improving lipid production from bagasse hydrolysate with trichosporon fermentans by response surface...

Research

Pap

er

RESEARCH PAPER New Biotechnology �Volume 29, Number 3 � February 2012

Improving lipid production from bagassehydrolysate with Trichosporon fermentansby response surface methodology

Chao Huang1, Hong Wu1, Ri-feng Li1 and Min-hua Zong2

1 Laboratory of Applied Biocatalysis, College of Light Industry and Food Sciences, South China University of Technology, Guangzhou 510640, China2 State Key Laboratory of Pulp and Paper Engineering, College of Light Industry and Food Sciences, South China University of Technology,

Guangzhou 510640, China

Oleaginous yeast Trichosporon fermentans was proved to be able to use sulphuric acid-treated sugar cane

bagasse hydrolysate as substrate to grow and accumulate lipid. Activated charcoal was shown as effective

as the more expensive resin Amberlite XAD-4 for removing the inhibitors from the hydrolysate. To

further improve the lipid production, response surface methodology (RSM) was used and a 3-level

4-factor Box–Behnken design was adopted to evaluate the effects of C/N ratio, inoculum concentration,

initial pH and fermentation time on the cell growth and lipid accumulation of T. fermentans. Under the

optimum conditions (C/N ratio 165, inoculum concentration 11%, initial pH 7.6 and fermentation time

9 days), a lipid concentration of 15.8 g/L, which is quite close to the predicted value of 15.6 g/L, could be

achieved after cultivation of T. fermentans at 25 8C on the pretreated bagasse hydrolysate and the

corresponding lipid coefficient (lipid yield per mass of sugar, %) was 14.2. These represent a 32.8%

improvement in the lipid concentration and a 21.4% increase in the lipid coefficient compared with the

original values before optimization (11.9 g/L and 11.7). This work further demonstrates that

T. fermentans is a promising strain for lipid production and thus biodiesel preparation from abundant

and inexpensive lignocellulosic materials.

IntroductionMicrobial oils, namely single cell oils (SCOs), have attracted more

and more attention as a promising feedstock for biodiesel produc-

tion because of their similarity to vegetable oils in fatty acid

composition [1]. However, the high cost of the culture medium

makes microbial oils less economically competitive. As a result,

production of microbial oils from wastes or renewable materials is

of significant importance. Up to date, several types of agro-indus-

trial residues, including sewage sludge, glycerol and monosodium

glutamate wastewater, have been used for this purpose [2–4], but

using lignocellulosic biomass seems to be a better strategy for cost-

effective preparation of lipids on a large scale because of its low

cost and most availability in nature.

Recent reports on lipid production with oleaginous microorgan-

ism on the synthetic medium containing xylose have suggested the

Corresponding authors: Zong, M.-h. ([email protected]), Wu, H. ([email protected])

372 www.elsevier.com/locate/nbt 1871-6784/$

possibility of microbial oils production from lignocellulosic

hydrolysate [3,5,6] and our work confirmed that Trichosporon fer-

mentans, a kind of oleaginous yeast belonging to the family of

Cryptococcaceae, could use the detoxified rice straw acid hydrolysate

for microbial lipid production although the lipid concentration and

lipid content were lower than that with glucose as the sole carbon

source (11.5 g/L vs. 17.5 g/L; 40.1% vs. 62.4%) [7]. And it is expect-

able that optimizing the fermentation conditions by some statistical

methods can improve the lipid production of T. fermentans on

lignocellulosic hydrolysate. Response surface methodology (RSM)

is a statistical technique for designing experiments, building mod-

els, evaluating the effect of the factors and searching for optimal

conditions [8]. During the past decades, RSM has been extensively

applied in the optimization of medium composition, fermentation

conditions and food manufacturing processes [9–11]. Conse-

quently, the influences of the key variables on the microbial oil

fermentation by T. fermentans could be examined by RSM.

- see front matter � 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.nbt.2011.03.008

mailto:[email protected]

mailto:[email protected]

http://dx.doi.org/10.1016/j.nbt.2011.03.008

New Biotechnology �Volume 29, Number 3 � February 2012 RESEARCH PAPER

TABLE 1

Coding and levels of experiment factors

Factor Symbol Code level

�1 0 1

C/N ratioa X1 130 170 210

Inoculum concentration X2 5% 10% 15%

Initial pH X3 7.0 7.5 8.0

Fermentation time X4 8 days 9 days 10 days

a The sugar concentration in the bagasse hydrolysate was fixed at 123.5 g/L. Yeast extract,

whose concentration was fixed at 0.5 g/L, and peptone were used as nitrogen source

(assuming that peptone contains 14% N (w/w) and 8% C (w/w), and yeast extract includes

7% N (w/w) and 12% C (w/w)). Different C/N ratio was obtained by altering the nitrogen

source concentration.

