Immobilization of penicillin G acylase on macro-mesoporous silica spheres

Immobilization of penicillin G acylase on macro-mesoporous silica spheres

Bioresource Technology 102 (2011) 529–535 Contents lists available at ScienceDirect Bioresource Technology journal homepage: www.elsevier.com/locate...

329KB Sizes 0 Downloads 10 Views

Recommend Documents

No documents
Bioresource Technology 102 (2011) 529–535

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Immobilization of penicillin G acylase on macro-mesoporous silica spheres Junqi Zhao, Yujun Wang ⇑, Guangsheng Luo, Shenlin Zhu The State Key Laboratory of Chemical Engineering, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China

a r t i c l e

i n f o

Article history: Received 16 July 2010 Received in revised form 18 September 2010 Accepted 20 September 2010 Available online 27 September 2010 Keywords: Enzyme immobilization Penicillin G acylase Mesopore Macropore Mass transfer resistance

a b s t r a c t In this study, macro-mesoporous silica spheres were prepared with a micro-device and used as the support for the immobilization of penicillin G acylase (PGA). To measure the enzymatic activity, the silica spheres with immobilized PGA were placed into a packed-bed reactor, in which the hydrolysis of penicillin G was carried out. The influences of the residence time, the initial concentration of the substrate, the accumulation of the target product 6-aminopenicillanic acid, and the enzyme loading amount on the performance of the immobilized PGA were investigated. The introduction of macropores increased the enzyme loading amount and decreased the internal mass transfer resistance, and the results showed that the enzyme loading amount reached 895 mg/g (dry support), and the apparent enzymatic activity achieved up to 1033 U/g (dry support). In addition, the immobilized PGA was found to have great stability. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction Penicillin G acylase (PGA; EC 3.5.1.11), an N-terminal nucleophile hydrolase with molecular dimensions of 7.0  5.0  5.5 nm3, is one of the key enzymes used in the pharmaceutical industry for the production of b-lactum antibiotics (Wei and Yang, 2003; Sio and Quax, 2004; Zhang et al., 2006; Cheng et al., 2007; Chandel et al., 2008). In recent years, many efforts have been devoted to using mesoporous silica materials as the supports for enzyme immobilization because of their large surface area and pore volume, tunable pore size and structures, openness to a wide variety of chemical modifications, convenience of reutilization, and environmental friendliness (Huang et al., 2003; Xu et al., 2003; Bootsma et al., 2008; Yang et al., 2008; Chouyyok et al., 2009; Salis et al., 2009; Bautista et al., 2010; Kumar et al., 2010; Wang et al., 2010). PGA enzymes have been immobilized on MCM-41 (He et al., 2000; Xue et al., 2004), MCM-48 (Xue et al., 2004), cubic Ia3d mesoporous silica (Lü et al., 2007, 2008), SBA-15 (Chong and Zhao, 2004a,b; Lü et al., 2008; Shah et al., 2008; Sun et al., 2009), and mesostructured cellular foams (Xue et al., 2008; Zhao et al., 2010) with different pore structures and surface properties. The experimental results showed that the mesoporous silica was a suitable support for the immobilization of PGA. However, morphologies of the supports used in these works were powdered, and it is well known that the submicrometer-sized powders are difficult to separate and recover in industrial processes. Thus, mesoporous silica spheres may be more promising supports. ⇑ Corresponding author. Tel.: +86 1062773017; fax: +86 1062770304. E-mail address: [email protected] (Y. Wang). 0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2010.09.076

However, compared with mesoporous silica powders, mesoporous silica spheres used as the supports for the immobilization of PGA may bring about larger internal mass transfer resistance for the hydrolysis of Pen G, which is a fast reaction with the diffusions of the substrate and products as the rate-determining steps. To solve this problem, macropores (>50 nm) were introduced into the mesoporous silica spheres, thus in this work, macro-mesoporous silica spheres were prepared with a micro-device based on our previous researches (Zhai et al., 2008, 2009). There, the performance of this kind of silica spheres as the adsorbents for two kinds of proteins – bovine serum albumin (BSA, 4  4  14 nm3) and lysozyme (LYS, 3  3  4.5 nm3) was investigated, and the results showed that the introduction of macropores increased the adsorbed amounts of BSA and LYS and significantly decreased the adsorption equilibrium time, implying that the macro-mesoporous silica spheres might be effective supports for the immobilization of PGA (7.0  5.0  5.5 nm3). This is because the macropores could make the PGA molecules penetrate into the internal mesopores of the silica spheres more easily, producing a high enzyme loading amount; moreover, the macropores could provide a highway for the mass transfer of the substrate and products toward and away from the active sites of the immobilized PGA molecules, obtaining a lower internal mass transfer resistance and a higher apparent activity. To improve the stability of the immobilized PGA, 3-aminopropyltriethoxysilane (APTS) was used as a chemical modifier to obtain aminopropyl-functionalized silica spheres, and the immobilization of PGA was performed through Schiff base reaction with glutaraldehyde. The hydrolysis of Pen G was carried out in a packed-bed reactor to determine the apparent activity of the

530

J. Zhao et al. / Bioresource Technology 102 (2011) 529–535

immobilized PGA on the macro-mesoporous silica spheres. A packed-bed reactor was adopted here to simplify the experimental procedure, avoiding filtration or centrifugation and the mass loss of the immobilized PGA in these separation processes. Moreover, compared with stirred tank reactors, packed-bed reactors can also avoid the damages to the support caused by the mechanical stirring and the deactivation of the immobilized PGA caused by the direct contact of the immobilized enzyme and the alkaline agent that was added to neutralize the by-product phenylacetic acid (PAA). The influences of the residence time, the initial concentration of Pen G, and the accumulation of the target product 6-aminopenicillanic acid (6-APA) on the performance of the immobilized PGA were investigated. In addition, the influence of the enzyme loading amount on the catalytic efficiency of the immobilized PGA was investigated.

