SciELO - Scientific Electronic Library Online

 
vol.15 issue3RELATIONSHIP BETWEEN MORPHOLOGY, RHEOLOGY AND GLUCOAMYLASE PRODUCTION BY Aspergillus awamori IN SUBMERGED CULTURESFLUIDDYNAMIC ASPECTS OF GAS-PHASE ETHYLENE POLYMERIZATION REACTOR DESIGN author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Article

Indicators

Related links

Share


Brazilian Journal of Chemical Engineering

Print version ISSN 0104-6632

Braz. J. Chem. Eng. vol. 15 no. 3 São Paulo Sept. 1998

http://dx.doi.org/10.1590/S0104-66321998000300005 

USE OF FRECTIONAL FACTORIAL DESIGN FOR SELECTION OF NUTRIENTS FOR CULTURING Paecilomyces variotii IN EUCALYPTUS HEMICELLULOSIC HYDROLYSATE

 

J.B. Almeida e Silva1*, U.A. Lima2, M.E.S. Taqueda3 and F.G. Guaragna2
1Departamento de Biotecnologia-FAENQUIL, 12600-000 Lorena-SP-Brazil
FAX- (0055) 012-553-3165 E-mail: feqlbio@eu.ansp.br
2Departamento de Tecnologia Bioquímico-Farmacêutica, Faculdade de Ciências Farmacêuticas da USP
3Departamento de Engenharia Química, Escola Politécnica da USP

 

(Received: March 5, 1998; Accepted: June 30, 1998)

 

Abstract - A eucalyptus hemicellulose fraction was hydrolysed by treating eucalyptus wood chips with sulfuric acid. The hydrolysate was used as the substrate to grow Paecilomyces variotii IOC-3764 cultured for 72 or 96 hours. The influence of the inhibitors, nutrients and fermentation time was verified by a 28-4 and, subsequently, a 25-1 fractional factorial design. The effects of the inhibitors (acetic acid and furfural), nutrients (rice bran, urea, potassium nitrate, ammonium sulfate, magnesium sulfate and sodium phosphate) and fermentation time were investigated. The highest yield (10.59 g/L of biomass) was obtained when the microorganisms were cultivated for 72 hours in a medium composed of 30 g/L rice bran, 9.4 g/L ammonium sulfate (2 g/L nitrogen) and 2 g/L sodium phosphate.
Keywords: Fermentation, experimental design, fractional factorial, xylose, glucose, acetic acid, furfural, nutrients.

 

 

INTRODUCTION

Paecilomyces variotii is a fungus commonly found in the air and in the soils of tropical countries. This fungus has been largely employed for the production of microbial protein, owing to its good performance in highly polluting residues, such as vinasse (Cabib et al., 1993), sulfite liquor (Romantschuk and Lehtomaki, 1978), rayon hydrolysate (Bajpai and Bajpai, 1987) and wood hydrolysate (Almeida e Silva et al., 1995a; 1995b, Almeida e Silva, 1996).

This paper presents the results of various attempts to optimize the conditions for culturing Paecilomyces variotii in eucalyptus hemicellulosic hydrolysate, with a view to producing microbial protein. This microorganism was selected from among twenty-one species of yeasts and filamentous fungi for its excellent performance in eucalyptus hemicellulosic hydrolysate (Almeida e Silva et al., 1995a). The protein biomass produced by this fungus during 72 hours of fermentation has all the amino acids necessary to feed humans and animals (Almeida e Silva et al., 1995b). The hemicellulosic hydrolysate employed in our assays presented high contents of acetic acid and furfural, which are regarded as the main inhibitors of microbial growth (Tran and Chambers, 1985).

According to Weigert et al. (1988), furfural exerts a negative influence on respiration and oxidative phosphorylation, retarding the growth of the microorganism. Acetic acid toxicity during the xylose-to-xylitol bioconversion was observed by Felipe et al. (1995), who found that the tolerance for this inhibitor depends on the species of microorganism and mainly on the pH of fermentation.

