Evaluation of fermentation conditions by Candida tropicalis for xylitol production from sago trunk cortex

Abstract

Xylitol production from sago trunk cortex hydrolysate using Candida tropicalis was evaluated in shake flasks and a bioreactor. The fermentation and kinetic behaviours of this microorganism were investigated using sago trunk cortex hydrolysate and commercial xylose as substrate. Results obtained for sago trunk hydrolysate were close to the commercial xylose with xylitol yield of 0.82 gg^-1 and productivity of 0.39 gL^-1h^-1. The maximum specific growth rate, µmax for sago trunk cortex was higher (0.24 h^-1) compared to commercial xylose (0.17 h^-1). The bioreactor study showed an increase of about 6% (w/v) of xylitol concentration and 10% (v/v) of volumetric productivity when compared to the results obtained under the shake flasks, keeping xylitol yield above 0.8 g g^-1.

Full Article

Evaluation of Fermentation Conditions by Candida tropicalisfor Xylitol Production from Sago Trunk Cortex

Nurul Lina Mohamad,^a Siti Mazlina Mustapa Kamal,^a,* Norhafizah Abdullah,^b and Iryane Ismail ^a

Xylitol production from sago trunk cortex hydrolysate using Candida tropicalis was evaluated in shake flasks and a bioreactor. The fermentation and kinetic behaviours of this microorganism were investigated using sago trunk cortex hydrolysate and commercial xylose as substrate. Results obtained for sago trunk hydrolysate were close to the commercial xylose with xylitol yield of 0.82 gg^-1 and productivity of 0.39 gL^-1h^-1. The maximum specific growth rate, µ_max for sago trunk cortex was higher (0.24 h^-1) compared to commercial xylose (0.17 h^-1). The bioreactor study showed an increase of about 6% (w/v) of xylitol concentration and 10% (v/v) of volumetric productivity when compared to the results obtained under the shake flasks, keeping xylitol yield above 0.8 g g^-1.

Keywords: Xylose; Xylitol; Sago trunk cortex; Candida tropicalis

Contact information: a: Department of Process and Food Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia; b: Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia; *Corresponding author: siti@eng.upm.edu.my

INTRODUCTION

Agricultural plant residues such as straw, bagasse, and trunk are disposed of after industrial, agricultural, and forestry activities. These materials represent an abundance of organic compounds and are composed of cellulose, hemicellulose, and lignin. These compounds can be a potential substrate if a process is available for their conversion into a value-added product (Foyle et al. 2007). Cellulosic material usually contains hexoses and hemicellulose that can easily be converted to glucose and xylose, respectively either by acid or enzymatic hydrolysis (Sun and Cheng 2002). Xylose can be used as carbon sources for the production of xylitol, a valuable alternative sweetener. Xylitol sweetness is similar to that of sucrose, but unlike sucrose it exhibits anticariogenic properties, which can promote oral health and prevent caries (Parajo et al. 1998). Owing to that advantage, xylitol has been used as a sweetener agent in a variety of food products, such as chewing gums, candies, and baked products (Winkelhausen et al. 2007). Bioconversion of xylitol is recognized as an alternative route to the chemical dehydrogenation process, which is claimed to be efficient and cost-effective (Nigam and Singh 1995; Parajo et al. 1998).

This study focused on the conversion of xylose obtained from hemicellulose of sago trunk cortex to xylitol. Malaysia is the highest sago producer in the world, exporting about 25,000 to 40,000 tonnes to Japan, Taiwan, Singapore, and other countries, and this value is increasing annually (Chew et al. 1999). Therefore, the production of sago wastes, such as sago trunk cortex (STC) will keep increasing proportionally along with the increase of sago starch production. Sago wastes, in the form of barks and husks, are largely composed of cellulose and lignin, and therefore, both are wastes and pollutants. This study evaluated the production of xylitol in sago trunk hydrolysate on shake flasks culture, bench bioreactor, and the kinetic behaviour of the yeast Candida tropicalis.

MATERIALS AND METHODS

Pretreatment

Sago barks were purchased from local cultivated sago in Melaka, Malaysia. The cortex (the outer layer) was removed from the barks and the pith (the inner portion), which was then chopped to smaller pieces (25 x 150 mm). The particles used in the experiments varied in size from 5 mm to 10 mm.

