Enzymatic Hydrolysis of Cassava Starch for Production of Bioethanol with a Colombian Wild Yeast Strain

A hidrólise enzimática do amido de mandioca para produção de xaropes de glucose foi avaliada usando alfa-amilase de Bacillus licheniformis e glucoamilase de Aspergillus níger. Também, uma mistura enzimática composta de α-amylase de Aspergillus kawachi e glucoamilase de Aspergillus níger foi testada. As condições da enzima para a hidrólise do amido foram otimizadas por um planejamento fatorial experimental (3×2) usando como variáveis a concentração do substrato, a relação enzima/ substrato e o tempo de reação. As condições ótimas de reação com 100 g de amido per L foram: α-amilase pH 5,0, 80 °C e 130,5 U g de amido; glucoamilase pH 4,5, 70 °C e 81,5 U g de amido. Adicionalmente, as condições ótimas da mistura enzimática foram pH 4,5, 46 °C e 16,4 U g de amido. Finalmente, a produção de álcool usando xaropes de glucose a partir do amido hidrolisado enzimaticamente foi realizada usando uma cepa selvagem de Candida sp isolada do caldo de cana de açúcar, obtendo produtividades em etanol volumétrico em torno de 1,8-3,2 g Lh.


Introduction
Increasing on energetic requirements and atmospheric contamination by combustion gases, has opened searching for new, safe, effective and more accessible energy sources.For this aim, biofuels production from different agricultural sources, such as agroindustrial by-products, vegetable materials as woods, among others, has recently started-up. 1,2Biofuels can be produced by many different types of substrates.Among these, cassava (Manihot esculenta Crantz), a plant with high starch content, is considered a cheap, abundant and renewable resource for production of fermentable glucose syrups and dextrins.Moreover, it is easily produced in tropical and sub-tropical zones, mainly in Asia, South-America and South-Africa. 35][6] These conventional processes are carried out in three steps: gelatinization, liquefaction and saccharification.][9] On the other hand, liquefaction is carried out by amylases that hydrolyze the chemical bond α-(1-4) of starch, producing dextrin, maltose, maltotriose and maltopentoses with a dextrose equivalent (DE) below 30.This enzyme processing of starch allows a rapid reduction in the viscosity of the solution. 10,11mylolyltic enzymes can be obtained from different sources.Among these, we can find thermostable bacterial enzymes from Bacillus licheniformis or B. amyloliquefaciens, whose are suited for liquefaction, because this step is performed at high temperatures (80-110 °C).Finally, saccharification process of partially hydrolyzed starch is carried out at lower temperatures (60-70 ºC).In this case are used enzymes such as glucoamylase (amyloglucosidase) obtained from Aspergillus niger or Rhizopus sp.][12] These amylolytic enzymes commonly are used in soluble solutions and they can be susceptible to inhibition for both reaction substrates and/or products.For this reason, identification and optimization of reaction conditions that affect enzymatic activity could improve economic and technological feasibility of this bioprocess.Among these different conditions are: temperature, pH, reaction time, enzyme concentration, viscosity, mixing rates, ionic strength, calcium concentration, etc. 11,15,16 Moreover, native starch is little soluble in water and resistant to hydrolysis at mild conditions.For this reason, it needs to carry out hydrothermal process for the gelatinization. 14However, a high energetic requirement for this process increases economical production costs of starch hydrolysis. 8In this work, we evaluated different amylolytic enzymes for obtaining the best conditions for hydrolysis of cassava starch, aimed to production of fermentable glucose syrups.

Cassava starch and amylolytic enzymes
Cassava starch was obtained from a Starch Flour Plant-Manzanarez Ltda.Co. (Bucaramanga, Colombia).This starch product had a moisture content of 7 and 93% of starch.Moisture content in the cassava starch was determined by drying a starch sample to constant weight.Enzymes used in this study were the following: Liquozyme ® SC DS, Spirizyme ® Fuel from Novozymes, and Stargen™ 001 from Genecor International.These amylases are specialized for hydrolyzing starch from cereals such as corn and wheat.

Pre-treatment of cassava starch
Starch was dissolved in water or 0.016 mol L -1 sodium acetate buffer, pH 4.0.This starch solution was heated at 66 ºC 17 under mechanical shaking 390 rpm by 30 min.