In the present study a fractional factorial design followed by the response surface methodology permitted the observation of the influences of these inhibitors and some nutrients (urea, potassium nitrate, ammonium sulfate, magnesium sulfate, sodium phosphate and rice bran), as well as the influence of the fermentation time necessary for P. variotii to grow in eucalyptus hemicellulosic hydrolysate.

 

MATERIALS AND METHODS

Raw Material

The hemicellulosic hydrolysate was obtained by cooking Eucalyptus grandis chips averaging 20 x 10 x 5 mm and containing 29% moisture. The ratio of chips (w/v) to 0.35% sulfuric acid solution

was 1: 4.5, the temperature was 156ºC and the time of hydrolysis was 27 minutes. The hydrolysate thus obtained was characterized with respect to pH, sugars, acetic acid and furfural.

The microorganism used was P. variotii IOC-3764 cultivated on a malt-agar slant.

Preparation of the Inoculum

The hydrolysate pH was adjusted to 4.0 with commercial CaO and flowing steam for 20 minutes. The nutrient solutions, except for that of urea, were sterilized separately at 121ºC for 20 minutes. The urea solution was sterilized by ultrafiltration. The rice bran suspension was sterilized at 120ºC for 15 minutes and aseptically centrifuged at 2,000 x G for 15 minutes. Only the supernatant was added to the medium.

A nongerminated-spore suspension obtained from P. variotii 3764 cultured for 96 hours at 30ºC on a malt-agar slant was utilized as the inoculum. Ten milliliters of 0.8% TWEEN 80 solution (v/v) was added to the tube containing the spore culture. After scraping the surface of the malt agar, the spore suspension was transferred to a tube and vigorously shaken. After counting the spores in a Neubawer chamber, the medium was inoculated with an amount of suspension suitable for initially containing 106 spores per milliliter.

Erlenmeyer flasks (250 mL) containing 100 mL of culture medium were agitated in a rotary incubator (model ETICA 430-R) at 200 rpm for 120 hours at 30ºC. Samples were taken at the end of the agitation time for analysis of residual sugar and dry matter.

The influence of inhibitors, nutrients and fermentation time on the growth of the microorganisms was studied using a 28-4 fractional factorial design (Box et al., 1978). The results showed that some of the factors did not affect the performance of the microorganisms significantly. As a consequence, a 25-1 fractional factorial design was employed. In addition, a rotary factorial design with a center point was used in order to obtain the response surface for the process.

The hydrolysate was diluted with distilled water until half the initial concentration was obtained for observation of the effects of the inhibitors (IN), namely acetic acid and furfural. Then it was adjusted for initial levels of xylose and glucose. The effects of fermentation time (FT) and nutrients, namely rice bran (RB) as a source of vitamins and amino acids, urea (UR), potassium nitrate (PN), ammonium sulfate (AS) as a source of nitrogen, potassium, sulfur, magnesium sulfate (MS) and sodium phosphate (SP), were also studied.

The factor levels used in both designs are shown in Table 1, where (-1) and (+1) represent the lowest and the highest levels, respectively. The data were analysed using the Statgraphics program, version 5.0.

Analytical Methods

pH was determined at 25ºC with a Procyon potentiometer, model PHD-10.

The concentration of total reducing sugars (TRS) was determined by using the Somogyi and Nelson method (Nelson, 1944).

Furfural content was determined by high performance liquid chromatography (HPLC) using a Shimadzu C-R7A chromatograph with a RP-18 column 200 mm in length and 10µ m particles. The eluent used was a mixture of acetonitrile water at a ratio of 1:8 with 1% acetic acid flowing at 0.8 mL/min and 25ºC. A UV detector, model SPD-10A was used. The volume of the injected sample was 20 m L.