ResearchPap

er

Sugar cane is widely planted in many countries, especially in

Brazil, India, Cuba, Mexico, Indonesia, Colombia and China [12]

and bagasse, the fibrous residue after extracting the juice from

sugar cane in the sugar production process, represents one of the

major lignocellulosic materials in Southern China. But to date,

little has been done on its value-added utilization and there has

been no report on efficient lipid fermentation from sugar cane

bagasse. Compared with other lignocellulosic biomass [7,13],

bagasse seems to be more competitive for lipid production

because of its low collection cost as a result that it can be used

directly for hydrolysis and the following fermentation in sugar

plants. Hence, in this work, the potential of T. fermentans in

accumulating lipid on bagasse hydrolysate was examined for the

first time. By contrast, resin was most commonly used for lig-

nocellulosic hydrolysate detoxification [7,14]; however, this

undoubtedly increased the whole process cost because of its

high price. So, a cheaper adsorbent, activated charcoal was

utilized in this work for evaluating its potential in removing

the inhibitors in lignocellulosic hydrolysate. For an enhanced

lipid production, RSM was adopted to optimize the fermentation

parameters.

Materials and methodsBagasse hydrolysate preparationBagasse from Guangdong Province in Southern China was

smashed to less than 0.5 mm and then mixed with dilute sulphuric

acid (1.5%, v/v) to give a mixture with a solid loading of 10% (w/v).

The mixture was treated in an autoclave at 121 8C for 90 min and

the liquid fraction was separated by vacuum filtration after cooling

and stored at 4 8C before use.

Detoxification procedureSulphuric acid-treated bagasse hydrolysate (SABH) was

detoxified before fermentation and the detoxification includes

overliming, concentration and adsorption. The procedure of

overliming and concentration was carried out as described pre-

viously [7]. As for adsorption, activated charcoal was used as the

adsorbent instead of resin Amberlite XAD-4 for its much lower

price. It was washed with water and equilibrated with 0.4 M HCl,

and then mixed with the concentrated hydrolysate (1/10, w/v).

The mixture was incubated at 30 8C, 200 rpm for 1 h and the

following filtration resulted in the detoxified hydrolysate, whose

pH was then adjusted to fermentation pH value with Ca(OH)2 or

5 M H2SO4.

Microorganism, media, precultivation and cultivationT. fermentans CICC 1368 was supplied by China Center of Indus-

trial Culture Collection and kept on wort agar at 4 8C. The pre-

culture was performed on precultivation medium (g/L, xylose 20,

peptone 10, yeast extract 10) in a 250 ml conical flask containing

50 ml fermentation broth at 28 8C and 160 rpm for 24 h. Then,

10% seed culture was inoculated to the culture medium. In

addition to the hydrolysate (detoxified SABH), the culture med-

ium also contained (g/L): yeast extract 0.5, peptone 1.8,

MgSO4�7H2O 0.4, KH2PO4 2.0, MnSO4�H2O 0.004, CuSO4�5H2O

0.0001. Cultivation was performed in a 250 ml conical flask

containing 50 ml fermentation broth in a rotary shaker at

25 8C and 160 rpm.

Optimization of lipid production by Box–Behnken design (BBD)A 3-level 4-factor Box–Behnken design was adopted to evaluate the

effects of C/N ratio (X1), inoculum concentration (X2), initial pH

value (X3) and fermentation time (X4) on the lipid production of T.

fermentans on SABH and a model was developed. In this study, the

experimental plan contained 29 trials and the independent vari-

ables were studied at three different levels, namely low (�1),

medium (0) and high (+1), whose values are shown in Table 1.

