Fermentation of cellulosic materials to mycoprotein foods

Biolech Adv. Vol It, pp. 4,'oq-.479, 1993 Printed m Great Bntam. ALl Rights Resen ed

0734--'9750,'q3 $24 00 ¢. Iqq3 Pergamon Press Lid

F E R M E N T A T I O N O F C E L L U L O S I C M A T E R I A L S TO MYCOPROTEIN FOODS MLIRRAY MOO-YOUNG, YUSUF CHISTI and DAGMAR VLACH

Departmentqf ChemtcalEngineenng, Univers.'ztyt~fWaterloo,Vv'aterloo,Ontario. Canada N2L 3G1

Abstract

A new bioprocess is described in which a cellulolytic, food-grade fungus Neurospora sitophila converts cellulosic materials to protein-rich products for food and fodder. The optimal conditions for the conversion are identified: 35-37 °C temperature, pH 5.5, 2.35 ms "t agitator tip speed. Scale-up of the production process to 1,300 L is reported. The mycoprotein production data on several types of cellulosic materials (sugarcane bagasse, corn stover, wood cellulose) are presented. The performance of N. sitophila is found to compare favourably with that of Chaetomhtm celhdoh,ticum, another cellulolytic organism previously reported on by us.

Key Words Mycoprotein, Neurospora s#ophila, cellulose, single cell protein, Chaetomh~ra cellulo~'ticum.

Introduction

Cellulosic residues (e.g., straw, corn stover, sugarcane bagasse) are under-utilized by-products of the agricultural indust~'. Much of these residues originate from plants used traditionally in food and feed production. Because they come from acceptable food sources, the residues could potentially be upgraded to food by improvements in digestibility, nutritive value and palatability. This paper describes a new process for converting various types of cellulosic solid residues to protein-rich products for food and fodder. The process in an extension of a recent invention for converting cereal-grain bran residues into proteinaceous products (Moo-Young et al., 1990). The process is based on the filamentous fungus Neurospora sitophila which has a long history of use as food in oriental preparations such as ontjorn (Hesseltine and Wang, 1967; Steinkraus, 1986; Wood and Yong, 1975). Additionally, N. sitophila has a processing advantage as being one of the faster growing

microfungi. With a maximum specific growth rate of 0.40 h "l it has a

doubling time which is shorter than that of some bacteria (Solomons, 1975). By comparison, the 469

r'..1 r',1C~,3-'~OU NG ~t ,~1

4-It

ma'~imum specific growth rates of other c o m m o n industrial fungi are half (e.g., for ~tt'ger) t~r exen les'~ than a third (e.g., f~r

.4~pe@ilh+s

Penicillium cho'sogem~m than that of N. sitophilu

i Solomons, 1'4;51.

The Process Concept Conceptually, the

N. siu~phila m y c o p r o t e m p r o d u c u o n process consists of the following step,,:

• Size reduction of the cellulosic residue by mdling or grinding: • Treatment of the residue with alkah, acid and'or steam to increase the accessiblli~ ~f the cellulose in the particles: • Fermentation ol the residue with N.

sitt;phila

either in submerged or surface culture:

• Solid-liquid ~eparauon and deh.~dration of the product tor direct use as todder: • Blending, possible nucleic acid reduction, textunzing and flavouring operauons for human food apphcatkm'~. The size reducuon, sohd-liquid separauon and dehydrahon steps use ~ell knossn chemical engineering operations discussed elsewhere (Chisti and Moo-Young,

1991). The alkah

pretreatment step (which, depending on the cellulosic material type. may not be needed'p ts described later in this article. Blending, te\turizing, colouring and fla,.ouring operations enhance the palatabilit~ and the organoleptic properties of the product. These operations are m common use m the rood pr,3ces~ing industry, and haxe been de~,eloped also for the fungal protein food "Outwn" being marketed m the Llnited Kingdom (Steinkraus, l~Sb). Nucleic acid reducuon i,, al,~o used m Ouorn manufacture [Steinkraus, ]986). The reduction of nucleic acid~ ma~ be reqmred if the product is used for human food. The dicta D le~,el of R N A should not exceed 2 g per da~: breakdo~n ol R N A in human hcMv leads to elevated le',els of uric acid ~hich nta~. c,tu,~e such metahohc dL,,orders as kidne~ stones. In ammals uric acid is readtl.v e~creted b,, con'~ersion to allantom, and R N A reduction in feeds is not nece'~saD'. The R N A reductton techmques ha~e been revie~ed before (Solomons, 1c)75, Sinske)and 3-annenbaum, ]9?5). Here v.e will fc,~cus only on the fermentaticm aspect'~ ot the N.

