World
Journal
of Microbiology
and Biotechnology
7. 419-427
Protein enrichment of sago starch by solid-state fermentation with Rhizopus spp.
E. Gumbira-Sa’id, D.A. Mitchell
Proteln enrichment of sago starch of three different diameters was Investigated both in flask culture and under forced aeration in a packed-bed fermenter using two strains of Rhlzopus. Protein production by R. oligosporus UQM 145F was superlor to Rhizopus sp. UQM 166F In the flask culture without aeratlon, with both preferring larger diameter (3 to 4 mm) spherical sago-beads. In the packed-bed fermenter with forced aeration, Rh/zopus sp. UQM 186F led to more rapid protein production compared to R. ollgosporus UQY 145F and produced equivalent final yields (about 10% protein on a dry wt basis). E. Gumbira-Sa’id, P.F. Greenfield and D.A. Mitchell are with the Department of Chemical Engineering, and H.W. Doelle is with the Department of Microbiology, University of Queensland, Queensland 4072, Australia. E. Gumbira-Sa’id is the corresponding author.
H.W.
Doelle,
P.F. Greenfield
and
In the year 2000, it is estimated that the world deficit in animal feed protein will be between 65 and 120 million tonnes (Senez 1985). Production of microbial protein from starchy materials has the potential for reducing the deficit in protein production from conventional sources. Sagopalm (Metroxylan sagus) is a starch source with significant potential in South-East Asian countries and the Pacific region. At present, there are an estimated 2 million hectares of natural or wild sagopalm, compared to only 200,000 hectares of cultivated sagopalm (Flach 1983). An economic analysis of protein production from starch revealed that a conventional, aseptic liquid fermentation system is not viable (Senez 1985; Carrizalez & Jaffe 1986; Daubresse et al. 1987; Yang 1988). Solid-state fermentation (SSF) on the other hand may have greater potential, because of its simpler production methods and, particularly, its reduced drying costs. The aim of this study is to produce a monogastric animal feed containing both starch (for its calorific value) and protein. Rhixopus was chosen because it has been used in the Indonesian diet for many centuries (Wang & Hesseltine 1982). This fungus contains a relatively high protein content of high biological value (Waliszewska et al. 1983) and exhibits significant protein productivity with cassava (Ramos-Valdivia et al. 1983; Sukara & Doelle 1988) as substrate.
Materials
and Methods
Micro-organisms Rh+pu.r oligosporus UQM 145F and Rh~~opus sp. UQM 186F were from the Culture Collection, Department of Microbiology, University of Queensland, Australia. They were maintained following the method of Mitchell et al. (1986) replacing cassava starch with sago starch.
@ 1991 Rapid
Communications
of Oxford
Ltd.
Preparation of Stibstrates Spherical beads, 2, 3 and 4 mm diameter, of sago substrate were supplied by Bogor Agricultural University, Indonesia. They were examined individually and also in combination: each sago-bead size comprised one-third of the total weight.
World
Journal
of Microbiology
and Biotechnology.
Vol 7, 1991
419
E. Gumbira-Sa’id
et
al. Substrates were soaked in the nutritive solution (100 g sample in 200 ml) for 60 min, drained and screened using a sheet of nylon mosquito-screen. The nutritive solution contained 8 g (NH&SO,, 2 g urea, 0.10 g KHJ’O,, 0.10 g KaHPO,, and 0.10 g Hortico trace elements fertilizer in 100 ml of distilled water. Flask-Culture Experiments Pre-soaked samples (approximately 10 g wet wt) were placed in pre-weighed 50 ml conical flasks with aluminium caps, and gelatinized at 70 to 80°C by steaming for 10 min. After cooling to 40°C substrates were inoculated with 1 ml of a 1 x lo6 spores/ml spore suspension (i.e. a spore density of 1 x lo5 spores/g of substrate weight), and re-weighed to give the initial weight of inoculated substrate. The flasks were put into air-tight IO-litre plastic boxes (20 flasks in each box), and incubated at 37°C for 72 h. At zero time followed by 12-h intervals, four flasks were removed at each time, three replicates for the determination of protein and pH, and one flask for the determination of moisture content and water activity. Packed-bed Experiments Pre-soaked samples (approximately 90 g wet weight) were placed in clean, dry 250 ml conical flasks, covered with aluminium foil, and then gelatinized at 70 to 80°C by steaming for 10 min. After cooling to 40°C substrates were inoculated with 10 ml of a 1 x lo6 spores/ml suspension to give a spore density of 1 x lo5 spores/g of substrate weight unless stated otherwise. The inoculated substrates were then transferred to the upper container of Sartorius-Polycarbonate Filter assemblies which were used as fermenters (Figure 1). Four fermenters were transferred into a 37°C waterbath. Each fermenter was aerated at 0.275 l/min with air bubbling through water in the lower chamber of the filter assembly. At 12 h, the contents of each fermenter were divided radially into six replicate samples of 10 g. The whole procedure was repeated in order to obtain samples at 24, 36, 48 and 60 h. For the study of optimal harvest time, samples were taken after 50 and 60 h, while for the study of inoculum spore density, samples were taken after 24, 36, 48 and 60 h fermentation.
Samples (about 10 g) were made up to 100 ml with distilled water, and the contents were homogenized for about 1 min at medium speed using a Virtis-23 Homogenizer (Virtis Research Equipment, New York, USA). Determination of protein was carried out using the Folin reaction after solubilization with NaOH (Gerhardt 1981). The pH was determined according to the AOAC method no. 140222 (AOAC 1984). Glucose was assayed following the method of Mitchell et al. (1986).
Flgure 1. Schematic diagram of the packedbed apparatus. AS-air supply; R-rotameter; H-heater; WB-waterbath; F-fermenter; S-stirrer; SC-spores collector; sp-sparger.
420
World Journal
of Microbiology
and Biotechnology,
Vol 7, 1997
Protein from sag0 starch Loss of total substrate weight was determined by calculating the difference between the initial substrate weight and the sample weight (net wt basis). Moisture content was determined using the loss of weight obtained by drying 2 g of sample at 105°C until constant weight was reached. Water activity was determined with 2 to 3 g of sample using a humidity meter (Novasina).
