A seguir serão apresentados dois artigos derivados da tese, submetidos para periódicos na área.
O primeiro artigo intitulado “Recovery and Purification of Chitosanase produced by
Bacillus cereus using Expanded bed adsorption (EBA) and Central Composite Design (CCD)”
foi submetido ao periódico Separation Science and Technology, e o segundo “Single-step purification of chitosanase from Bacillus cereus using expanded bed chromatography” foi submetido ao Journal of Cromathography B.
Recovery and Purification of Chitosanase produced by Bacillus cereus using Expanded Bed Adsorption (EBA) and Central Composite Design (CCD)
Nathália Kelly de Araújo*, Maria Luiza Oliveira Xavier, Vanessa Carvalho Pimentel, Carlos Eduardo de Araújo Padilha, Nayane Macedo Portela da Silva, Gorete Ribeiro de Macedo, Everaldo Silvino dos Santos.
*Corresponding author at: Laboratório de EngenhariaBioquímica, Departamento de Engenharia Química, Universidade Federal do Rio Grande do Norte – UFRN/Brazil. Phone: +55 84 3215 3769 R: 214. Fax: +55 84 3215 3770. E-mail address: nakar_rn@hotmail.com (N. K.Araújo).
Abstract
This study presents a system for EBA to purification of chitosanase from broth extract in a single step. A chitosanase-producing strain was isolated and identified as Bacillus cereus C- 01. Statistical analyses based showed that distributor height had strong influence on the process affecting considerably both purification factor (P) and enzyme yield (Y). The optimization of purification variables resulted in an approximately 3.66-fold increase in P compared with that observed under not optimized conditions. This system is promising for recovery of chitosanase from B. cereus C-01 and it is economically viability since it promotes the reduction steps.
Keywords: Biocatalysis; Enzyme Activity; Chitosanase; Bioseparations; Chromatography; Expanded bed adsorption.
INTRODUCTION
Derived from the hydrolysis of Chitosan, the Chitooligosaccharides (COS) have several biofunctional properties, including anti-inflammatory [1], prebiotic [2], antimicrobial [3], antitumor [4, 5], and immunopotentiating effects [6]. Thereby, the nutraceutical and medical industries have been given special attention to production of chitooligosaccharides from chitosan [7].
The depolymerization of chitosan is mainly conducted using acid hydrolysis and enzymatic degradation. The acid hydrolysis is not usually employed due to the complexity of control of the reaction. Moreover the use of acid promotes in the formation of secondary compounds that are elimination difficult [8]. Enzymes exhibit high specificity and operate under mild conditions. Additionally, the use of enzymes allows control of reaction and product formation by adjusting the concentration of the enzyme, pH, temperature and reaction time. Thus, enzyme technology probably is the most promising approach [9]. Several enzymes can hydrolyze the chitosan; in this process are employed specific (chitosanases) and non-specific enzymes (carbohydrases, proteases, lipases etc.) [10].
For applications requiring high purity (use in therapy and diagnosis), enzyme purification may be necessary. The chromatography process conventional to remove product and impurities are usually designed with steps of clarification, precipitation with salts and a series of capture and polishing [11]. Expanded bed adsorption (EBA) is an integrated process which combines solid-liquid separation, volume reduction by protein adsorption and recovering the product in a single unit operation [11, 12]. This technique has been successfully employed to biomolecules purification [13-16].
Optimization of the purification process can be performed using experimental design. The estimate of the influential variables during the adsorption process can be determined by correct selection of design and optimization models. The experimental design methodology can be used to determine the effects of individual factors and their interactions [17]. Box and Wilson developed a technique named is Central composite design (CCD) which use response surface methodology. An advantage of CCD is the reduction of number of experiments in the studies with a wide number of factors and levels [18]. Furthermore, by use of response surface methodology (RSM), the main and interaction effects of variables can be explored and optimal value of each variable can be definite [19].
The present study attempted to optimize the process of chitosanase purification by EBA using a CCD with attention to both the purification factor (P) parameters as well as
enzyme yield (Y). A batch shake-flask culture of Bacillus cereus was used to enzyme production.