~#ophila

mycoprotein process tor

upgrading cellulomc~.

Fermentation Process Development Development of the process through shake flask and pdot scale fermentations ~as needed tc~ ans~er st~me fundamental questions: What are the optimum pH and temperature tor protem production and cellulose utihzation? What are the rates and the maximal levels of protein tormatton and cellulose consumption? How does the utihzation orcellulose - a difficult to tmhze •,olid substrate - compare with microbial bsomass production on more readil,~ accessible

FERMENTAT1ONI TO MYCOPROTEIN FOODS

471

molasses? i.e., ho~ does the "best ease" protein production performance of the fungus, not limited b) substrate accessibilib, compare with that on less expensh, e cellulosic substrates? Ho~ would some of the common cellulosic residues such as corn stover and bagasse fare in terms of the ability to support protein biosynthesis? Would the fungus perform as well in fermenters as in shake flasks? What is its susceptibility to mechanical and other damage in fermenters? And, hog' does ,\,'euro~pora sitophih~ compare with the capabilities of Chaetomhtm celhtloh.'ticton, one of the better known cellulol~tic organisms? These questions ~ere answered by experimentation which proved the process concept and its ,~calabilit3 as discussed below.

Experimental Cultures and inocula. The filamentous fungi ,Veurospon~ sitophila (ATCC 36935) and Chaetomittm celhdoh'ticttm

(ATCC 32319) were maintained separatel.~ at 4 ~C in submerged cuhures on

glucose (It) kgm 3 ) supplemented ~ith )'east extract (Dill'co) f2 kgm 3) and the follo~ ing nutrient salts (per litre): (NH4)2SO 4. 0.47 g; urea, 1/.86 g: KH2PO 4. I),714 g: MgSO4.7H,O, ().2 g: CaCl,. 0.2 g; FeCI 3. 3.2 rag: ZnSO4.TH20. 4.4 mg: HsBO 3. 0.114 rag; (NH4)0Mo,O24.4H20. 0.4S rag: CuSO4.5H20, 1).7S rag: MnCI~.4H,O. fl. 144 rag. lnocula were grown at 26 ~C on the specified carbon source (5 kgm 3) supplemented ~ith 0.5 kgm -3 molasse~ (Holfman Feeds Ltd. Heidelberg. OntarioJ and the earher specified sahs. The fermentation media contained a carbon source ("Solka Floe" wood cellulose, sugarcane bagasse, corn stover, or molasses). The Solka Floc cellulose (a-cellulose) was made from wood pulp (James Ri~er Corporation. Berlin. Ne~ Hampshire). The KS1010 and the BW300 grades used in this ~ork had an a,.erage particle (fibre) length of290 p.m and 22 p.m, re,~pecti~el). The BW300 grade had a degree of crs.stallinit,~ of 62-65 % cDstalline, whereas the KSI016 had a greater proportion of cD'stalline cellulose at 75-7-', %. Apart front the Solka Floc cellulose, all other residues were pretreated v, ith sodium h)droxide (0.15 kgkg residue) at 121 '~C for 30 minutes, .Although the media were supplemented v, ith the full complement of the earlier specified nutrient salts, only ammonium sulfate and phosphates were essential requirement ~ t h certain naturall)-occurrmg cellulosic residues such as straw and corn sto'.er.

