Jumal Kejuruteraan 7 (1995) 33-48 Coconut-Water as Fermentation Medium for Enzyme Production (I): Factor Analysis in Batch Shake-Flask Fermentation Jailani Salihon Jamaliah Md. Jahim Ahmad Jaril Asis ABSTRACf Factor analysis lVas applied to batch shake-flask fermentations for the production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae N.C.r.C. 754 on a medium based on coconut-water allowing a screening of the experimental variables temperature (T), pH, shaking rate (RPM) and the initial concentrations of glucose (G), yeast extract (YE), peptone (P) and phosphoric acid (PAl supplements. Both Ihe enzyme specific yield (nmol g") and the biomass yield (gL·t) were found to be strongly influenced by all the above experimental variables except P. Previous media formulations for the fermentation have shown rhat peptone is very important in ensuring good grolVth and enzyme production. The possibility of formulating a coconut-lVater based medium which needs less peptone supplement would lower medium costs as well as partly solve the waste- water treatment problem of the coconut processing plants. ABSTRAK AnaUsis faktor lelah digunakan Un/uk menapis pembolehubah-pembolehubah suhu (T), pH, kadar goncangan (RPM) dan kepekatan-kepekatan awal tambahan glukos (G), eks/rak yis (Y£), pepton (P) dan asid fosforik (PA) dalam fermentasi sesekumpul kelalang goncang Un/uk penghasilan enzim sitokrom p-450 oleh yis Saccharomyces cerevisiae N.C. Y.c. 754 dalam medium berasaskan air kelapa. Didapati kedua-dua hasil enzim tentu (nmol g.t) dan hasil biojisim (gL·t) kuat dipengaruhi oleh semUG pembolehubah tersebut kecuali P. Formulasi-formulasi medium sebelum in; untuk Jermentasi tersebut telah menunjukkan bahawa pepton amat penting untuk mempastikan tumbesaran yang baik dan penghasilan enzim yang banyak. Kemungkinan wujud untuk memformulasi medium berasaskan air kelapa yang memerlukall kurang tambahan pepton yang akan mengurangkan kos medium di samping menyumbang dalam menyelesaikan masalah rawatan air buangan kilang pemprosesan kelapa. INTRODUCTION The production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae has been studied mostly from the biochemical angle where the interests lie in finding out more about its characteristics [24, 25, 261, the ...
16
Embed
Coconut-Water as Fermentation Factor Analysis in Batch ... · Factor Analysis in Batch Shake-Flask Fermentation Jailani Salihon Jamaliah Md. Jahim Ahmad Jaril Asis ABSTRACf Factor
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Jumal Kejuruteraan 7 (1995) 33-48
Coconut-Water as Fermentation Medium for Enzyme Production (I):
Factor Analysis in Batch Shake-Flask Fermentation
Jailani Salihon Jamaliah Md. Jahim
Ahmad Jaril Asis
ABSTRACf
Factor analysis lVas applied to batch shake-flask fermentations for the production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae N.C.r.C. 754 on a medium based on coconut-water allowing a screening of the experimental variables temperature (T), pH, shaking rate (RPM) and the initial concentrations of glucose (G), yeast extract (YE),
peptone (P) and phosphoric acid (PAl supplements. Both Ihe enzyme specific yield (nmol g") and the biomass yield (gL·t) were found to be strongly influenced by all the above experimental variables except P. Previous media formulations for the fermentation have shown rhat peptone is very important in ensuring good grolVth and enzyme production. The possibility of formulating a coconut-lVater based medium which needs less peptone supplement would lower medium costs as well as partly solve the wastewater treatment problem of the coconut processing plants.
ABSTRAK
AnaUsis faktor lelah digunakan Un/uk menapis pembolehubah-pembolehubah suhu (T), pH, kadar goncangan (RPM) dan kepekatan-kepekatan awal tambahan glukos (G), eks/rak yis (Y£), pepton (P) dan asid fosforik (PA)
dalam fermentasi sesekumpul kelalang goncang Un/uk penghasilan enzim sitokrom p-450 oleh yis Saccharomyces cerevisiae N.C. Y.c. 754 dalam medium berasaskan air kelapa. Didapati kedua-dua hasil enzim tentu (nmol g.t) dan hasil biojisim (gL·t) kuat dipengaruhi oleh semUG pembolehubah tersebut kecuali P. Formulasi-formulasi medium sebelum in; untuk Jermentasi tersebut telah menunjukkan bahawa pepton amat penting untuk mempastikan tumbesaran yang baik dan penghasilan enzim yang banyak. Kemungkinan wujud untuk memformulasi medium berasaskan air kelapa yang memerlukall kurang tambahan pepton yang akan mengurangkan kos medium di samping menyumbang dalam menyelesaikan masalah rawatan air buangan kilang pemprosesan kelapa.