Results
and Discussion
Flask-culture Experiments The two strains of Rhi.yopus generated differently shaped profiles for protein production. Linear production of protein for the firt 60 h was observed for all bead diameters with R. oligosporus UQM 145F (Figure 2). Rhiropus sp. UQM 186F (Figure 3), on the other hand, only gave linear protein production until 48 h. In the case of 4 mm sago-beads, growth was initially slow but accelerated from 24 to 36 h. After a peak at 36 h the protein level began to decrease. Linear increases in protein content were also observed by Mitchell et al. (1986) during SSF of cassava by R. oligosporus UQM 145F. The rate of protein production by R. oligosporus UQM 145F was consistently faster than that of Rhiyopus sp. UQM 186F (Table 1). Also, for both fungi there was a tendency for faster growth on larger sago-beads. Smaller sago-beads or a mixture of sizes results in closer packing of the particles and a reduced void space between particles (Mudgett 1986), which restricts oxygen availability and therefore growth. The highest protein content on substrates fermented by R. oligospows UQM 145F was 53 to 77 mg/g initial dry substrate (IDS) which was achieved in 60 to 72 h, while Rhixo$ws sp. UQM 186F produced only 34 to 49 mg/g IDS in 36 to 60 h of fermentation. This result is different to that obtained in liquid culture with growth on cassava, where Rhixopns sp. UQM 186F produced a higher protein content (9 g protein/100 g initial substrate) compared to R. ohgosporu UQM 145F (8 g protein/l00 g initial substrate) on cassava (Sukara 1987). The loss of total substrate weight (Table 2) is largely due to the conversion of substrate to COa. The loss of total substrate weight by R. oligosporus UQM 145F (7.9%) was two times higher than that by Rhixopus sp. UQM 186F (3.8%), while the protein production by R. ohgosporu UQM 145F was only about 1.5 times (Table 1) higher than that by Rhixopus sp. UQM 186F. Starch measurements made during a study of the effect of nitrogen on growth (unpublished data) confirm the
loo1 80
60
Figure 2. Protein production (mg/g initial dry substrate) by R. oligosporus UQM 145F on different diameters of sago-bead substrate during 72 h fermentation in flask culture. O-2mm; l -3mm; A4mm; A-
0
12
24
mixed.
World
Journal
36 Time (hours)
of Microbiology
48
and Biotechnology,
60
Vol 7, 1991
72
421
E. Gzmbira-Sa’id
et
al.
Figure 3. Protein production (mg/g initial dry substrate) by Rhizopus sp. UQM 186F on different diameters of sago-bead substrate during 72 h fermentation in flask culture. O-2mm; l -3mm; A4mm; Amixed.
0
12
24
36 Tlme (hours)
Table 1. Protein production rates and the highest protein I?. o//gosporus UOM 145F and Rhkopus sp. UQM 188F sago starch in the flask culture. Sago starch diameter (mm)
R. oligosporus
UOM
Rate*
Protein
2 3 4 Mixed
0.60 0.63 1.04 0.44
53 59 77 54
48
production on different
145F
0.6 2 4 4
(60) (72) (60) (72)
72
(with standard error) substrate diameters
Rhizopus
productiont f + + f
60
sl. UQM
Rate*
Protein
0.39 0.48 0.36
36 43 49 34
by of
188F
productiont k + f +
1 4 5 0.6
(48) (60) (36) (48)
*The first 60 h of fermentation for R. oligosporus UQM 145F and the first 48 h of fermentation for Rhizopus sp. UQM 186F. Given as mg/g initial dry substrate per hour. THighest protein production (mg/g initial dry substrate). Numbers in parentheses indicate the time of fermentation (h).
Table 2. Fermentation of sago-beads 188F in the flask culture.*
by R. oligosporus
Time
145F
R. ollgosporus
UQM
UQM
145F
Rhizopus
and
Rhbopus
sp. UQM
sp. UQM
188F
(h) a 0 12 24 36 48 60 72
0.0 0.03 0.06 0.20 0.40 0.54 0.79
b 0.95 0.95 0.94 0.95 0.91 0.92 0.92
C
d
e
a
b
C
d
e
53 54 54 57 59 61 62
6.6 6.6 4.4 7.0 7.2 7.0 7.1
0 0 18 17 40 48 28
0.0 0.03 0.05 0.09 0.12 0.12 0.38
0.96 0.94 0.94 0.93 0.93 0.94 0.93
53 54 55 55 55 55 55
6.6 6.8 4.9 3.6 3.6 3.7 3.7
0 0 12 7 24 16 0
* Data taken from the average of three diameters a-Loss of total substrate weight (g wet wt basis b-Water activity. c-Moisture content (%). d-pH. e-Glucose (mg/g initial dry substrate).
422
World
Journal
of Microbiology
and Biotechnology,
Vol 7, 1991
and one mixture of diameters. different from the initial substrate
weight).
Protein from sag0 starch difference in substrate utilization. Rhixopus sp. UQM 186F utilized only 20 to 22% of the available starch compared to 30 to 43% utilization by R. ohgosporus UQM 145F. Therefore, although R. oligosporus UQM 145F produced higher protein contents (Table l), it is less efficient in converting starch to protein compared to Rhixopus sp. UQM 186F since a greater proportion of hydrolysed starch remained as residual glucose (Table 2). The presence of glucose in the medium indicates a limitation in the system, such as intra-particle mass transfer (Moo-Young et a/. 1983) or fungal packing density (Laukevics et al. 1985). Raimbault et al. (1985), using Rhixopw sp. and cassava, found 15% of protein and more than 39% of total sugars in the final product. However, the presence of sugar in animal feed is not disadvantageous since it can be utilized by the animal directly. Water activity decreased during growth (Table 2). This decrease was greater in the case of R. oIigo,poruJ UQM 145F due to higher glucose production (Table 2). This is in agreement with observations made by Narahara et al. (1982) in koji fermentation, where the accumulation of reducing sugars caused the water activity to decrease. The moisture content of the sago-beads fermented by R. oljgosporus UQM 145F increased from 53 to 62% throughout the growth period, while the moisture content changed insignificantly in those samples fermented by Rhi7optr.r sp. UQM 186F (Table 2). The decrease in water activity with an increase in moisture content is due to the accumulation of glucose in the aqueous phase (Narahara et al. 1982). An increase in moisture content during fermentation was also observed by Yang (1988) in the protein enrichment of sweet potato residue with amylolytic microorganisms, and is claimed to be due to the production of metabolic water. Changes in moisture content during fermentation represent a significant factor in designing a control system for the humidity in an aerated system. The pH patterns during growth of the two organisms differed significantly (Table 2). With an initial pH of 6.7, R. oligosporus UQM 145F caused the pH to drop to 4.2 after 24 h of fermentation where it then increased again to 6.9 or 7.0 between 36 and 72 h. This increase is due to the hydrolysis of urea which liberates ammonia and counteracts the rapid acidification due to the uptake of the ammonium ion (Raimbault & Alazard 1980). In the case of Rhixopus sp. UQM 186F, the pH decreased to 3.6 or 3.7 between 36 and 72 h and did not increase again, indicating that Rhiropus sp. UQM 186F may not have an urease enzyme. The decrease in the pH value to below 4.0 (Table 2) on sago-bead substrates fermented by Rhixopus sp. UQM 186F is likely to be responsible for the lower production of protein. Mitchell et al. (1988) found that the optimal pH for R. oligosporus UQM 145F is around 7.0. The control of pH in a SSF system can represent a major problem. These preliminary flask experiments indicate that R. o&osporus UQM 145F may have a greater potential for SSF of sago starch, owing to its better pH profile and higher protein productivity. Packed-bed Experiments Protein production by both fungi. The patterns of protein production by both fungi (Table 3) on sago-beads of mixed sizes were identical to those observed in the flask culture, with only R. oligosporus UQM 145F showing linear growth (Figure 2). However, Rhiyopzls sp. UQM 186F exhibited a higher yield and a faster protein production rate than in flask culture. The improved growth of Rhi7opn.r sp. UQM 186F might be due to the constant supply of humid air during fermentation, which maintains aerobic conditions, and regulates the temperature and moisture content of the substrate (Durand & Chereau 1987). The highest protein contents produced were 107 mg protein/g dry sample by R. oIigo!porus UQM 145F which was achieved in 60 h, and 99 mg/g dry sample
World
Journal
of Microbiology
and Biotechnology,
Vol 7, 1991
423
E.