Fermentation conditions. Fermentations were conducted either in shake flaskx or in a 75 L (nominal) stirred tank fermenter ( M B R Sulzer, Switzerland). The shake flask runs v,ere performed in 250 mL (nominal) flasks containing lfl0 mL medium including an specified carbon source and the nutrient salts. The flasks ~,ere sterilized at 121 °C for 3fl minutes, co~71ed to ambient, inoculated and held at the specified temperature on a D'rato D' shaker at 250 rpm. Unless otherwise indicated, the pH at inoculation was 6.0. At desired times, the flasks were

472

M. ~.f(~-~-'~ O L i N G

~t ,tl

rapidly cooled and stored at 4 ~C ~f necessar 3. T h e flasks `.~ere analyzed for total d ~ solids. crude protein a n d cellulo';e.

Crude protein and celhdose. For crude protein and cellulose determinations, the f e r m e n t a u o n broth was filtered under s u c u o n through a 25 p,m "Nitex" nylon cloth ( T h o m s o n Co., Scarbrough, Ontario), the filter cake was ~`.ashed ~ith several broth v o l u m e s of deionized `.`.ater and dried o v e r m g h t at 9(~ '~C. The d~' btomass was g r o u n d to l m m particle size and u ~ortton ,.,.as analvzed for total m t r o g e n using a mlcroKjeldahl technique (Lang, 1958). T h e crude protein c o n t e n t ol the b i o m a ~ `.`.ere calculated as 6.25 ,, total nitrogen, and percent I~`..'~s) protein as g r a m protein per 11)(I g total dr). sohds. The cellulose content ~`.ere d e t e r m i n e d b) the s p e c t r o p h o t o m e t r i c anthrone-sulfuric acnd m e t h o d (Llpdegrafl, 196t41: percent cellulose ~`.a,, calculated on the s a m e bas~s as crude protein.

Shear effects. T h e influence ~3f shear on protein production ~sa~ investigated m the 75 L f e r m e n t e r (,,essel d i a m e t e r = 1).31~ m; working a,;pect ratio = l,t)) w~th a ~ o r k m g ~olume latter inoculatton) of 51) L. T h e t e m p e r a t u r e and pH v, ere controlled at 26, ' C and pH ¢~.1), respecu`.el,,. ]'he dissolved oxygen level ~`.as nc~t allo~`.ed to drop belo~ 20 % of air ~aturation, Aeratton rate ',aricd ((1.4 - 11.,',¢,',~.['n) in re~pon~e to the tlissol~ed o \ 3 g e n le`.el. A O-blade d~.',c turbine `.`.as used Ior agitation I d i a m e t e r ot impeller.'tank d i a m e t e r = 0.57; li3catlon above b o t t o m of tank = q).B impeller d i a m e t e r l at 25ql, 30ql or 350 rpm c o r r e s p o n d i n g respecti`.el.~ t~ tip s p e e d s tlf 2.35, 2.~2 and 3.29 n3~l. ,V. situphila was grov, n on Solka Floc ( K S l t l l h i 15 kgnt3'l s u p p l e m e n t e d ~`.sth mola~se~ ((I.5 kgm-~). (NH41_~SO ~ ¢1).28 gL-II: urea (0.52 gL lJ. K H 2 P O 4 I l.(t gL l ) and other, pre`. uou~l.v listed, nutrient s a h s at half the concentrations spec~tied earher.

Results and Discussion

Temperature effects. The effect of t e m p e r a t u r e on cellulose milizatton and crude protein production i~ sho~`.n m Figure 1. T h e result,; in Figure 1 `.`.ere obtained in shake fla.,ks `.`.Lth ,V.

sitophila grown on industrial cellulose slur% m e d m m t ll'l kgm -3 Solka Floc grade KS1010) ~ u p p l e m e n t e d v, ith molasse~ t l kgm x) and the nutHem ~alt~,. E'tch d a m point in the figure (Figure 1) c o r r e s p o n d e d to the m a x i m u m cellulose utilization and protein productson at the given t e m p e r a t u r e at 3S hour~, since inoculauon (1() ~ `.,`., c~a. 0.3 kgm -3 initial protem) of the flasks. T h e f e r m e n t a t i o n s p e a k e d ca. 3S hours after initiation. A m a x i m u m cellulose utilizauon of - 86 ~c of orLginal cellulose and protein production ol - 35 ¢; of the total dR?,. ~`.eight `.`.ere o b s e ~ ' e d at 37 ~C. The optimal fermentation t e m p e r a t u r e range was 35-37 °C- t e m p e r u t u r e s higher than - 3~ ~C caused s h a r p decline m btomass production ( M o o - Y o u n g et al., 1992'1.