INTRODUCTION
The production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae has been studied mostly from the biochemical angle where the interests lie in finding out more about its characteristics [24, 25, 261, the
...
.. 34
mechanisms of its production and its role in the yeast [14, 22, 8, 23, 13, 2] with a view to findiqg potential applications. As interests moved from studying its synthesis and its role in the yeast to actually testing the purified form of the enzyme in various applications [6, 20, 4] the need arose for relatively large amounts of the enzyme to be available readily and cheaply. This has led to the pursuit of the course of optimisation of the production of the enzyme prior to scaling up.
Coconut-water which is normally thrown away as waste in coconut processing [I] has shown positive results when tested as the fermentation medium for the production of the insect toxin by Bacillus thuringiensis [7]. A survey of the analyses of its components [3, 18, 19, I] support these results. For a waste coconut-water is relatively rich in sugar (38.8 g V' according to Jayatissa et at. [19]) and it also contains significant amounts of amino acids [3] as well as the elements calcium, magnesium and iron [20] all of which are required by S. cerevisiae for growth and fermentation to occur [12]. An optimised medium for cytochrome p-450 fermentation based on glucose, mycological peptone, yeast extract and sodium chloride costs RM 3.54 per litre [21, 12].
A successful fermentation medium based on coconut-water should reduce fermentation raw material costs whilst at the same time partly solve the effluent treatment problem of discarded coconut-water [I].
This paper describes preliminary studies on the suitability of coconutwater as the base of a fermentation medium for the production of the enzyme cytochrome p-450 by the yeast S. cerevisiae N.C.Y.C. 754. The method of Factor Analysis [9, 10, 15] is employed to select experimental variables which are relevant to the fermentation.
THEORY
AN INTRODUCTION ON FACTOR ANALYSIS
The method of Factor Analysis [9, 10, 15] will be used to screen the experimental variables which are most relevant to the fermentation. This method has been shown to allow an efficient screening of the experimental variables which are most relevant to the biomass yield in a particular type of batch fermentation [16].
The method of Factor Analysis [9, 10, 15] enables us to describe the various experimental variables in terms of mutually orthogonal factors which are uncorrelated to each other but which have the same mean and the same variance as the standardised form of the experimental variables. Mutually orthogonal factors are important in that only such factors may be used to construct linear models, where the interactions between factors are not taken into account. Empirical models are constructed to describe the yields in terms of these mutually orthogonal factors. The significance of each factor in its effect on the yield is then detennined by removing the particular factor from the model involving all the factors and comparing the mean square difference between the actual data and the prediction of the resulting model with the mean square difference between the actual data and the predictions of the model involving all the factors using the statistical F-test [II]. These factors can then be classified into catagories according to how
35
significantly each of them affect the yield. If an experimental variable contributes only to factors which do not affect the yield signjficantly then it can be concluded that it is not relevant to the yield and can be dropped from subsequent experiments. If an experimental variable contributes to one or more factors which have significant effects on the yield then the experimental variable is relevant to the yield and should be retained for further investigation and optimisation.
TIlE SCREENING OF EXPERIMENTAL V ARlABLES RELEVANT TO TIlE FERMENTATION
To construct the models, a new table of the experimental results consisting of the yields and the factors has to be evaluated by calculating the experimental values of each factor. A linear regression between these factors and the yields is then constructed. Other models are constructed in the same manner but employing successively less factors. The evaluation of the significance of each factor is then done by the statistical F-test [5].
In this work the estimation of the regression coefficients of all models were done by using a Numerical Algorithm Group (NAG) routine (available from Pusat Komputer, Universiti Kebangsaan Malaysia) which operates by minimising the sum of squared differences between the actual and predicted yields. The results of the regression analysis were tested for significance at three levels of confidence namely 99%, 95% and 90%. The breakdown of the significance of the results into these three confidence levels were considered sufficient for this preliminary work. The F-values for these confidence levels for relevant degrees of freedom are given in Table 13.