MATERIALS AND METHODS Materials
Chitosan (85 % deacetylated; molecular weight, 90–190 kDa) acquired from Sigma- Aldrich Co. (St. Louis, Missouri, USA) was solubilized using 0.1 M HCl [20]. Streamline DEAE resin was purchased from GE healthcare (Uppsala, Sweden). PC33 Siemens Kit was provided by Maria Alice Hospital, Natal/Brazil. Buffer solutions and other chemicals were reagent grade.
Isolation and identification of chitosanolytic bacterium
The microorganism was isolated from soil samples (Natal/Brazil, S 05º52’11’’, Wo 35º13’08.4’’). Serially diluted soil samples were inoculated on plates containing peptone (6.0 gL-1), chitosan (2.0 gL-1), Magnesium sulfate (0.5 gL-1), Potassium dibasic phosphate (1.0 gL- 1) (basic medium) and agar (15 gL-1) and incubated at 30 ºC for 2 days. A single colony showing prominent chitosanolytic activity was selected and subjected to taxonomic analysis, Gram-staining and pathogenicity test with PC33 Siemens Kit. To further identify the bacterium, polymerase chain reaction (PCR) was performed to amplify part of the bacterial 16S rRNA gene. The forward and reverse primers were 27F (5’-AGA GTT TGA TCC TGG CTC AG-3’) and 1525R (5’-AAG GAG GTG ATC CAG CC-3’), respectively. The PCR products were analyzed by electrophoresis on agarose gel and after they were purified, quantified and used in sequencing reaction. Four sequencing reactions were performed for each sample using the primers 518F (5’-CCA GCA GCC GCG GTA ATA CG-3’), 800R (5’-TAC CAG GGT ATC TAA TCC-3’) and 1492R (5’-TAC GGY TAC CTT GTTA CGA CTT-3’). The nucleotide sequence of the 16S rRNA gene of C-01 was determined by an ABI 3730 DNA Sequencer (Applied Biosystems, USA) and compared with published 16S rRNA sequences in a NCBI BLAST search.
Chitosanase production
The cells from the stock cultures were transferred to 50 mL aliquots of pre-cultivation medium dispensed into 250 mL Erlenmeyer flasks, then incubated at 120 rpm for 30 h at 30 ºC. The pre-cultivation medium consisted of (g.L−1): peptone 6.0, chitosan 2.0, Magnesium sulfate 0.5 and Potassium dibasic phosphate 1.0. This was then aseptically inoculated at a level of 10% (v/v) into a 250 mL Erlenmeyer flask containing 50 mL of same medium. The
flasks were then placed on a rotary shaker at 120 rpm for 24 h at 30 ºC (Tecnal, TE421). Samples of this cultivation were used for purification assay.
Analytical methods
The chitosanase activity of the crude enzyme and chromatography samples were measured via the incubation of 250 μL of enzyme sample with 250 μL of 1% chitosan solution (pH 6.0). The mixture was heated in a water bath for 30 minutes, at a temperature adjusted to 55 ºC. The reaction was interrupted by boiling for 10 minutes [20]. The formation of reducing sugars was analyzed in spectrophotometer (Thermo Spectronic) using the 3,5-Dinitrosalicylic acid method, using D-glucosamine as standard [21]. A unit (U) of chitosanase was defined as the amount of enzyme capable of generating 1 µmol of D- glucosamine per minute under the conditions above described.