FERMENTATION TO MYCOPROTEIN FCKDD5

473

50

90 -~

40

E o~

0

Z

o

/

20

t---

7C,,~

I'M .._J

ZC, Z

")-' /

PRC,T;IFd<

b4 -,

Z

Ld 20 I.-0 ['y O_

0 a:

/ 2"

.z) N

S

c,'

Lxl k-)

30

35

TEMPERATURE

40

45

(°C)

E.O

/

LO 0

.=..2, -., -J .-3 I

4

2,2, .~

I C'

25

I--

I

1,2' U

pH

• Figure 1. Effect of fermentation temperature on N. sitophila protein production and cellulose utilization. • Figure 2. Effect of fermentation pH on crude protein production (,,V.sitophih~)and cellulose utilization.

Effect ofptt. The influence of pH on N. sitophila fermentations of cellulose (Solka Floe BW300) is shown in Figure 2. These fermentations were conducted in the 75 L fermenter at 37 ° C . The concentration of cellulose and the supplements were the same as used in the pe~'ious set of experiments on temperature effects. The inoculum for these runs was grown on KS1016 grade of Solka Floc cellulose (10 kgm -3) supplemented in the same way as the fermentation medium. All fermentations were inoculated at pH 6.(1, the pH was allowed to fall to one of the set points shown in Figure 2. and was controlled at the set point. The protein production and cellulose utilization reported in Figure 2 ~,ere the maximum values which occurred ~ 38 hours into the fermentation. A pH optimum of pH - 5,5 was identified. At this pH - 80 % of the cellulose ~hlch was originally present had been used up by 38 hours, and - 2 kgm "3 protein had been produced ~hJch represented - 33 % of dry weight of the product (Moo-Young et al.. 19921. Typically, in the ammonium/urea containing media as used in this work, the pH in fungal fermentations without pH control, initially declines due to consumption of NH4 + as the nitrogen source. Upon exhaustion of NH4 +, urea supplies the required nitrogen and the pH of the broth rises. This behaviour is observed in

N. sitophila fermentations (e.g., Oguntimein et al., 19921 as

well as those of other fungi. The ammonium,'urea ratio in the medium can be manipulated for rough pH control particularly in shake flasks: however, superior pH control, e.g., by acid,'alkali addition, in large scale fermentations can enhance production rate and product }4eld sufficientl} that such control is worthwhile e~en for Io~ value products.

4-4

M. MOO-', OLiNG ,-t al

Agitation. Mechanical agttation in stirred fermenters i.,, kno~n to danrage mycelial biomass and affect the yield oi: the product (Chisn and Moo-5oung, 1989, Moo-Young and Ch~sti, 19gS: Ujcova et al., 1981]). Characterization o{ the influence of the impeller ~peed on N.

~it,'~phila

protein producnon ~as reqmred to identify the ~uitable operational condiuons, an},' scale-up hmltation';, and the senblti',it,, of this parncular fermentation to impeller induced shear. The protein producnon profiles at tanou.,, agitation rates (tip speeds) are s h o ~ n in Figure 3. For otheD.~,i,.e identical conditions, increa';ing tip speed ol the Rushton dt.,,c turbine impeller loitered the rate ol/ protein production tFigure 31, and the maxm]um protein yield. Thus, a, shou.n in Table l, the ma',:lmum specihc protein productton rate 1~1 decreased fronl a high ot 11.09 h l at 2511 rpm to a Io,*, of 0.05 h -I at 35(I rpm. In relative term.,,, the protein production rate (~UR) at tile htghest rpm v,a~ onl) 55 r;: ol that at the Iov.,est agitatton. Data on peak protein productton and cellulose utdtzation lat 3g hours into tile lermentationl are shown in Table 1 in absolute and relative terms. At the hlghe,,t tip speed used ( 3.29 m,."l ) a dt.,,tinct lag pha',e ~as nouced ( Ftgure 3) in protein prcMuct~on compared to tile results at Iot~.er agitation mtensme~,.

