4 microbial lipids

8/3/2019 4 Microbial Lipids

http://slidepdf.com/reader/full/4-microbial-lipids 1/10

4 Microbial Lipids

COLIN RATLEDGEHull, United Kingdom

1 Introduction 1351.1 Lipid Nomenclature and Major Lipid Types 138

2 Accumulation of Lipid 1392.1 Patterns of Accumulation 1392.2 Efficiency of Accumulation 142

2.3 Biochemistry of Accumulation 1433 Triacylglycerols and Fatty Acids 146

3.1 Bacteria 1473.1.1 Polyunsaturated Fatty Acids in Bacteria 147

3.2 Yeasts 1483.2.1 Production of a Cocoa Butter Equivalent Yeast Fat 148

3.2.1.1 Direct Feeding of Stearic Acid 1503.2.1.2 Inhibition of Stearoyl Desaturase 1513.2.1.3 Mutation 1523.2.1.4 Metabolic Manipulation 1543.2.1.5 Conclusions 154

3.3 Molds 155

3.3.1 g -Linolenic Acid (GLA, 18:3 v -6) 1593.3.2 Dihomo-g -Linolenic Acid (DHGLA, 20:3 v -6) 1603.3.3 Arachidonic Acid (ARA, 20: 4 v -6) 1613.3.4 Eicosapentaenoic Acid (EPA, 20 :5 v -3) 1623.3.5 Docosahexaenoic Acid (DHA, 22: 6 v -3) 1623.3.6 Eicosatrienoic Acid (ETA, 20 :3 v -9, “Mead Acid”) 1633.3.7 Conclusions 163

3.4 Algae 1643.4.1 g -Linolenic Acid (GLA, 18:3 v -6) 1673.4.2 Arachidonic Acid (ARA, 20: 4 v -6) 1673.4.3 Eicosapentaenoic Acid (EPA, 20 :5v -3) 1683.4.4 Docosahexaenoic Acid (DHA, 22: 6 v -3) 169

3.4.5 Conclusions 170



134 4 Microbial Lipids

4 Sterols, Carotenoids, and Polyprenes 1704.1 Sterols 1704.2 Carotenoids 1724.3 Polyprenoids 175

5 Wax Esters and Polyesters 1765.1 Wax Esters 1765.2 Polyesters – Poly-b -Hydroxyalkanoates 177

6 Other Lipids 1806.1 Biosurfactants 1806.2 Ether (Archaebacterial) Lipids 1816.3 Phospholipids and Sphingolipids 1836.4 Prostanoid-Type Lipids 184

7 Conclusions 1858 References 186



1 Introduction 135

1 Introduction

Since the publication of the 1st Edition of “Biotechnology” and the earlier chapter onthe biotechnology of lipids in 1986, a consid-erable number of developments have takenplace in this field. Some microbial lipid prod-ucts have now been produced commerciallyand prospects for other developments appearto be not too far away. In some cases, as e.g.with the bacterial lipid poly-b -hydroxybuty-rate, no counterpart exists from plant or ani-mal sources and consequently the economics

of producing this product lie outside the nor-mal oils and fats domain. With most other mi-crobial lipids, these are the equivalent in com-position to plant-derived oils and consequent-ly must compete against these in any potentialmarket place. Only the highest valued oilshave any chance of being produced by bio-technological means as it is impossible for mi-croorganisms to produce oils and fats ascheaply as the main commodity oils are pro-duced from plant and animal sources. Howev-er, there is always the possibility of producing

a microbial oil as an adjunct to some wastetreatment process in a way similar to that of-ten used to produce microbial proteins (SCP– single cell protein) for animal feed fromsome unwanted substrate. Microbial oils –which could then be referred to as single celloils (SCO) – would have the double advan-tage over SCP in that they could probably sellfor a higher price than SCP and, moreover,could be used for a technical purpose shouldthe nature of the substrate prevent the prod-uct being returned into the food chain.

