layer is hard to form when PES content of top layer is 5% because of less PES in top layer. The top PES layer is inclined finger-like pores when PES content of top layer is 10%, and the top PES layer is straight finger-like pores when PES content of top layer is 15%. The formation of finger-like
36
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
Fig. 3. SEM pictures of top surface of single layer MMMs (A) and double layer PES/HAP MMMs with 10% PES as top layer (B).
3.2. Pure water flux and protein adsorption of double layer mixed matrix membrane 3.2.1. PES/HAP double layer MMMs Table 1 shows pure water fluxes and BSA adsorption capacities of single layer MMMs and double layer MMMs with different PES top layer. It can be seen from Table 1 that pure water flux of double layer MMMs decreases gradually with increasing PES solution concentration from 5 to 15% because top PES layer of double layer MMMs gradually become denser. The pure water flux of double layer MMMs with top layer with 15% PES (116.7 ± 3.0 l m−2 h−1 ) decreases a lot in comparison to that of single layer MMMs (217.7 ± 7.2 l m−2 h−1 ). As shown in Table 1, single layer MMMs keeps good BSA adsorption capacity (26.0 ± 2.1 mg/g membrane and 43.3 ± 2.1 mg/g HAP) compared to pure HAP particles (48.2 ± 3.2 mg/g HAP). BSA adsorption capacity of double layer MMMs decreases gradually with increasing PES concentration. BSA adsorption capacity appears a little decrease when PES concentration of top layer is 5% and 10%, which means that double layer MMMs keep good BSA adsorption capacity when top PES layer is prepared with low concentration of PES dope. BSA adsorption capacity appears abrupt decrease when PES concentration of top layer is 15% and BSA adsorption rate of HAP in double layer MMMs decreases sharply to 59.1%, which means
denser top PES layer inhibits adsorption efficiency of HAP in double layer MMMs. Fig. 4 shows BSA static adsorption capacity of HAP in double layer PES/HAP MMMs and double layer PES/HAP MMMs at different equilibrium time. The top PES layer of double layer MMMs was prepared with 10% PES. The BSA adsorption capacities of HAP in double layer PES/HAP MMMs and double layer PES/HAP MMMs increase in time during 10 h then equilibrium is reached, as shown in Fig. 4. The obtained equilibrium capacity after 24 h is 21.8 mg BSA/g membrane which is equal to 39.5 mg BSA/g HAP in the double layer MMMs. The BSA adsorption lines of HAP and double layer MMMs have similar tendency, which means HAP keeps good adsorption performance in the double layer MMMs and the PES top layer does not inhibit BSA from freely transporting in the matrix of double layer MMMs. 3.2.2. Effect of pH value on BSA adsorption The adsorption of BSA on the pure HAP particles, HAP in double layer MMMs and double layer PES/HAP MMMs as a function of pH (pH = 5–9) are shown in Fig. 5. The top PES layer of double layer MMMs was prepared with 10% PES. The BSA adsorption capacities of pure HAP particles, HAP in double layer MMMs and double layer MMMs increase with increasing pH value of solutions and reach the maximum at pH 7. Compared to the BSA adsorption capacity of pure HAP particles, the BSA adsorption capacity of HAP in double layer MMMs decreases a little and the maximum value decreases from 54.7 mg/g HAP to 41.3 mg/g HAP at pH 7 after HAP is embedded in double layer MMMs. BSA adsorption capacities of HAP in double layer MMMs (41.3 mg/g HAP) decrease slightly compared to that 50
Double layer MMMs HAP in MMMs Adsorbed BSA (mg/g)
pores means that the top PES solution starts to demix instantaneously [44]. With increasing PES content, the demixing delays and a thicker sublayer forms, which affect the formation and size of pores inside membrane. It also means that the phase separation speed decreases and PES molecular chains have more time to relax with increasing PES content of top layer, which results in the formation of different finger-like pores when different PES content is used. Moreover the thickness of top PES layer of double layer MMMs increases with increasing PES concentration for top PES layer. The filtration resistance of double layer MMMs increases with increasing PES concentration for top PES layer. So it is better to choose low PES concentration for top PES layer. The thickness of top layer prepared with 5% PES is very thin because of low concentration of casting solution, which can lead to more macropores in PES base and easily results in the leakage of HAP particles from MMMs. The thickness of top layer prepared with 10% PES concentration is better for double layer MMMs. Fig. 3 shows the surface structure of single layer PES/HAP MMMs and double layer PES/HAP MMMs. The top PES layer of double layer MMMs was prepared with 10% PES. Fig. 3A shows that HAP particles uniformly distribute at the surface of single layer MMMs and the surface of single layer MMMs is rough. Whereas the surface of double layer MMMs (Fig. 3B) is smooth. It is suggested that the top PES layer covers double layer MMMs completely.
