Enzyme immobilisation using siliceous mesoporous molecular sieves

Enzyme immobilisation using siliceous mesoporous molecular sieves

Microporous and Mesoporous Materials 44±45 (2001) 763±768 www.elsevier.nl/locate/micromeso Enzyme immobilisation using siliceous mesoporous molecula...

205KB Sizes 2 Downloads 92 Views

Microporous and Mesoporous Materials 44±45 (2001) 763±768


Enzyme immobilisation using siliceous mesoporous molecular sieves Humphrey H.P. Yiu, Paul A. Wright *, Nigel P. Botting School of Chemistry, University of St. Andrews, The Purdie Building, St. Andrews, Fife, KY16 9ST Scotland, UK Received 16 February 2000; accepted 12 April 2000

Abstract The use of mesoporous molecular sieves in enzyme immobilisation has been studied. Three di€erent types of mesoporous sieves (MCM-41, MCM-48 and SBA-15) were selected because of the di€erences in their pore dimensions and structures. Commercially available porous silica gel was chosen for comparison. The model enzyme chosen in this study was trypsin. The samples of immobilised trypsin were active for the hydrolysis of N-a-benzoyl-D L -arginine-4-nitroanilide (BAPNA). The amount of enzyme adsorbed on the supports was found to be related to the pore size of the molecular sieves. Moreover, the pore size and the structure of the support also a€ected the activity of the supported enzymes. Ó 2001 Elsevier Science B.V. All rights reserved. Keywords: Enzyme immobilisation; Trypsin; Mesoporous molecular sieves; SBA-15; MCM-41

1. Introduction In 1992, a new class of mesoporous material M41S was developed by the Mobil research group [1,2]. This type of material possesses high surface area (700±1000 m2 g 1 ), high pore volume (approximately 1.0 ml g 1 ) and a well ordered pore structure. The unique structural properties of M41S materials were thought to provide excellent supports for catalysts [3], such as metal oxides, organometallic compounds and even enzymes. Immobilisation of enzymatic catalysts using M41S type materials was studied by Dõaz and Balkus [4]


Corresponding author. Fax: +44-1334-463-808. E-mail address: [email protected] (P.A. Wright).

and it was found that the eciency of immobilisation was dependent on the molecular size of enzyme. The larger the enzyme molecule, the lower the eciency. In 1998, Zhao et al. [5] reported a route for synthesising mesoporous molecular sieves with pore sizes considerably larger than traditional MCM-41, that still possess a highly ordered hexagonal array of channels, which involved the use of a coblock polymeric surfactant as template. The  should make increase in pore size up to 300 A these materials more suitable for enzyme immobilisation. In this work, we demonstrate the use of large pore mesoporous molecular sieves in enzyme immobilisation. The activity of immobilised enzymes was tested using a common assay of trypsin activity.

1387-1811/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 7 - 1 8 1 1 ( 0 1 ) 0 0 2 5 8 - X


H.H.P. Yiu et al. / Microporous and Mesoporous Materials 44±45 (2001) 763±768

2. Experimental 2.1. Materials All chemicals were used as received without further puri®cation. Silica gel (Fisher) used was 30±70 mesh chromatographic silica gel with aver surface area 600 m2 /g. age pore size of 40 A, 2.2. Preparation of pure siliceous MCM-41 Pure siliceous MCM-41 was prepared based on the procedure reported by Cheng et al. [6]. A mixture of an aqueous solution of 25% tetramethylammonium hydroxide (TMAOH, 6.92 g, Aldrich), cetyltrimethylammonium bromide (CTMABr, 9.84 g, 98%, Aldrich) and deionised water (67 g) was stirred at 30°C until it turned clear. Fumed silica (6.0 g, Sigma) was then dissolved in the mixture and stirred for 2 h at room temperature. The molar ratio of the gel was 1 SiO2 : 0.27CTMABr:0.19TMAOH:40H2 O. The gel was then allowed to age for 24 h on the bench before being transferred to a stainless steel pressure autoclave and heated to 150°C for 48 h. The precipitate was ®ltered before being washed with distilled water and dried in air. The dry precipitate was calcined at 550°C for 4 h under ¯owing nitrogen and the calcination procedure was repeated under ¯owing oxygen. 2.3. Preparation of MCM-48 The preparation procedure of MCM-48 is similar to previous publications [7]. The molar ratio of the gel was 1SiO2 :0.65CTMABr:0.5NaOH: 165H2 O. Tetraethyl orthosilicate (TEOS, Aldrich) was used as silica source. The gel was stirred in a polypropylene bottle at room temperature for 1 h and then heated to 100°C for 4 days. The solid was ®ltered, air-dried and calcined at 550°C under ¯owing nitrogen for 4 h. Finally, the calcination procedure was repeated under ¯owing oxygen. 2.4. Preparation of SBA-15 Large pore mesoporous silicate SBA-15 was prepared based on the procedure reported by