Sago Trunk Cortex Hydrolysate and Detoxification

The sago trunk used in this study were pretreated with 8% sulphuric acid for 60 min of hydrolysis time, in an autoclave at 121°C at solid to liquid ratio of 1:10 in 250 mL flasks (Mohamad et al. 2011). The solid and liquid fraction was separated by using vacuum filtration and the remaining hydrolysate was neutralized using CaCO₃. The detoxification conditions were carried out with 2.5% concentration of activated charcoal at pH 5.5 and treated for 60 min at room temperature. The completely detoxified hydrolysates were recovered by using a vacuum filtration. All the experiments were conducted in duplicate.

Microorganism and Fermentations

The strain of Candida tropicalis was selected due to its ability to ferment xylose, and it has been shown to be a xylitol producer by researchers in the past (Soleimani et al. 2006; Winkelhausen and Kuzmanova 1998). Pure culture was donated by the Department of Biomedical Sciences, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, stored at 4 °C on Sabouraud dextrose agar and subcultured twice a month.

A single colony of Candida tropicalis was subcultured in a 500 mL Erlenmeyer flask containing 23 g L^-1 D-xylose, 10 g L^-1 yeast extract, 15 g L^-1 KH₂PO₄, 3 g L^-1 (NH₄)₂HPO₄, and 1 g L^-1MgSO₄•7H₂O (Walther et al. 2001). The flasks were incubated at 30 °C for 18 to 24 h to obtain an initial cell concentration with OD600nm of 0.4 to 0.6 and subsequently inoculated into the production medium at 10% (v/v). The batch cultivations were conducted in shake flasks and bioreactor. The shake flask fermentations were conducted in 500 mL Erlenmeyer flasks and contained 200 mL of sago trunk hydrolysate medium as carbon sources at 34 ºC on a rotary shaker at 250 rpm for 72 h at pH 4.

Fermentation in the bioreactor was carried out using a 3 L stirred tank bioreactor (Fermentec Resources Sdn. Bhd., Malaysia) containing 1.5 L fermentation medium of sago trunk hydrolysate with agitation, aeration, temperature, pH, and dissolved oxygen control. A single six-blade turbine impeller mounted on the agitator shaft above the air sparger was used for agitation at 250 rpm and air flow at 0.4 Ll/min The initial pH was adjusted to 4 and was not controlled during fermentation. The low pH value is the optimum condition for Candida tropicalis as an acidogenic type of yeast (Sampaio et al. 2006; Ke et al. 2009). The fermentation broth was sampled every 6 h to monitor the xylose and xylitol concentration.

Xylitol production was controlled by the fermentation conducted with pure xylose as a carbon source. The medium containing 23 g L^-1 D-xylose, 10 g L^-1 yeast extract, 15 g L^-1 KH₂PO₄, 3 g L^-1(NH₄)₂HPO₄, and 1 g L^-1 MgSO₄•7H₂O. The experiments were conducted with the same conditions as the fermentation, with hydrolysate as the produc-tion medium. The experimental values were the means of two independent samples.

Analysis

Samples from the production medium were centrifuged at 10 000 x g for 10 min. The decanted supernatant was separated and used for sugar concentration analysis. Xylose, glucose, and xylitol were analysed using High Performance Liquid Chromatography (HPLC, Class VP (Shimadzu, Japan) with refractive index detector and the Inertsil NH₂ 5 µm column (GLSciences, USA). The mobile phase used was 75% acetonitrile and 25% water with a flow rate of 0.5 mL/min and an oven temperature of 30 °C. The cell concentration was estimated by measuring the absorbance at 600 nm using the UV-VIS spectrophotometer (Ultrospec 3100 pro, Amersham Biosciences, USA). The dry cell weight was determined by drying in an oven at 95 °C until a constant weight was achieved.

Growth Kinetics for Fermentation Process

The rates of xylose consumption, xylitol production, and yeast population growth can be predicted from a kinetic model. During yeast growth in fermentation medium, new cells are formed catalytically from substrate with specific growth rate, µ (h^-1) and this can be expressed as,

 (1)

or

 (2)

where r_x = rate of cell growth, µ is the specific growth rate (h^-1), xis the cell concentration (g L^-1), and t is time.

Monod Equation

A relationship between the specific growth rate, µ, and the limiting substrate concentration, S, was proposed by Monod (1979). The Monod equation states that,

 (3)

where S is the concentration of the limiting substrate (g L^-1), µ_maxis the maximum specific growth rate (h^-1), and K_s is the saturation constant (g L^-1).