Enzymatic hydrolysis of cassava starch
Starch hydrolysis was performed by two different methodologies.In the first one, it was carried out in two consecutive steps: liquefaction and saccharification.This procedure was performed in a batch reactor stirred at 390 rpm.In the liquefaction step, starch was gelatinized by thermal treatment and treated with alpha-amylase (Liquozyme ® SC DS de Novozymes) for obtaining dextrins from cassava starch.Dextrins were determined as dextrose equivalent (DE) (equation 1) and reducing sugars, and reported in terms of units of enzymatic activity U (defined as 1 µmol of reducing sugar released per min), using the 3,5-di-nitro-salicylic acid (DNS) colorimetric method. 14he saccharification step was carried out with glucoamylase (Spirizyme ® Fuel from Novozymes) producing glucose syrups from dextrins obtained in the previous liquefaction step.Additionally, bioprocess efficiency was determined quantifying glucose concentration by spectrophotometry using a glucose oxidase method (Glucose Kit, BioSystems ® , S.A). 19,20n the second methodology, a simultaneous liquefaction and saccharification starch hydrolysis was performed in batch mode.In this case, it was used non-gelatinized starch, adding an enzymatic mixture of α-amylase and glucoamilase (Stargen™ 001) and the same reaction conditions described previously. (1)

Effect of temperature on enzymatic activity of amylases
The enzymatic activity was determined for each amylolytic enzyme at different temperatures.Liquozyme ® SC DS and Spirizyme ® Fuel were evaluated at 66, 70 and 80 °C, while Stargen™ 001 was determined at 30, 47 and 56 °C.Periodically, samples (1 mL) from enzyme reactions were withdrawn for determination of reducing sugars and quantification of DE.

Effect of pH and substrate concentration on enzymatic activity of amylases
The influence of pH on enzymatic activity was determined in a range of 3.0-6.0.Enzymatic activity was tested using substrate concentrations in the range of 100-300 g L -1 , determining enzyme activity and DE parameters as previously described.

Optimization of cassava hydrolysis
Once reaction conditions for enzymatic hydrolysis of cassava starch were determined, starch hydrolysis using non-diluted enzymes and high substrate concentrations was evaluated.In this case, enzymatic hydrolysis was performed at constant pH and temperature conditions previously established for these enzymes, 8,23 similar values for these parameters to those reported in the technical sheet for these enzymes. 21,22,24An experimental design of 3 3 using 2 measurements by run (see Table 1) was carried out for optimization of saccharification process for analyzing and validating the most important variables affecting this enzymatic process.
The experimental design for three variables, including enzyme/substrate, substrate concentration and time at three levels (low (−1), medium (0) and high ( 1)) (Table 1) were used for screening based on our previously study.The total number of experiments to be performed was (n k )r, where n is the number of levels, k is the number of variables, and r is the number of replicates for the runs obtaining a total of (27) × 2 = 54 experiments.To avoid bias, the 54 runs were performed in a totally random order.Finally, the statistical significance was determined by P-value, and the proportion of variance explained by the model obtained was given by the multiple coefficient of determination, R 2 .

Culture medium and yeast fermentation conditions
Inoculum was constituted by wild Candida sp yeast pre-cultured for 12 h at 37 ºC and under microaerophilic conditions (without shaking), using a liquid medium solution YPG, containing glucose (150 g L -1 ), peptone and yeast extract (10 g L -1 ).Afterwards, a fermentation medium YG containing glucose (150 g L -1 ), yeast extract (10 g L -1 ) was used for ethanol production.