Gumbiru-Ju’id
et
al. Table 3. Protein production (with standard error) (mglg of sago-beads of mlxed-size diameters by f?. oligosporus 166F in the packed-bed fermenter. Time
(h)
R. oligosporus
0 12 24 36 48 60
12 4 38 52 92 107
UQM f * If: If: f f
dry sample) during 60 h fermentation UQM 146F and Rh/zopus sp. UQM
145F
Rhizopus
1 0.3 2 2 2 2
sp. UQM 12 4 12 64 99 65
by Rh&pl/s sp. UQM 186F in 48 h. It is obvious therefore 186F can be harvested in a shorter time (48 h).
that Rh&pt/s
f * + + f +
166F
1 0.4 0.6 2 3 5
sp. UQM
Determination of optimal harvest time. To find whether the growth period of R. ohgosporm UQM 145F can be decreased to between 48 and 60 h, two fermentation times, 50 and 60 h, were specifically investigated. The results obtained for protein production by R. oligosporus UQM 145F using four different particle sizes as substrate are shown in Figure 4. Protein production on all sago-bead sizes was higher after 60 h fermentation. The highest protein content was found on 3 mm diameter, followed by mixed-size, 4 and 2 mm diameters. This result differs from those obtained in the flask culture, where larger diameters gave higher protein contents because of looser packing of the substrate particles. Forced aeration should overcome the problem of oxygen depletion which occurred with small particle sizes in static culture. Under forced aeration, linear growth of Aspergih koppan on rice was reported by Huang et al. (1985). It was also observed that A. koppun grew best on the smaller rice particle sizes. However, the poorer growth of R. oligosporm UQM 145F on the sago-beads of 2 mm diameter could be due to the agglomeration of particles and the excessive moisture absorbed during fermentation, which inhibits mycelial development.
1201
100
80 60
4. Protein production (mg/g dry after 50 and 60 h fermentation by R. oligosporus UQM 145F on different diameters of sago-bead substrate in the packed-bed fermenter.
60
Figure sample)
424
World
Journal
of Microbiology
and Biotechnology,
Time (hours) ISSI
Vol 7, 1991
2mm
Ii3
3mm
BE4
4mm
lXl
mixed
Protein from sag0 starch Studies of inoculum spore densities. The effect of inoculum spore density of R. oligosporus UQM 145F on protein production was investigated using three 1eveIs of spore density: 3 x 105, 1.2 x IO6 and 6 x lo6 spores/g substrate. The patterns of protein production by R. oligosporus UQM 145F on the mixed-sizes diameter of sago-beads substrate with three initial spore densities are presented in Figure 5. A spore density of 3 x IO5 spores/g substrate gave the highest protein production followed by densities of 1.2 and 6 x IO” after 60 h fermentation. This is in contrast to the study by Huang et al. (1985) in which a density of 1 x 10’ spores/g substrate gave the best results. However, a lower spore density is preferable since it simplifies the preparation of the spore suspension. Euaiuation of Performance With forced aeration in packed-bed culture the performance of both fungi improved, with the effect being quite marked for Rhixopus sp. UQM 186F. Compared with flask culture Rhixopus sp. UQM 186F produced higher protein content in a shorter or in the same time. The superior performance of R. oligosporus UQM 145F over Rhiyopus sp. UQM 186F which was observed in flask culture was no longer apparent. In the packed-bed experiment R&opus sp. UQM 186F attained almost the same maximum protein content as R. oligosporus UQM 145F, and in fact did so in a shorter time. The packed-bed system has produced about 10% protein on a dry wt basis. This is much better than the 5.0% protein obtained by Czajkowska & Ilnicka-Olejniczak (1988) for growth of Aspergihs oryrae on coarse rye meal and beet pulp, and as good as the protein-enriched cassava (7 to 10% protein) obtained by R&opus or Aspergihs (Hutagalung & Tan 1976) or the 11% protein obtained with Rhixopus oryiae (Daubresse et a/. 1987). However, protein production was lower than that reported for protein enrichment (18% protein) of cassava with Aspergilhs niger (Raimbault & Alazard 1980) or R. oIigosporus NRRL 2710 (23% protein) (RamosValdivia et al. 1983). This paper has described the use of spherical sago-beads. In practice, it is more likely that the sago pith would simply be chopped or grated into irregular pieces. These studies were carried out with regular beads because this allows even dispersion within the substrate mass of the inoculum, aeration and nutrients for
Figure
5.
Protein
production
(mg/g
dry
sample) by R. oligosporus UQM 145F on the mixed-size diameters of sago-bead substrate using three initial inoculum spore densities.