Lineweaver-Burk Method

The cell growth saturation coefficient, K_s and µ_max can be estimated by rearranging the Monod equation in Lineweaver-Burk form, by taking reciprocals of both sides of the equality sign,

1/μ=K_s/μ_maxS +1/μ_max (4)

where the slope of a plot of 1/µ versus 1/S will allow K_s to be evaluated, while the intercept is 1/µ_max

RESULTS AND DISCUSSION

Composition of Sago Trunk Cortex and Hydrolysate

Analysis of the sago trunk cortex composition was carried out in the previous study (Mohamad et al. 2011). The sago trunk was composed of 44.13% cellulose, 21.09% hemicellulose, 23.30% lignin, and 1.53% ash. Meanwhile, the sago trunk cortex hydrolysate was composed of 24.1 gL^-1 xylose, 2.5 gL^-1 glucose, 2.53 gL^-1 furfural, and 2.15 gL^-1 phenolic compounds. The detoxification treatment has minimal effect on sugar recovery, with only 4% of sugar loss of xylose and glucose (Mustapa Kamal et al.2011).

Inoculum Preparation and Xylitol Fermentation in Shake Flasks and Bioreactor

The preparation of inoculums was done in the reference medium (pure xylose and nutrient) and sago trunk hydrolysate, without adding nutrient. The reason to use xylose derived from sago trunk hydrolysate without adding nutrient was to identify the potential of this xylose as a substrate for the growth of Candida tropicalis. In addition, the growth of Candida tropicalis and xylitol production from sago trunk hydrolysate was compared with the pure xylose medium. Notably, the use of pure xylose for Candida tropicalisgrowth in the reference medium is a limiting factor, since this substrate has a high cost compared to other carbon sources. Besides, its availability in the market is also limited (Silva et al. 2006). Hence, the use of sago trunk cortex hydrolysate for inoculum preparation can overcome this problem and could reduce fermentation costs.

The growth of Candida tropicalis in sago trunk cortex hydrolysate and reference medium was studied. The results showed that there were no significant differences between the two media (Figs. 1 and 2). The maximum xylitol concentration was obtained at 48 h for both media. However, the reference medium showed faster xylose consump-tion and earlier xylitol accumulation than the sago trunk cortex medium. This situation could be due to the presence of glucose in the sago trunk cortex hydrolysate, whereby the yeast consumed glucose rather than xylose. Xylitol started to accumulate once the glucose had been depleted at 20 h in the sago trunk hydrolysate medium. Canilha et al. (2008) found a similar situation with the cultivation of Candida guilliermondii in the wheat straw hydrolysate. Likewise, similar results were also found by Silva et al. (2006), whereby the researchers found the use of Candida guilliermondii was greater in the rice straw hemicellulose hydrolysate than in the reference medium with a maximum xylitol concentration of 51.5 gL^-1 and 64% xylitol yield relative to the theoretical value. Moreover, recent studies of Wang et al. (2012) obtained a xylitol yield of 71% from corncob hydrolysate by using immobilized Candida tropicalis.

Fig. 1. Fermentation profiles of xylitol production on the shake flasks by C. tropicalis, in reference medium. Legends: (♦) xylose, (▲) xylitol, and (x) cells

Fig. 2. Fermentation profiles of xylitol production on the shake flasks by C. tropicalis, in the shake flasks of sago trunk hydrolysate medium. Legends: (♦) xylose, (▲) xylitol, (x) cells, and (■) glucose

The inoculum grown in the previous sago trunk cortex hydrolysate was used in this experiment under the same conditions of temperature, pH, and agitation speed. The results obtained are shown in Fig. 3. Consumption of carbon source was faster with xylose depleting at about 54 h, and the maximum xylitol production was 20.938 g L^-1, achieved at the same time required for the shake flask, i.e. 48 h. Hence, the maximum xylitol concentration increased 6% by using a bioreactor. Applying the same conditions to Candida guilliermondii, Silva et al. (2006) attained the maximum xylitol concentration, which decreased by 14% in a bench bioreactor compared to aerated shake flasks.

Fig. 3. Fermentation profiles of xylitol production by C. tropicalis, within 3-L stirred tank bioreactor. Legends: (♦) xylose, (▲) xylitol, (x) cells, and (■) glucose

The kinetic parameters obtained in the media are presented in Table 1. The results confirmed that the use of Candida tropicalis grown in the sago trunk hydrolysate (shake flasks and bioreactor) had no significant differences when compared to the reference medium. These results proved that the sago trunk hydrolysates implied an effective xylose source for the xylitol production.

Table 1. Kinetic Parameters of the Experiment Conducted under the Optimum Conditions with Candida tropicalis Grown in the Reference Medium and the Sago Trunk Hydrolysate

P_m, maximum xylitol production; Y_P/S, xylitol yield on the consumed xylose; Y_X/S, biomass yield on the consumed xylose Q_P, volumetric productivity; µ_max, specific growth rate, Ks, Monod cell growth saturation coefficient.