 

Table 1: Factor levels used for the 28-4 and 25-1 fractional factorial designs

FACTORS

ABBREVIATIONS

LEVELS

(-)

(+)

Furfural (g/L)

IN

1.00

2.4

Acetic acid (g/L)

IN

2.5

5.4

Urea (g/L)*

UR

0.0

2.0

Potassium nitrate (g/L)*

PN

0.0

2.0

Ammonium Sulfate(g/L)*

AS

0.0

2.0

Magnesium sulfate (g/L)

MS

0.0

1.5

Sodium phosphate (g/L)

SP

0.0

2.0

Rice bran (g/L)

RB

10.0

30.0

Fermentation time (h)

FT

72.0

96.0

* Concentration in g/L of nitrogen

 

Xylose, glucose and acetic acid contents were determined by HPLC using a Shimadzu C-R7A chromatograph with an Aminex 87H column, 300 x 7.8 mm. The eluent used was H2SO4O.02N flowing at 0.6 mL/min and 45ºC. The volume of the injected sample was 20m L.

The mass of microorganisms formed during the fermentative process was assayed by direct weighing. A 325 mesh Tyler sieve, 0.044 mm opening, was used to separate the microorganisms from the medium. The microorganisms were retained by the sieve. The mycelium was washed in flowing water to eliminate the residual fermentation medium, dried at 80ºC at atmospheric pressure and weighed.

A statistical analysis was carried out using the Statgraphics program, version 6.0.

 

RESULTS AND DISCUSSION

Table 2 shows a comparison of the chemical compositions of the hydrolysate used in previous work and the hydrolysate used in this study.

Although the hydrolysate employed in this study apparently had better characteristics than others previously tested under identical conditions, the microorganisms did not grow satisfactorily. For this reason, the effects of the inhibitors, nutrients and fermentation time on the growth of the fungi were investigated.

The design matrix with the amounts of biomass produced and the levels of inhibitors, nutrients and fermentation time are displayed in Table 3.

 

Table 2: Chemical composition of the eucalyptus hemicellulosic hydrolysate

Components

Concentration (g/L)

Xylose

14.6*

25.6

Glucose

10.2*

1.7

TRS

29.1*

30.0

Furfural

5.0*

2.1

Acetic acid

6.8*

5.4

* Almeida e Silva et al., 1995b

 

Table 3: Experimental matrix used in the 28-4 fractional factorial design and amounts of biomass produced

RUN

FACTORS

BIOMASS

IN

RB

UR

MS

AS

PN

SF

FJ

(g/L)*

1

-1

-1

-1

-1

-1

-1

-1

-1

5.75

2

+1

-1

-1

-1

-1

+1

+1

+1

6.70

3

-1

+1

-1

-1

+1

-1

+1

+1

11.12

4

+1

+1

-1

-1

+1

+1

-1

-1

10.67

5

-1

-1

+1

-1

+1

+1

+1

-1

4.92

6

+1

-1

+1

-1

+1

-1

-1

+1

5.35

7

-1

+1

+1

-1

-1

+1

-1

+1

2.81

8

+1

+1

+1

-1

-1

-1

+1

-1

10.83

9

-1

-1

-1

+1

+1

+1

-1

+1

6.08

10

+1

-1

-1

+1

+1

-1

+1

-1

7.27

11

-1

+1

-1

+1

-1

+1

+1

-1

9.68

12

+1

+1

-1

+1

-1

-1

-1

+1

4.20

13

-1

-1

+1

+1

-1

-1

+1

+1

3.90

14

+1

-1

+1

+1

-1

+1

-1

-1

3.78

15

-1

+1

+1

+1

+1

-1

-1

-1

11.57

16

+1

+1

+1

+1

+1

+1

+1

+1

7.39

* Average of two replicates

 

Table 4 shows estimates of the effects, standard errors and Student’s t test for the 28-4 fractional factorial design. Apparently the inhibitors (furfural and acetic acid) have no significant effects on the production of biomass at a probability level lower than 5% (P< 0.05). This also applies to magnesium sulfate (MS) and potassium nitrate (PN). Studies conducted by Almeida e Silva et al. (1995a) showed that P. variotii presents satisfactory development in hydrolysate containing up to 1.98 g/L of furfural. Roberto et al. (1991) noticed that the presence of furfural in sugar cane bagasse hemicellulosic hydrolysate does not influence the development of yeasts. With regard to acetic acid, Cabib et al. (1983) found that 95% of the organic acids present in the culture medium is consumed by Candida utilis and P. variotii. Felipe et al. (1995) also found that Candida guilliermondii is able to assimilate acetic acid when cultivated in sugar cane bagasse hemicellulosic hydrolysate.