24
m
36
3x105
World
Journal
48
Time
(hours)
BEa
1.2 x 106
of Microbiology
and Biotechnology,
60
I-x3
6x
Vol 7, 1991
lo6
425
E. Gumbiru-Ju ‘id et al. growth. This enables rapid even growth, penetration, and maximum substrate utilization (Hesseltine 1972; Forage 1979; Mitchell et al. 1988). It can be concluded that in the absence of forced aeration, both Rhi~opus species grew better on sago-beads of larger diameters. However, protein production by R. oligosporus UQM 145F was faster and gave higher values than did Rhizopus sp. UQM 186F. Continuous supply of humid air during fermentation improved the growth of both fungi, especially Rhi~opus sp. UQM 186F. A simple air fluidizedbed fermenter, which enables homogeneous growth of fungi, is currently being developed in our laboratory.
Acknowledgements We would like to thank the Inter University Centre of Biotechnology of Indonesia and the International Development Program of Australian Universities and Colleges for providing a scholarship for E. Gumbira-Sa’id, and Bogor Agricultural University, Indonesia, for providing sago starch samples.
References AOAC 1984 Official Methods of Analysis, 13’h edn, Washington, DC: Association of the Official Analytical Chemists. CARRIZALEZ, V. & JAFFE, W. 1986 Solid-state fermentation: an appropriate biotechnology for developing countries. Interciencia 11, ‘h-15. CZAJKOWSKA, D. & ILNICKA-OLEJNICZAK, 0.1988 Biosynthesis of protein by microscopic fungi in solid-state fermentation. I. Selection of Aspergillus strain for enrichment of starchy materials in protein. Acta Biotecbnologica 8, 407-413. DAUBRESSE, P., NTIBASHIRWA, S., GHEYSEN, A. & MEYER, J.A. 1987 A process for the protein enrichment of cassava by solid substrate fermentation in rural conditions. Biotechnology
and Bioengineering
29, 962-968.
DURAND A. & CHEREAU, D. 1987 A new pilot reactor for solid-state fermentation: application to the protein enrichment of sugar beet pulp. Biotechnology and Bioengineering 31, 476-481.
FLACH, M. 1983 The sagopalm plant protection paper 47. pp. l-85. FAO: Rome. FORAGE, A.J. 1979 Utilization of agricultural and food processing wastes containing carbohydrates. Chemical Socieg Review 8, 3OS314. GERHARDT, P. 1981 Manual of Methods for General Bacteriology. pp. 358359. Washington, DC: American Society of Microbiology. HESSELTINE, C.W. 1972 Solid-state fermentations. Biotechnology and Bioengineering 14, 517532.
HUANG, S.Y., WANG, H.H., WEI, C.-J., MALANEY, G.W. & TANNER, R.D. 1985 Kinetic responses of the koji solid-state fermentation process. Topics in Eqyme and Fermentation Biotechnology
10, 88-108.
HUTAGALUNG, R.I. & TAN, P.H. 1976 Utilization of nutritionally improved cassava by nutrient supplementation and microbial enrichment in poultry and pigs. In Proceedings of the Fourth Symposium oj tbe International Society of Tropical Root Crops, eds Cock, J., MacIntyre, R. & Graham, M. pp. 255-262. Cali: CIAT. LAUKEVICS, J.J., APSITE, A.F., VIESTURS, U.S. & TENGERDY, R.P. 1985 Steric hindrance of growth of filamentous fungi in solid substrate fermentation of wheat straw. Biotecbnologr and Bioengineering
27, 1687-l
691~
MITCHELL, D.A., GREENFIELD, P.F. & DOELLE, H.W. 1986 A model substrate for solid-state fermentation. BiotecbnoloD Letters 8, 827-832. MITCHELL, D.A., DOELLE, H.W. & GREENFIELD, P.F. 1988 Agar plate growth studies of Rbixopus ohgosporus and Aspergillus orJ?ae to determine their suitability for solid-state fermentation. Applied MicrobioLogy and Biotechnology 8, l-5. MOO-YOUNG, M., MOREIRA, A.R. & TENGERDY, R.P. 1983 Principles of solid-substrate fermentation. In The Fiiamentous Fungi, eds Smith, J.E., Berry, D.R. & Kristiansen, B. pp. 117-144. London: Edward Arnold. MUDGETT, R.E. 1986 Solid-state fermentation. In Manual oj Industrial Microbiology, eds Demain, A.L. & Solomon, N.A. pp. 6683. Washington, DC: American Society of Microbiology. NARAHARA, H., KOYAMA, Y., YOSHIDA, PICHYANGKURA, S., UEDA, R. & TAGUCHI, H. 1982 Growth and enzyme production in a solid-state culture of Aspergillus ory?ae. Journal of Fermentation Technology 60, 311-319.
426
World Journal
of Microbiology
and Biotechnology,
Vol 7, 1991
Protein
from
sag0 starch
RAIMBAULT, M. & ALAZARD, D. 1980 Culture method to study fungal growth in solid fermentation. European Journal of Applied Microbiology 9, 199-209. RAIMBAULT, M., REVAH, S., PINA, F. & VILLALOBOS, P. 1985 Protein enrichment of cassava by solid substrate fermentation using molds isolated from traditional foods. JournaL oj Fermentation Technology 63, 395-399. RAMOS-VALDIVIA, A., DE LA TORRE, M. & CASAS-CAMPELO, C. 1983 Solid-state fermentation of cassava with R&opus. oligosporus NRRL 2710. In Production and Feeding of Single Cell Protein, ed. Ferranti, M.P. pp. 104-111. London: Applied Science Publishers. SENEZ, J.C. 1985 Microbial food and feed. In Biotechnology and Bioprocess Engineering, ed. Ghose, T.K. pp. 3099319. New Delhi, India: Indian Institute of Technology. SUKARA, E. 1987 Production of single cell protein from cassava by microfungi. PhD Thesis. Brisbane: University of Queensland. SUKARA, E. & DOELLE, H.W. 1988 Cassava starch fermentation pattern of Rhi~opus ohgosporus. MIRCEN JouraL oj Applied Microbiology and Biotecbnolog), 4, 465471. WALISZEWSKA, A., GARCIA, H.S. & WALISZEWSKI, K. 1983 Nutritional evaluation of R&opus okgosporus biomass propagated on potato. Nutritional Report International 28, 197-202.
WANG, H.L. & HESSELTINE, C.W. 1982 Oriental fermented foods. In Prescott and Dunn’s Industrial Microbiology, 4th edn, ed. Reed, G. pp. 492-538. Westport, Connecticut: AVI Publishing Company. YANG, S.S. 1988 Protein enrichment of sweet potato residue with amylolytic yeasts by solid-state fermentation. Biotechnology and Bioengineering 32, 886890.
(Received 30 October 1990; revised 19 December 1990; accepted 10 January
World
Journal
of Microbiology
and Biotechnology,
1990)
Vol 7, 1991
427
Journal
of Microbiology
and Biotechnology
7. 419-427
Protein enrichment of sago starch by solid-state fermentation with Rhizopus spp.