The major commercial plant oils continueto be dominated by soybean oil (current 1993production is about 187106 t); palm oil,though, continues to be the fastest growingmarket with 147106 t now being producedcompared to 67106 t in 1983. If the presentrate of expansion in palm oil production con-tinues in Malaysia and Indonesia (BASIRONand IBRAHIM, 1994; LEONARD, 1994), thenpalm oil will overtake soybean oil productionby the end of this decade. Rapeseed oil (now97106 t in 1993) is also expanding mainly dueto increased cultivation in Europe and Cana-

da. The variety now under cultivation is the

low- (or zero-) erucic acid (20:1) oil which isthen a permitted oil for food manufacture.

Overall production of plant and animal oilsis increasing at about 3% per annum; produc-tion in 1992/93 was about 857106 t and is ex-pected to reach 1057106 t by the year 2000(MIELKE, 1992). Pricing of these materials re-mains highly competitive as most products us-ing oils can switch between the various typesaccording to the price of the day. The averageprice index for the major commodity oils isabout US$ 500–550 per t though groundnutoil, e.g., is always significantly higher than theaverage at $ 800–850 per t. The highest priced

commodity oil, excluding the speciality mate-rials, is always olive oil at $ 1,500–2,000 per t.Its price depends on its quality which includesminor, but very important, flavor compo-nents. Animal fats (tallow and lard) havesteadily declined in consumption over thepast decade and are likely to fall even furtherto about 20% of the total market by 2001(SHUKLA, 1994). Their prices are thereforeusually at or below the average index level.

The trends in world oil and fats suppliesare under constant surveillance and are fre-

quently reviewed in various publications: theextensive reviews by SHUKLA (1994) andMIELKE (1992) can be recommended thoughfor current information journals such as LipidTechnology (P. T. Barnes & Associates), Oilsand Fats International (Chase Webb, St IvesPLC), INFORM (American Oil Chemists’Society, Illinois) provide invaluable and con-tinuously up-dated information in most areas.There are, in addition, a number of special-ized trade reviews that provide weekly pricesof the traded oils.

The fatty acid composition of the majorcommercial oils is given in Tab. 1. The nom-enclature of lipids is given in Sect. 1.1. It willbe appreciated that, unlike say animal feedprotein, the composition of the fats variesconsiderably from species to species. In allcases, however, the oil or fat is composed al-most entirely (`98%) of triacylglycerols(formerly known as triglycerides) – Sect. 1.1.For edible purposes, the oil or fat is retainedin this form though the individual fatty acylgroups on the glycerol can be modified –usually by chemical means – without affecting

the triacylglycerol structure per se. Some en-




T

a b . 1 .

F a t t y A c i d C o m p o s i t i o n o f F

a t s a n d O i l s o f A n i m a l a n d P l a n t O r i g i n

F

a t s / O i l s

R e l a t i v e P r o p o r t i o n o f F a t t y A c y l G r o u p s [ % ( w / w ) ]

4 : 0 – 1 0 : 0

1 2 : 0

1 4 : 0

1 6 : 0

1 6 :

1

1 8 : 0

1 8 : 1

1 8 : 2

1 8 : 3

2 0 : 0

2 0 : 1

O t h e r s

A

n i m a l F a t s

B

u t t e r f a t

B

e e f t a l l o w

L

a r d

1 0

P P

3 P P

1 1 3 2

2 7

2 4

2 6

2 4 3

1 2

1 9

1 4

2 9

4 3

4 4

2 3 1 0

P 1

P

P P P

P P P

1 5 : 0 c 1 7 : 0 ,

1 5 : 0 c 1 7 : 0 ,

1 4 : 1 c 1 7 : 1 ,

3 % 2 %

;