All the experiments were done in triplicate and the average lipid

concentration obtained after fermentation was taken as the

response variable (Y). The experimental design used in this work

is shown in Table 2. The response variable was fitted by a second-

order model to correlate the response variables to the independent

variables. The second order polynomial coefficients were calcu-

lated and analyzed using the ‘Design Expert’ software (Version 7.0,

Stat-Ease Inc., Minneapolis, USA). The general form of the second-

degree polynomial equation is:

Y ¼ b0 þX

biXi þ bi jXiX j þX

biiX2i : (1)

where Y is the predicted response; b0 the intercept, bi the linear

coefficient, bij the quadratic coefficient, bii is the linear-by-linear

interaction between Xi and Xj regression coefficients and Xi, Xj are

input variables that influence the response variable Y.

Statistical analysis of the model was performed to evaluate the

analysis of variance (ANOVA). This analysis included Fisher’s F-

test (overall model significance), its associated probability p(F),

correlation coefficient R, determination coefficient R2 which

measure the goodness of fit of regression model. For each variable,

the quadratic models were represented as contour plots (3D)

and response surface curves were generated using Design Expert

software.

Analytical methodsBiomass, lipid content and fatty acid profile of the lipid were

determined as described by Zhu et al. [6]. The hydrolysate samples

were analyzed by HPLC, using a method described before [7] and

the total reducing sugar concentration was measured by the DNS

method [15].

Results and discussionMicrobial oil production on pretreated bagasse hydrolysateThere are four main kinds of monosaccharide in SABH, namely,

xylose, glucose, arabinose and galactose and pentose was about

www.elsevier.com/locate/nbt 373


TABLE 2

Box–Behnken design arrangement and responses

Run Code level Lipid yield (g/L)

X1 X2 X3 X4 Actual value Predicted value

1 0 �1 1 0 9.50 9.21

2 �1 1 0 0 10.55 10.83

3 1 0 0 1 7.95 7.58

4 1 0 �1 0 5.74 6.28

5 0 0 0 0 15.30 15.31

6 1 �1 0 0 7.29 6.57

7 0 1 0 1 8.27 8.28

8 0 1 0 �1 10.78 10.93

9 �1 �1 0 0 6.52 6.97

10 0 0 0 0 14.95 15.31

11 0 0 1 �1 13.22 12.37

12 0 0 1 1 9.32 9.52

13 0 0 0 0 15.73 15.31

14 �1 0 �1 0 11.39 10.89

15 0 �1 0 �1 7.06 7.47

16 1 1 0 0 8.41 7.51

17 �1 0 0 �1 10.63 11.03

18 0 1 1 0 9.39 9.54

19 0 0 0 0 15.10 15.31

20 �1 0 1 0 10.18 10.06

21 0 0 �1 �1 9.83 9.18

22 0 �1 0 1 6.68 6.95

23 1 0 1 0 10.04 10.95

24 1 0 0 �1 9.08 9.62

25 0 0 0 0 15.49 15.31

26 �1 0 0 1 10.41 9.90

27 0 1 �1 0 9.37 9.69

28 0 �1 �1 0 5.33 5.22

29 0 0 �1 1 8.46 8.86

B

A

12111098765432100

20

40

60

80

100

120

Suga

r con

cent

ratio

n (g

/L)

Fermentation time (d)

1211109876543210

0

5

10

15

20

25

30

35

40

15

20

25

30

35

40

Lip

id c

onte

nt (%

)

Bio

mas

s (g/

L) /

lipid

yie

ld (g

/L)

Fermentation time (d)

FIGURE 1

Microbial oil production on sulphuric acid-treated sugar cane bagasse

hydrolysate by T. fermentans. (a) Time course of cell growth and lipid

accumulation. (~) Biomass; (&) lipid yield; (*) lipid content; (b) time courseof sugar utilization. (&) Total sugars; (~) xylose; (!) glucose; (5) arabinose;

(^) galactose. Fermentation conditions: inoculum concentration 10%, initial

pH 6.5, 25 8C, 160 rpm.