Optimization of enzymatic saccharification process of cassava starch and ethanol production by wild yeasts strain
For these experiences, it was carried out starch hydrolysis at optimal conditions, and hydrolyzed starch containing fermentable sugars was used as carbon source for ethanol production by a native strain of Candida sp.Culture medium was inoculated with 0.3 g DCW L -1 of wild yeast, working at 0.2 mol L -1 sodium acetate buffer, pH 5, 35 ºC and 120 rpm of stirring, using a Shaker Max Q Mini 4450.
Ethanol production was performed by different methodologies, working with 250 g of starch per L as carbon source for ethanol fermentation.In the first one, starch hydrolysis and ethanol fermentation were carried out in consecutive steps, using 9.3 × 10 3 U g -1 of starch of α-amylase Liquozyme SC DS for starch hydrolysis and 3.7 × 10 3 U g -1 of starch of the glucoamylase Spirizyme fuel from Novozymes.In the second one, a simultaneous saccharification and fermentation (SSF) process was performed adding 39.3 U g -1 of starch of an enzymatic mixture of α-amylase and glucoamilase (Stargen™ 001) from Genecor.At different times (7, 20, 30 and 45 h of fermentation), samples were withdrawn from the fermentation broth.The samples were centrifuged at 10.000 g for 15 min.Supernantants were used for determination of ethanol, reducing sugars and biomass concentrations.Ethanol was determined by headspace-gas chromatography (HS-GC) using a gas chromatograph-Hewlett Packard 5890 according to Hailong et al. 25 Biomass concentration was determined by dry cell weight (DCW) methodology.Samples from fermentation broths were withdrawn at different times and centrifuged at 10.000 g and subsequently washed with distilled water 3 times (3×).Subsequently were dried at 105 °C until to obtain constant weight.Alternatively, dilutions from fermentation broths (in exponential phase) were made for determination of cell concentration by turbidimetry at 650 nm and correlated with in g DCW L -1 determinations.

Effect of enzyme/substrate ratio on enzymatic hydrolysis
In this study, different enzyme/substrate ratios (E/S) were tested.Higher DE values were used as criterion for selecting an enzyme for hydrolysis of cassava starch.Best results were obtained with a ratio of 130.5 U g -1 of starch for enzyme from Novozymes (Figure 2A) 81.6 U g -1 of starch for saccharification process (Figure 2B) and 16.4 U g -1 of starch for enzymatic mixture from Genecor (Figure 2C).The enzyme reaction conditions were the following: liquefaction and saccharification as described in pre-treatment starch, while in simultaneous liquefaction-saccharification were carried out at pH 4.0 and 37 °C.For both process we used a starch concentration of 100 g L -1 .Additionally, at higher E/S ratios, a both increase on released DE became and reaction rates were achieved.Nevertheless, E/S ratios reached in this study were lower than reported from providers 21,22,24 and Zhao et al. 23

Effect of temperature on enzymatic activity
This study was achieved at the same reaction conditions used to study effect of enzyme/substrate ratio on enzymatic hydrolysis.In Figure 3 are shown results of effect of temperature on different amylolytic enzymes hydrolyzing cassava starch.Optimal reaction temperatures varied and were dependent on type of enzyme.In general, initial hydrolysis rate was increased with temperature, obtaining optimal hydrolysis rates for Liquozyme ® SC DS, Spirizyme ® Fuel (amylase and glucoamylase from Novozymes) at 80 and 70 °C, respectively.These results are according to results obtained by Zhao et al. 23 On the other hand, we obtained an optimal temperature of 47 °C for Stargen™ 001, achieving similar data to those reported by Wang et al. 26 However, starch hydrolysis was decreased at high temperatures, due to thermal inactivation of amylolytic enzymes (Wu and Miao, 27 Morales et al. 12 ).

Effect of pH on enzymatic activity
In these studies, we worked in range of pH values with high catalytic activity. 13This study was achieved at the same reaction conditions used to study effect of enzyme/ substrate ratio on enzymatic hydrolysis.In Figure 4 are shown results of pH effect on enzymatic activity.In this figure we can see that optimal pH were 5.0, 4.5 and 4.0, for liquefaction, saccharification and simultaneous liquefaction and saccharification, respectively.These values were in the range to those found by Zhao et al. 23 for enzymes from  Novozymes, and Stargen™ 001 obtained by Wang and co-workers. 23,26