40
30
20
10
0
0
5
10
15
20
25
Time (hour) Fig. 4. BSA static adsorption of double layer PES/HAP MMMs (n = 3) at different equilibrium time. Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%).
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
37
Table 1 Pure water flux, adsorption to BSA of single layer MMMs and double layer PES/HAP MMM (n = 3) (PES content of top layer for double layer MMMs is 10%). PES/%
Pure water flux (l m−2 h−1 )
0 5 10 15
217.7 192.4 151.4 116.7
± ± ± ±
Adsorbed BSA (mg/g HAP)
Adsorbed BSA (mg/g membrane)
7.2 5.4 4.6 3.0
26.0 24.5 21.8 15.1
± ± ± ±
2.1 2.8 2.4 1.4
43.3 42.4 39.5 28.5
± ± ± ±
BSA adsorption rate/%
2.1 2.8 2.4 1.4
89.8 88.0 82.0 59.1
*Adsorbed BSA of pure HAP particles is 48.2 ± 3.2 mg/g HAP [41]. SD indicates standard deviations.
in single layer MMMs (49.8 mg/g HAP), as shown in previous work [41]. It is suggested that HAP particles keep good BSA adsorption efficiency in PES/HAP double layer MMMs at different pH value. BSA adsorption capacity of double layer MMMs reaches the maximum (22.8 mg/g membrane) at pH 7. It was reported that BSA as a “soft protein” has lower structural stability and is adsorbed even under unfavorable conditions of a hydrophilic and electrostatically repelling surface by a structural rearrangement with a higher adsorption affinity [45,46]. The isoelectric point of BSA is around 4.8 [47]. When pH value is lower than 4.8, BSA carries positive charges and BSA adsorption is enhanced due to the electrostatic affinity interaction between HAP with negative charges and BSA with positive charges. When pH value is higher than 4.8, BSA carries negative charges. The pH used in this study is from 5 to 9, higher than the isoelectric point of BSA. The electrostatic interactions are dependent on the surface charges between the negatively charged COOH group of BSA and positively charged calcium ion of HAP. The electrostatic interactions generating between BSA and HAP is the maximum at pH 7, which makes the adsorbed amount of BSA reach the maximum. 3.2.3. BSA desorption of HAP in double layer MMMs Fig. 6 shows BSA desorption capacity of HAP in double layer PES/HAP MMMs at different time. The top PES layer of double layer MMMs was prepared with 10% PES. The BSA desorption capacity increase quickly within the first 7 h and BSA desorption capacity reaches 19.9 mg/g HAP (50.4% BSA release form HAP), which may attribute to the BSA weakly adsorbed on the outer surface of HAP. The BSA desorption capacity increases slowly with extending time after 7 h and BSA desorption capacity reaches 34.5 mg/g HAP (87.3% BSA release form HAP) after 24 h, as shown in Fig. 6. The BSA desorption capacity reached 41.5 mg/g HAP and 94.4% BSA released from HAP for single layer MMMs, as shown in previous work [41]. It means that PES top layer of double layer MMMs prevents adsorbed BSA from releasing from HAP quickly. Not 100% BSA release from HAP in MMMs is maybe because some BSA is adsorbed on the inner surface of pores in HAP. It takes more time for BSA adsorbed on the