Zhao et al. [5]. Poly(alkylene oxide) block copolymer surfactant Pluronic P-123 (BASF) was used as template. The molar ratio of the gel was 1 SiO2 :0.017Pluronic P-123:2.9HCl:202.6H2 O. TEOS was used as silica source. The surfactant was dissolved in distilled water and followed by the addition of hydrochloric acid (35% in water, Fluka). The solution was stirred for an hour. TEOS (98%, BDH) was added dropwise to the reaction solution and stirred for further 24 h. The reaction mixture is then transferred into a sealed polypropylene bottle and heated at 100°C for 48 h. The white precipitate was ®ltered, washed with distilled water and air dried. The dry precipitate was calcined at 550°C for 4 h under ¯owing nitrogen and the calcination procedure was repeated under ¯owing oxygen. Addition of mesitylene (98%, Aldrich) into the surfactant solution equal to 20 wt.% of surfactant was introduced to increase the pore size of SBA15. The sample is named SBA-15M2 [5]. 2.5. Characterisation of mesoporous materials The ordered structure of the mesoporous molecular sieves was analysed on a Philips XÕPert System di€ractometer using CuKa radiation …k ˆ  The scans were taken from 2h ˆ 1:5° 1:5418 A†. to 8°. The BET surface area and pore size of calcined MCM-41 and calcined MCM-48 were measured by nitrogen adsorption and desorption at 196°C using a Coulter Omnisorp 100CX instrument. For calcined SBA-15, the adsorption and desorption of nitrogen was measured on a Micromeritics ASAP2010 instrument (in collaboration with University of Edinburgh). The pore size of these samples was calculated by the BJH method. A Philips CM20 200 kV microscope was used to record the TEM micrographs of SBA-15 and SBA15M2 materials. 2.6. Immobilisation of enzymes The enzyme immobilisation method was based on that of Dõaz and Balkus [4]. Trypsin from bovine pancreas (Sigma, EC was used. Solid support (0.25 g) was suspended in 5.0 ml of 15 lM

H.H.P. Yiu et al. / Microporous and Mesoporous Materials 44±45 (2001) 763±768


Scheme 1. BAPNA hydrolysis.

trypsin solution in pH 6.0 Tris±HCl bu€er (50 mM, Sigma) for 2 h at 4°C under stirring. The supernatant was separated from the solid materials by centrifugation. The amount of enzyme immobilised on the solid supports was determined using the Bradford assay for protein concentration [8].

pH ˆ 8 buffer). A 1.0 cm3 sample was taken every 15 min. The supernatant absorbance at 405 nm was recorded after the sample has been centrifuged for 5 min at 7000 rpm. The yellow colour of the supernatant is due to the formation of p-nitroaniline.

2.7. Bradford assay for protein content For every sample, including standard solutions, 0.10 cm3 of the supernatant was added to 1.0 cm3 Bradford reagent (Sigma). After 30 min, the protein content in the supernatant was measured from the absorbance at 595 nm. A calibration using standard protein solution (from 0 to 100 lg cm 3 ) was carried out for calculating the protein content in solution. The decrease in this absorbance was calculated as the amount of enzyme adsorbed on the support surface.

3. Results and discussions 3.1. Characterisation of mesoporous molecular sieves The X-ray di€raction patterns of MCM-48, MCM-41 and SBA-15 are depicted in Fig. 1. The unit cell dimensions of these materials were found  ahex…MCM-41† ˆ 50:26 A  to be acub…MCM-48† ˆ 86:15 A,  and ahex…SBA-15† ˆ 112:42 A.