The protein content was determined using the Bicinchoninic Acid method with a BCA Protein Assay Kit (Pierce, Rockford, IL) and Bovine Serum Albumin (BSA) as standard. Experimental Designs
The optimization strategy that was used is based on the concept of simultaneous variation of possible influencing factors (independent variable) on the quality criteria (dependent variable) of the system considered. This strategy was used to evaluate the influence of three factors on the enzyme purification. Dependents variable were defined on the quality criteria purification factor (P) and enzyme yield (Y) to optimize the chromatographic process:
𝑃 =𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑝𝑢𝑟𝑖𝑓𝑖𝑒𝑑 𝑠𝑎𝑚𝑝𝑙𝑒𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑐𝑟𝑢𝑑𝑒 𝑠𝑎𝑚𝑝𝑙𝑒 Eq.1
𝑌(%) =𝐸𝑛𝑧𝑦𝑚𝑎𝑡𝑖𝑐 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑝𝑢𝑟𝑖𝑓𝑖𝑒𝑑 𝑠𝑎𝑚𝑝𝑙𝑒𝐸𝑛𝑧𝑦𝑚𝑎𝑡𝑖𝑐 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑐𝑟𝑢𝑑𝑒 𝑠𝑎𝑚𝑝𝑙𝑒 Eq.2 A CCD was employed to examine the combined and individual effect of three independent variables for chitosanase purification. The CCD is usually characterized by three operations: 2k axial runs, 2k factorial runs and central point runs [19]. The number of experiments (N), represented by Equation 3, depends on how many factors (k) are in the CCD:
N=2k + 2k + central points Eq. 3
The CCD in this study was designed based in 6 axial points, 8 factorial points and 3 replicates at the center. Initially, it was constructed the factorial and central parts, followed by the axial one. The quantitative factors were load flow velocity (X1), settled bed height (X2)
and distributor height (X3) utilized as independent variables for the CCD. The experimental design with coded and uncoded values is presented in Table 1.
Table 1: Matrix of CCD to coded and uncoded level of independent variables.
Variables
Levels
-α -1 0 +1 +α
Load flow velocity (cm/h) 66 100 150 200 234
Settled bed height (cm) 2.64 4.0 6.0 8.0 9.36 Distributor height (cm) 0.64 2.0 4.0 6.0 7.36
From a second-order polynomial equation is possible to explain the behavior of each variable, their interactions, and statistical analysis and acquire predicted responses [7]. The following second-order polynomial Equation was used:
Y = b0 + ∑ bixi + ∑bijxixj + ∑biix2i Eq.4
Where Y is the predicted response, b0 is the model constant, bi is the linear
coefficients, bii are the quadratic coefficients and bij is the interaction coefficients. Xi
represents the independent variables (known for each run). The experimental design analysis and their subsequent regression analysis were evaluated using STATISTICA 7.0 (StatSoft, USA). The analysis of variance (ANOVA) was also carried out. The p-values less than 0.05 (p <0.05) were considered statistically significant.
The optimized conditions from statistical model were validated and the chitosanase purification was evaluated. The validation assay was conducted with 150 cm.h-1 load flow velocity, 6.0 cm settled bed height and 7.36 cm distributor height. The samples were withdrawn at appropriate intervals, which were subjected to further analysis in terms of chitosanase activity and protein concentration.
Operation of EBA
The ligand Diethylaminoethyl (DEAE) of the streamline ion exchanger (GE Healthcare, Uppsala, Sweden) was applied to EBA chromatography. This resin is a weak anionic exchanger. A homemade EBA glass column (2.6 cm x 30.0 cm) was used. A different amount of glass microspheres was added to work as a distributor, i.e., to enhance fluid distribution at the column inlet. The peristaltic pump was used to transport the fluid (model Perimax 12, Spetec). The correct column vertical alignment was guaranteed in all assays. All experiments were performed at the room temperature.
Runs were conducted without clarification. The column with different settled-bed height of resins was equilibrated with a 50mM Tris-HCl pH 8.5 buffer (buffer A) to give a stable height, where H/H0=1.5. The broth was then loaded (flow-through step) onto this
column at the superficial velocity required according to Table 1, following the washing with buffer A (150 mL). The elution was performed by step-wise gradient mode with 50 mM Tris- HCl pH 8.5 buffer containing 0.3 M (buffer B), 0.7 M (buffer C), and 1 M (buffer D) NaCl. Both washing and elution were conducted at the superficial velocity of 100 cm.h−1. In all purification steps, several fractions were collected for protein quantification and enzyme assay.