There are also some second-order interactions with significant effects at a probability level lower than 5% (P< 0.05). However, they are unclear in this design and cannot be easily identified. Finally, there are some interaction effects that have no significant influence.

Table 4 also reveals that urea (UR) and fermentation time (FT) have significant but negative main effects (P< 0.05) at a probability level lower than 5%. This means that there is a tendency towards better results after 72 hours of fermentation when the levels of these factors are lower, i. e., when urea is absent. A biomass loss of 1.06 g/L was observed when the culture medium was supplemented with urea, whereas a biomass increase of 2.40 g/L occurred when the medium was supplemented with ammonium sulfate. The influence of the nitrogen source was studied in more detail by Almeida e Silva et al., 1995a.. The same authors observed that the cultivation in ammonium sulfate not only gives better results than in urea, but also keeps the pH values unaltered. The cultivation in urea tends to raise the pH value, probably owing to the consumption of organic acids present in the hydrolysate.

Based on the results of eight assays using the 2 8-4 fractional factorial design, eight more assays were performed to complete half of a fraction of the 25-1 fractional factorial design to obtain additional information about factors that really have significant effects on the growth of microorganisms (Table 5).

 

Table 4: Estimated effects, standard errors and Student’s t test for biomass production by P. variotii in the 28-4 fractional factorial design, as a function of inhibitor and nutrient concentrations and fermentation time

EFFECTS

ESTIMATES

STANDARD ERRORS

t

Average

6.845

± 0.173

-

Inhibitor (IN)

0.354

± 0.346

1.02

Rice bran (RB)

3.371

± 0.346

9.74*

Urea (UR)

-1.057

± 0.346

3.05*

Sodium phosphate (SP)

1.757

± 0.346

5.07*

Magnesium sulfate (MS)

-0.227

± 0.346

0.66

Potassium nitrate (PN)

-0.688

± 0.346

1.99

Ammonium sulfate (AS)

2.398

± 0.346

6.93*

Fermentation time (FT)

-1.808

± 0.346

5.22*

INxRB+URxSP+MSxFT+ASxPN

-0.877

± 0.346

2.53*

INxUR+RBxSP+MSxPN+ASxFT

0.686

± 0.346

1.98

INxMS+RBxFT+URxPN+ASxSP

-2.499

± 0.346

7.22*

INxAS+RBxPN+URxFT+MSxSP

-1.104

± 0.346

3.19*

INxPN+RBxAS+URxMS+SPxFT

0.912

± 0.346

2.63*

INxSP+RBxUR+MSxAS+PNxFT

0.292

± 0.346

0.84

INxFT+RBxMS+URxAS+PNxSP

-0.418

± 0.346

1.21

* Significance at 5% probability level

 

Table 6 shows the estimated effects, standard errors and Student’s t test for the production of biomass by P. variotii in a medium containing the factors mentioned in Table 1. The main effects of urea (UR), ammonium sulfate (AS) and sodium phosphate (SP), as well as the effects of second-order interactions between rice bran and fermentation time, urea and ammonium sulfate, urea and fermentation time, and ammonium sulfate and sodium phosphate exerted significant influences on the growth of the microorganisms at a 5% probability level, according to the Student’s t test results.

 

Table 5: Experimental matrix used in the 25-1 fractional factorial design and amounts of biomass produced

ASSAYS

FACTORS

BIOMASS

RB

UR

AS

SP

FT

(g/L)*

1

-1

-1

-1

-1

+1

3.37

2

+1

-1

-1

-1

-1

3.55

3

-1

+1

-1

-1

-1

3.78

4

+1

+1

-1

-1

+1

2.81

5

-1

-1

+1

-1

-1

5.53

6

+1

-1

+1

-1

+1

10.43

7

-1

+1

+1

-1

+1

5.35

8

+1

+1

+1

-1

-1

11.57

9

-1

-1

-1

+1

-1

2.93

10

+1

-1

-1

+1

+1

7.23

11

-1

+1

-1

+1

+1

3.90

12

+1

+1

-1

+1

-1

10.83

13

-1

-1

+1

+1

+1

11.69

14

+1

-1

+1

+1

-1

10.59

15

-1

+1

+1

+1

-1

4.92

16

+1

+1

+1

+1

+1

7.39

* Average of two replicates.