inner surface of pores in HAP to release. In Boonsongrit’s study [48], approximately 80% BSA released from HAP after 24 h in physiological buffer solution and 10 mM phosphate buffer solution. In Wu’s study [49], 75–98.6% BSA released from HAP with different pore volume and pore size. In Sarkar’s group, BSA adsorption/release on HAP nanoparticles [50] and HAP-gelatin nanobiocomposite [51] was studied. Rod HAP nanoparticles exhibited relatively higher BSA adsorption capacity (28 mg/g) compared to the counterpart spherical and fibroid nanoparticles and around 75% BSA released within 96 h. The adsorption capacity of HAP-gelatin nanobiocomposite was 51.3 mg/g and 60% BSA released within 24 h. The specificity of HAP–protein interactions in MMMs provides a potential application in the field of protein delivery due to adsorption–desorption activity of HAP and proteins. 3.2.4. BSA diffusion through double layer MMMs For estimating the transport of BSA retention solutes through MMMs, we tested diffusion through and adsorption onto double layer MMMs using a two-compartment diffusion device with a 2 mg/ml BSA feed solution. Fig. 7 shows the relative contribution of adsorption and diffusion to the total removal of BSA for double layer PES/HAP MMMs. The top PES layer of double layer MMMs is prepared with 10% PES. The BSA removal by adsorption contributes to far more than that by diffusion and contributes 62.6% of the total removal after 24 h when double layer MMMs were used during BSA diffusion test, as shown in Fig. 7. Double layer MMMs have more BSA adsorption rate in the total BSA removal compared to single layer MMMs (BSA adsorption rate in the total BSA removal is 38.4% when one piece of single layer MMMs was used), as shown in previous work [41]. It is suggested that the resistance of BSA diffusing across double layer MMMs increases when double layer MMMs are used. More BSA is absorbed by double layer MMMs, not by diffusion because top PES layer makes MMMs surface denser and limit more BSA to diffuse through double layer MMMs. BSA stays in MMMs for longer time due to the resistance of BSA diffusing across MMMs increasing, which makes HAP particles in MMM have enough time to capture BSA. The MMMs combine BSA retention solutes removal
60
40
Desorption of BSA (mg/g HAP)
Adsorbed BSA (mg/g)
50 40 30 20
Pure HAP particles HAP in double layer MMMs PES/HAP double layer MMMs
10 0
4
5
6
7
8
9
10
pH value Fig. 5. BSA adsorption of pure HAP particles, HAP in double layer MMMs, double layer PES/HAP double layer MMMs (n = 3) at different pH value. Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%).
30
20
10
0
0
5
10
15
20
25
Time (h) Fig. 6. BSA desorption of HAP in double layer PES/HAP MMMs (n = 3) at different time. Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%).
38
J. Sun, L. Wu / Colloids and Surfaces B: Biointerfaces 123 (2014) 33–38
Removel (mg/g membrane)
40
Total removel Diffusion Adsorption
30
20
10
0
0
5
10
15
20
25
Time (hour) Fig. 7. BSA total removal, diffusion and adsorption plotted vs. time (n = 3). Error bars indicates standard deviations (PES content of top layer for double layer MMMs is 10%; one piece of double layer PES/HAP MMM was used).