Table I. Effect o{ Impeller Speed on Protein Producuon and Cellulose Utdization Inlpeller Speed (rpm)

Tip Speed (ms 1 I

/1 i h 1 )

MR {I )

Crude Protein (~:?)

Cellulose Uttlization (e~,)

250

2.35

O.()q

1.0

31.1 ( 11"

79.8 I 1)*

3t,t

2.S2

().H7

11.78

2-'..7 (o..~g)

60.0 (O.g6)

35q)

3.29

0.115

0.55

21.2 111.67 )

55.6 ( 0.7111

" Values in parentheses are relatl~.e to the ',alue at 2511 rpm.

Z.

]

T ':'= :TE,r.

•

Z

•

L- Jp.-

¢

2

~'

i

/

,'

TII:

JFCP_-I_-' • "

- 'tr

i,,,' -

i

Tir,lE

-

1

_,

Ell,i £

m

~

-92 ~ --'3 2

~ Pi ; 71r.1~

F i g u r e 3. Effect

of agttation

on protein

production. Impeller speed (rpm I:L g l 250: (e~ 3(10: and (o) 350.

, r', I

Figure 4. Compartson ol protein production m shake flasks and the "/5 L tern]enter.

FERMENTATIONTO M5 COPRO1T_INFCY-DDS

475

Clearly, the N. sitophiht fermentations were quite sensitive to excessive agitation, and low agitation rates, consistent with adequate mbfing and ox3,gen supply were indicated for the successful production process (Moo-Young et al., 1992). Further experiments (Figure 41 confirmed that an agltauon speed of 251) rpm in the 75 L fermenter was optimal for protein production. At th~s speed the protein production m the fermenter agreed closely with that m shake flasks (Figure 4). The data in Figure 4 were obtained ~ith N. sitophila grown on Chinese (Peoples Republic of China) bagasse (10 kgm 3) which had been pretreated w~th sodium h>droxide. These fermentations were conducted at 37 °C and pH 5.5. ZC' t.I,',L ~ ~:;z

-

.'

,~ IC

.%

, ? II-

~2~,3.:, 5'3E i Iij. ~,~ rt-,- ] I

'-

'"

b /' ,7 ,,'

Z ~l h -I .'

Ck

•

I

' 10

I 15

.~

(~

O 41

I 2~'

IID

"

,

.

(.'

I

4 ~,r-.pr,,I.a ~, : c , ~ z ~ ,

"_- ,-

25

TflYIE th) Figure 5. Comparison of N. sitophila protein production on molasses and bagasse in 75 L fermenter (37 °C. pH 5.5, 250 rpm impeller speed).

'o, ._-.co,.

C~llul,2,1 l I ,,: urr,

: '-' t ~ h -1 3r ] 7 : C

~ ~t / ' TrME

~,h)

Figure 6. Protein production: N. sitophila vs. C. celluh~h.,ticum.

Substrate characteristics. The protein production perlormance of N. sitophila on sugarcane bagasse and molasses was compared as sho~n in Figure 5. A specific growth rate of 0.26 h -t was obtained on bagasse, a less accessible sohd substrate. On the more readily accessible molasses. the specific growth rate (= 0.41 h l ) ~as - 1.6-tbld greater. The growth rate on molasses v,as comparable to the maximum ,,alue of 0.40 h 1 reported for N. sitophila growing on glucose at 30 °C (Anderson et al., 1975). These observations confirmed that the protein production process could potentia[I.~ be improved significantly by intpro~,ing the accessibflib of the substrate to the fungus (Moo-Young et al., 1992). Substrate a',ailabili b ~as limited either b~ restricted physical access of the fungal cellulases to the solid particle and,'or by inherent limitations in the rate of hydrolysis of cellulose. The latter could be due to either a limited rate of production of cellulases or due to limitauons in the kmeucs of the hydrolytic reaction itself. The possibility that secretion of N ~#ophila cellulases and their inherent hydrol)tic capability combined, ~,,ere less than that of other microfungi was dJscnunted in ~iew of the results sho~n in Figure 6, ~,here protein production