2. Methods 2.1. Chemicals Tetraethyl orthosilicate (TEOS) was produced by Xilong Chemical Company (Shantou, China). Sorbitan monooleate (Span 80) was produced by China Medicine Shanghai Chemical Reagent Corporation (Shanghai, China). Trioctylamine (TOA) was purchased from Feixiang Chemical Company (Zhangjiagang, China). Poly(ethylene glycol) (PEG 20,000), sodium dihydrogen phosphate dihydrate (NaH2PO42H2O), and sodium phosphate dibasic dodecahydrate (Na2HPO412H2O) were produced by Yili Fine Chemical Company (Beijing, China). Methylcellulose (MC) and 3-aminopropyltriethoxysilane (APTS) were purchased from Acros Organics Company (New Jersey, USA). Glutaraldehyde was purchased from Bodi Chemical Company (Tianjin, China). Penicillin G acylase (PGA; EC 3.5.1.11) was purchased from Shunfeng Haideer Limited Company (Zhejiang, China). Penicillin G potassium (Pen G) was produced by North China Pharmacy Factory (Hebei, China). 6-aminopenicillanic acid (6-APA) was purchased from Alfa Aesar Chemical Limited Company (Tianjin, China). All the chemicals were used as received without further purification. 2.2. Synthesis of macro-mesoporous silica spheres Macro-mesoporous silica spheres were synthesized with a coaxial micro-device, using a method based upon our previous researches (Zhai et al., 2008, 2009). The coaxial microdevice was fabricated on two polymethyl methacrylate (PMMA) plates (40 mm  40 mm  3 mm). The main channel was about 2.0 mm in diameter with a long polytetrafluoroethylene (PTFE) tube (3 m in length, 1.5 mm id  2.0 mm od) embedded in it. A needle (0.2 mm id  0.7 mm od) was inserted into the PTFE tube coaxially to introduce a dispersed aqueous phase. There were two side channels (1.6 mm id) with two needles (1.3 mm id  1.6 mm od) embedded in them, and the side channels were fixed perpendicularly to the main channel to introduce the continuous phase. The dispersed phase was prepared as follows: 0.25 g of MC and 1.0 g of PEG 20,000 were dissolved in 5.0 g of HCl aqueous solution with a concentration of 0.01 mol/L under stirring. Then 2.5 g of TEOS was added, and the resulting mixture was stirred for 3 days at room temperature. The continuous phase was a mixture containing 2 wt.% span 80, 30 wt.% TOA, and 68 wt.% octane. The dispersed and continuous phase solutions were injected in by syringe pumps at flow rates of 0.02 and 0.4 mL/min, respectively, to form monodisperse silica sol droplets. The PTFE tube was kept at 40 °C to realize the gelation of the silica sol droplets. Solidified silica spheres were obtained at the exit of the PTFE tube. Then

the silica spheres together with ca. 20 mL of the continuous phase solution were transferred into an autoclave and treated at 100 °C for 24 h. After that, the silica spheres were collected by filtration, washed with water and ethanol, dried at 80 °C for 12 h, and calcined at 550 °C for 6 h. The white silica spheres obtained were designated as macro-mesoporous silica spheres (MMSS). 2.3. Grafting aminopropyl and glutaraldehyde groups and immobilizing PGA The methods of grafting aminopropyl and glutaraldehyde and immobilizing PGA on MMSS were based upon our previous research (Zhao et al., 2010). The grafting of aminopropyl was accomplished as follows: 0.05 g of APTS was added to 20 g of HCl aqueous solution with a concentration of 0.1 mol/L, and the resulting mixture was stirred at room temperature for 30 min to allow the sufficient hydrolysis of APTS, after which, 0.2 g of MMSS was added. The mixture was shaken at 160 rpm and 25 °C for 8 h, transferred into an autoclave, and kept at 100 °C for 24 h. Then the resulting solid was filtrated, washed with 500 mL of deionized water and 500 mL of ethanol, and air-dried at 80 °C for 12 h. The samples obtained were designated as NH2–MMSS. The grafting of glutaraldehyde was carried out as follows: 0.02 g of NH2–MMSS was packed into a glass column with a length of 1 cm and internal diameter of 0.4 cm using the dry packing method, and the column was kept at 25 °C. Glutaraldehyde aqueous solution with a concentration of 5 wt.% flowed through the column at a flow rate of 0.083 mL/min for 30 min, before and after which, deionized water flowed through the column at the same flow rate for 30 min. The samples obtained were designated as CHO–NH2–MMSS. To quantify the amounts of grafted aminopropyl and glutaraldehyde groups, the mass of carbon, hydrogen, and nitrogen of MMSS, NH2–MMSS, and CHO–NH2–MMSS was measured. Amounts of grafted aminopropyl (qA, lmol/g dry MMSS) and glutaraldehyde (qG, lmol/g dry MMSS) were calculated as follows:

NNH2—MMSS  NMMSS  106 14 C CHO—NH2—MMSS  C NH2—MMSS qG ¼  106 12  5 qA ¼

where, NNH2–MMSS (wt.%) is the mass of the nitrogen element of NH2–MMSS; NMMSS (wt.%) is the mass of the nitrogen element of MMSS; CCHO–NH2–MMSS (wt.%) is the mass of the carbon element of CHO–NH2–MMSS; CNH2–MMSS (wt.%) is the mass of the carbon element of NH2–MMSS. Here, an apparent grafted amount is used for glutaraldehyde because the glutaraldehyde group is considered as a single molecule to calculate the grafted amount, but it is actually in multimer form when grafted onto the support. Thus, the available amount of glutaraldehyde is much lower than the apparent grafted amount. The covalent immobilization of PGA was accomplished as follows: 15 mL of PGA solution obtained by diluting 1 mL of enzyme liquid with 14 mL of phosphate buffer (0.2 mol/L, pH 7.9) flowed through the column packed with CHO–NH2–MMSS at varying flow rates, before and after which, blank phosphate buffer (0.2 mol/L, pH 7.9) flowed through the column at the same flow rate for 30 min. For all these operations, the column was kept at 30 °C. The samples obtained were designated as PGA–CHO–NH2–MMSS, and the resulting column was stored in phosphate buffer (0.2 mol/L, pH 7.9) at 4 °C when not in use. The concentration of PGA in the inlet and outlet solutions was measured by Bradford assay. The enzyme loading amount (qE, mg/g dry MMSS) was calculated as follows:

qE ¼

mPGA;

in  mPGA; W support

out

J. Zhao et al. / Bioresource Technology 102 (2011) 529–535

where, mPGA, in (mg) is the amount of PGA in the inlet solution; mPGA, out (mg) is the amount of PGA in the outlet solution; Wsupport is the dry weight of the support (g). 2.4. Hydrolysis of Pen G The hydrolysis of Pen G was carried out as follows: Pen G solution in phosphate buffer (0.2 mol/L, pH 7.9) flowed through the column packed with PGA–CHO–NH2–MMSS, which was kept at 37 °C. The concentrations of Pen G and 6-APA were determined using a HP series 1100 HPLC system, which consisted of an 1100 variable wavelength detector, an Agilent Chemstation, and an Agilent ZORBAX SB-C18 column (4.6  250 mm). The mobile phase employed was a mixture of phosphate buffer (0.1 mol/L, pH 3.5) and methanol (60: 40, v/v). The flow rate of the mobile phase was 1.0 mL/min, and the detector was set at 225 nm. Once any experimental condition was changed, a steady state was checked, and the concentrations were the average values of three measured values. The apparent activity (AIME, U/g dry support) of the immobilized PGA, the conversion rate of Pen G (XPen G, %), and the rate of the reaction (v, mmol/L min) were calculated according to the following formulas:

 C 6-APA; in Þ  F W support ðC Pen G; in  C Pen G; in Þ X Pen G ¼  100% C Pen G; in ðC C ÞF v ¼ 6-APA; out 6-APA; in V AIME ¼

ðC 6-APA;

out

where, C6-APA, out (mmol/L) is the concentration of 6-APA in the outlet solution; C6-APA, in (mmol/L) is the concentration of 6-APA in the inlet solution, i.e., the initial concentration of 6-APA (apart from the experiment about the influence of the accumulation of 6-APA, this term equals to zero); CPen G, out (mmol/L) is the concentration of Pen G in the outlet solution; CPen G, in (mmol/L) is the concentration of Pen G in the inlet solution, i.e., the initial concentration of Pen G; F (mL/min) is the flow rate; V (cm3) is the reactor volume, which was calculated as follows:

V ¼ Vt  Vs Vt (cm3) is the total reactor volume, and Vs (cm3) is the support volume, which were calculated as follows:

Vt ¼ Vs ¼

p

D2 Lc 4 c W support

qsupport

where, Dc (cm) is the inside diameter of the column; Lc (cm) is the length of the column; qsupport (g/cm3) is the skeletal density of the silica spheres. The residence time of Pen G solution (s, min) was calculated as follows:



V F

To examine the stability of the immobilized PGA, Pen G solution with an initial concentration of 2.7 mmol/L flowed through a column (qE = 639 mg/g) with a residence time of 1.687 min for 24 h. To investigate the influence of the residence time, Pen G solution with an initial concentration of 13.4 mmol/L flowed through a column (qE = 895 mg/g) with varying residence time of 0.045– 1.798 min. To investigate the influence of the initial concentration of Pen G, the hydrolysis reaction was carried out in a column (qE = 639 mg/g) with a residence time of 0.011 min and varying initial concentrations of Pen G of 2.8–417.3 mmol/L. To investigate the influence of the accumulation of 6-APA, mixed solution of Pen G (with an initial concentration of 13.4 mmol/L) and 6-APA (with