MATERIALS AND METHOD
THE STRAIN
The yeast S. cerevisiae N.C.Y.C 754 was grown in batch in conical flasks shaken in a refrigerated incubator shaker (L.H. Fermentation uK/Syarikat Harapan, Malaysia) at constant temperatures and shaking speeds. The inoculum was prepared by growing the yeast from the slope culture in a medium of sterilised coconut-water. 100 ml of the medium is shaken in 250 ml conical flasks at 30'C with a shaking rate of 200 r.p.m. for 24 hours. Inocculation of the experimental flasks was done using a sterile pipette at 1.0 ml per flask .
THE MEDIUM
A fresh coconut-water sample was obtained from Restoran Universiti, Universiti Kebangsaan Malaysia. It was filtered to remove all suspended matter. 300 ml portions in 500 ml flasks were sterilised at 15 p.s.i for 20 minutes followed by storage at 4' C. As this was a preliminary study, no attempt was made to analyse the composition of the coconut water sample. lt would be done at a later stage if the medium proved to be useful.
The cu1ture medium in the experiment consisted of this coconut water sample with varying initial concentration of glucose, yeast extract, mycological peptone and phosphoric acid supplements and fixed initial concentra~ons of sodium chloride (5 g L·I) and ammonium sulphate (10
g L·I) supplements. These supplements have been obtained from various sources at prices as in Table 1.
100 ml of medium was shaken in 250 ml flasks. In all cases the varying amounts of glucose was dissolved in a fixed amount of coconut-water and sterilised at 15 p.s.i. for IS minutes separately from the rest of the components which were also dissolved in a fixed amount of coconut-water and sterilised at 15 p.s.i. for IS minutes to avoid the formation of pigments which masks the yeast and prevents the detection of cytochrome p-450 in whole yeast cells [14]. The two parts of the medium were then left to cool down before being mixed. The initial pH of each experimental flask was adjusted to specific values by addition of either 2M NaOH or 2M HC!. The details of each experiment are given in Table 2.
CYTOCHROME P-450 ASSAY AND BIOMASS DETERMINATION
The experiments were harvested at 48 hours after inoculation. Cytochrome p-450 was assayed by a modification of the procedure of Omura and Sato [17]. A 50 ml sample of the fermentation broth was taken using a sterile pipette and then centrifuged at 4640 g for 15 minutes. After decanting the supernatant, a phosphate buffer of pH 7.2 was added to the yeast cells to make a 10% (w/v) suspension on which the spectrophotometric determination of cytochrome p-450 was made. Another 50 ml sample is centrifuged at 4640 times gravity for 15 minutes followed by decanting of the supernatant, resuspending the yeast in distilled water, recentrifuging at 4640 times gravity for 15 minutes and decanting the supernatant again. The yeast is then resuspended in distilled water and dried at !05'e for 24 hours, cooled in a dessicator and its weight determined.
EXPRESSIONS OF YIELDS
Three expressions of yields were investigated namely the enzyme specific yield (nmol g~l) the enzyme volume-yield (nmol L·I) and the biomass dry weight (gL~').
EXPERIMENTAL DESIGN
In this work the value of each experimental variable was varied over as wide a range as possible so as to gain the maximum knowledge of the behaviour of the system over wide-ranging conditions. In most cases practical difficulties prevented the use of too large a variation in the values of the
TABLE 2. Experimental Results
Flask RPM T P YE G H,PO, pH Enzyme Yield Biomass No. r.p.m °C gL-' gL-l gL-' cmJL-1 pH nmolL-1 nmol g-I gL- '
experimental variables. For example most flasks incubated at higher than 34"<: failed to show any presence of cyctochrome p-450. In cases of zero yields the experiment could not be treated statistically since a zero yield is not sensitive to further cbanges in the values of the variables which would otherwise have decreased the yield.
O~--~~~r-----~~~---r----~----~r---~ 10 20 30 40 so
PREDICTED YIELD (nmol g. I)
FIGURE 1. The Performance of the Linear Model Involving All the Factors for the Enzyme Specific Yield
60
" .J .e9
" .J w ;: .J
'" ~
28
26 o
24
22
20 D
18
16
14
12
10
8
6
4
2 a
o
25% DEVIATION BAND
o
o
o
D.
a 0
a
00 2 .4 6 8 10 12 14 16 18 20 22 24 26 28
PREDICTED YIELD (g L -!)
