Microsoft Word - TOC_R.doc HUNGARIAN JOURNAL OF INDUSTRIAL CHEMISTRY VESZPRÉM Vol. 37(2) pp. 83-88 (2009) IMMOBILIZED ENZYMES AVAILABILITY FOR GLYCEROL – 1,3 PROPANEDIOL BIOCONVERSION A. BALASSY , Á. NEMETH, B. SEVELLA Budapest University of Technology and Economics, Department of Applied Biotechnology and Food Science Faculty of Chemical Technology and Biotechnology, H-1111 Budafoki street 6-8, Budapest, HUNGARY E-mail: lyra1111a@gmail.com The increasing abundance of glycerol, its renewability, and attractive pricing make it an appealing platform chemical. Glycerol can be utilized in various ways of biological transformation to industrially valuable products. There are some important derivatives among these such as 3-hydroxypropionaldehyde ((3-HPA) commercial name: reuterin), dihydroxy- acetone (DHA), and 1,3-propanediol (PDO). PDO has the largest potential need. Several bacteria are able to ferment glycerol to PDO under anaerobic conditions. A new recombinant technology was worked out with the leadership of DuPont and Genecor in the last decade, in which a recombinant Escherichia coli converts glucose directly to PDO. Microbiological production of a molecule has always lower yield than an enzymatic method and usually produces a lot of by-products that can be avoided using enzymatic bioconversion. For years we have been working on a new enzymatic process in order to produce PDO and DHA simultaneously from glycerol. For this enzymatic way three key enzymes are needed: Glycerol-dehydratase (GDHt), 1,3-propanediol-oxydoreductase (PDOR), glycerol-dehydrogenase (GDH). These enzymes must be produced by microbial fermentation. We would like to present the results of developing the fermentation media in economic aspect. Our pleriminary results (not publicated) indicated that the 2xYT medium is perfect for our Clostridium strain. However in economic aspect the concentrations of yeast extract and bactotryptone are too high to scale-up this fermentation procedure. The first purpose of this study was to optimize the fermentation media. Optimized media concentrations are yeast extract, 6 g·l-1 and bactotryptone, 2 g·l-1 (instead of 10 and 16 g·l-1 respectively). The other aim of our work was to develop immobilized enzymes suitable for glycerol to propanediol bioconversion and develop a method to measure the immobilized enzymes activity. We tried a covalent method applying chitosan matrix activated with glutaraldehyde. Chitosan – a natural polyaminosaccharide obtained from chitin – was chosen as support material to bind the three key enzymes from the crude enzyme solution. We could detect that PDOR and GDH enzyme bind to chitosan beads. Immobilized enzymes activity was higher than of the soluble enzymes. Keywords: Glycerol; 1,3-propanediol; enzyme activities; immobilization Introduction As glycerol is a nontoxic, edible, biodegradable compound, it will provide important environmental benefits to platform products. Glycerol can be converted into industrially valuable derivatives through various ways of biological transformation. There are some important derivatives among these such as 3-hydroxy- propionaldehyde (3-HPA), dihydroxyacetone (DHA) and 1,3-propanediol (PDO). PDO has the largest potential with more than 100 000 t/yr world production capacity. Two examples of large scale applications of 1,3-propanediol are SORONA made by DuPont applying BioPDO produced via biological way and CORTERRA [1], trade mark of Shell utilising terephtalic acid (PTA) and PDO produced by chemical synthesis (Fig. 1). A new recombinant technology [2] was worked out with the leadership of Dupont and Genencor in the last decade, in wich a recombinant Escherichia coli converts glucose directly to PDO. Although the fermentative PDO production of DuPont is much more environmental friendly than the chemical way, an enzymatic process would be more advantageous. It would not have an upper limit of product yield (which is also decreased by the biomass formation), and would not produce any unwanted metabolites. That is why