Research

Pap

er

five times more than hexose (Table 3). As seen from Table 3, before

detoxification, the concentration of each tested inhibitor in

bagasse hydrolysate was nearly the same as that in rice straw

hydrolysate and using cheap activated charcoal as adsorbent

yielded a hydrolysate of similar inhibitor concentration to that

of rice straw hydrolysate treated with resin Amberlite XAD-4 and

TABLE 3

Composition of bagasse hydrolysate after each step of detoxificatio

Compound (g/L) Untreated hydrolysate O

Glucose 5.2

Xylose 30.2 2

Galactose 1.5

Arabinose 3.9

Acetic acid 1.7

Furfural 0.72

5-HMF 0.04

374 www.elsevier.com/locate/nbt

suitable for microbial oil fermentation [7]. This can remarkably

decrease the cost of hydrolysate detoxification. The time courses of

cell growth and lipid accumulation of T. fermentans on the detox-

ified SABH are shown in Fig. 1a, and the sugar concentration

during the fermentation is presented in Fig. 1b. Similar to the

fermentation on rice straw hydrolysate in our previous work [7],

there was a clear lag phase at the beginning of fermentation as

indicated by the phenomenon that on the 1st day, the growth and

lipid accumulation were hardly detected. The presence of some

inhibitors in the fermentation broth partly contributes to this.

After that, the biomass began to increase, with the specific growth

n treatment

verliming Concentration Adsorption

4.8 18.1 16.8

8.2 116.1 92.9

1.3 5.1 2.4

3.4 14.2 11.4

1.4 4.0 3.7

0.39 0.00 0.00

0.03 0.11 0.02


ResearchPap

er

rate increasing from 0.03 h�1 during the first day to 0.08 h�1

during the second day, indicating that the cells had gradually

adapted to the fermentation environment. From day 2 to day 3,

hexoses (glucose and galactose) were gradually used up and the

cells began to use pentoses (xylose and arabinose) simultaneously

as the carbon source. Obviously, the four sugars can be completely

utilized by T. fermentans cells although the cells used them in

different orders and at varying rates. From the 2nd day, biomass

increased steadily and reached the maximal at day 11. By contrast,

lipid accumulation was also delayed with quite a low lipid content

during the first three days, in contrast to our previous study using

glucose as the sole carbon source where no obvious delay was

recorded [6]. There are several possible reasons for this. First, the

nitrogen source concentration was quite high at the beginning of

fermentation, and generally only under nitrogen deficiency con-

ditions could the lipid be biosynthesized and accumulated in a

large quantity [16]. Second, the inhibitors in the hydrolysate

greatly inhibited the growth and lipid accumulation of oleaginous

yeast [17]. From day 3, the lipid content of the yeast cells increased

sharply and reached the maximum of 39.9% on day 9. After that, a

clear decline in lipid content was observed. However, the biomass

went up continuously. The similar phenomena were also reported

previously [6,7,18] and the use of the accumulated lipid for cell

growth accounts for this. The maximum lipid concentration of

11.9 g/L was achieved on the 9th day and the corresponding lipid

coefficient (lipid yield per mass of sugar, %) was 11.7. Compared

with our previous works [6,7], these results are close to those on the

pretreated rice straw hydrolysate (11.5 g/L and 11.9), but still

much lower than those on the synthetic nitrogen-limited medium

without any inhibitors (17.5 g/L and 20.6) under the same fer-

mentation conditions.

TABLE 4

Analysis of variance (ANOVA) for the quadratic model

Source DF Sum of squares Mean squar

Model 14 242.73 17.34

X1 1 10.40 10.40

X2 1 17.26 17.26

X3 1 11.08 11.08

X4 1 7.54 7.54

X1X2 1 2.12 2.12

X1X3 1 7.59 7.59

X1X4 1 0.21 0.21

X2X3 1 4.31 4.31

X2X4 1 1.13 1.13

X3X4 1 1.60 1.60

X12 1 62.67 62.67

X22 1 116.52 116.52

X32 1 45.92 45.92

X42 1 46.27 46.27

Residual 14 6.06 0.43

Lack of fit 10 5.67 0.57

Pure error 4 0.38 0.10

Total 28 248.78

R2 = 0.9757; Adj. R2 = 0.9513.

Optimization of fermentation parameters with RSMAmong different optimization method, RSM has attracted more

and more attention because of its non-requirement on the calcu-

lation of the local sensitivity of each design variable and being

effective for both the single- and multi-disciplinary optimization

problems. So RSM was adopted to further improve the lipid

production in this work.

The single factor experimental results (data not shown) sug-

gested that the major factors affecting the lipid production on

pretreated SABH were C/N ratio, inoculum concentration, initial

pH, and fermentation time. Thus, a 3-level 4-factor BBD was used

to evaluate the effects of the above-mentioned four factors on the

lipid production. The corresponding BBD and experimental data

are shown in Table 2. It can be seen from Table 2 that the lipid

concentration was significantly influenced by the fermentation

conditions. After the analysis of variance which gave the level of

response as a function of four independent variables by employing

multiple regression analysis, the regression equation was obtained.