M. M(X3-YOUNGet al

476

by N. sitophila was compared with that by C. celluloh'ticvm. The fungi were grown in shake flasks on alkah pretreated corn sto~er (10 kgm -3) supplemented with the nutrient salts. As shown in the figure, N. sitophila protein production was comparable to that obtained with (2

celhdoh'ticum, e~,en though at 26 °C the cultivation temperature for Neurospora was less than optimum. The ma.xtmum specific growth rate of C. celhdoh'ticum was about 12 % greater than that of N. sitophila. Both fungi had utilized - 93 c~- of the cellulose by 24 hours into fermentation (Moo-Younget al., 1992). Despite these results, we believe that cellulase secretion by N. sitophila can be enhanced to further improve its cellulolytic potential. This ~ie~ ts supported by the obse~'ed two-fold increase in cellulase activity per unit biomass upon disruption of N. sitophila (Baldwin and Moo-Young, 1991) which implies that onl.~ about 5(3 % of the available cellulol}tic acti,,'it)' is normally secreted. More detailed work on the production of cellulases by N. sitophila, and on the characteristics of those cellulases was reported b.~ Oguntimein et al. (1992).

The

product. The N. sitophila raw protein product had a pleasant almond smell. ?dmond or

minced-meat flavour occurs also in ontjom produced by fermentation of peanut press cake by

At. sitophila (Hesseltine and Wang, 1967). The average composition of the fungus was (% v...Iw): 45 % crude protein, 40 % carbohydrates. 10 % fats, 5 % minerals, vitamins, etc. The amino acid composition (g per 100 g protein) of the fungal protein is shown in Table 2 where it is compared v, ith that of fodder yeast (Candida utilis), soyabean meal and the FAO reference protein.

Table 2. ,~nino Acid Composition of Various Proteins

N. sitophila

C. utilis

Soyabean Meal

FAO Reference

Threonine

3.8

5.5

4.0

2.8

ValJne

6.9

6.3

5.0

4.2

Cystine

0.8

0.7

1.4

2.0

Methionine

1.8

1.2

1.4

2.2

lsoleucine

5.0

5.3

5.4

4.2

Amino Acid

Leucine

8.0

7.0

7.7

4.8

Tyrosine

2.1

3.3

2.7

2.8

Phenylalanine

4.2

4.3

5.1

2.8

Lysine

5.3

6.7

6.5

4.2

FERMENTATION TO NP(COPRO'FEIN FOODS

477

Process Scale-up Having identified the optimal fermentation conditions and quantified the fungal susceptibilit) to mechanical damage in fermenters, the fermentation process was further de,,eloped at the 75 L scale and scaled-up to 1,300 L pilot plant (M'oo-Young et al., 1987). The Rushton disc turbine agitators were replaced with axial flow "Prochem" hydrofoil-type impellers (Chisti and MooYoung, 1991) of equal diameter. A draft tube ~as added to enhance axial flow of the highly viscous, non-Newtonian fermentation broth (Moo-Young and Chisti, 1988) and hence the bulk mixing in the fermenter was improved. Air was sparged in the annulus between the draft tube and the walls of the fermenter. This airlift-stirred tank hybrid bioreactor proved superior to either a basic airlift or a stirred tank. :Mthough the airlift configuration without mechanical agitation could be used (Moo-Young et al., 1987), the bulk mixing of the broth was not as effective as in the hybrid dc~ice. Unhke many ~iscous fermentations which have been successfully performed in airlift bioreactors (Chisti, 1989), N. sitophila broths containing cellulosics particulates are more ', iscous and pseudoplastic. For example, C. cellulog.'licum broths are rheologicall.v easier to handle than those of N. sitophila (Moo-Young et al.. 1987). The bioreactor scale-up considerations such as gas-liquid mass transfer and gas holdup effects in those broths have been reported on previously (Chisti and Moo-Young, 1988: Moo-Young et al., 1987). Other general aspects of fermentation plant design which apply in different degrees to food and feed plants ha~,e been detailed elsewhere (Chisti, 1992: Chisti, 1992a: Chistl and Moo-Young, 1991 ).