industrial enzymatic process is usually called “Clean technology”. Figure 1: Basic unit of CORTERRA 84 Our research group has been working for several years on a process [3], where glycerol is converted into PDO and DHA simultaneously by the enzymes of Clostridium butyricum. For this enzymatic way three key enzymes are needed [4]. Glycerol-dehydratase (GDHt; E.C.4.2.1.30) that produces 3-HPA from glycerol with or without B12 as coenzyme (our Clostridium butyricum origin enzyme is B12-independent); 1,3-propanediol-oxydoreductase (PDOR; E.C.1.1.1.202) that converts HPA into 1,3-PD using NADH2. In our process the NAD+ produced by PDOR can be regenerated by a series of oxydoreductases (alcohol- dehydrogenase, formate dehydrogenase etc.), but we usually use glycerol-dehydrogenase (GDH, E.C. 1.1.1.6) which produces DHA (as the third key enzyme). Anaerobic conditions (Biebl et al. 1999). Clostridium butyricum is to our knowledge the best natural 1,3- producer from glycerol [6] and the only microorganism identified so far to use a coenzyme B12-independent glycerol dehydratase. Our pleriminary results (not publicated) indicated that the 2xYT medium described by Reynauld et al [7] is perfect for the Clostridium strain and suitable for enzyme production. However from economic aspect the concentrations of yeast extract and bactotryptone are too high to scale-up this fermentation. One of the purposes of this study was to optimize the fermentation media. Immobilization has been revealed as a very powerful tool to improve almost all enzyme properties, if properly designed, like stability, activity, specificity and selectivity, reduction of inhibition effects. Immobilized biocatalysts offer some further advantages, like the reusability of the enzymes, the availability of the product in higher purity. Chitosan is known as an ideal support material for enzyme immobilization because it is non-toxic, and not harmful to the enzymes. In addition to these, chitosan appears economically attractive since chitin is the second most abundant biopolymer in nature beside cellulose. To carry out enzymatic bioconversions, we have already used successfully a membrane reactor system retaining the enzymes [8-9]. To compare it to other immobilization technics we tried a covalent immobilization method with chitosan matrix. The other aim of our work was to develop immobilized enzymes suitable for glycerol to propanediol bioconversion. Materials and methods Organism and medium Clostridium butyricum NCIMB 8082 (the same as VPI 1718) was used as enzyme source. The strain were stored at 4 °C on Petri dish on modified Reinfored Clostridial Medium (Difco, agar: 12.5 g). Batch cultures of C. butyricum were grown anaerobically at 37 °C in 2xYT medium (yeast extract, 10 g·l-1; bactotryptone, 16 g·l-1; NaCl, 5 g·l-1), supplemented with 20 g·l-1 glycerol or glucose as carbon source [5]. However, as described under “Results”, the concentrations of yeast extract, bactotryptone, glycerol and glucose were varied. Analytical methods The cell concentrations were estimated as cell dry weight using a predetermined correlation factor (K: 1.496 g·l-1) between optical density at 600 nm (Pharmacia LKB- Ultrospec Plus spectrophotometer, Pharmacia Co., USA) and cell dry weight. Substrates and products concentrations were determined by Waters Breeze HPLC System described previously [10]. Fermentation experiments Batch cultivations are described previously [10, 12] (shown in Table 1). Samplings (in the cases of test tubes) were realized from different parallel test tubes at the corresponding timepoint, because the need of anaerobic cultivation performed in a closed ANAEROSTAT system (Merck). The growth temperature was 37 °C, the agitation speed was 200 rpm. Preparation of cell-free extracts Fermentation broths (from 1-l Bioreactor (B. Braun Biostat® Q DCU) and 2-l Bioreactor (B. Braun Biostat® M)) were centrifuged at 6000 rpm for 40 min. For resuspending the harvested cells, HEPES buffer pH = 7.4 was applied. This centrifugation/resuspendation