531

varying initial concentrations) flowed through a column (qE = 639 mg/g) with a residence time of 0.045 min. To investigate the influence of the enzyme loading amount on the catalytic efficiency of the immobilized PGA, Pen G solution with varying concentrations of 2.7–26.8 mmol/L flowed through three columns (qE = 895, 639, 256 mg/g, respectively) with a residence time of 0.045 min. 2.5. Characterization Scanning electron microscope (SEM) (JEOL JSM 7401F) experiments were conducted at 1.0 kV. Transmission electron microscopy (TEM) experiments were performed on a JEOL JEM 2010 microscope operating at 120 kV. The isotherms of nitrogen adsorption and desorption were measured at 77 K using a Quantachrome Autosorb-1-C Chemisorption-Physisorption Analyzer. Before measurements were taken, the samples were outgassed at room temperature for 1.5 h. The BET surface area was calculated from the adsorption branch in the relative pressure range of 0.10–0.30, and the total pore volume was calculated at a relative pressure of about 0.995. The pore size distribution was calculated from the adsorption branch using the Barrett–Joyner–Halenda (BJH) method. Macropore size distributions and the skeletal density of the silica spheres were measured with a Micromeritics Autopore IV 9500 Porosimeter using the mercury intrusion method. The mass of carbon, hydrogen, and nitrogen was measured using a PE 2400 C, H, N elementary analyzer. The enzyme concentration was analyzed by a UV Spectrophotometer (HP 8453, Agilent). 3. Results and discussion 3.1. Characterization of macro-mesoporous silica spheres SEM and TEM images (not shown) were taken to observe the external and internal structures of MMSS, and it turned out that the silica spheres were solid and monodispersed with a diameter of ca. 700 lm. The external surface of the silica spheres was distributed with mesopores and submicrometer-sized macropores, and the internal structure of the silica spheres was less compact and presented as a network of mesopore-containing silica walls and micrometer-sized interconnected macropores. The isotherms of nitrogen adsorption and desorption and pore size distributions of MMSS, NH2–MMSS, CHO–NH2–MMSS, and PGA–CHO–NH2–MMSS are shown in Fig. 1, and their physicochemical properties are listed in Table 1. The isotherm curves of nitrogen adsorption and desorption showed that, after the grafting of aminopropyl and glutaraldehyde and the immobilization of PGA, the samples could still maintain their mesoporous structures; however, the surface area and the total pore volume gradually decreased, indicating the presence of aminopropyl, glutaraldehyde, and PGA molecules on the surface of the internal pores of MMSS. The pore size distribution of MMSS was also determined using the mercury intrusion method, and the result showed that there was a considerable quantity of macropores in MMSS (Fig. 1c). The amounts of grafted aminopropyl and glutaraldehyde groups and the enzyme loading amount are shown in Table 1, confirming the success of grafting aminopropyl, glutaraldehyde and immobilizing PGA on the surface of the internal pores of MMSS. The enzyme loading amount on CHO–NH2–MMSS was as high as 895 mg/g, which was much higher than that on some reported spherical supports (Bianchi et al., 1996; Janssen et al., 2001). Furthermore, the enzyme loading amount was even higher than that on some reported silica powdered supports (Chong and Zhao, 2004a,b; Lü et al., 2007, 2008; Shah et al., 2008; Xue et al., 2008; Sun et al., 2009; Zhao et al., 2010), showing that compared with

532

J. Zhao et al. / Bioresource Technology 102 (2011) 529–535

1200

a

MMSS Ads MMSS Des NH 2-MMSS Ads NH 2-MMSS Des

3

Volume adsorbed (cm /g)

1000 800

CHO-NH2-MMSS Ads CHO-NH2-MMSS Des PGA-CHO-NH2-MMSS Ads

600

PGA-CHO-NH2-MMSS Des

400 200 0 0.0

0.2

0.4 0.6 Relative pressure P/P0

0.8

1.0

3.2. Stability of the immobilized PGA

2.0

b

MMSS Des NH 2-MMSS Des

1.6 dV/dlgD (cm 3/g)

the silica powders, the increase of the diameter of the silica spheres did not bring about the decrease of the enzyme loading amount. And this occurred thanks to the presence of macropores. For the silica spheres only with mesopores, it was difficult for the PGA molecules to get immobilized on the surface of the internal mesopores because some already immobilized PGA molecules on the external surface of the silica spheres might have blocked the entrance of the mesopores, and then the surface of the internal mesopores would be wasted; for the macro-mesoporous silica spheres, however, mesopores were connected with macropores, and the macropores constructed a highway between the external surface of the silica spheres and the surface of the internal mesopores to facilitate the PGA molecules getting immobilized, thus the utilization efficiency of the surface of the internal mesopores would be increased and a high enzyme loading amount would be obtained.

Continuous experiments of the hydrolysis of Pen G were carried out in the packed-bed reactor, and the results showed that the conversion rate of Pen G remained more than 99% after 24 h, indicating a great stability of the immobilized PGA on the macro-mesoporous silica spheres. Meanwhile, the apparent activity of the immobilized PGA had no obvious change after being stored at 4 °C for 30 days, showing a great storage stability of the immobilized PGA on the macro-mesoporous silica spheres.

CHO-NH 2-MMSS Des PGA-CHO-NH2-MMSS Des

1.2

0.8

3.3. Influences of the residence time, the initial concentration of Pen G, and the accumulation of 6-APA on the performance of the immobilized PGA

0.4

0.0 1

0.25

10 Pore diameter (nm)

In the experiments, the residence time was found to have a significant influence on the performance of the immobilized PGA on the macro-mesoporous silica spheres, as shown in Fig. 2. When

100

c

300

100

250

80

0.15 200

XPen G (%)

qE=895 mg/g

0.10

60

CPen G, in=13.4 mmol/L

150 40

AIME (U/g)

dV/dlgD (cm3/g)

0.20

100

0.05 20

50

0.00 10

100 Pore diameter (nm)

1000

10000 0

Fig. 1. Nitrogen adsorption/desorption isotherms (a) and BJH desorption pore size distributions (b) of the tested samples, pore size distribution of MMSS determined by the mercury intrusion method (c) (Ads: adsorption; Des: desorption).

0.0

0.4

0.8

1.2

1.6

0 2.0

T (min) Fig. 2. Influence of the residence time on the performance of the immobilized PGA.