2.8. Test for leaching of enzyme from supports For every sample of supported enzyme (0.175 g), 14 cm3 of Tris±bu€er solution (50 mM, pH ˆ 8) was added and stirred at 25°C for 2 h. The protein content of the solution was measured using the Bradford assay as previously mentioned. The percentage leaching of enzyme from the supports was then calculated from these protein content data. 2.9. Enzymatic activity measurement Hydrolysis of N-a-benzoyl-D L -arginine-4-nitroanilide (BAPNA, Sigma, see Scheme 1) was applied to assess the activity of the immobilised enzymes. For each activity measurement, all 0.25 g of supported enzyme sample was suspended in 20 cm3 of 1 mM BAPNA solution (prepared in a

Fig. 1. X-ray di€raction pattern of three mesoporous molecular sieves from 2h ˆ 1:5±8°.


H.H.P. Yiu et al. / Microporous and Mesoporous Materials 44±45 (2001) 763±768

Table 1 Physical properties of mesoporous molecular sieves MCM-41, MCM-48 and SBA-15 Sample

Pore diameter  (A)

BET surface area (m2 g 1 )

Pore volume (cm3 g 1 )

MCM-41 MCM-48 SBA-15

35 24 56

1069 1206 918

1.28 1.16 1.15

The BET surface area and pore size of calcined MCM-41, MCM-48 and SBA-15 samples are shown in Table 1. Two di€erent types of image were discerned for the SBA-15 sample using conventional bright ®eld TEM. Fig. 2 shows an ordered hexagonal mesoporous structure. From the TEM micrographs, the pore diameter of the SBA-15 sample was measured  The addition of mesitylene during to be 80 A. preparation (SBA-15M2) increased the pore size of  However, the channels became SBA-15 to 100 A. curled as shown in Fig. 2(b). 3.2. Quantifying the enzyme adsorption capacity of supports and the leaching test Table 2 shows the percentage of trypsin adsorbed on the supports and determined using the Bradford assay. The results from the test for leaching are also included. All supports adsorbed more than 90% of the trypsin. However, 35±72% of the trypsin was observed to leach out from the support after stirring in a bu€er solution …pH ˆ 8† for 2 h. The area coverage estimated for these amounts of trypsin was less than 1% of the BET surface area of the supports. Therefore, it was unlikely that the surface of the supports were saturated with trypsin molecules. This indicated that the leaching of trypsin was due to the weakness of the interaction between the trypsin molecules and the surface of the supports. MCM-48 showed the highest amount of leaching. The  appears small pore diameter of MCM-48 (24 A) not to be large enough for trypsin molecules  [4]) to enter the struc(spherical diameter ˆ 38 A ture. The pore diameter of MCM-41 was measured  which is slightly smaller than the to be 35 A, trypsin molecules. The decrease in leaching could

Fig. 2. TEM images of (a) SBA-15 and (b) SBA-15M2. SBA-15 showed a highly ordered hexagonal array of straight, parallel channels. SBA-15M2 also showed an ordered hexagonal array of channels but these channels were curled as shown.

indicate that trypsin molecules were able to di€use into the channels of MCM-41. In the case of larger pore SBA-15 and SBA-15M2, trypsin molecules should have little diculty in entering the channels. 3.3. Catalytic activities of immobilised enzymes As mentioned in the previous section, a considerable amount of leaching from each support was found. These leached trypsin molecules should

H.H.P. Yiu et al. / Microporous and Mesoporous Materials 44±45 (2001) 763±768 Table 2 The percentage of trypsin found in the supernatant solution after immobilisation process and the % of trypsin leached from support during the leaching test  % Residue Support Pore size (A) % Leached

MCM-41 MCM-48 SBA-15 SBA-15M2 Silica gel

30 24 56 100b 40c


from support

8 0a 0a 0a 0a

46 72 52 35 41


Under the measureable limit. Measured from the TEM micrograph. c Pore diameter value from the manufacturer. b

be active in hydrolysis of BAPNA. However we found a signi®cant di€erence in activity between trypsin on MCM-48 and trypsin on silica gel (see Fig. 3). This might be explained by the possibility of BAPNA molecules being strongly adsorbed inside the channels of MCM-48. As the trypsin molecules were too large to get inside these small channels, the BAPNA molecules were e€ectively separated from the enzyme. A low activity was therefore recorded. The three-dimensional channel system of MCM-48 was thought to be excellent

Fig. 3. The measured product yield of p-nitroaniline in the solution from the hydrolysis of BAPNA using immobilised trypsin. Data is shown for trypsin on SBA-15 (), SBA-15M2 (s), MCM-41 (m) and MCM-48 ( ). The product yield of pnitroaniline of trypsin on silica gel (- - -) is much higher.