Process Economics The economics of fungal protein manufacture were discussed in detail by Moo-Young et al. ( 1979: 1986) for a production processes based on C. celhdo~.'ticum. Those economic analyses are equally valid for the N. sitophila based process because of the similarities between the t~o production schemes: same cellulosic substrates, media supplements, and pretreatment operations: identical downstream processing of the product and similar growth and protein production characteristics of the two fungi. For conversion of Kraft paper pulpmill clarifier sludge (95 % cellulose) to mycoprotein, Moo-Young et al. (1986) determined that a minimum processing capacity of 6.5 tonne per day of sludge was required to break-even. A 96 % conversion of the cellulose to a product containing 38 % protein was assumed with so)meal-based protein being the reference selling price. The major contributors to production cost were the utilities (at 37.2 % of total cost), the nutrients (at 36.4 c~ of total cost) and the equipment depreciation /at 18.1 c~, of total cost).

M.b.lL~3-', OUNG ,'t al.

47,.,

In speciality, toods industr~ the production cost cons~derauons are of lesser map,)rtance thau in leeds manufacture. For example, the Q u o r n m s c o p r o t e m is commerc~all.,, produced using hydrolysed starch - a m,~re expen,,ixe .,,ub,~trute than celluloaic re~.~due - toe fungal culti',au~m t Steinkraus, 198bl. Hence, the ,'\., fft,',l#tda protein proce,,,; is expected ~o be ec~momicall,. ,. iable in spemalit} foods m a r k e t , , e.g.. Mr vegetariun,~,

Conclusion T h e food-grade fungus ,\'euro~pr,ra fft~phila

~a~

cultured on ~ariou.,, solid celluMslc sub.,,trate,,

(v~ood cellulose, :,ugarcane bagasse, c~rn sto~erl f~,r mycoprotein production. The optml,.I conditions for cellulose utihzati~m and protein prc, duction .,tere found t,~ he 35-37 :C t e m p e r a t u r e , pH 5.5 and agitator tip speed n,,t exceeding 2.35 nl.,,-I m a -',5 k , t i r r e d lermenter.

Lip to c,.

tank

9~1 c~. utdizauon cff cellulo,,e could be achle,.ed. T h e cellulol,.uc

perforntance of ,V. silophila c~mlpared latourabl~ ~[th that ot (lu,'temffum c~,lhtloh'ticum ~hich is well knov. n for m, ability, to degrade cellulo,,.e. The nlycoprotein production process ~.a,, pro'.en to 1,311tl L tqlot ,,tale. The pr,~ce',~, is an etfecte, e Illeth~d for Fro,tern-enrichment ol cellulo~,~c ,,ub~,tance,, lot lo~d and fi,dder u,.e.

References Anderson, C., Longton, J., Maddix, C., Scammel], G. W. and Solomons, G. L. (Iq75). The gro~th of mlcrofungi on carbohydrates. In Sin,qle-Cell Protein 11, ¢Tannenbaum. S. R. and Wang, D. [. C., editorsl, The M | T Pre~, Cambridge. MA, pp. 314-329.

Baldsin, C. V. and Moo-Young. M. { 10911, Disruption of a tllamentou,, lungal organism (,\".