Table 1 Physicochemical properties of MMSS before and after the immobilization of PGA. Samples

SBET (m2/g)

Vtotal (cm3/g)

DA (nm)

DBJHD (nm)

qA (lmol/g)

qG (lmol/g)

qE (mg/g)

MMSS NH2–MMSS CHO–NH2–MMSS PGA–CHO–NH2–MMSS

370 350 253 178

1.75 1.67 1.28 0.92

18.9 19.1 20.3 20.6

17.1 12.1 17.7 17.6

– 208 (1.2)a – –

– – 221 (8.0)a –

– – – 895 (53.7)a

SBET, multipoint BET surface area; Vtotal, total pore volume; DA, average pore diameter; DBJHD, BJH method desorption pore diameter. a The values in the brackets: standard deviation of the experimental results.

533

J. Zhao et al. / Bioresource Technology 102 (2011) 529–535

the residence time increased from 0.045 to 1.798 min, the conversion rate of Pen G increased from 17.7% to 96.8%, but the apparent activity of the immobilized PGA decreased from 261 to 40 U/g, which indicated the influence of the external mass transfer. To lower the influence of the external mass transfer, a short residence time (no longer than 0.045 min) was adopted in the following experiments. It can be seen from Fig. 2 that a complete conversion of Pen G could be obtained by prolonging the residence time, but this was at the cost of the decrease of the apparent activity of the immobilized PGA. According to this result, several columns can be cascaded to carry out the hydrolysis of Pen G. While the front columns transform most of the substrate with a high apparent activity by adopting a shorter residence time, the last one transforms the remaining substrate completely by adopting a longer residence time. Meanwhile, PAA can be neutralized by adding alkaline agent between these columns, and this can avoid the deactivation of the immobilized PGA caused by the direct contact of the immobilized enzyme and the alkaline agent that was added to neutralize PAA. Besides the residence time, the initial concentration of Pen G is another important factor influencing the performance of the immobilized PGA on the macro-mesoporous silica spheres. As shown in Fig. 3, with the increase of the initial concentration of Pen G, the conversion rate of Pen G decreased, and the apparent activity of the immobilized PGA initially increased and then decreased. In the experimental range, it is found that with an initial Pen G concentration of 308.7 mmol/L, the apparent activity of the immobilized PGA on our support achieved up to 1033 U/g (dry support), which was higher than that on some reported spherical supports, as shown in Table 2. Although the sphere diameter of our support was larger and the surface area of our support was not the largest, the enzyme loading amount and the apparent activity

7

1200

6

1000

5

qE=639 mg/g

600

=0.011 min

3

AIME (U/g)

XPen G (%)

800 4

400 2 200

1

0

0 0

100

200 300 CPen G, in (mmol/L)

400

Fig. 3. Influence of the initial concentration of Pen G on the performance of the immobilized PGA.

of the immobilized PGA on our support were higher. And this occurred thanks to the pore structure of the macro-mesoporous silica spheres. As discussed before, the presence of macropores made the PGA molecules get immobilized on the surface of the internal mesopores more easily, then the utilization efficiency of the surface of the mesopores was higher, and a higher enzyme loading amount was obtained. Also, the macropores provided a highway for the mass transfer of Pen G, 6-APA, and PAA toward and away from the active sites of the immobilized PGA molecules, thus the influence of the internal mass transfer was weakened. All these were favorable for obtaining a higher apparent activity of the immobilized PGA on the macro-mesoporous silica spheres. It is known that the main reaction of the hydrolysis of Pen G resulting in the formation of 6-APA and PAA is commonly accompanied by: (1) noncompetitive enzyme inhibition by Pen G; (2) noncompetitive enzyme inhibition by 6-APA; (3) competitive enzyme inhibition by PAA. As shown in Fig. 3, when the initial Pen G concentration was higher than 308.7 mmol/L, the apparent activity of the immobilized PGA decreased, which indicated the influence of the inhibition by Pen G. In Fig. 3, when the initial Pen G concentration was at a lower level (2.8–56.6 mmol/L), the inhibitions by Pen G, 6-APA, and PAA were weak, and the kinetic behaviors of the immobilized PGA could be investigated with the Michaelis–Menten equation:



V m  C Pen K m þ C Pen

G; in

ð1Þ

G; in

where, Km (mmol/L) is the apparent Michaelis constant; Vm (mmol/ L min) is the maximum of reaction velocity. The Eadie–Hofstee plot for Eq. (1) can be expressed as follows:

v ¼ K m

v C Pen

þ Vm

ð2Þ

G; in

According to the linear plot of v to v/CPen G, in (not shown) and Eq. (2), the values of Km and Vm were determined as 27.9 mmol/L and 102.2 mmol/L min, respectively, for the immobilized PGA under the experimental conditions. In Fig. 3, when the initial Pen G concentration was at a higher level (309.1–417.3 mmol/L), the inhibitions by 6-APA and PAA were relatively weak compared with the inhibition by Pen G, and the apparent Michaelis–Menten equation could be expressed as follows:

1

v

¼

1 C Pen G; in þ V m V m  K i; Pen

ð3Þ G

where, Ki, Pen G is the apparent constant of Pen G inhibition. According to the linear plot of 1/v to CPen G, in (not shown) and Eq. (3), the value of Ki, Pen G was determined as 480.9 mmol/L for the immobilized PGA under the experimental conditions. Thus, the value of Km was about a factor of 16 lower than the value of Ki, Pen G, indicating a relatively weak inhibition by Pen G. In industrial processes, the adopted initial concentration of Pen G is about 8% (w/v) considering the inhibition by Pen G (Katchalski-Katzir and Kraemer, 2000), and the results in Fig. 3 showed that the apparent activity of the immo-