RESULTS AND DISCUSSION
Identification of chitosanase-producing strain
The microorganism was isolated from soil samples using procedure described in section 2.2. Among over 15 strains were isolated in the laboratory and screened for chitosanase activity, the Bacillus C-01 strain was selected for further study. This strain showed the highest chitosanase activity. Strain C-01 is a gram-positive, which grows in both aerobic and anaerobic environments. The culture showed colonies with rounded, irregular border, average size of 4 mm and white coloring. According to the results of 16S rDNA partial nucleotide sequence analysis, strain C-01 was identified as B. cereus.
CCD and Model fitting
As observed with EBA systems, several factors have to be considered for the optimization of protein purification. Such as, load flow velocity, settled bed height and distributor height [22, 23]. The chitosanase purification was optimized using statistical experimental design involving three variables, namely load flow velocity (X1,) settled bed height (X2)and distributor height (X3), at five levels. Table 2 presents the Central Composite design variables (coded levels) and effect estimates for screening of factors affecting P and Y. The best values of the P were obtained on the assay 17 and 14. In other assays, it can be observed low values purification factor. This fact can be related to the strong interaction of biomass with the resin. Similar results were also observed by [24], whereby the purification factor varied in function of feedstock. The performance of many direct-contact methods for downstream processing of bioproducts can be affected by biomass adhesion or interaction. Biological feedstock can also severely affect this process by particles co-adhesion or aggregation [25].
Table 2. Central Composite design variables (coded factor and levels) and effect estimates for screening of factors affecting purification factor (P) and enzyme yield (Y).
Coded Variable levels
Run number. X1 X2 X3 P Y 1 -1 -1 -1 1.18 45.12 2 +1 -1 -1 0.51 17.03 3 -1 +1 -1 0.92 43.47 4 +1 +1 -1 0.90 39.67 5 -1 -1 +1 1.24 46.92 6 +1 -1 +1 1.27 47.27 7 -1 +1 +1 0.64 27.26 8 +1 +1 +1 0.74 29.99 9 (C) 0 0 0 0.54 22.44 10 (C) 0 0 0 0.53 21.35 11 (C) 0 0 0 0.59 23.26 12 -α 0 0 0.89 39.61 13 +α 0 0 1.01 22.61 14 0 -α 0 1.40 35.40 15 0 +α 0 0.85 28.63 16 0 0 -α 1.10 31.79 17 0 0 +α 1.43 34.79 Effect estimates I* -0.05(L) 0.18(Q) -0.28(L) 0.31(Q) 0.13(L) 0.39(Q)
Effect estimates II**
-8.40(L) 8.12(Q) -4.00(L) 8.76(Q) 1.64(L) 9.67(Q)
X1, load flow velocity (cm.h-1); X2, settled bed height (cm); X3, distributor height (cm); P, purification factor; Y, enzyme yield (%); *Effect
As can be observed in Table 2, intermediary conditions do not improve either P or Y. Even though all factors play role in both P and Y it can be seen that Distributor Height (X3) and its interaction with other factors were the most important. The main effects of variables on P and Y are also shown in Table 2. On the basis of statistical analysis for quadric model, the variables evidencing the most significant effects on P and Y were settled bed height and distributor height as well as their interactions.
Analysis of variance (ANOVA)
Table 3 presents the results of the analysis of variance (ANOVA). The variables with confidence levels exceeding 95% (p < 0 .05) were considered to be significant. The goodness of model fit was checked by the determination coefficient (R2). The ANOVA of regression model demonstrated that the R2 to P and Y were 0.835 and 0.9147, respectively, thus indicating that about 90% of the variability in the response could explained by the model. The value of the adjusted determination coefficient (Adj. R2 = 0.623 for P and 0.805 for Y) is also acceptable, which indicates a high significance of the model.
Table 3: Analysis of variance (ANOVA) for central composite design of purification factor and enzyme yield.
Variable Adj R2 (%) R2 (%) F
cal Ftab Fcal/Ftab
Purification Factor (P) 62.3 83.5 4.171 3.68 1.13
Enzyme Yield (Y) 80.5 91.4 48.02 3.68 13
By ANOVA and regression coefficients can be deduced that the quadratic model showed high adequacy for the explanation of adsorption behavior. For purification factor, the most influential variables were the X3 (Q) (p value 0.001977), X2 (L) (p value 0.003015) and X2 (Q) (p value 0.003052). The interaction X2X3 also presents strong influence about P.