A quadratic model for the lipid concentration of T. fermentans is

given below (in terms of coded factors):

Y ¼ 15:31 � 0:93X1 þ 1:20X2 þ 0:96X3 � 0:79X4 � 0:73X1X2

þ 1:38X1X3 � 0:23X1X4 � 1:04X2X3 � 0:53X2X4

� 0:63X3X4 � 3:11X12 � 4:24X2

2 � 2:66X32 � 2:67X4

2; (2)

where Y is the predicted lipid concentration (g/L) and X1, X2, X3

and X4 were C/N ratio (mol/mol), inoculum concentration (%),

initial pH and fermentation time (days), respectively.

Table 4 shows the results of the quadratic polynomial model in

the form of analysis of variance (ANOVA), which was done by the

software Design-Expert. As shown in Table 4, the model was highly

e F-value P-value Coefficient estimate

40.08 <0.0001

24.04 0.0002 �0.9339.89 <0.0001 1.20

25.61 0.0002 0.96

17.42 0.0009 �0.794.89 0.0441 �0.73

17.55 0.0009 1.38

0.48 0.5004 �0.239.95 0.0070 �1.042.62 0.1277 �0.533.70 0.0750 �0.63

144.88 <0.0001 �3.11269.37 <0.0001 �4.24106.17 <0.0001 �2.66106.97 <0.0001 �2.67

5.93 0.0505



Research

Pap

er

significant with a very high model F-value (40.08) and a very low P-

value (P < 0.0001). The R2 value (0.9757) indicated a good agree-

ment between the experimental and the predicted values and

showed that the model was reliable for lipid production in this

work. The value of adj-R2 (0.9513) suggested that the total varia-

tion of 95.13% for the lipid concentration was attributed to the

independent variables and only about 4.87% of the total variation

could not be explained by the model. The lack-of-fit value was not

FIGURE 2

Response surface plots showing binary interaction of different variables. The interafermentation time and C/N, (d) initial pH and inoculum concentration, (e) fermentat


significant (P = 0.0505), indicating that the equation was adequate

for predicting lipid concentration under all conditions. The model

coefficients for each variable are also shown in Table 4 and F-value

and P-value were employed to check the significance of each

coefficient of the model. The larger F-value and smaller P-value

suggested higher significance of the corresponding coefficient.

Among the model terms, X1, X2, X3, X4, X1X3, X2X3, X12, X2

2,

X32, X4

2 were significant at the 99% probability level and X1X2 was

ction between (a) inoculum concentration and C/N, (b) initial pH and C/N, (c)ion time and inoculum concentration and (f) fermentation time and initial pH.


TABLE 5

Fatty acid composition of plant oils and lipid from T. fermentansa

Lipid source Fatty acid composition of lipid (%)

C14:0 C16:0 C18:2 C18:1 C18:0 Others

Canola NDb 4–5 20–31 55–63 1–2 10–12

Corn NDb 7–13 39–52 30.5–43 2.5–3 1

Olive 1.3 7–18.3 4–19 55.5–84.5 1.4–3.3 NDb

Palm 0.6–2.4 32–46.3 6–12 37–53 4–6.3 NDb

Peanut 0.5 6–12.5 13–41 37–61 2.5–6 1

Safflower (high oleic) NDb 4–8 11–19 73.6–79 2.3–8 NDb

Soybean NDb 2.3–11 49–53 22–30.8 2.4–6 2–10.5

T. fermentans 0.7 27.5 10.1 54.2 5.8 1.7a The composition of different plant oils was described as the previous work [20].b ND means not detected. R

esearchPap

er

at the 95% probability level. By contrast, other terms were not

significant.

The relationship between the response and experimental

levels of each variable can be demonstrated by three-dimen-

sional response surface plots which represented the regression

equation mentioned above. And the optimal levels of the vari-

ables can also be determined visually be these plots [19]. The

response surface curves are shown in Fig. 2. As shown in the

surface plots, there were obvious interactions between each pair

of variables. The interaction of C/N ratio and initial pH value, C/

N ratio and inoculum concentration, and inoculum concentra-

tion and initial pH value was significant (P-value less than 0.05).