Table 2 Comparisons of spherical supports for the immobilization of PGA. Support Amberlite XAD7 beads ETMA–EGDM beads Poly(HEMA-co-diviyl benzene) beads Eupergit C beads Macro-mesoporous silica spheres

DA (nm) a

9 – – 25a 20.3

SBET (m2/g)

Ds (lm)

qE (mg/g)

AIME (U/g dry support)

Reactor

Literature

450a 108 – – 253

100–200 250–420 150–200 170a 700

145.4 – – 100 639

400 191 825.8 748b 1033

Stirred tank reactor Stirred tank reactor Stirred tank reactor Stirred tank reactor Packed-bed reactor

Bianchi et al. (1996) Skaria et al. (1997) Wang et al. (2007a) Janssen et al. (2001) This work

Ds, sphere diameter; HEMA, b-hydroxyethyl methacrylate; ETMA, epithio-propyl methacrylate; EGDM, ethylene glycol dimethacrylate. a Data reported by Kallenberg et al. (2005). b Activity based on the wet weight of support.

534

J. Zhao et al. / Bioresource Technology 102 (2011) 529–535

bilized PGA and the conversion rate of Pen G did not change much (changing in the range of 907–1033 U/g and 1.0–1.2%, respectively) when the initial Pen G concentration was at a higher level of 254.3–417.3 mmol/L (9.4–15.4% (w/v)), showing the benefit of using the macro-mesoporous silica spheres as the supports for the immobilization of PGA. The inhibition by 6-APA was also investigated in this work. 6-APA was added into the inlet solution to simulate the accumulation of 6-APA, and the results are shown in Fig. 4. It can be seen that when the concentration of 6-APA in the inlet solution increased from 0 to 90.2 mmol/L, the apparent activity of the immobilized PGA decreased from 293 to 163 U/g, and the conversion rate of Pen G decreased from 14.1% to 11.3%. Here, the inhibitions by Pen G and PAA were relatively weak compared with the inhibition by 6-APA, and the apparent Michaelis–Menten equation could be expressed as follows:

1

v

¼

K m þ C Pen V m  C Pen

  C 6-APA; in 1þ K i; 6-APA in

G; in G;

ð4Þ

where, Ki, 6-APA is the apparent constant of 6-APA inhibition. According to the linear plot of 1/v to C6-APA, in (not shown) and Eq. (4), the value of Ki, 6-APA was determined as 109.6 mmol/L for the immobilized PGA under the experimental conditions. Pen G solution with varying concentrations of 2.7–26.8 mmol/L flowed through the column with the same residence time, and according to the linear plot of v to v/CPen G, in (not shown) and Eq. (2), the values of Km and Vm were determined as 21.8 mmol/L and 116.5 mmol/L min, respectively, for the immobilized PGA under the experimental conditions. Thus, the value of Km was about a factor of 4 lower than the value of Ki, 6-APA, and considering that the value of Km was about a factor of 16 lower than the value of Ki, Pen G, the inhibition by 6-APA was stronger than the inhibition by Pen G.

400

15

350

qE=639 mg/g =0.045 min CPen G, in=13.4 mmol/L

13

300

12

250

11

200

10 0

20

40

60

80

AIME (U/g)

XPen G (%)

14

150 100

C6-APA, in (mmol/L) Fig. 4. Influence of the accumulation of 6-APA on the performance of the immobilized PGA.

In addition, the accumulation of PAA could also influence the performance of the immobilized PGA. Besides the competitive enzyme inhibition by PAA (not investigated in this work), the accumulation of PAA can lower the pH value, which is unfavorable for obtaining a high apparent activity, and the immobilized PGA may even get deactivated if the pH value is too low. In this work, phosphate buffer with a high concentration of 0.2 mol/L was adopted to get a large buffering capacity, and there was no significant change of the pH value after the Pen G solution flowed through the packed-bed reactor. To weaken the inhibitions by 6-APA and PAA, many efforts have been devoted to in situ separating them from the hydrolysis system (Den Hollander et al., 2002; Diender et al., 2002; Wyss et al., 2004; Wang et al., 2007b). 3.4. Influence of the enzyme loading amount on the catalytic efficiency of the immobilized PGA It is known that the enzyme loading amount is one of the most important factors which should be considered for the immobilization of enzyme. This is because, generally speaking, a high enzyme loading amount is favorable for obtaining a high apparent activity of the immobilized PGA. On the other hand, according to the results of our early research (Zhao et al., 2010), a too high enzyme loading amount is unfavorable for obtaining a high specific activity of the immobilize PGA. In this work, the influence of the enzyme loading amount on the catalytic efficiency of the immobilized PGA was also investigated. The kinetic behaviors of the immobilized PGA with different enzyme loading amount were investigated, and the values of Vm and Km of the immobilized PGA were obtained according to the linear plots of v to v/CPen G, in (not shown) and Eq. (2), as shown in Table 3. It can be seen that when the immobilization time of PGA was prolonged, the enzyme loading amount increased greatly. This is because at the beginning of the covalent immobilization process, the PGA molecules were inclined to get immobilized on the external surface of the silica spheres, and with the prolonging of the immobilization time, the enzyme molecules gradually entered into the internal mesopores and got immobilized inside the silica spheres. In this process, the outer already immobilized PGA molecules could not change their position because of the presence of the covalent bond, and the other PGA molecules could only get across the already immobilized PGA molecules to get immobilized deeper in the silica spheres. And this made a long immobilization time of PGA become necessary to get a high enzyme loading amount in the covalent immobilization. However, with the increase of the enzyme loading amount, the value of Km increased, and the catalytic efficiency (kcat/Km) of the immobilized PGA decreased. And these changes might have some relationship with the distribution position of the immobilized PGA molecules. As discussed before, with a shorter immobilization time, the enzyme loading amount was lower, and the immobilized PGA molecules were distributed near the external surface of the silica spheres; with the prolonging of the immobilization time, the enzyme loading amount increased, and the immobilized PGA molecules were distributed deeper in the silica spheres, and then the internal mass transfer distance between the active sites of