cycle was repeated twice. Cells were disrupted by ultrasonic desintegration (Labsonic-P, 300W; Sartorius) [10]. The enzyme activity of supernatant was determined. Enzyme assays 1,3-propanediol-oxydoreductase (PDOR), glycerol- dehydratase (GDH) and glycerol-dehydratase (GDHt) activity was determined photometrically using the method described previously [13]. Immobilization method Chitosan was chosen as support material and enzyme solution including the three enzymes was applied as enzyme mixture. Support was prepared by dropping aqueous chitosan solution into 1 N NaOH solution containing 25% ethanol. Cross linking was carried out with 5% glutaraldehyde in K-phosphate buffer pH 7 (shaked in room temperature, 4 hours). 85 Immobilization of the enzymes was performed by shaking the activated chitosan beads with enzyme solution at 10 °C (pH 7.4; support/enzyme solution = 1:2). However, as described under “Results”, the time of immobilization were varied. At first chitosan beads are white, after cross-linking green and in the end of immobilization yellow. Results Fermentation process development To carry out enzymatic bioconversions a large amount of enzyme solution and high enzyme activity is needed. For this, it was neccessary to eliminate the economical bottleneck (costly media) of the scale up process and to select sufficient carbon source and pH regulation. The concentrations of yeast extract and bactotryptone are too high to scale-up this fermentation procedure. At first fermentation experiments were made in shaked test tubes and shaking flasks to determine the optimal concentrations (data shown at CHEMPOR2008 [10]). Carbon source was glucose or glycerol, pH regulated with K-phosphates or CaCO3. Beside the maximum level of bactotryptone (BT; 16 g·l-1) and two types of substrates we analyzed the effect of yeast extract (YE) concentration (Fig. 2a). Biomass production decreased significantly below 6 g·l-1 YE. While BT concentration was reduced to 4 g·l-1 at constant YE (10 g·l-1), the biomass was not fallen remarkably (Fig. 2b). 0 2 4 6 8 10 0 2 4 6 8 10 Yeast extract (g/l) X (g/l) Figure 2a: −x− substrate: 20 g/l Glucose ··x·· substrate: 20 g/l Glycerol 0 2 4 6 8 10 0 4 8 12 16 Bactotryptone (g/l) X (g/l) Figure 2b: −x− substrate: 20 g/l Glucose ···x·· substrate: 20 g/l Glycerol Further fermentation experiments resulted that the BT concentration can be reduced to 2 g·l-1 beside pH regulation with CaCO3 [13]. Optimized media concentrations are yeast extract, 6 g·l-1 and bacto- tryptone, 2 g·l-1 with glycerol as carbon source and pH regulation with CaCO3. For enzyme production we carried out several fermentations in lab scale bioreactor. 1,3-propanediol fermentation scale-up procedure results obtained on glycerol batch cultivation with optimized YE and BT are shown in Table 2. Table 1:Fermentation experiments scale-up (reference) Runs Test type YE (g·l-1) BT (g·l-1) Glu (g·l-1) Gly (g·l-1) Ferm.1 BQ (1 l) 10 16 20 20 Ferm. 2 BQ (1 l) 10 16 - 20 Runs X (g·l-1) YPD (mol·mol-1) YBP (g·g-1) J (g·l-1·h-1) Ferm.1 6.87 0.57 0.24 0.3 Ferm. 2 7.94 0.61 0.28 0.4 pH regulation: 20% NH4OH and 25% H2SO4 BQ (1 l): 1 l Bioreactor (B. Braun Biostat® Q DCU) YE: yeast extract; BT: bactotryptone; Glu: glucose; Gly: glycerol PD: 1,3-propanediol; BP: by products Table 2:Fermentation experiments scale-up (optimized) Runs Test type YE (g·l-1) BT (g·l-1) Gly (g·l-1) 1 SF(400) 6 2 20 2 BQ (1 l) 6 2 20 3 BM (2 l) 6 2 20 Runs X (g·l-1) YPD (mol·mol-1) YBP (g·g-1) J (g·l-1·h-1) 1 4.99 0.90 0.42 0.7 2 7.08 0.99 0.34 0.5 3 8.99 0.96 0.28 0.9 pH regulation: CaCO3, 15.489 g·l -1 SF: shaking flask BQ (1 l):Bioreactor (B. Braun Biostat® Q DCU) BM (2 l): 2 l Bioreactor (B. Braun Biostat® M) YE: yeast extract; BT: bactotryptone; Gly: glycerol PD: 1,3-propanediol; BP: by products All the features of optimized fermentations are better than results of reference fermentations shown in Table 1. Reference enzyme activities from Ferm. 1. and Ferm. 2. are shown in Table 3. Enzyme activities and protein concentrations from