Among them, the interaction of X1X3 was positive to the lipid

concentration whereas the other two were negative. Surpris-

ingly, the interaction between fermentation time (X4) and

each other three variances was not significant, indicating that

fermentation time showed little influence on the other three

variables. According to the analysis by the ‘Design-expert’ soft-

ware, the optimal values of the four key variables for lipid

fermentation of T. fermentans were C/N ratio 164.82, inoculum

concentration 10.73%, initial pH 7.57 and fermentation time

8.83 days, respectively.

TABLE 6

Lipid production on different agro-industrial residues by various m

Strain Carbon source Initial casource co(g/L)

Y. lipolytica Industrial fats 10.0

M. isabellina Glycerol 26.8

Mucor sp. RRL001 Tapioca starch ND a

C. echinulata Glycerol 26.7

L. starkeyi Sewage sludge 40.0b

R. glutinis Monosodium glutamate wastewater 35.0

T. fermentans Molasses 150.0

T. fermentans Rice straw hydrolysate 116.9

T. fermentans Sugar cane bagasse hydrolysate 123.5

a ND means not detected.b About 40 g/L glucose was added into sewage sludge.

Lipid production under optimal conditionsFor operation convenience, the optimal fermentation para-

meters were set as follows: C/N ratio 165, inoculum concentra-

tion 11%, initial pH 7.6 and fermentation time 9 days. Under

these conditions, a lipid concentration of 15.8 g/L, which is

quite close to the predicted value of 15.6 g/L (relative error

1.41%), could be obtained after cultivation of T. fermentans at

25 8C on the pretreated SABH whose sugar concentration was

123.5 g/L. And the corresponding lipid coefficient was 14.2.

These represented a 32.8% improvement in the lipid concentra-

tion and a 21.4% increase in the lipid coefficient compared with

the original values before optimization with RSM (11.9 g/L and

11.7), indicating that RSM is a powerful tool for optimization of

fermentation process.

As can be seen from Table 5, the lipid from T. fermentans mainly

contained palmitic acid, stearic acid, oleic acid and linoleic acid,

and the unsaturated fatty acids amounted to about 65%. A com-

parative result of the fatty acid composition between the lipid

from T. fermentans and most commonly used plant oils in biodiesel

production is also shown in Table 5. Apparently, the fatty acid

composition of the lipid is similar to that of vegetable oils, and the

lipid is a promising feedstock for biodiesel production.

icroorganisms

rbonncentration

Lipidconcentration(g/L)

Volumetricproductivity(g/L day)

Reference

3.8 2.4 [21]

3.3 0.3 [3]

5.0 ND a [22]

2.0 0.1 [3]

6.4 0.8 [2]

5.0 1.6 [4]

12.8 1.8 [6]

11.5 1.4 [7]

15.8 1.8 This work



Research

Pap

er

The results of this work and those with different agro-industrial

residues as feedstock are depicted in Table 6. Although the volu-

metric productivity of T. fermentans on bagasse hydrolysate is not

as high as that of Y. lipolytica on industrial fats, the higher lipid

concentration and moderate volumetric productivity make olea-

ginous yeast T. fermentans very promising for lipid production

from abundant and inexpensive lignocellulosic materials.

ConclusionsSulphuric acid-treated sugar cane bagasse hydrolysate can be effi-

ciently used for the cell growth and lipid accumulation of T.

fermentans. Optimization of fermentation parameters by RSM

resulted in a 32.8% increase in the lipid concentration of T. fermen-

tans on SABH and a lipid concentration of 15.8 g/L, which is quite

close to the predicted value of 15.6 g/L, could be obtained after

cultivation of T. fermentans under the optimal culture conditions.


Also, this represented a 21.4% improvement in the lipid coefficient.

The lipid from T. fermentans has similar fatty acid composition to

that of vegetable oils, thus it is promising for biodiesel production.

AcknowledgementsWe acknowledge the National Natural Science Foundation of

China (Grant No. 31071559), the Guangdong Province

Cooperation Project of Industry, Education and Academy (Grant

No. 2008A010700006), the Science and Technology Project of

Guangdong Province (Grant No. 2009B080701085), the Open

Project Program of the State Key Laboratory of Pulp and Paper

Engineering, SCUT (Grant No. 200823), the Fundamental

Research Funds for the Central Universities, SCUT (Grant No.

2009zm0199, 2009zz0026, 2009zz0018) and Major State Basic

Research Development Program ‘973’ (Grant No. 2010CB732201)

for financial support.