Table 3 Influence of the enzyme loading amount on the catalytic efficiency of the immobilized PGA. Column

tPGA (h)

qE (mg/g)

Vm (mmol/L min)

Km (mmol/L)

kcat (L/min)

kcat/Km (L/mmol min)

1 2 3

1 3 20

256 (22.6)a 639 (31.1)a 895 (53.7)a

99.8 [11.1]b 116.5 [4.8]b 145.1 [9.4]b

19.2 [3.3]b 21.8 [1.3]b 29.7 [2.6]b

3.90 1.82 1.62

0.203 0.083 0.055

tPGA, immobilization time of PGA; kcat, turnover frequency; kcat/Km, catalytic efficiency. a The values in the round brackets: standard deviation of the experimental results. b The values in the square brackets: error of the linear fitting.

J. Zhao et al. / Bioresource Technology 102 (2011) 529–535

the immobilized PGA molecules and the external surface of the silica spheres increased. As a result, the internal mass transfer resistance for the substrate and products increased, and then the catalysis efficiency of the immobilized PGA decreased. By adjusting the immobilization time of PGA, the enzyme loading amount and the influence of the internal mass transfer resistance could be controlled, and a high catalysis effiency of the immobilized PGA could be obtained. According to this result, a high enzyme loading amount conflicted with a high catalysis effiency of the immobilized PGA. To get them compatible with each other, solid core-porous shell structured silica spheres may be a promising support for the immobilization of PGA, which may be investigated in our future work. 4. Conclusions In this study, macro-mesoporous silica spheres were used as the supports for the immobilization of PGA. The silica spheres were easier to separate and recover compared with the silica powders, and the macropores increased the enzyme loading amount (up to 895 mg/g dry support) and lowered the internal mass transfer resistance. The hydrolysis of Pen G was carried out in a packedbed reactor to determine the apparent enzymatic activity, which could reach 1033 U/g (dry support). In addition, it is found that adjusting the immobilization time of PGA could control the influence of the internal mass transfer. Acknowledgements We would like to gratefully acknowledge the financial support of the National Natural Science Foundation of China (20976096, 20490200, 20525622) and the National Basic Research Plan (2007CB714302). References Bautista, L.F., Morales, G., Sanz, R., 2010. Immobilization strategies for laccase from Trametes versicolor on mesostructured silica materials and the application to the degradation of naphthalene. Bioresource Technology 101, 8541–8548. Bianchi, D., Golini, P., Bortolo, R., Cesti, P., 1996. Immobilization of penicillin G acylase on aminoalkylated polyacrylic supports. Enzyme and Microbial Technology 18, 592–596. Bootsma, J.A., Entorf, M., Eder, J., Shanks, B.H., 2008. Hydrolysis of oligosaccharides from distillers grains using organic–inorganic hybrid mesoporous silica catalysts. Bioresource Technology 99, 5226–5231. Chandel, A.K., Rao, L.V., Narasu, M.L., Singh, O.V., 2008. The realm of penicillin G acylase in b-lactam antibiotics. Enzyme and Microbial Technology 42, 199–207. Cheng, S., Song, Q., Wei, D., Gao, B., 2007. High-level production penicillin G acylase from Alcaligenes faecalis in recombinant Escherichia coli with optimization of carbon sources. Enzyme and Microbial Technology 41, 326–330. Chong, A.S.M., Zhao, X.S., 2004a. Design of large-pore mesoporous materials for immobilization of penicillin G acylase biocatalyst. Catalysis Today 93–95, 293–299. Chong, A.S.M., Zhao, X.S., 2004b. Functionalized nanoporous silicas for the immobilization of penicillin acylase. Applied Surface Science 237, 398–404. Chouyyok, W., Panpranot, J., Thanachayanant, C., Prichanont, S., 2009. Effects of pH and pore characters of mesoporous silicas on horseradish peroxidase immobilization. Journal of Molecular Catalysis B: Enzymatic 56, 246–252. Den Hollander, J.L., Zomerdijk, M., Straathof, A.J.J., Van der Wielen, L.A.M., 2002. Continuous enzymatic penicillin G hydrolysis in countercurrent water-butyl acetate biphasic systems. Chemical Engineering Science 57, 1591–1598. Diender, M.B., Straathof, A.J.J., Van der Does, T., Ras, C., Heijnen, J.J., 2002. Equilibrium modeling of extractive enzymatic hydrolysis of penicillin G with concomitant 6-aminopenicillanic acid crystallization. Biotechnology and Bioengineering 78, 395–402. He, J., Li, X.F., Evans, D.G., Duan, X., Li, C.Y., 2000. A new support for the immobilization of penicillin acylase. Journal of Molecular Catalysis B: Enzymatic 11, 45–53.