via both diffusion as well as adsorption in one single step. The top PES layer of double layer MMMs works as an isolation layer to prevent HAP from contacting protein solution directly and at the same time allows HAP in the matrix to adsorb protein. 4. Conclusions In this study, the double layer PES/HAP mixed matrix membranes (MMMs) were prepared by an immersion phase separation process. A particle free PES membrane layer was co-cast with the single layer MMMs which led to formation of double layer MMMs with good transport capacity and BSA adsorption capacity. The BSA adsorption capacity of double layer MMMs reaches the maximum at pH 7. 87.3% BSA releases form HAP after 24 h during desorption process. Double layer MMMs have high BSA removal rates and high BSA adsorption rate during the process of BSA diffusion through double layer MMMs. The double layer MMMs with top PES layer keep good filtration, adsorption and desorption properties. Acknowledgements The authors thank National Natural Science Foundation of China (51203020); Scientific Research Starting Foundation for Returned Overseas Chinese Scholars, Ministry of Education of China; Donghua University central university scientific research special fund (2232012D3-32). References [1] X. Su, Y. Tian, W. Zuo, J. Zhang, H. Li, X. Pan, Water Res. 50 (2014) 267. [2] K. Plate, S. Beutel, H. Buchholz, W. Demmer, S. Fischer-Frühholz, O. Reif, R. Ulber, T. Scheper, J. Chromatogr. A 1117 (2006) 81. [3] T. Muschin, S. Han, T. Kanamoto, H. Nakashima, T. Yoshida, J. Polym. Sci. Polym. Chem. 49 (2011) 3241.
[4] S.A. Hashemifard, A.F. Ismail, T. Matsuura, J. Colloids Interface Sci. 359 (2011) 359. [5] D.Q. Vu, W.J. Koros, S.J. Miller, J. Membr. Sci. 211 (2003) 311. [6] M.E. Avramescu, Z. Borneman, M. Wessling, J. Membr. Sci. 218 (2003) 219. [7] M.E. Avramescu, Z. Borneman, M. Wessling, J. Chromatogr. A 1006 (2003) 171. [8] M.E. Avramescu, Z. Borneman, M. Wessling, J. Membr. Sci. 322 (2008) 306. [9] S. Saiful, Z. Borneman, M. Wessling, J. Membr. Sci. 280 (2006) 406. [10] K.K. Kopec, S.M. Dutczak, L. Bolhuis-Versteeg, M. Wessling, D.F. Stamatialis, Sep. Purif. Technol. 80 (2011) 306. [11] S.M. Saufi, C.J. Fee, J. Membr. Sci. 444 (2013) 157. [12] S.M. Saufi, C.J. Fee, Sep. Purif. Technol. 77 (2011) 68. [13] S.M. Saufi, C.J. Fee, J. Chromatogr. A 1218 (2011) 9003. [14] M.M. Khan, V. Filiz, G. Bengtson, S. Shishatskiy, M.M. Rahman, J. Lillepaerg, V. Abetz, J. Membr. Sci. 436 (2013) 109. [15] J. Yin, G. Zhu, B. Deng, J. Membr. Sci. 437 (2013) 237. [16] J.H. Li, Y.Y. Xu, L.P. Zhu, J.H. Wang, C.H. Du, J. Membr. Sci. 326 (2009) 659. [17] S. Yu, X. Zuo, R. Bao, X. Xu, J. Wang, J. Xu, Polymer 50 (2009) 553. [18] S. Khodadoust, M. Ghaedi, R. Sahraei, A. Daneshfar, J. Ind. Eng. Chem. 20 (2014) 2663. [19] F. Motahari, M.R. Mozdianfard, M. Salavati-Niasari, Process Saf. Environ. (2014), http://dx.doi.org/10.1016/j.psep.2014.06.006. [20] A.T.N. Lam, J. Yoon, E.