Run 3 are shown in Fig. 3. Table 3: Enzymes from reference fermentation REFERENCE Ferm 1. Ferm 2. C, protein (mg/ml) 1,16 1.62 PDOR (U/ml enzyme sol.) 0.25 0.26 GDH (U/ml enzyme sol.) 0.33 0.026 GDHt (U/ml enzyme sol.) 0.63 0.31 sol.: solution 86 PDOR activity was 10-times higher than the reference (2.64 U·ml-1 enzyme solution; and 0.25–0.26 U·ml-1 enzyme solution, respectively). GDH activity was not detected from the optimized fermentation broth (reference: 0.33 and 0.026 U·ml-1 enzyme solution; carbon source glucose and glycerol or only glycerol). However GDH is of commercial trade, thus it can be a supplement for the crude enzyme solution. 1,85 2,64 0 0,14 0 0,5 1 1,5 2 2,5 3 C, protein (mg/ml) PDOR (U/ml enzyme sol.) GDH (U/ml enzyme sol.) GDHt (U/ml enzyme sol.) Figure 3: Enzyme activities and protein concentration from optimized fermentation Enzymes immobilization To carry out enzymatic bioconversions, we already used a membrane reactor system retaining the enzymes [8]. To compare it to other immobilization technics we tried a covalent immobilization method with chitosan matrix. Chitosan was chosen as support material and crude enzyme solution from optimized fermentation broth including PDOR and GDHt enzymes supplemented with commercial GDH was applied as enzyme source. Immobilization was carried out as described in “Materials and methods”. The enzyme activity measurement method we developed for immobilized enzyme preparations was reported at Younghem2008 Conference [11]. At first NAD+-dependent Glucose dehydrogenase (GlcDH, E. C. 1.1.1.47) from Bacillus megaterium was used as a test enzyme (it has commercial trade). After the process was successfully adapted, the enzyme was changed to the two key enzymes of glycerol metabolism (NAD+-dependent PDOR and GDH). The production of NADH was monitored by the increase of absorbance at 340 nm and 25 °C. Firstly we used one bead in a cuvette (without shaking). In this case we couldn’t detect enzyme activity (data shown Fig. 4). As a next step we used not one but ten chitosan beads (with shaking), and measured the absorbance during the reaction in discrete points from given portion of reaction mixture (data shown Fig. 5). After volumetric correction the line gradient determined the NADH production rate and the immobilized enzyme activity (data shown Fig 6). -0,15 -0,1 -0,05 0 0,05 0,1 0,15 0,2 0 200 400 600 Time (s) A340 Figure 4: Immobilized enzyme activity (one bead) - x -: background absorbance - ♦ -: NADH absorbance 0 0,4 0,8 1,2 1,6 0 5 10 15 20 25 30 Time (min) A340 Figure 5: Immobilized enzyme activity (ten beads) - x -: background absorbance - ♦ -: NADH absorbance y = 5,6895x - 59,402 R2 = 0,9719 0 40 80 120 160 0 5 10 15 20 25 30 Time (min) A340,MOD Figure 6: Immobilized enzyme activity (after volumetric correction) - ♦ -: NADH absorbance Soluble enzyme activity (U·ml-1) was measured before and after immobilization. Enzyme activity on chitosan beads (U·gl-1) was measured in the end of immobilization. The effect of immobilization time was determined. Relative activity (%) of immobilized PDOR was measured at various immobilization time (1.5; 3; 4.5 and 24 hours; data shown Fig. 7). After 24 hours the relative PDOR enzyme activity decreased to 25 % of the original activity. 1.5 hours semms to be enough for immobilization. After 1.5 hours imobilization we realised that the immobilized PDOR enzyme activity was 100 times higher and GDH enzyme activity was 160 times higher than the soluble enzyme activity (primary activity; data shown Fig. 8). 87 Relative immobilized PDOR activity (%) 0 25 50 75 100 1,5 3 4,5 24 Time (hour) Figure 7: Effect of immobilization time 0,07 10,49 0,113 12,218 0 4 8 12 16 Primary activity (U/ml) Immobilized enzyme activity (U/g) PDOR GDH Figure 8: PDOR and GDH enzyme activity Binding of the third enzyme, GDHt to chitosan beads applying the conditions mentioned above has not been proven, yet. Only the activity of the soluble enzyme was detectable after the immobilization procedure (data shown Fig. 9 and 10).This shows that in case of GDHt the applied immobilization process did not assured a certain binding, the process needed further optimization. 