2 %

P

l a n t O i l s

C

o c o n u t o i l

P

a l m

k e r n e l o i l

C

o c o a b u t t e r

O

l i v e o i l

R

a p e s e e d o i l

G

r o u n d n u t o i l a

S u n f l o w e r o i l

S o y b e a n o i l

C

o r n o i l

C

o t t o n s e e d o i l

1 5 8

P P P P P P P P

4 7

4 8

P P P P P P P P

1 8

1 6

P P P P P P P 1

9 8 2 6

1 3 4

1 1 7

1 1

1 1

2 2

PPP 1 P P P P P 1

3 3 3 5 3 2 2 5 4 2 3

6 1 5

3 5

7 1

6 2

4 8

1 9

2 4

2 8

1 9

2 2 3 1 0

2 2

3 2

6 8

5 4

5 8

5 4

PPP 1

1 0

P 1 7 1 1

PP 1 1 P 1 P P P P

P P P P P 2 P P P P

P P P P P 2 2 : 0 c 2 4 : 0 ,

PPPP

5 %

“ E x o t i c ” P l a n t O i l s

B

o r a g e s e e d o i l

E

v e n i n g p r i m r o s e s e e d o i l

B

l a c k c u r r a n t s e e d o i l

P P P

P P P

P P P

1 1 8 6

P P P

4 2 1

1 6 9

1 0

3 9

7 0

4 8

2 2 b

9 b

1 7 b

P P P

4 . 5

PP

2 2 : 1 ,

2 4 : 0 ,

a - 1 8 : 3 ,

2 .

5 %

1 .

5 %

1 3 %

a

A l s o k n o w n a s p e a n u t o i l .

b

g - L i n o l e n i c a c i d ,

1 8 : 3

( 6 ,

9 ,

1 2 ) .



1 Introduction 137

zymatic reformulation of triacylglycerols oc-curs on an industrial scale using stereospecificlipases to transesterify palm oil fractions intothe much more expensive cocoa butter-liketriacylglycerols (OWUSU-ANSAH, 1993).

With some technical applications of oils, itis the fatty acid that is required; consequentlysaponification (hydrolysis) of the triacylglyc-erol is carried out and the fatty acid usedeither as such, e.g., with soap manufacture, oris modified to an appropriate derivativewhich is then used in a multitude of products:from detergents to adhesives.

The aim of all biotechnological processes is

to produce products that are either cheaperthan can be obtained from other sources, in-cluding possible chemical synthesis, or are notavailable by any other means. Within the fieldof lipids, the opportunities to produce triacyl-glycerol lipids are limited to the highest val-ued materials. The highest priced bulk (com-modity) oil is cocoa butter whose price hasvaried between $ 8,000 to $ 3,000 per t overthe past decade. At the higher price level, theprospects of producing a cocoa butter equi-valent oil by yeast technology have looked fa-

vorable. This topic is specifically reviewed lat-er (see Sect. 3.2.1).Other very high valued oils are those in the

health care market and which have had var-ious claims made on their behalf for the ame-lioration of various diseases and conditions.Of current interest are oils containing the po-lyunsaturated fatty acids: g -linolenic acid,18:3 (v -6); arachidonic acid, 20:4 (v -6); eico-sapentaenoic acid, 20:5 (v -3); and docosa-hexaenoic acid, 22:6 (v -3). Oils containingsuch fatty acids are found in a number of mi-croorganisms and are reviewed in Sects. 3.3and 3.4.

The very highest priced lipids though areprobably the prostanoid compounds encom-passing the prostaglandins, leukotrienes, andthromboxanes. These are mainly used fortreatment of uncommon disorders or for ex-perimental purposes. Consequently, theamounts required per annum are probably atthe kilogram stage rather than the ton (orkiloton) stage with other lipid products. Pros-pects for producing such materials are brieflymentioned in Sect. 6.4.

Thus, if we view microorganisms as a po-

tential source of the widest types of lipidsthen it is possible to identify a number of po-tentially attractive products. For the purposesof this article, I have therefore used the broaddefinition of a lipid as any material that is de-rived from a (micro)organism, is directly sol-uble in organic solvents, and is essentially awater-insoluble material. However, as there isstill considerable interest in the manner inwhich microorganisms synthesize large quan-tities of lipids, much of the review will be tak-en up with the more conventional types of oils and fats that they produce. The entiresubject of microbial lipids, encompassing all

aspects and not just biotechnology, has beenthe above subject of a two-volume mono-graph by RATLEDGE and WILKINSON (1988a,1989). The industrial applications of microbiallipids have also been the subject of a mono-graph edited by KYLE and RATLEDGE(1992). Details concerning the degradation of fats, oils, and fatty acids, including the actionof lipases and phospholipases, which are notcovered here have been recently reviewedelsewhere by the author (RATLEDGE, 1993).