References

1 Li, Q. et al. (2008) Perspectives of microbial oils for biodiesel production. Appl.

Microbiol. Biotechnol. 80, 749–756

2 Angerbauer, C. et al. (2008) Conversion of sewage sludge into lipids by Lipomyces

starkeyi for biodiesel production. Bioresour. Technol. 99, 3051–3056

3 Fakas, S. et al. (2009) Evaluating renewable carbon sources as substrates for single

cell oil production by Cunninghamella echinulata and Mortierella isabellina. Biomass

Bioenergy 33, 573–580

4 Xue, F.Y. et al. (2008) Studies on lipid production by Rhodotorula glutinis

fermentation using monosodium glutamate wastewater as culture medium.

Bioresour. Technol. 99, 5923–5927

5 Zhao, X. et al. (2008) Medium optimization for lipid production through co-

fermentation of glucose and xylose by the oleaginous yeast Lipomyces starkeyi. Eur.

J. Lipid Sci. Technol. 110, 405–412

6 Zhu, L.Y. et al. (2008) Efficient lipid production with Trichosporon fermentans and

its use for biodiesel preparation. Bioresour. Technol. 99, 7881–7885

7 Huang, C. et al. (2009) Microbial oil production from rice straw hydrolysate by

Trichosporon fermentans. Bioresour. Technol. 100, 4535–4538

8 Myers, R. et al. (2009) Response Surface Methodology: Product and Process Optimization

Using Designed Experiments. John Wiley & Sons, New York

9 Cui, F.J. et al. (2006) Optimization of the medium composition for production of

mycelial biomass and exo-polymer by Grifola frondosa GF9801 using response

surface methodology. Bioresour. Technol. 97, 1209–1216

10 Li, X. et al. (2009) Optimization of culture conditions for production of yeast

biomass using bamboo wastewater by response surface methodology. Bioresour.

Technol. 100, 3613–3617

11 Wang, X. et al. (2008) Optimization of methane fermentation from effluent of bio-

hydrogen fermentation process using response surface methodology. Bioresour.

Technol. 99, 4292–4299

12 Cardona, C.A. et al. (2010) Production of bioethanol from sugarcane bagasse:

status and perspectives. Bioresour. Technol. 101, 4754–4766

13 Chen, X. et al. (2009) Screening of oleaginous yeast strains tolerant to

lignocellulose degradation compounds. Appl. Biochem. Biotechnol. 159,

1–14

14 de Mancilha, I.M. and Karim, M.N. (2003) Evaluation of ion exchange resins for

removal of inhibitory compounds from corn stover hydrolyzate for xylitol

fermentation. Biotechnol. Prog. 19, 1837–1841

15 Miller, G. (1959) Use of dinitrosalisylic acid (DNS) for determination of reducing

sugars. Anal. Chem. 31, 426–428

16 Ratledge, C. (2004) Fatty acid biosynthesis in microorganisms being used for Single

Cell Oil production. Biochimie 86, 807–815

17 Kimura, K. et al. (2006) Inhibition of lipid accumulation and lipid body formation

in oleaginous yeast by effective components in spices, carvacrol, eugenol, thymol,

and piperine. J. Agric. Food Chem. 54, 3528–3534

18 Fakas, S. et al. (2007) Compositional shifts in lipid fractions during lipid turnover

in Cunninghamella echinulata. Enzyme Microb. Technol. 40, 1321–1327

19 Haider, M. and Pakshirajan, K. (2007) Screening and optimization of media

constituents for enhancing lipolytic activity by a soil microorganism using

statistically designed experiments. Appl. Biochem. Biotechnol. 141, 377–390

20 Bajpai, D. and Tyagi, V.K. (2006) Biodiesel: source, production, composition,

properties and its benefits. J. Oleo Sci. 55, 487–502

21 Papanikolaou, S. et al. (2001) Kinetic profile of the cellular lipid composition in an

oleaginous Yarrowia lipolytica capable of producing a cocoa-butter substitute from

industrial fats. Anton. Van Leeuw. Int. J. Gen. Mol. Microbiol. 80, 215–224

22 Ahmed, S.U. et al. (2006) Effects of various process parameters on the production of

gamma-linolenic acid in submerged fermentation. Food Technol. Biotechnol. 44,

283–287

improving lipid production from bagasse hydrolysate with trichosporon fermentans by response surface...

Documents