The results showed that there was an increase in the volumetric productivity when compared to the results obtained in the shake flasks. By performing the process using the bioreactor, the xylitol production of 20.938 g L^-1 was attained at 48 h with a volumetric productivity of 0.43 g L^-1h^-1 and xylitol yield of 0.83 g g^-1. These values were increased up to 10% from the reference medium in bioreactor.

Xylitol is produced when there is limited oxygen supply, and such conditions lead to an increase in xylitol productivity (Aguiar et al.2002). In the bioreactor fermentations, the availability of oxygen was increased, leading to cell growth. The higher cell densities in the medium yielded limited oxygen availability in the vessel and this situation occurred were increased the xylitol accumulation as well as substrate consumption.

Growth Kinetics of Candida tropicalis

Growth kinetics of Candida tropicalis were determined by using pure xylose (reference medium) as the limiting factor. Specific growth rates of Candida tropicalis were determined according to Equation (1). The variation of specific growth rate, µ, with time is shown in Fig. 4 (A-C). From a very low value, it increases dramatically to a maximum value, µ_max. The same trend was observed with sago trunk hydrolysate in shake flasks and bioreactor.

Fig. 4. Specific Growth Rate according to Monod equation (A) reference medium, (B) sago trunk hydrolysate in shake flasks, and (C) sago trunk hydrolysate in bioreactor

The saturation constant, K_s, and the specific growth rate, µ_maxconstant were determined using the Lineweaver-Burk method. This is shown in Fig. 5 (A-C). The values of the kinetic parameters µ_maxand K_s characterize the organisms and growth substrate. If an organism has very high affinity for the limiting substrate (a low K_svalue), then the growth rate will not be affected until the substrate concentration has declined to a very low level with a corresponding short decelerating phase for the culture. However, if the microorganism has a low affinity for the substrate (a high K_svalue), then the growth rate will be deleteriously affected at a relatively high substrate level. Thus, the deceleration phase for the culture would be relatively long, and the yield of biomass will be reduced (Ahmad and Holland 1995). In this present study, K_s was found to be lower than the substrate concentration for reference medium (2.26 g L^-1).

Fig. 5. Evaluation of growth constant by using the Lineweaver-Burk method (A) reference medium (B) sago trunk hydrolysate in shake flasks (C) sago trunk hydrolysate in bioreactor

The growth rate of Candida tropicalis was not affected until the decrease of xylose concentration. The growth kinetics study also indicated the decrease in saturation constant, K_s for growth in shake flasks (1.96) and bioreactor (1.64) for sago trunk hydrolysate. This is due to the increase in the affinity that Candida tropicalis has for xylose obtained from sago trunk hydrolysate. The effects of µ_max and K_s value were identified for sago trunk hydrolysate medium in the shake flasks and bioreactor. The value of µ_max and K_s are presented in Table 1. The maximum specific growth rate, µ_max, for xylose obtained from sago trunk hydrolysate in shake flasks (0.25 h^-1) and bioreactor (0.24 h^-1) was slightly higher than specific growth rate in the reference medium. This situation could be due to the complex medium of sago trunk hydrolysate, which consists of other unknown substances that are consumed by the microorganisms. Aguiar et al. (2002) estimated a high value of maximum specific growth rate for Candida guilliermondii as high as 3.259 h^-1 using D-xylose as the main substrate. This author explained that high estimated value of affinity constant counterbalanced the decrease in specific growth rate because of the low substrate concentration rather than high cell mass inhibition.

CONCLUSIONS

1. The production of xylitol from sago trunk hydrolysate was obtained in shake flask with xylitol yield (0.82 g g^-1) and productivity (0.39 g L^-1h^-1). A slightly higher value of xylitol yield (0.84 g g^-1) and productivity (0.44 g L^-1h^-1) were achieved in the bioreactor.
2. The sago trunk cortex hydrolysate appeared to be an alternative to the production of xylitol based on values of the selective fermentative parameters (YP/S = 0.85 g g^-1, QP = 0.39 g L^-1h^-1) that have been achieved without the necessity of adding nutrients to the media formulated by the detoxified hydrolysate.
3. Performance of the process in the bioreactor under optimum fermentation conditions allowed for an increase of about 6% in xylitol concentration and 10% in volumetric productivity when compared to the results obtained under the shake flasks, keeping xylitol yield above 0.8 g g^-1.
4. Specific growth rate was found to be higher when fermentation was done in the bioreactor compared to shake flasks. The controlling pH and temperature in the bioreactor may provide better aeration and mixing. This leads to better mass and energy transfer, resulting in faster and higher growth.