Effect of substrate concentration
In Figure 5 is shown the effect of substrate concentration on enzymatic activity at optimal pH and temperature determined in previous studies.In this figure is observed that increasing in substrate concentration produced a decrease in enzymatic activity on starch hydrolysis.This effect could be due to diffusional constraints during the enzymatic hydrolysis of starch, 16 caused mainly by increasing in viscosity by gelatinized starch.Similar observations have been reported by Baks et al., 28 working with high starch concentrations (650 g of strach L -1 ).On the other hand, it is possible that enzymatic inhibition of amylases could be produced by high substrate concentrations.
For this reason, it was selected 200 g of starch L -1 of cassava starch for enzymes from Novozymes and 250 g of starch per L) for the enzyme from Genecor. 24,25owever, these enzymes are not saturated at these starch concentrations.On the other hand, stirring rate was kept at 390 rpm during all starch hydrolysis process, and it was not detrimental for enzyme stability, 9 allowing god mass transfer and mixing performances for enzymatic starch hydrolysis. 16

Optimization of starch hydrolysis by factorial design
Analysis of variance (Table 1) indicates that the secondorder polynomial model (equation 2) was significant and adequate to represent the actual relationship between the response (glucose released by enzymatic starch hydrolysis in units of concentration (g L -1 )) and the significant variables, with a very small P-value (p < 0.05) and a coefficient of determination R 2 = 0.7977.Y = 194.8+ 40.3 X 1 + 39.0 X 2 + 29.8 X 3 + 5.6 X 1 2 + 22.9 X 1 X 2 + 8.9 X 1 X 3 -9.4X 2 2 + 0.86 X 2 X 3 -12.1 X 3

2
(2 where Y is glucose concentration (g L -1 ) and X 1 , X 2 and X 3 are enzyme/substrate ratio (U g -1 of starch), starch (g L -1 ) and time (min), respectively.This equation was statistically adjusted and could explain 75.6% of variability on glucose concentration which means that 24.4% is due to factors not considered in this investigation (uncertainty), with 23% of mean absolute error.
From these regression coefficients obtained in the equation 2, it is possible to determine that the most linear important independent variables for an efficient production of glucose from cassava starch are: X 1 , X 2 and X 3 variables.This equation corroborates that these variables are very significant (with a very small P-value (p < 0.05); see Table 2), followed by the interaction of linear variables of X 2 X 1 , for finding finally other second order combinations, with minimal contribution to the response variable or indirect relation over response.The optimal saccharification parameters were obtained by solving the regression equation (equation 2).The optimal parameters were: 5.9 U g -1 of starch of spirizyme fuel, 400 g L -1 dry matter and 45 min for saccharification.Under optimal conditions obtained by design we reached a glucose yield of 320.7 g L -1 .

Ethanol production by wild yeast using hydrolyzed cassava starch as substrate
The results obtained during fermentation of hydrolyzed starch (Table 3) were favorable for both hydrolytic processes used in this study.We used two approaches for ethanol production: first one, starch hydrolysis and then yeast fermentation, and second one, simultaneous saccharification and fermentation (SSF).
Based on the obtained results the time of the overall process of ethanol production may increase in the simultaneous saccharification and fermentation.The energy savings could be attained since the SSF process was effectively performed at 35 ºC, which is a lower temperature than the optimal temperature from the action of enzymes for starch hydrolysis and then yeast fermentation (> 60 ºC).
In addition, the SSF process bring some advantages such as prevention of inhibition by substrate concentration, and that as the sugars are produced, these are fermented also avoids previous process of starch hydrolysis and the use of other process.

Figure 3 .
Figure 3.Effect of temperature on enzyme activity during enzymatic hydrolysis of cassava starch.

Figure 2 .
Figure 2. Dextrose equivalent production at different enzyme/substrate ratios during enzymatic hydrolysis of cassava starch.

Figure 4 .
Figure 4. Effect of pH on enzyme activity during the starch hydrolytic bioprocess.

Figure 5 .
Figure 5.Effect of substrate concentration on enzyme activity.

Table 1 .
Matrix for the experimental design of saccharification process of cassava starch using spirizyme fuel from Novozymes Figure 1.The schematic flow diagram of the enzymatic hydrolysis procedure.

Table 2 .
Analysis of variance for synthetic variables pertaining to response glucose concentration (g L -1 ) a X 1 , X 2 and X 3 are enzyme/substrate ratio (U g -1 of starch), starch (g L -1 ) and time (min), respectively.

Table 3 .
Kinetic parameters of ethanol production obtained during fermentation of a wild yeast strain using different fermentable sugar as carbon sources