 

Table 6: Estimated effects, standard errors and Student’s t test for biomass production by P. variotii using the 25-1 fractional factorial design as a function of the inhibitor and nutrient concentrations and fermentation time

EFFECTS

ESTIMATES

STANDARD ERRORS

t

Average

6.614

± 0.219

-

RB

2.865

± .0438

6.55*

UR

-0.596

± 0.438

1.36

AS

3.633

± 0.438

8.30*

SP

1.634

± 0.348

3.73*

FT

-0.191

± 0.348

0.44

RbxUR

0.798

± 0.348

1.82

RbxAS

0.256

± 0.348

0.58

RBxSP

0.285

± 0.348

0.65

RBxFT

-1.978

± 0.348

4.52*

URxAS

-1.653

± 0.348

3.78*

URxSP

-0.754

± 0.348

1.72

URxFT

-2.721

± 0.348

6.22*

ASxSP

-1.210

± 0.348

2.77*

ASxFT

0.755

± 0.348

1.73

SPxFT

0.429

± 0.348

0.98

* Significance at 5% probability level

 

Increasing the rice bran concentration in the culture medium from 10 to 30 g/L increases the biomass concentration by 2.86 g/L, while adding ammonium sulfate and sodium phosphate increases the biomass production by 3.63 and 1.63 g/L, respectively.

Rice bran is an important source of vitamins and amino acids that greatly furthers the development of microorganisms. Almeida e Silva et al. (1995b) used different media for culturing P. variotii and found that the fungus had a good performance in media containing rice bran. Roberto et al. (1995) studied the influences of this factor both on the growth of Candida guilliermondii and on the production of xylitol and found that the presence of this nutrient enhances xylose consumption by 4.75 g/L and xylitol production by 4.69 g/L.

On the other hand, among the four second-order interactions with significant effects, urea and fermentation time stand out. The fact that their estimated effects are negative confirms the tendencies observed for their estimated main effects.

This paper is an attempt to demonstrate the applicability of statistical theories to the study of fermentative processes. Twenty-four assays provided a better understanding of the behavior of P. variotii cultivated in eucalyptus hemicellulose hydrolysate. The development of the microorganism is not affected by the contents of the compounds considered as inhibitors of microbial growth, especially furfural and acetic acid which are normally present in hydrolysate. The fungi develop better in a medium with no special treatment other than the adjustment of pH to 4.0. Only the following nutrients are required by the microorganism: a carbon source, ammonium sulfate and sodium phosphate. These characteristics are essential for the microorganism to be used as a microbial protein (Johnson and Harris, 1948).

The highest yield (10.59 g/L of biomass) was obtained when the microorganisms were cultivated for 72 hours in a medium composed of 30 g/L rice bran, 9.4 g/L ammonium sulfate (2.0 g/L nitrogen) and 2.0 g/L sodium phosphate.

Additional experiments are being carried out using mean levels of the factors selected (rice bran, ammonium sulfate and sodium phosphate), in accordance with an orthogonal factorial design, to obtain the process response surface.

 

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the financial support of CNPq/RHAE. The authors especially thank Dr. Roy Edward Bruns for his assistance and advice in the development of this work. Appreciation is also extended to Ms. Maria Eunice Machado Coelho for translating the manuscript into English.