E. Gumbira-Sa’id, D.A. Mitchell
Proteln enrichment of sago starch of three different diameters was Investigated both in flask culture and under forced aeration in a packed-bed fermenter using two strains of Rhlzopus. Protein production by R. oligosporus UQM 145F was superlor to Rhizopus sp. UQM 166F In the flask culture without aeratlon, with both preferring larger diameter (3 to 4 mm) spherical sago-beads. In the packed-bed fermenter with forced aeration, Rh/zopus sp. UQM 186F led to more rapid protein production compared to R. ollgosporus UQY 145F and produced equivalent final yields (about 10% protein on a dry wt basis). E. Gumbira-Sa’id, P.F. Greenfield and D.A. Mitchell are with the Department of Chemical Engineering, and H.W. Doelle is with the Department of Microbiology, University of Queensland, Queensland 4072, Australia. E. Gumbira-Sa’id is the corresponding author.
H.W.
Doelle,
P.F. Greenfield
and
In the year 2000, it is estimated that the world deficit in animal feed protein will be between 65 and 120 million tonnes (Senez 1985). Production of microbial protein from starchy materials has the potential for reducing the deficit in protein production from conventional sources. Sagopalm (Metroxylan sagus) is a starch source with significant potential in South-East Asian countries and the Pacific region. At present, there are an estimated 2 million hectares of natural or wild sagopalm, compared to only 200,000 hectares of cultivated sagopalm (Flach 1983). An economic analysis of protein production from starch revealed that a conventional, aseptic liquid fermentation system is not viable (Senez 1985; Carrizalez & Jaffe 1986; Daubresse et al. 1987; Yang 1988). Solid-state fermentation (SSF) on the other hand may have greater potential, because of its simpler production methods and, particularly, its reduced drying costs. The aim of this study is to produce a monogastric animal feed containing both starch (for its calorific value) and protein. Rhixopus was chosen because it has been used in the Indonesian diet for many centuries (Wang & Hesseltine 1982). This fungus contains a relatively high protein content of high biological value (Waliszewska et al. 1983) and exhibits significant protein productivity with cassava (Ramos-Valdivia et al. 1983; Sukara & Doelle 1988) as substrate.
Materials
and Methods
Micro-organisms Rh+pu.r oligosporus UQM 145F and Rh~~opus sp. UQM 186F were from the Culture Collection, Department of Microbiology, University of Queensland, Australia. They were maintained following the method of Mitchell et al. (1986) replacing cassava starch with sago starch.
@ 1991 Rapid
Communications
of Oxford
Ltd.
Preparation of Stibstrates Spherical beads, 2, 3 and 4 mm diameter, of sago substrate were supplied by Bogor Agricultural University, Indonesia. They were examined individually and also in combination: each sago-bead size comprised one-third of the total weight.
World
Journal
of Microbiology
and Biotechnology.
Vol 7, 1991
419
E. Gumbira-Sa’id
et
al. Substrates were soaked in the nutritive solution (100 g sample in 200 ml) for 60 min, drained and screened using a sheet of nylon mosquito-screen. The nutritive solution contained 8 g (NH&SO,, 2 g urea, 0.10 g KHJ’O,, 0.10 g KaHPO,, and 0.10 g Hortico trace elements fertilizer in 100 ml of distilled water. Flask-Culture Experiments Pre-soaked samples (approximately 10 g wet wt) were placed in pre-weighed 50 ml conical flasks with aluminium caps, and gelatinized at 70 to 80°C by steaming for 10 min. After cooling to 40°C substrates were inoculated with 1 ml of a 1 x lo6 spores/ml spore suspension (i.e. a spore density of 1 x lo5 spores/g of substrate weight), and re-weighed to give the initial weight of inoculated substrate. The flasks were put into air-tight IO-litre plastic boxes (20 flasks in each box), and incubated at 37°C for 72 h. At zero time followed by 12-h intervals, four flasks were removed at each time, three replicates for the determination of protein and pH, and one flask for the determination of moisture content and water activity. Packed-bed Experiments Pre-soaked samples (approximately 90 g wet weight) were placed in clean, dry 250 ml conical flasks, covered with aluminium foil, and then gelatinized at 70 to 80°C by steaming for 10 min. After cooling to 40°C substrates were inoculated with 10 ml of a 1 x lo6 spores/ml suspension to give a spore density of 1 x lo5 spores/g of substrate weight unless stated otherwise. The inoculated substrates were then transferred to the upper container of Sartorius-Polycarbonate Filter assemblies which were used as fermenters (Figure 1). Four fermenters were transferred into a 37°C waterbath. Each fermenter was aerated at 0.275 l/min with air bubbling through water in the lower chamber of the filter assembly. At 12 h, the contents of each fermenter were divided radially into six replicate samples of 10 g. The whole procedure was repeated in order to obtain samples at 24, 36, 48 and 60 h. For the study of optimal harvest time, samples were taken after 50 and 60 h, while for the study of inoculum spore density, samples were taken after 24, 36, 48 and 60 h fermentation.
Samples (about 10 g) were made up to 100 ml with distilled water, and the contents were homogenized for about 1 min at medium speed using a Virtis-23 Homogenizer (Virtis Research Equipment, New York, USA). Determination of protein was carried out using the Folin reaction after solubilization with NaOH (Gerhardt 1981). The pH was determined according to the AOAC method no. 140222 (AOAC 1984). Glucose was assayed following the method of Mitchell et al. (1986).
Flgure 1. Schematic diagram of the packedbed apparatus. AS-air supply; R-rotameter; H-heater; WB-waterbath; F-fermenter; S-stirrer; SC-spores collector; sp-sparger.
420
World Journal
of Microbiology
and Biotechnology,
Vol 7, 1997
Protein from sag0 starch Loss of total substrate weight was determined by calculating the difference between the initial substrate weight and the sample weight (net wt basis). Moisture content was determined using the loss of weight obtained by drying 2 g of sample at 105°C until constant weight was reached. Water activity was determined with 2 to 3 g of sample using a humidity meter (Novasina).