FIGURE 2. The Performance of the Linear Model Involving All the Factors for the Biomass Yield
43
It was assumed that the response surface did possess a maximum. If the area that has been examined did not cover the maximum. a linear model might be sufficient to represent the surface since there would not be as much curvature as there would be if it had contained the maximum. Therefore, a linear model approach was tried first. If it did not fit the data of the orthogonal factors and yields, a non-linear model would have to be fitted instead so as to account for the curvature that is likely to occur near the maximum yield.
RESULTS
The results of the factor analysis are given in tables as follows:
Table 2
Table 3 Table 4 Table 5
gives the details of all experiments and the experimental results. gives the correlation matrix of the standardised variables. gives the eigen values of the correlation matrix. gives the .coefficients of the eigen vectors associated with the eigen values of the correlation matrix.
44
C'
.J 0 e .5 Q .J OJ ;: .J
'" E '"
900
800
700
0
600 40% DEVIATION BAND
500 0
400 00
300 0 0
0 0
200
o o
o
..
o
0
0
CO
0
40% DEVIATION BAND
0
o
O~----~~~~------r-----~----~ o 100 200 300 400
PREotCTED YIELD (nmol L .1)
FIGURE 3. The Perfonnance of the Linear Model Involving All Ihe Factors for the Enzyme Volume Yield
Table 6 gives the values of the factors. Tables 7 10 12 give the evaluation of the linear model involving all the
factors, for each of the three yield expressions. Figures I to 3 show the performance of the linear models for the enzyme
specific yield, the biomass yield and the enzyme volume yield respectively.
DISCUSSION AND CONCLUSIONS
The experimental variables that characterise a factor are defined as those that have absolute values of coefficients greater than 0.4 (thus allowing for 40% error) in the equation describing that factor in terms of the experimental variables. The following conclusions can be made from Table 5.
I. Factor F, and F, are the pH factors, being characterised by pH and the initial concentration of phosphoric acid.
2. Factor F, is the oxygenation factor, being characterised by RPM, initial glucose concentration and temperature. The solubility of oxygen is
45
affected 'by temperature and glucose concentration in the medium, while RPM affects the mass transfer of oxygen from liquid surface into bulk medium.
3. Factors F3 and F, are the nutritional factors, F3 being characterised by the iqitial concentrations of yeast extract and peptone and F, being characterised by the initial concentrations of glucose and peptone.
4. The factor F, is characterised by temperature and the initial concentration of peptone. The correlation between temperature and peptone is not known except that both of them affect the solubility of oxygen in the medium.
5. The factor F, is characterised by RPM and temperature.
Figure I on the performance of the linear equation involving all the factors for the enzyme specific yield shows that 66% of the data points lie within the 40% deviation band. Figure 2 on the performance of the linear equation involving all the factors for the biomass yield shows that 60% of the data points lie within the 25% deviation band. Figure 3 on the performance of the linear equation involving all the factors for the enzyme volume yield shows that 66% of the data points lie within the 40% deviation band. Thus the model for the biomass yield is the most accurate with the model for the enzyme volume yield and the model for the enzyme specific yield showing bigger errors. This is to be expected because while the biomass weights can be determined quite accurately because they level off at the end of the fermentations (stationary phase of growth), the enzyme levels go up sharply and then decrease after reaching a peak [21]. More readings of the enzyme levels would be required to achieve greater accuracy of peak enzyme content determination, but this is not possible in shake-flask work. Although errors as high as 40% may be involved, meaningful discussion and conclusions on the results can be achieved by employing the statistical approach to analysis using the F-test [5].
Table 7 for the evaluation of the linear model for the enzyme specific yield shows that removing all the factors except factor F, do not make the linear model incur errors which are significant at 99% confidence level compared to the error incurred by the model involving all the factors. Therefore, at 99% confidence level the best model (i.e. the simplest model which does not incur significant errors compared to the errors incurred by the model involving all the factors) for the enzyme specific yield is the model involving only factor F, and a constant. Table 8 gives the coefficient of the factor and the constant for this linear model. Table 5 shows that the factor F, is the pH factor, being characterised mainly by contributions from the experimental variables PA and pH. Thus it can be concluded with 99% certainty that the enzyme specific yield is strongly influenced by the experimental variables PA and pH.