535

Huang, H.Y., Yang, R.T., Chinn, D., Munson, C.L., 2003. Amine-grafted MCM-48 and silica xerogel as superior sorbents for acidic gas removal from natural gas. Industrial & Engineering Chemistry Research 42, 2427–2433. Janssen, M.H.A., van Langen, L.M., Pereira, S.R.M., van Rantwijk, F., Sheldon, R.A., 2001. Evaluation of the performance of immobilized penicillin G acylase using active-site titration. Biotechnology and Bioengineering 78, 425–432. Kallenberg, A.I., van Rantwijk, F., Sheldon, R.A., 2005. Immobilization of penicillin G acylase: the Key to optimum performance. Advanced Synthesis & Catalysis 347, 905–926. Katchalski-Katzir, E., Kraemer, D.M., 2000. Eupergit C, a carrier for immobilization of enzymes of industrial potential. Journal of Molecular Catalysis B: Enzymatic 10, 157–176. Kumar, A.G., Perinbam, K., Kamatchi, P., Nagesh, N., Sekaran, G., 2010. In situ immobilization of acid protease on mesoporous activated carbon packed column for the production of protein hydrolysates. Bioresource Technology 101, 1377–1379. Lü, Y.J., Guo, Y.L., Wang, Y.Q., Liu, X.H., Wang, Y.S., Guo, Y., Zhang, Z.G., Lu, G.Z., 2008. Immobilized penicillin G acylase on mesoporous silica: the influence of pore size, pore volume and mesophases. Microporous and Mesoporous Materials 114, 507–510. Lü, Y.J., Lu, G.Z., Wang, Y.Q., Guo, Y.L., Guo, Y., Zhang, Z.G., Wang, Y.S., Liu, X.H., 2007. Functionalization of cubic Ia3d mesoporous silica for immobilization of penicillin G acylase. Advanced Functional Materials 17, 2160–2166. Salis, A., Pisano, M., Monduzzi, M., Solinas, V., Sanjust, E., 2009. Laccase from Pleurotus sajor-caju on functionalized SBA-15 mesoporous silica: immobilisation and use for the oxidation of phenolic compounds. Journal of Molecular Catalysis B: Enzymatic 58, 175–180. Shah, P., Sridevi, N., Prabhune, A., Ramaswamy, V., 2008. Structural features of penicillin acylase adsorption on APTES functionalized SBA-15. Microporous and Mesoporous Materials 116, 157–165. Sio, C.F., Quax, W.J., 2004. Improved b-lactam acylases and their use as industrial biocatalysts. Current Opinion in Biotechnology 15, 349–355. Skaria, S., Rao, E.S., Ponrathnam, S., 1997. Porous thiiranyl polymers: newer supports for immobilization of penicillin G acylase. European Polymer Journal 33, 1481–1485. Sun, H., Bao, X.Y., Zhao, X.S., 2009. Immobilization of penicillin G acylase on oxirane-modified mesoporous silicas. Langmuir 25, 1807–1812. Wang, F., Guo, C., Yang, L.-R., Liu, C.-Z., 2010. Magnetic mesoporous silica nanoparticles: fabrication and their laccase immobilization performance. Bioresource Technology 101, 8931–8935. Wang, W., Deng, L., Peng, Z.H., Xiao, X., 2007a. Study of the epoxydized magnetic hydroxyl particles. Enzyme and Microbial Technology 40, 255–261. Wang, Z., Wang, L., Xu, J.-H., Bao, D., Qi, H., 2007b. Enzymatic hydrolysis of penicillin G to 6-aminopenicillanic acid in cloud point system with discrete countercurrent experiment. Enzyme and Microbial Technology 41, 121–126. Wei, D.-Z., Yang, L., 2003. Effects of ethylene glycol on the synthesis of ampicillin using immobilized penicillin G acylase. Journal of Chemical Technology and Biotechnology 78, 431–436. Wyss, A., Von Stockar, U., Marison, I.W., 2004. Production and characterization of liquid-core capsules made from cross-linked acrylamide copolymers for biotechnological applications. Biotechnology and Bioengineering 86, 563–572. Xu, X.H., Song, C.S., Andrésen, J.M., Miller, B.G., Scaroni, A.W., 2003. Preparation and characterization of novel CO2 ‘‘molecular basket” adsorbents based on polymermodified mesoporous molecular sieve MCM-41. Microporous and Mesoporous Materials 62, 29–45. Xue, P., Lu, G.Z., Guo, Y.L., Wang, Y.S., Guo, Y., 2004. A novel support of MCM-48 molecular sieve for immobilization of penicillin G acylase. Journal of Molecular Catalysis B: Enzymatic 30, 75–81. Xue, P., Xu, F., Xu, L.D., 2008. Epoxy-functionalized mesostructured cellular foams as effective support for covalent immobilization of penicillin G acylase. Applied Surface Science 255, 1625–1630. Yang, L.M., Wang, Y.J., Luo, G.S., Dai, Y.Y., 2008. Preparation and functionalization of mesoporous silica spheres as packing materials for HPLC. Particuology 6, 143–148. Zhai, Z., Chen, Y., Wang, Y.J., Luo, G.S., 2009. Chiral separation performance of micrometer-sized monodispersed silica spheres with high protein loading. Chirality 21, 760–768. Zhai, Z., Wang, Y.J., Chen, Y., Luo, G.S., 2008. Fast adsorption and separation of bovine serum albumin and lysozyme using micrometer-sized macromesoporous silica spheres. Journal of Separation Science 31, 3527–3536. Zhang, M., Shi, M., Zhou, Z., Yang, S., Yuan, Z., Ye, Q., 2006. Production of Alcaligenes faecalis penicillin G acylase in Bacillus subtilis WB600 (pMA5) fed with partially hydrolyzed starch. Enzyme and Microbial Technology 39, 555–560. Zhao, J.Q., Wang, Y.J., Luo, G.S., Zhu, S.L., 2010. Covalent immobilization of penicillin G acylase on aminopropyl-functionalized mesostructured cellular foams. Bioresource Technology 101, 7211–7217.