sitophila) using a bead ntfll of nmcl design II. Increased recove D ot celiuk~ses. BiowclmoL Tectmique~'. 5. 337-34'. Chisti. Y. { 1¢~8q),Airlift Bioreactor~, Elsevier, Nex~ York, pp. 355. Chisti, Y. (Iqq2), Build better industrial bioreactor~, Chem. Ettg. Progress, 88t I ). 55-5,'4. Chisti, Y. { 1'492a), A~:,ure bloreactor stenhty. Chem. Eng. Pn@ress, 88(91. ,%t)-N5. Chisti, Y. and l%hm-Young, M. (lO&"g), Ga,,, holdup beha~ iour in f e r m e n t a t i o n broths and other n o n - n e w m m a n I]uids in p n e u m a n c a l l y a g m u e d reactors. Chem. Eng. J.. 39, B31-B3r,. Chisti, Y. and Moo-Young, M. (1989), O n the calculauon of s h e a r rate a n d a p p a r e n t ~ isc,~sit~ in airlift and bubble c o l u m n bloreactor~. Bioteclmol. Bioeng., 34, 1301-1302. Chisti, Y. and M~m-Young, M. (1901), F e r m e n t a u o n technolttg,), bioprocessing, ~cale-up and m a n u f a c t u r e . In Bi,technol~o': The Science and the Buffness, (Moses, V. and Cape. R. E., edttors), Hardwood A c a d e m i c Publishers, Ne~ York, pp. 107-200.

FERMENTATION TO ~'COPRO]T_IN FOODS

4TM

Hesseltine, C. W. and Wang, H. L. (1967). Traditional fermented foods. Biotechnol. Biewng.. 9. 275-288. Lang, C. A. (1958), Simple microdetermination of Kjeldahl nitrogen in biological materials. Anal

Chem.. 30, 1692-1694. Moo-Young, M., Burrell, R. E. and Michaelides, J. (1990), Process for upgrading cereal milling by-products into protein-rich food products. US Patent 4,q38,q72. Moo-Young, M. and Chisti, Y. (1988), Considerations for desigmng bioreactors for shearsensitive culture. BiotechnoloD', 6(11 ), 12ql- 12q6. Moo-Young, M., Chisti, Y. and ~qaeh, D. (1992). Fermentative conversion of cellulosic substrates to microbial protein b,,' A.'eurospora sitophihz. Biotechnol. Lett.. 14, 863-868. Moo-Young, M., Halard, B., Allen, D. G., Burrell, R. and Kawase. Y. (1987), Ox3gen transfer to m.vcelial fermentation broths in an airlift fermentor. Bioteehnol. Bioeng., 30, 746-753. Moo-Young, M., Lamptey, J. and Girard, P. (1980), Bioconverslon of cellulosic ssaste into protein and fuel products: A case studs' of the technoeconomic potential. In Perspectives in Biotechnolc,~. and Applied MicrobioloD', fMoo-Young. M. and Alani. D. I., editors), Elsevier. London, pp. 183-201. Moo-Young, M., Moreira, A. R. and Daugulis, A. J. (1979), Economics of termentation processes for SCP production from agricultural wastes. Cana~L J. Chem. Eng., 57. 741-749.

Oguntimein, G., "vlach, D. and Moo-Young. M. (1992), Production of cellulolytic enz3mes bs Neurospora sitophila grown on cellulosic matermls. Biore~ource Technolo~', 39, 277-283. Sinskey, A. J. and Tannenbaum, S. R. (1075), Removal of nucleic acids in SCP. In Sbzgle-Cell Protebt H, (Tannenbaum, S, R. and Wang, D. I, C.. edm~rs), The MIT Press. Cambridge, MA. pp. 158-178. Solomons, G. L. (1975), Submerged culture production of mycelial biomass. In The Filamen/otL~ Fungi, vol. 1, (Smith, J. E. and BerD', D. R., editors), Ed~ard Arnold, London. pp. 249-264. Steinkraus, K. H. (1986), Microbial biomass protein grown on edible substrates: The indigenous fermented foods. In Microbial Biomass Protehtr., (Moo-Young, M. and Gregory, K. F., editors). Elsevier, London, pp, 33-45. Ujeova, E., Fencl, Z. Musilkova, M. and Seichert, L. (1980). Dependence of release ot nucleotides from fungi on fermentor turbine speed. BiotechnoL Bioeng., 22. 237-241. Updegraff, D. M. (1969), Semimicro determination of cellulose in biological materials. ,4haL

Chem., 32, 420-424. Wood, J. B. and Yong, F. M. ( 1975 ). Oriental food fermentations. In The Filamentotl~ Fungi, vol. 1, (Smith, J. E. and Berry, D. R., editors), Edward Arnold, London. pp. 265-280.