Results
and Discussion
Flask-culture Experiments The two strains of Rhi.yopus generated differently shaped profiles for protein production. Linear production of protein for the firt 60 h was observed for all bead diameters with R. oligosporus UQM 145F (Figure 2). Rhiropus sp. UQM 186F (Figure 3), on the other hand, only gave linear protein production until 48 h. In the case of 4 mm sago-beads, growth was initially slow but accelerated from 24 to 36 h. After a peak at 36 h the protein level began to decrease. Linear increases in protein content were also observed by Mitchell et al. (1986) during SSF of cassava by R. oligosporus UQM 145F. The rate of protein production by R. oligosporus UQM 145F was consistently faster than that of Rhiyopus sp. UQM 186F (Table 1). Also, for both fungi there was a tendency for faster growth on larger sago-beads. Smaller sago-beads or a mixture of sizes results in closer packing of the particles and a reduced void space between particles (Mudgett 1986), which restricts oxygen availability and therefore growth. The highest protein content on substrates fermented by R. oligospows UQM 145F was 53 to 77 mg/g initial dry substrate (IDS) which was achieved in 60 to 72 h, while Rhixo$ws sp. UQM 186F produced only 34 to 49 mg/g IDS in 36 to 60 h of fermentation. This result is different to that obtained in liquid culture with growth on cassava, where Rhixopns sp. UQM 186F produced a higher protein content (9 g protein/100 g initial substrate) compared to R. ohgosporu UQM 145F (8 g protein/l00 g initial substrate) on cassava (Sukara 1987). The loss of total substrate weight (Table 2) is largely due to the conversion of substrate to COa. The loss of total substrate weight by R. oligosporus UQM 145F (7.9%) was two times higher than that by Rhixopus sp. UQM 186F (3.8%), while the protein production by R. ohgosporu UQM 145F was only about 1.5 times (Table 1) higher than that by Rhixopus sp. UQM 186F. Starch measurements made during a study of the effect of nitrogen on growth (unpublished data) confirm the
loo1 80
60
Figure 2. Protein production (mg/g initial dry substrate) by R. oligosporus UQM 145F on different diameters of sago-bead substrate during 72 h fermentation in flask culture. O-2mm; l -3mm; A4mm; A-
0
12
24
mixed.
World
Journal
36 Time (hours)
of Microbiology
48
and Biotechnology,
60
Vol 7, 1991
72
421
E. Gzmbira-Sa’id
et
al.
Figure 3. Protein production (mg/g initial dry substrate) by Rhizopus sp. UQM 186F on different diameters of sago-bead substrate during 72 h fermentation in flask culture. O-2mm; l -3mm; A4mm; Amixed.
0
12
24
36 Tlme (hours)
Table 1. Protein production rates and the highest protein I?. o//gosporus UOM 145F and Rhkopus sp. UQM 188F sago starch in the flask culture. Sago starch diameter (mm)
R. oligosporus
UOM
Rate*
Protein
2 3 4 Mixed
0.60 0.63 1.04 0.44
53 59 77 54
48
production on different
145F
0.6 2 4 4
(60) (72) (60) (72)
72
(with standard error) substrate diameters
Rhizopus
productiont f + + f
60
sl. UQM
Rate*
Protein
0.39 0.48 0.36
36 43 49 34
by of
188F
productiont k + f +
1 4 5 0.6
(48) (60) (36) (48)
*The first 60 h of fermentation for R. oligosporus UQM 145F and the first 48 h of fermentation for Rhizopus sp. UQM 186F. Given as mg/g initial dry substrate per hour. THighest protein production (mg/g initial dry substrate). Numbers in parentheses indicate the time of fermentation (h).
Table 2. Fermentation of sago-beads 188F in the flask culture.*
by R. oligosporus
Time
145F
R. ollgosporus
UQM
UQM
145F
Rhizopus
and
Rhbopus
sp. UQM
sp. UQM
188F
(h) a 0 12 24 36 48 60 72
0.0 0.03 0.06 0.20 0.40 0.54 0.79
b 0.95 0.95 0.94 0.95 0.91 0.92 0.92
C
d
e
a
b
C
d
e
53 54 54 57 59 61 62
6.6 6.6 4.4 7.0 7.2 7.0 7.1
0 0 18 17 40 48 28
0.0 0.03 0.05 0.09 0.12 0.12 0.38
0.96 0.94 0.94 0.93 0.93 0.94 0.93
53 54 55 55 55 55 55
6.6 6.8 4.9 3.6 3.6 3.7 3.7
0 0 12 7 24 16 0
* Data taken from the average of three diameters a-Loss of total substrate weight (g wet wt basis b-Water activity. c-Moisture content (%). d-pH. e-Glucose (mg/g initial dry substrate).
422
World
Journal
of Microbiology
and Biotechnology,
Vol 7, 1991
and one mixture of diameters. different from the initial substrate
weight).
Protein from sag0 starch difference in substrate utilization. Rhixopus sp. UQM 186F utilized only 20 to 22% of the available starch compared to 30 to 43% utilization by R. ohgosporus UQM 145F. Therefore, although R. oligosporus UQM 145F produced higher protein contents (Table l), it is less efficient in converting starch to protein compared to Rhixopus sp. UQM 186F since a greater proportion of hydrolysed starch remained as residual glucose (Table 2). The presence of glucose in the medium indicates a limitation in the system, such as intra-particle mass transfer (Moo-Young et a/. 1983) or fungal packing density (Laukevics et al. 1985). Raimbault et al. (1985), using Rhixopw sp. and cassava, found 15% of protein and more than 39% of total sugars in the final product. However, the presence of sugar in animal feed is not disadvantageous since it can be utilized by the animal directly. Water activity decreased during growth (Table 2). This decrease was greater in the case of R. oIigo,poruJ UQM 145F due to higher glucose production (Table 2). This is in agreement with observations made by Narahara et al. (1982) in koji fermentation, where the accumulation of reducing sugars caused the water activity to decrease. The moisture content of the sago-beads fermented by R. oljgosporus UQM 145F increased from 53 to 62% throughout the growth period, while the moisture content changed insignificantly in those samples fermented by Rhi7optr.r sp. UQM 186F (Table 2). The decrease in water activity with an increase in moisture content is due to the accumulation of glucose in the aqueous phase (Narahara et al. 1982). An increase in moisture content during fermentation was also observed by Yang (1988) in the protein enrichment of sweet potato residue with amylolytic microorganisms, and is claimed to be due to the production of metabolic water. Changes in moisture content during fermentation represent a significant factor in designing a control system for the humidity in an aerated system. The pH patterns during growth of the two organisms differed significantly (Table 2). With an initial pH of 6.7, R. oligosporus UQM 145F caused the pH to drop to 4.2 after 24 h of fermentation where it then increased again to 6.9 or 7.0 between 36 and 72 h. This increase is due to the hydrolysis of urea which liberates ammonia and counteracts the rapid acidification due to the uptake of the ammonium ion (Raimbault & Alazard 1980). In the case of Rhixopus sp. UQM 186F, the pH decreased to 3.6 or 3.7 between 36 and 72 h and did not increase again, indicating that Rhiropus sp. UQM 186F may not have an urease enzyme. The decrease in the pH value to below 4.0 (Table 2) on sago-bead substrates fermented by Rhixopus sp. UQM 186F is likely to be responsible for the lower production of protein. Mitchell et al. (1988) found that the optimal pH for R. oligosporus UQM 145F is around 7.0. The control of pH in a SSF system can represent a major problem. These preliminary flask experiments indicate that R. o&osporus UQM 145F may have a greater potential for SSF of sago starch, owing to its better pH profile and higher protein productivity. Packed-bed Experiments Protein production by both fungi. The patterns of protein production by both fungi (Table 3) on sago-beads of mixed sizes were identical to those observed in the flask culture, with only R. oligosporus UQM 145F showing linear growth (Figure 2). However, Rhiyopzls sp. UQM 186F exhibited a higher yield and a faster protein production rate than in flask culture. The improved growth of Rhi7opn.r sp. UQM 186F might be due to the constant supply of humid air during fermentation, which maintains aerobic conditions, and regulates the temperature and moisture content of the substrate (Durand & Chereau 1987). The highest protein contents produced were 107 mg protein/g dry sample by R. oIigo!porus UQM 145F which was achieved in 60 h, and 99 mg/g dry sample
World
Journal
of Microbiology
and Biotechnology,
Vol 7, 1991
423
E.