For Y, the most influential variables were the interaction X2X3 (p value 0.002195), X3 (Q) (p value 0.003483) and X1 (L) (p value 0.003805).
By regression analysis of CCD, it was possible to obtain the following Equation for P and Y:
YP= 0.570 -0.027X1 +0.091X12 – 0.144X2 + 0.158X22 + 0.064X3 + 0.196X32 + 0.090X1X2 +
0.120X1X3 – 0.159X2X3 Eq.5
YR= 22.073 – 4.203X1 + 4.064X12 – 2.00X2 + 4.384X22 + 0.820X3 + 4.836X32 + 3.333X1X2 +
Where YP is the response for purification factor and YR is the response for enzyme
yield. X1, X2 and X3 are the coded values of the test variables (load flow velocity, settled bed height and distributor height, respectively).
By F test can be seen that the proposed models (Equations 5 and 6) are statistically significant. As for P, Fcal value was 1.3 times greater than Ftab. Already for the Y, the
observed difference was much higher: 13 times larger than Ftab (Table 3).
Response surface methodology (RSM)
CCD, Box-Behnken design and three-level factorial design are classes of RSM and have different characteristics. Among the most popular techniques, the CCD allows optimization and estimation of the main and interaction of variables with least number of experiments [19].
Figure 1 shows the response surface and contour plots for variation in P and Y versus significant variables. The surface plots confirm that the interactions between the independent variables have a strong effect on response variables. The combined decrease in load flow velocity and settled bed height contribute to the increase of P (Figure 1a). The load flow velocity and distributor height need to be at the same level to improve the purity. High values of speed and distribution height combined increase the purity as well as low values of speed combined with low values of distributor height also raise the purity (Figure 1b). While an increase in distributor height of the combined with a decrease in settled bed height contributes to the increase in purity (Figure 1c).
The decrease in load flow velocity leads to significant increase in Y (Figures 2d and 2e). Whereas that increase in distributor height and settled bed height lead to significant increase in Y (Figures 1e and 1f).
Figure 1: Response surfaces for 23 central composite designs (a) X
1 – X2 (purification factor); (b) X1 – X3 (purification factor); (c) X2 – X3 (purification factor); (d) X1 – X2 (Enzyme yield); (e) X1 – X3 (Enzyme yield); (f) X2 – X3 (Enzyme yield). X1, load flow velocity; X2, settled bed height; X3, distributor height. Other variables, except for two variables in each figure, were maintained at zero in coded units.
In this study the parameters settled bed height and distributor height show different behaviors. We believe that biomass, resin and distributor interaction is so strong that determine the effect on the response variable. The influence of biomass on the hydrodynamic stability of expanded beds has been demonstrated by several researches [22, 24, 26]. The distributor has a great action on the flow structure [23]. For responses Y the distributor height has a positive effect. When the distributor height is increased, there is less back-mixing effect. The flow is more homogeneous, and it promotes the adsorption. A smaller mixture favors the plug-flow. Therefore, it can be conclude that the optimum conditions are far from the central point, as can be seen in Figure 1.
To check the normality assumption in fitted model, It was plotted the predicted versus the observed for P and Y (Figure 2). The figures exhibit linear relationship between predicted and observed. The measured response was really close to the predicted.
Figure 2: The experimental data versus the predicted data of normalized of purification factor (a) and enzyme yield (b).
Test of the model
Validation experiments were conducted to ensure the adjustment of the model for chitosanase purification. This assay was carried out under following conditions: load flow velocity 150 cm.h-1, 6.0 cm of settled bed height and 7.36 cm distributor height. Figure 3 shows the chromatogram for chitosanse purification. The elution step was carried out with the NaCl gradient. In the first increase in ionic strength, a peak of enzyme activity and protein concentration was obtained.
Figure 3: Chromatographic profile of crude extract from Bacillus cereus in streamline DEAE ion exchange column. The column was previously equilibrated with 50 mM Tris-HCl, pH 8.5 (Buffer A) and the sample was eluted from column with 50 mM Tris-HCl buffer, pH 8.5 containing 0.3, 0.7 and 1.0M NaCl (Step wise).