At 95% confidence level, the best model for the enzyme specific yield is the model involving the factors Fl' F" F3, F" and a constant. Table 5 shows that these factors consisted of contributions from all the experimental variables with the exception of P which contributes only weakly. Thus it can be concluded with 95% certainty that the enzyme specific yield is strongly influenced by all the experimental variables with the e~ception of P which influences it only weakly. The situation is still very much the same at 90%
."
46
confidence level where although the best model consisted of all the factors F, to F, and a constant, P does not have a sizeable contribution to F,.
Table 9 for the evaluation of the linear model for the biomass yield shows that removing all the factors except F, do not make the linear model for the biomass yield incur errors which are significant at 99% confidence level. Thus, like the enzyme specific yield, the best model for the biomass yield at 99% confidence level also consists of only the factor F, and a constanl, Table 10 gives the coefficient of the factor and the constant for this linear model. Therefore, it can be concluded with 99% certainty that the biomass yield is strongly influenced by the experimental variables PA and pH.
At 95% confidence level, the best model for the biomass yield consists of the factors F, and F, and a constant. Table 5 shows that the factor F, is characterised mainly by contributions from the experimental variables RPM
and G with T contributing weakly. At 90% confidence level, the best model consisted of the factors Fl' Fz' F) and a constant. The factor FJ is characterised mainly by contributions from the experimental variable YE with P contributing only marginally.
The enzyme volume yield is derived by multiplying the enzyme specific yield and the biomass yield. It is this yield that shows the total amount of enzyme present in a volume of fermentation broth. Table I for the evaluation of the linear model for the enzyme volume yield shows that removing the factors F" F" F, and F, do not make the linear model for the enzyme volume-yield incur errors which are significant at 99% confidence level compared to the errors incurred by the model involving all the factors. Therefore the best model for the enzyme volume-yield consisted of the factors F
" F" F, and a constant. Table 12 gives the coefficients of these
factors and the constant for this linear model. We can thus say with 99% certainty that the enzyme volume yield is strongly influenced by the experimental varibles AF. pH. RPM. G and YE with T and P each being of marginal influence.
The best models for the enzyme volume-yield at 95% and 90% confidence levels both include the factor F, and F, in 'addition to the factors F, to F, which already have been found to be important at 99% confidence level. The experimental variable T has an important contribution to F, and F" but in both factors the contribution of the experimental variable P is marginal.
Previous media formulations for cyctochrome p-450 fermentation [14), [21) have shown that P is very important in ensuring good yeast growth and enzyme production. This work has shown that with coconut-water as the base of the medium, the variations in the level of P as in Table 2 are not relevant to the enzyme specific yield at 99% confidence level and not relevant to the biomass yield at 95% confidence level or higher, besides being only of marginal influence to the enzyme volume-yield at 90% confidence level. This shows that with coconut-water as the base of the medium, the requirement of the fermentation for the contribution by peptone has hargely been provided for by the coconut-water itself.
This work bas also shown that although coconut-water is rich in sugar [19) the variations in the level of G, as in Table 2, is relevant to the enzyme specific yield and the biomass yield at 95% confidence level and relevant to the enzyme volume-yield at 99% confidence level.
47
Peptone is by far the most expensive component as shown in Table 1. The possibility of formulating a coconut-water based medium which needs less peplone supple men I than previous formulations [14. 21) would lower medium costs as well as partly solve the wastewater treatment problem of the coconut processing plants [I].
Symbols
RPM
YE P PA T G
NOTATION
Description
Shaking speed Initial concentration of yeast extract Initial concentration of peptone Initial concentratlon of phosphoric acid Temperature Initial concentration of glucose
REFERENCES
Unils
r.p.m gL" gL" cm-1L' "C gL"
I. Azman Firdaus Shafiee. 1986. University-Industry R&D Collaboration: A Chemical Engineering Perspective. Proceeding of the 2nd Symposium of Malaysian Chemical Ellgineer.f.
2. Blatiak. A .. King. 0.1 .. Wiseman. A .. Salihon. J. and Winkler. M.A .. 1985. Enzyme Induction by o"ygen in the accumulation of Cytochrome p-450 during Batch Fermentation in 20% D-gJucose with Saccharomyces cerevisiae, 1985. Enzyme Microb. Techno!. 7. 553.