ACKNOWLEDGMENTS

The authors wish to thank Ministry of Science, Technology and Innovation, Malaysia (MOSTI) under SCIENCE-FUND (Project: 03-01-04-SF0815) and Ministry of Agriculture (MOA) under AGRI-FUND (Project: 05-01-04-SF1127) for financial support.

REFERENCES CITED

Aguiar, W. B., Faria, L. F. F., Couto, M. A. P. G., Araujo, O. Q. F., and Pereira, N. (2002). “Growth model and prediction of oxygen transfer rate for xylitol production from D-xylose by C. guilliermondii,” Biochemical Engineering Journal 12(1), 49-59.

Ahmad, M. N., and Holland, C. R. (1995). “Growth kinetics of single-cell protein in batch fermenters,” Journal of Food Engineering 26, 443-452.

Canilha, L., Carvalho, W., Giulietti, M., and Almeida e Silva, J. B. (2008). “Xylitol bioproduction from wheat straw: Process engineering and technical feasibility,” Journal of Biotechnology136(Supplement 1), S51-S51.

Chew, T. A., Isa, A. H. M., and Mohayidin, M. G. (1999). “Sago (Metroxylon sagu Rottboll),” Journal of Sustainable Agriculture14(4), 5-17.

Foyle, T., Jennings, L., and Mulcahy, P. (2007). “Compositional analysis of lignocellulose materials: Evaluations of methods used for sugar analysis of waste paper and straw,” Bioresource Technology 98, 3026-3036.

Ke, K. C., Jian, A. Z., Hong, Z. L., Wen, X. P., Huang, W., Jing, P. G., and Jing, M. X., (2009). “Optimization of pH and acetic acid concentration for bioconversion of hemicellulose from corncobs to xylitol by Candida tropicalis,” Biochemical Engineering Journal 43, 203-207.

Mohamad, N. L., Mustapa Kamal, S. M., and Abdullah, A. G. L. (2011). “Optimization of xylose production from sago trunk cortex by acid hydrolysis,” African Journal of Food Science and Technology 2(5), 102-108.

Monod, J. (1979). “The growth of bacterial cultures,” Ann. Rev. Microbiol. 3, 371-393

Mustapa Kamal, S. M., Mohamad, N. L., Abdullah, A. G. L., and Abdullah, N. (2011). “Detoxification of sago trunk hydrolysate using activated charcoal for xylitol production,” Procedia Food Science __, 908-913.

Nigam, P., and Singh, D. (1995). “Processes of fermentative production of xylitol – A sugar substitute,” Process Biochemistry30(2), 117-124.

Parajó, J. C., Domínguez, H., and Domínguez, J. (1998a). “Biotechnological production of xylitol. Part 1: Interest of xylitol and fundamentals of its biosynthesis,” Bioresource Technology65(3), 191-201.

Sampaio, F.C., de Moraes, C.A., De Faveri, D., Perego, P., Converti, A., and Passos, F.M.L., (2006). “Influence of temperature and pH on xylitol production from xylose by Debaryomyces hansenii UFV-170,” Process Biochemistry 41, 675-681.

Soleimani, M., Tabil, L., and Panigrahi, S. (2006). “Bio-production of a polyol (xylitol) from lignocellulosic resources: A review,” Paper presented at the CSBE/SCGAB 2006 Annual Conference.

Sun, Y., and Cheng, J. (2002). “Hydrolysis of lignocellulosic materials for ethanol production: A review,” Bioresource Technology 83, 1-11.

Walther, T., Hensirisak, P., and Agblevor, F. A. (2001). “The influence of aeration and hemicellulosic sugars on xylitol production by Candida tropicalis,” Bioresource Technology 76(3), 213-220.

Wang, L., Wu, D., Tang, P., Fun, X., and Yuan, Q. (2012). “Xylitol production from corncob hydrolysate using polyurethane foam with immobilized Candida tropicalis,” Carbohydrate Polymers 90, 1106-1113.

Winkelhausen, E., and Kuzmanova, S. (1998). “Microbial conversion of D-xylose to xylitol,” Journal of Fermentation and Bioengineering 86(1), 1-14.

Winkelhausen, E., Malinovska, R.J., Velickova, E., Kusmanova, S. (2007). “Sensory and microbiological quality of a baked product containing xylitol as an alternative sweetener,” International Journal of Food Properties 10, 639-649.

Article submitted: December 18, 2012; Peer review completed: March 3, 2013; Revised version received: March 22, 2013; Accepted: March 23, 2013; Published: April 1, 2013.