Gumbiru-Ju’id
et
al. Table 3. Protein production (with standard error) (mglg of sago-beads of mlxed-size diameters by f?. oligosporus 166F in the packed-bed fermenter. Time
(h)
R. oligosporus
0 12 24 36 48 60
12 4 38 52 92 107
UQM f * If: If: f f
dry sample) during 60 h fermentation UQM 146F and Rh/zopus sp. UQM
145F
Rhizopus
1 0.3 2 2 2 2
sp. UQM 12 4 12 64 99 65
by Rh&pl/s sp. UQM 186F in 48 h. It is obvious therefore 186F can be harvested in a shorter time (48 h).
that Rh&pt/s
f * + + f +
166F
1 0.4 0.6 2 3 5
sp. UQM
Determination of optimal harvest time. To find whether the growth period of R. ohgosporm UQM 145F can be decreased to between 48 and 60 h, two fermentation times, 50 and 60 h, were specifically investigated. The results obtained for protein production by R. oligosporus UQM 145F using four different particle sizes as substrate are shown in Figure 4. Protein production on all sago-bead sizes was higher after 60 h fermentation. The highest protein content was found on 3 mm diameter, followed by mixed-size, 4 and 2 mm diameters. This result differs from those obtained in the flask culture, where larger diameters gave higher protein contents because of looser packing of the substrate particles. Forced aeration should overcome the problem of oxygen depletion which occurred with small particle sizes in static culture. Under forced aeration, linear growth of Aspergih koppan on rice was reported by Huang et al. (1985). It was also observed that A. koppun grew best on the smaller rice particle sizes. However, the poorer growth of R. oligosporm UQM 145F on the sago-beads of 2 mm diameter could be due to the agglomeration of particles and the excessive moisture absorbed during fermentation, which inhibits mycelial development.
1201
100
80 60
4. Protein production (mg/g dry after 50 and 60 h fermentation by R. oligosporus UQM 145F on different diameters of sago-bead substrate in the packed-bed fermenter.
60
Figure sample)
424
World
Journal
of Microbiology
and Biotechnology,
Time (hours) ISSI
Vol 7, 1991
2mm
Ii3
3mm
BE4
4mm
lXl
mixed
Protein from sag0 starch Studies of inoculum spore densities. The effect of inoculum spore density of R. oligosporus UQM 145F on protein production was investigated using three 1eveIs of spore density: 3 x 105, 1.2 x IO6 and 6 x lo6 spores/g substrate. The patterns of protein production by R. oligosporus UQM 145F on the mixed-sizes diameter of sago-beads substrate with three initial spore densities are presented in Figure 5. A spore density of 3 x IO5 spores/g substrate gave the highest protein production followed by densities of 1.2 and 6 x IO” after 60 h fermentation. This is in contrast to the study by Huang et al. (1985) in which a density of 1 x 10’ spores/g substrate gave the best results. However, a lower spore density is preferable since it simplifies the preparation of the spore suspension. Euaiuation of Performance With forced aeration in packed-bed culture the performance of both fungi improved, with the effect being quite marked for Rhixopus sp. UQM 186F. Compared with flask culture Rhixopus sp. UQM 186F produced higher protein content in a shorter or in the same time. The superior performance of R. oligosporus UQM 145F over Rhiyopus sp. UQM 186F which was observed in flask culture was no longer apparent. In the packed-bed experiment R&opus sp. UQM 186F attained almost the same maximum protein content as R. oligosporus UQM 145F, and in fact did so in a shorter time. The packed-bed system has produced about 10% protein on a dry wt basis. This is much better than the 5.0% protein obtained by Czajkowska & Ilnicka-Olejniczak (1988) for growth of Aspergihs oryrae on coarse rye meal and beet pulp, and as good as the protein-enriched cassava (7 to 10% protein) obtained by R&opus or Aspergihs (Hutagalung & Tan 1976) or the 11% protein obtained with Rhixopus oryiae (Daubresse et a/. 1987). However, protein production was lower than that reported for protein enrichment (18% protein) of cassava with Aspergilhs niger (Raimbault & Alazard 1980) or R. oIigosporus NRRL 2710 (23% protein) (RamosValdivia et al. 1983). This paper has described the use of spherical sago-beads. In practice, it is more likely that the sago pith would simply be chopped or grated into irregular pieces. These studies were carried out with regular beads because this allows even dispersion within the substrate mass of the inoculum, aeration and nutrients for
Figure
5.
Protein
production
(mg/g
dry
sample) by R. oligosporus UQM 145F on the mixed-size diameters of sago-bead substrate using three initial inoculum spore densities.