 

REFERENCES

Almeida e Silva, J.B., Aplicação de modelagem matemática na produção de proteína microbiana por Paecilomyces variotii em hidrolisado hemicelulósico de eucalipto. Ph.D. diss., FCF, Universidade de São Paulo, São Paulo, Brazil (1996).         [ Links ]

Almeida e Silva, J.B., Mancilha, I.M., Teixeira, M.A. and Vanetti, M.C.D., Utilization of Hydrolysate of Eucalyptus Hemicellulose for Production of Microbial Protein, Arquivos de Biologia e Tecnologia, Curitiba, 38, 147-156 (1995a).         [ Links ]

Almeida e Silva, J.B., Mancilha, I.M., Vanetti, M.C.D. and Teixeira, M.A., Microbial Protein Production by Paecilomyces varioti Cultivated in Eucalyptus Hemicellulosic Hydrolysate, Bioresource Technology, 52, 197-200 (1995b).         [ Links ]

Bajpai, P. and Bajpai, P.K., Single Cell Protein Production from Rayon Pulp Mill Waste by Paecilomyces variotii, Journal of Fermentation Technology, 65, 349-351 (1987).         [ Links ]

Barros Neto, B., Scarminio, I.S. and Bruns, R.E., Planejamento e otimização de experimentos. Editora UNICAMP, Campinas, 299p. (1995).         [ Links ]

Box, G.E.P., Hunter, W.G. and Hunter, J.S., Statistics for Experimenters: An Introduction to Design, Data Analysis and Model Building. John Wiley & Sons, New York (1978).

Cabib, G., Silva, H.J., Giulietti, A. and Ertola, R., The Use of Sugar Cane Stillage for Single Cell Protein Production, Journal of Chemical Technology and Biotechnology, 33B, 21-28 (1983).         [ Links ]

Chen, H.C., Response Surface Methodology for Optimizing Citric Acid Fermentation by Aspergillus foetidus, Process Biochemistry, 29, 399-405 (1994).         [ Links ]

Felipe, M.G.A., Vieira, D.C., Vitolo, M., Silva, S.S., Roberto, I.C. and Mancilha, I.M., Effect of Acetic Acid on Xylose Fermentation to

Xylitol by Candida guilliermondii, Journal of Basic Microbiology, 35, 171-177 (1995).

Houng, J.Y. Chen, K.C. and Hsu, W.H., Optimization of Cultivation Medium Composition for Isoamilase Production, Applied Microbiology and Biotechnology, 31, 61-64 (1989).         [ Links ]

Johnson, M.C. and Harris, E.E., Acclimatization of Various Yeasts to Wood Sugar, Journal of the American Chemical Society, 70, 2961-2963 (1948).         [ Links ]

Nelson, N.A., Photometric Adaptation of the Somogyi Method for the Determination of Glucose, Journal of Biological Chemistry, 153, 375-380 (1944).         [ Links ]

Roberto, I.C., Felipe, M.G.A., Lacis, L.S., Silva, S.S. and Mancilha, I.M., Utilization of Sugar Cane Bagasse Hemicellulosic Hydrolyzate by Candida guilliermondii for Xylitol Production, Bioresource Technology, 36, 271-275 (1991).         [ Links ]

Roberto, I.C., Sato, S., Mancilha, I.M. and Taqueda, M.E.S., Influence of Media Composition on Xylitol Fermentation by Candida guilliermondii Using Response Surface Methodology, Biotechnology Letters, 17, 1223-1228 (1995).         [ Links ]

Romantschuk, H. and Lehtomaki, M., Operational Experiences of the First Full-Scale "Pekilo" SCP Mill Application, Process Biochemistry, 3, 16-19, 29 (1978).         [ Links ]

Tran, A.V. and Chambers, R.P., Red Oak Wood Derived Inhibitors in the Ethanol Fermentation of Xylose by Pichia stipitis CBS 5776, Biotechnology Letters, 7, 841-846 (1985).         [ Links ]

Weigert, B. Klein, C., Rizzi, M., Lauterbach, C. and Dellveg, H., Xylose Fermentation by Yeast Influence of Furfural on the Aerobic Growth of the Yeast, Biotechnology Letters, 10, 895-900 (1988).         [ Links ]

 

 

* To whom correspondence should be addressed.