3. Bienvenido, C. Sison Jr., 1976. Disposal of Coconul Processing Waste. Phillipine 1. Coconut Studies 2. 39-41.
4. Coulson. C.1 .. King, D.1. and Wiseman, A., 1984, Chemotherapeutic and Agrochemical Applications of Cytochrome p-450 Ligands 1984, Trends in Bi()l:hemical Sciences.
5. Cochran. W.O. and Cox.. G.M .. 1957. Experimelllal Design, John Wiley and Sons, New York.
6. Dorr. A.W .. Winkler. M.A. and Wiseman, A. , 1984, Computer-linked on-line Sensing of Carbon Dioxide and Oxygen in Fermentation Trends Anal. Chern. 3. 15·18.
7. Femendez, W.L. Ocampo, T.A. , and Perez, D.C., 1975. Coconut-water in Three Media Reduces Cell Yield of Bacillus thur;ngiensis Var. thuringiensis borliner. Phil. Agri .. 58. 273-279.
8. Gondal. J.A .. 1979. Ph.D Thesis. University of Surrey. 9. Harman. H.H .. 1967, Modern Factors Analysis, 2nd Edition (Revised) - The
Uoiv. Chicago Press. Chicago. 10. Harman. H.H .. 1967 Factor Analysis: in Mathematical Methods for Digitci/
Compucers. Eds. Ralston. A. and Wil!'. H.S. Vol. 1. 204. John Wiley and Sons Inc. New York.
II. Himmelblau, n.M., 1970, Process Analysis by Statistical Methods. John Wiley and Sons Inc. New York.
12. Jamaliah Md. Jahim. 1987. Tesis Sm. Kej . UKM . 13. King. OJ .• 1983. Ph.D Thesis, University of Surrey. 14. Lim. T.K., 1977. Ph.D Thesis, University of Surrey. 15. Mulaik, S.A., 1972. Factor Analysis. McGraw Hill, New York. 16. Moresi. M., Nacca. c.. Nardi. R. and Palleschi. C .. 1979. Factor Analysis in a
Whey Fennentalion by Kluyvermyces fragilis , Eur J. Appl. Mic:robiaJ. BiolechlloJ.
49·61. .,
48
17. Omura, T and Sato. R., 1964, The Carbon Monoxide-binding pigment of Liver Microsomes 1. Evidence for its Haemprotein Nature. J. Biul. Chern. 239, 2370-2379.
18. PutTanon, R. t Bhumiratana, S. , Tantiwechwuttikul. P. and Patamayothin, N. 1983, Concentrated Coconut-water by Membrane Process Alln. Rep. of ASEAN Workshop Group 011 the Management and Urili.'wt;on of Food Waste Material.
19. Jayatissa. P.M .. Jeya Raj E.E.. Tirimanna. A.S,L. and Senanayake. V.M .. 1972. Work study on Water Recovery b.\' Microorganism. UNESCO/ICRO, University Malaya.
20. Azari. M.R .. 1984, Ph.D Thesis. University of Surrey. 21. Salihon. J., 1984, Optimisation of Yield of An Intracellular Enzyme from Yeast
in Batch Fermentation Ph.D Thesis. University of Surrey. 22. Wiseman, A .. Lim, T.K., and Woods, L.F.J ., 1978, Regulations of the Biosynthesis
of Cytochrome p-450 in Brewer 's Yeast: Role of cyclic AMP, Biochemica et Biophysica Acta, 544. 615-623.
23. Woods. L.F.1 .. 1979. Ph.D Thesis. Universi.y of Surrey. 24. Yoshida, Y. , Kumaoka, H. and Sato. R.. 1974a. Studies on the Microsomal
Electron-transport System of Anaerobically Grown Yeast I . Intracellular Localisation and Characterisation. J. Biochem. (Tokyo), 75. 1201-1210.
25. Yoshida. Y .. Kumaoka. H. and Sato. R., 1974b. Studies on the Microsomal Electron-transport System of Anaerobically Grown Yeast. 2. Purification and characterisation of cytochrome bs' 1. Biochem. (Tokyo), 7S. 1211-1219.
26. Yoshida. Y.. Kumaoka. H. and Sato. R .• 1975. Smdies on the Microsomal E\ect.ton-\.rans?On System o( Anaerobically Gmwn Yeas\. 3. Spect.t'a' Characterisation of cyctochrome p-450. J. Bioc:hem. (Tokyo), 75. 785-794.
Department of Chemical and Process Engineering Universiti Kebangsaan Malaysia 43600 UKM Bangi Malaysia