-O. Ganbold, D.K. Singh, D. Kim, K.-H. Cho, S.J. Son, J. Choo, S.Y. Lee, S. Kim, S.-W. Joo, J. Colloids Interface Sci. 425 (2014) 96. [21] A. Jaworska, K. Malek, J. Colloids Interface Sci. 431 (2014) 117. [22] F. Vázquez-Hernández, C. Mendoza-Barrer, V. Altuzar, Mater. Sci. Eng. B 174 (2010) 290. [23] M. Espanol, I. Casals, S. Lamtahri, M.T. Valderas, M.P. Ginebra, Biointerphases 7 (2012) 37. [24] J.M. Thomann, M.J. Mura, M.S. Behr, J.D. Aptel, A. Schmitt, E.F. Bres, J.C. Voegel, Colloids Surf. 40 (1989) 293. [25] T. Kawasaki, S. Takahashi, K. Ideda, Eur. J. Biochem. 152 (1985) 361. [26] K. Kandori, A. Masunari, T. Ishikawa, Calcified Tissue Int. 76 (2005) 194. [27] F. Mohandes, M. Salavati-Niasari, Chem. Eng. J. 252 (2014) 173. [28] F. Mohandes, M. Salavati-Niasari, Mater. Sci. Eng. C 40 (2014) 288. [29] F. Mohandes, M. Salavati-Niasari, Z. Fereshteh, M. Fathi, Ceram. Int. 40 (2014) 1222–1227. [30] T.Y. Liu, S.Y. Chen, D.M. Liu, S.C. Liou, J. Control. Release 107 (2005) 112. [31] K. Kandori, S. Oda, M. Fukusumi, Y. Morisada, Colloids Surf. B: Biointerfaces 73 (2009) 140. [32] S. Dasgupta, A. Bandyopadhyay, S. Bose, Langmuir 26 (2010) 4958. [33] K. Ozeki, T. Hoshino, H. Aoki, T. Masuzawa, Biomed. Mater. Eng. 23 (2013) 239. [34] K. Kandori, H. Hamazaki, M. Matsuzawa, S. Togashi, Adv. Powder Technol. 25 (2014) 354. [35] E. Fujii, M. Ohkubo, K. Tsuru, S. Hayakawa, A. Osaka, K. Kawabata, C. Bonhomme, F. Babonneau, Acta Biomater. 2 (2006) 69. [36] S. Takemoto, Y. Kusudo, K. Tsuru, S. Hayakawa, A. Osaka, S. Takashima, J. Biomed. Mater. Res. A 69 (2004) 544. [37] D.F. Hagen, S.J. St. Mary, L.A. Errede, US patent 4,906,378. [38] K.K. Kar, S. Pramanik, US patent 8,652,373. [39] A. Kozlowski, M.J. Roberts, US patent 8,372,422. [40] F. Mohandes, M. Salavati-Niasari, RSC Adv. 4 (2014) 25993. [41] J. Sun, L. Wu, Appl. Surf. Sci. 308 (2014) 155. [42] M.S.L. Tijink, M. Wester, J. Sun, A. Saris, L.A.M. Bolhuis-Versteeg, S. Saiful, J.A. Joles, Z. Borneman, M. Wessling, D.F. Stamatialis, Acta Biomater. 8 (2012) 2279. [43] M.S.L. Tijink, M. Wester, G. Glorieux, K.G.F. Gerritsen, J. Sun, P.C. Swart, Z. Borneman, M. Wessling, R. Vanholder, J.A. Joles, D.F. Stamatialis, Biomaterials 34 (2013) 7819. [44] C.A. Smolders, A.J. Reuvers, R.M. Boom, I.M. Wienk, J. Membr. Sci. 73 (1992) 259. [45] W. Norde, C.I. Anuisem, Colloids Surf. 66 (1992) 73. [46] W. Norde, T. Arai, H. Shirahama, Biofouling 4 (1991) 37. [47] T.P.B. Nguyen, J.W. Lee, W.G. Shim, H. Moon, Microporous Mesoporous Mater. 110 (2008) 560. [48] Y. Boonsongrit, H. Abe, K. Sato, M. Naito, M. Yoshimura, H. Ichikawa, Y. Fukumori, Mater. Sci. Eng. Solid-State Mater. Adv. Tech. 148 (2008) 162. [49] X. Wu, X. Song, D. Li, J. Liu, P. Zhang, X. Chen, J. Bionic Eng. 9 (2012) 224. [50] S. Swain, D. Sarkar, Appl. Surf. Sci. 286 (2013) 99. [51] B. Basu, S.K. Swain, D. Sarkar, RSC Adv. 3 (2013) 14622.