0 20 40 60 80 100 0 1,5Time (hour) Relative GDHt activity (%) GDHt controll GDHt residual Figure 9: GDHt enzyme activity (experiment 1) 0 20 40 60 80 100 0 1,5Time (hour) Relative GDHt activity (%) GDHt controll GDHt residual Figure 10: GDHt enzyme activity (experiment 2) Conclusion Using Clostridium butyricum as an enzyme source our preliminary results indicated that the 2xYT medium is perfect for the cultivation of the bacterium. However from economic aspects the concentrations of organic components are too high to scale-up this fermentation process. Examining the culture medium resulted in an optimal composition: 6 g·l-1 yeast extract and 2 g·l-1 bactotryptone with glycerol as carbon source and pH regulation with CaCO3. To carry out enzymatic bioconversions a large amount of enzyme solution and high enzyme activity is needed. Optimized PDOR activity was 10-times higher than the reference (2.64 U·ml-1 enzyme solution; and 0.25–0.26 U·ml-1 enzyme solution, respectively). Immobilization has been revealed as a very powerful tool to improve almost all enzyme properties. The effect of immobilization time was determined. Too long immobilization time caused significant activity lost, 1.5 hours seemed to be enough for immobilization of the PDOR enzyme. Applying the the 1.5 hour imobilization process the immobilized PDOR enzyme activity was 100 times higher and GDH enzyme activity was 160 times higher than that of the soluble enzyme. Binding the third enzyme, GDHt to chitosan beads has not been proven, yet. REFERENCES 1. BIEBL H., MENZEL K., ZENG A. P., DECKWER W. D.: Microbial production of 1,3-propanediol. Applied Microbiology and Biotechnology (1999) 52, 289– 297 2. Dupont Tate & Lyle Bio Products begins bio- propanediol production. (2007) DOI: 10. 1016/S1351-4210(07)70046-X Focus on Surfactants, 2, 4-5 3. SEVELLA B., KUPCSULIK B., NÉMETH Á., NOVÁK L., POPPE L., DUKAI J., NAGY F.: P 05 00961 (2005) 4. PAPANIKOLAOU S., RUIZ-SANCHEZ P., PARISET B., BLANCHARD F., FICK M.: High production of 1,3- propanediol from industrial glycerol by a newly isolated Clostridium butyricum strain. Journal of Biotechnology (2000) 77, 191–208 5. SAINT-AMANS S., PERLOT P., GOMA G., SOUCAILLE P.: High production of 1,3-propanediol from glycerol by Clostridium butyricum VPI 3266 in a simply controlled fed-batch system. Biotechnology Letters (1994) 16, 831–836. 6. BIEBL H., MENZEL K., ZENG A. P., DECKWER W. D.: Microbial production of 1,3-propanediol. Applied Microbiology and Biotechnology (1999) 52, 289–297 88 7. REYNAUD C., SARC. P., MEYNIAL-SALLES I., CROUX C., SOUCAILLE P.: Molecular characterization of the 1,3-propanediol (1,3-PD) operon of Clostridium butyricum. Applied Biological Sciences (2003) 100(9), 5010–5015 8. NÉMETH Á., SEVELLA B.: Development of a New Bioprocess for 1,3-propandiol Production I.: Modeling of Glycerol bioconversion to 1,3- propanediol with Klebsiella pneumoniae Enzymes. Applied Biochemistry and Biotechnology (2008) 144(1), 47–58. 9. NÉMETH Á., BALÁSSY A., SEVELLA B.: Difficulties and solutions for the assays of the key enzymes of a new enzymatic glycerol bioconversion. Periodica Polytechnica Chemical Engineering (2008) 52, (1), 1–6. 10. BALÁSSY A., NÉMETH Á., SEVELLA B.: A New Enzymatical Process for Glycerol Utilization 10th International Chemical and Biological Engineering Conference (CHEMPOR2008), 2008 11. BALÁSSY A., NÉMETH Á., SEVELLA B.: Enzyme immobilization for utilization of biodiesel byproduct glycerol 6th International Congress of Young Chemists (Youngchem2008) 12. BALÁSSY A., NÉMETH Á., SEVELLA B.: Enzyme production with Clostridium butyricum IET BioSysBio (Synthetic Biology, Systems Biology and Bioinformatics) Conference Cambridge, 2009 13. NÉMETH Á., KUPCSULIK B., SEVELLA B.: 1,3- Propanediol oxidoreductase production with Klebsiella pneumoniae DSM2026. World Journal of Microbiology and Biotechnology (2003) 19, 659–663.