The validation experiment with three replicates resulted in an average P of 1.46 (± 0.12), while the P predicted for this assay by the regression model was 1.23. Under not optimized conditions (central point), the P mean was 0.55 (± 0.032) therefore the optimization resulted in an approximately 2.65-fold increase. The average Y for validation experiment was 34% (± 1.6), the predicted value of Y by the regression model was 37%. The average Y in central point was 22.5 (± 0.96), what indicates 1.65-fold increase.
This study was generally consistent with the results to packed purification of another chitosanases that were reported to a final yield of 23.6% [27], 30% [28] and 17.1% [29]. To applications on unpacked bed purifications, these results are similar to those found in the purification of a C-phycocyanin [14], an E. coli-Based therapeutics [18] and β-galactosidase [30]. It is important to highlight that the procedure used here has just one-step, and can be coupled to another EBA procedure such as cationic as well as hydrophobic interaction.
CONCLUSION
In the present study a CCD was used to optimize the application of the EBA, based on anionic exchange, to recover as well as to purify a chitosanase from Bacillus cereus. This
statistical experimental approach confirmed to be effective for the optimization of chitosanase purification. CCD analyses showed that the optimum conditions for maximum P were 150 cm.h-1 (load flow velocity), 6.0 cm for the settled bed height and 7.36 cm for the distributor height. The validation experiments proved that optimum conditions enhance the purification factor and also evidenced a good applicability of model. This study provides knowledge about hydrodynamics as well as biomass interaction in the purification protocol of biomolecules by EBA. Furthermore, the system is promising for recovery of chitosanase from
B. cereus C-01 and it is economically viability since it promotes the reduction steps.
Acknowledgements
The authors thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior and the Conselho Nacional de Desenvolvimento Científico e Tecnológico for financial support.
Conflicts of Interest
The authors declare no conflict of interest.
REFERENCES
[1] K. Azuma, T. Osaki, S. Kurozumi, M. Kiyose, T. Tsuka, Y. Murahata, T. Imagawa, N. Itoh, S. Minami, K. Sato, Y. Okamoto, Anti-inflammatory effects of orally administered glucosamine oligomer in an experimental model of inflammatory bowel disease, Carbohydrate polymers, 115 (2015) 448-456.
[2] X.F. Kong, X.L. Zhou, G.Q. Lian, F. Blachier, G. Liu, B.E. Tan, C.M. Nyachoti, Y.L. Yin, Dietary supplementation with chitooligosaccharides alters gut microbiota and modifies intestinal luminal metabolites in weaned Huanjiang mini-piglets, Livestock Science, 160 (2014) 97-101.
[3] I. Younes, S. Sellimi, M. Rinaudo, K. Jellouli, M. Nasri, Influence of acetylation degree and molecular weight of homogeneous chitosans on antibacterial and antifungal activities, International Journal of Food Microbiology, 185 (2014) 57-63.
[4] C.F. de Assis, L.S. Costa, R.F. Melo-Silveira, R.M. Oliveira, M.G. Pagnoncelli, H.A. Rocha, G.R. de Macedo, E.S. Santos, Chitooligosaccharides antagonize the cytotoxic effect of glucosamine, World J Microbiol Biotechnol, 28 (2012) 1097-1105.
[5] S. Masuda, K. Azuma, S. Kurozumi, M. Kiyose, T. Osaki, T. Tsuka, N. Itoh, T. Imagawa, S. Minami, K. Sato, Y. Okamoto, Anti-tumor properties of orally administered glucosamine
and N-acetyl-d-glucosamine oligomers in a mouse model, Carbohydrate Polymers, 111 (2014) 783-787.
[6] P. Zhang, W. Liu, Y. Peng, B. Han, Y. Yang, Toll like receptor 4 (TLR4) mediates the stimulating activities of chitosan oligosaccharide on macrophages, International Immunopharmacology, 23 (2014) 254-261.
[7] Y.-J. Wee, L.V.A. Reddy, K.-C. Chung, H.-W. Ryu, Optimization of chitosanase