feature as a support for catalysts [9], but, in our case, such an advantage was outweighed by the small channel size. On increasing the channel size, trypsin on MCM-41 showed a signi®cant improvement in terms of activity. A further increase in the channel size of the support (trypsin supported on SBA-15) produced a higher activity than trypsin on MCM41. This can be accounted for by the improved accessibility of the substrate to the active site of enzyme molecules. The large pores of SBA-15 allow faster di€usion for both substrate and products. Trypsin on MCM-41 was ®rst studied by Balkus [4]. The problem associated with leaching was also mentioned in that report, and use of 3-aminopropyltriethoxysilane to close the channel ends of MCM-41 was suggested to minimise leaching of enzyme. The use of 3-aminopropyltriethoxysilane would be expected to be less successful for SBA-15 as the channel size is more than twice that of MCM-41. Therefore, the decrease in channel size may not be enough to stop the enzyme from leaching. Notably, the increased pore size in SBA-15M2 sample did not improve the activity of supported trypsin. The curled pore structure of SBA-15M2 might be less e€ective for substrate and product di€usion, so that the straight channel structure of SBA-15 could be more suitable for enzyme immobilisation. Trypsin supported on porous silica gel showed the highest activity. Although the pore size of the  on average, the external silica gel used was 40 A surface area of this material is considerably higher than other mesoporous molecular sieves. For example, the pore surface area of MCM-41 is around 95% of its total surface area. In case of using silica gel to support enzyme, a relatively large proportion of enzyme would be supported on the non-porous surface compared with the enzyme supported on mesoporous molecular sieves. Furthermore, since very low loadings of enzymes are being used in this study, there is still a large unoccupied surface area on the support. It is possible that products remain adsorbed on the support surface, which would distort the activity measurement. The lower activity measured from trypsin on mesoporous molecular sieves than trypsin on silica


H.H.P. Yiu et al. / Microporous and Mesoporous Materials 44±45 (2001) 763±768

gel could therefore also be explained by the lower overall surface area of the silica gel.

4. Conclusions This work shows large pore mesoporous molecular sieves such as SBA-15 have a higher potential in immobilising enzymes than traditional MCM-41 materials although both have a similar hexagonal pore array structure. The increased pore size can resolve the problems associated with poor di€usion and pore blocking. From this point of view, SBA-15 can also be used in supporting other, larger molecules, which MCM-41 cannot achieve. However, the e€ects of adsorption for the products are still need to be examined. A di€erent solvent system can be bene®cial in the study of this direction. Although not shown in this work, the high surface areas of mesoporous molecular sieves mean that more enzyme molecules could be supported on a ®xed amount of solid to increase activity. Moreover, the narrow pore size distribution of mesoporous molecular sieves can in principle provide higher shape selectivity towards smaller products, whereas externally supported enzyme may be less shape selective for certain classes of reaction. Leaching of enzyme during applications is still a major problem. This should be minimised by surface modi®cation or covalent attachment of the

enzyme to the solid. Further study in this direction is being carried out. Acknowledgements The authors would like to thank EPSRC for funding the project, Dr. C. Park of University of Leeds for TEM images and Dr. R. Brown of Edinburgh for nitrogen adsorption isotherm of SBA15 sample. Thanks are also due to Dr. D.H. Fitzgerald of St. Andrews for her advice on enzyme characterisations. References [1] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.-W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, J. Am. Chem. Soc. 114 (1992) 10834. [2] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710. [3] A. Corma, Chem. Rev. 97 (1997) 2373. [4] J.F. Dõaz, K.J. Balkus Jr., J. Mol. Catal. B 2 (1996) 115. [5] D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120 (1998) 6024. [6] C-F. Cheng, D.H. Park, J. Klinowski, J. Chem. Soc. Faraday Trans. 93 (1997) 193. [7] Q. Huo, D.I. Margolese, G.D. Stucky, Chem. Mater. 8 (1996) 1147. [8] M.J. Dunn, in: E.L.V. Harris, S. Angel (Eds.), Protein Puri®cation Methods ± a Practical Approach, Oxford University Press, Oxford, 1994, pp. 17±18. [9] M.S. Morey, A. Davidson, G.D. Stucky, J. Porous. Mater. 5 (1998) 195.