24
m
36
3x105
World
Journal
48
Time
(hours)
BEa
1.2 x 106
of Microbiology
and Biotechnology,
60
I-x3
6x
Vol 7, 1991
lo6
425
E. Gumbiru-Ju ‘id et al. growth. This enables rapid even growth, penetration, and maximum substrate utilization (Hesseltine 1972; Forage 1979; Mitchell et al. 1988). It can be concluded that in the absence of forced aeration, both Rhi~opus species grew better on sago-beads of larger diameters. However, protein production by R. oligosporus UQM 145F was faster and gave higher values than did Rhizopus sp. UQM 186F. Continuous supply of humid air during fermentation improved the growth of both fungi, especially Rhi~opus sp. UQM 186F. A simple air fluidizedbed fermenter, which enables homogeneous growth of fungi, is currently being developed in our laboratory.
Acknowledgements We would like to thank the Inter University Centre of Biotechnology of Indonesia and the International Development Program of Australian Universities and Colleges for providing a scholarship for E. Gumbira-Sa’id, and Bogor Agricultural University, Indonesia, for providing sago starch samples.
References AOAC 1984 Official Methods of Analysis, 13’h edn, Washington, DC: Association of the Official Analytical Chemists. CARRIZALEZ, V. & JAFFE, W. 1986 Solid-state fermentation: an appropriate biotechnology for developing countries. Interciencia 11, ‘h-15. CZAJKOWSKA, D. & ILNICKA-OLEJNICZAK, 0.1988 Biosynthesis of protein by microscopic fungi in solid-state fermentation. I. Selection of Aspergillus strain for enrichment of starchy materials in protein. Acta Biotecbnologica 8, 407-413. DAUBRESSE, P., NTIBASHIRWA, S., GHEYSEN, A. & MEYER, J.A. 1987 A process for the protein enrichment of cassava by solid substrate fermentation in rural conditions. Biotechnology
and Bioengineering
29, 962-968.
DURAND A. & CHEREAU, D. 1987 A new pilot reactor for solid-state fermentation: application to the protein enrichment of sugar beet pulp. Biotechnology and Bioengineering 31, 476-481.
FLACH, M. 1983 The sagopalm plant protection paper 47. pp. l-85. FAO: Rome. FORAGE, A.J. 1979 Utilization of agricultural and food processing wastes containing carbohydrates. Chemical Socieg Review 8, 3OS314. GERHARDT, P. 1981 Manual of Methods for General Bacteriology. pp. 358359. Washington, DC: American Society of Microbiology. HESSELTINE, C.W. 1972 Solid-state fermentations. Biotechnology and Bioengineering 14, 517532.
HUANG, S.Y., WANG, H.H., WEI, C.-J., MALANEY, G.W. & TANNER, R.D. 1985 Kinetic responses of the koji solid-state fermentation process. Topics in Eqyme and Fermentation Biotechnology
10, 88-108.
HUTAGALUNG, R.I. & TAN, P.H. 1976 Utilization of nutritionally improved cassava by nutrient supplementation and microbial enrichment in poultry and pigs. In Proceedings of the Fourth Symposium oj tbe International Society of Tropical Root Crops, eds Cock, J., MacIntyre, R. & Graham, M. pp. 255-262. Cali: CIAT. LAUKEVICS, J.J., APSITE, A.F., VIESTURS, U.S. & TENGERDY, R.P. 1985 Steric hindrance of growth of filamentous fungi in solid substrate fermentation of wheat straw. Biotecbnologr and Bioengineering
27, 1687-l
691~
MITCHELL, D.A., GREENFIELD, P.F. & DOELLE, H.W. 1986 A model substrate for solid-state fermentation. BiotecbnoloD Letters 8, 827-832. MITCHELL, D.A., DOELLE, H.W. & GREENFIELD, P.F. 1988 Agar plate growth studies of Rbixopus ohgosporus and Aspergillus orJ?ae to determine their suitability for solid-state fermentation. Applied MicrobioLogy and Biotechnology 8, l-5. MOO-YOUNG, M., MOREIRA, A.R. & TENGERDY, R.P. 1983 Principles of solid-substrate fermentation. In The Fiiamentous Fungi, eds Smith, J.E., Berry, D.R. & Kristiansen, B. pp. 117-144. London: Edward Arnold. MUDGETT, R.E. 1986 Solid-state fermentation. In Manual oj Industrial Microbiology, eds Demain, A.L. & Solomon, N.A. pp. 6683. Washington, DC: American Society of Microbiology. NARAHARA, H., KOYAMA, Y., YOSHIDA, PICHYANGKURA, S., UEDA, R. & TAGUCHI, H. 1982 Growth and enzyme production in a solid-state culture of Aspergillus ory?ae. Journal of Fermentation Technology 60, 311-319.
426
World Journal
of Microbiology
and Biotechnology,
Vol 7, 1991
Protein
from
sag0 starch
RAIMBAULT, M. & ALAZARD, D. 1980 Culture method to study fungal growth in solid fermentation. European Journal of Applied Microbiology 9, 199-209. RAIMBAULT, M., REVAH, S., PINA, F. & VILLALOBOS, P. 1985 Protein enrichment of cassava by solid substrate fermentation using molds isolated from traditional foods. JournaL oj Fermentation Technology 63, 395-399. RAMOS-VALDIVIA, A., DE LA TORRE, M. & CASAS-CAMPELO, C. 1983 Solid-state fermentation of cassava with R&opus. oligosporus NRRL 2710. In Production and Feeding of Single Cell Protein, ed. Ferranti, M.P. pp. 104-111. London: Applied Science Publishers. SENEZ, J.C. 1985 Microbial food and feed. In Biotechnology and Bioprocess Engineering, ed. Ghose, T.K. pp. 3099319. New Delhi, India: Indian Institute of Technology. SUKARA, E. 1987 Production of single cell protein from cassava by microfungi. PhD Thesis. Brisbane: University of Queensland. SUKARA, E. & DOELLE, H.W. 1988 Cassava starch fermentation pattern of Rhi~opus ohgosporus. MIRCEN JouraL oj Applied Microbiology and Biotecbnolog), 4, 465471. WALISZEWSKA, A., GARCIA, H.S. & WALISZEWSKI, K. 1983 Nutritional evaluation of R&opus okgosporus biomass propagated on potato. Nutritional Report International 28, 197-202.
WANG, H.L. & HESSELTINE, C.W. 1982 Oriental fermented foods. In Prescott and Dunn’s Industrial Microbiology, 4th edn, ed. Reed, G. pp. 492-538. Westport, Connecticut: AVI Publishing Company. YANG, S.S. 1988 Protein enrichment of sweet potato residue with amylolytic yeasts by solid-state fermentation. Biotechnology and Bioengineering 32, 886890.
(Received 30 October 1990; revised 19 December 1990; accepted 10 January
World
Journal
of Microbiology
and Biotechnology,
1990)
Vol 7, 1991
427