It will be appreciated, of course, that al-

though microorganisms remain a potentialsource of oils and fats, there is considerableeffort being put into the production of oilsand fats from conventional plant sources.Such efforts include the modification of pea-nut oils (groundnut oil) to produce changes inthe fatty acid composition so that the moredesirable oils can be produced more cheaply.The application of genetic engineering is nowgathering pace as a means of producing “tail-or-made” oils and fats in plants and is likelyto supersede the traditional plant breedingapproach as a means of creating what iswanted more quickly and with greater cer-tainty. This review, however, will not includeany detailed review of the current develop-ments in plant genetic engineering as appliedto the commodity oils and fats. Readersshould though be aware that such advancesare now likely to be a major influence in theavailability of “improved” oils for everydayuse and will undoubtedly ensure that thesematerials remain highly competitively pricedfor many years to come. The recent reviewsby HARWOOD (1994a, b), MURPHY (1994a)

and RATTRAY (1994) and the monographs




Tab. 2. Developments in Genetic Engineering (by in vitro Mutagenesis andGene Transfer) Being Applied for the Modification of Plant Oils (adapted

from RATTRAY, 1994)

Plant Target FattyAcid

Objective Application

Soybean andrapeseed

16:016:018:0

18:1a-18:3

increasedecreaseincrease

increasedecrease

margarineedible oilmargarine; cocoa buttersubstitute(?)improved edible oilimproved stability andodor

Rapeseed 22:1 increase erucic acid for

oleochemicalsSunflower 18:1

18:2c18:3increasedecrease 6 olive oil

substitute

Linseed 18:3 increase oleochemicals

Groundnut 18:1 increase improved edible oil

edited by RATTRAY (1991) and by MURPHY

(1994b) will be found particularly useful inthis respect. Tab. 2 summarizes some of the

current developments that are now takingplace in this area.The opportunities for microorganisms to

produce oils and fats of commercial value forthe bulk markets remain doubtful but wherethe product cannot be obtained from else-where then this provides a much better op-portunity for a microbial oil than attemptingto replicate what is already available fromplant sources. Some opportunities neverthe-less do exist but they have to be identifiedwith some care. Hopefully, some of the fol-lowing material may indicate to the astutereader where such opportunities may lie.

1.1 Lipid Nomenclature and MajorLipid Types

Fatty acids are long chain aliphatic acids(alkanoic acids) varying in chain length from,normally, C12 to C22 though both longer andshorter chain-length acids are known. In mostcells (microbial, plant, and animal), the pre-

dominant chain lengths are 16 and 18. Fatty

acids may be saturated or unsaturated withone or more double bonds which are usuallyin the cis (or Z ) form. The structure of a fatty

acid is represented by a simple notation sys-temP X : Y , where X is the total number of Catoms and Y is the number of double bonds.Thus, 18:0 is octadecanoic acid, that is stearicacid; 16:1 is hexadecenoic acid, that is palmit-oleic acid, with one double bond, and 18:3would represent octadecatrienoic acid, a C18

acid with three double bonds. The position of the double bond(s) is indicated by designat-ing the number of the C atom, starting fromthe COOH terminus, from which the doublebond starts: oleic acid is thus 18:1 (9) signify-ing the bond is from the 9th (to the 10th) Catom. If it is necessary to specify the isomer,this is added as “c” (for “cis”pZ ) or “t ” for“trans”pE). In this review, the cis/trans sys-tem is used. Thus, cis, cis-linoleic acid is 18:2(c 9, c 12). Most naturally-occurring unsatu-rated fatty acids are in the cis configurationand, unless it is stated otherwise, this configu-ration may be assumed.

With polyunsaturated fatty acids (PUFA),the double bonds are normally methylene-in-terrupted: UCHuCHUCH2UCHuCHU.Thus, once the position of one bond is speci-

fied all the others are also indicated. In num-



2 Accumulation of Lipid 139

bering PUFAs, a “reverse” system is usedwhere the position of only the last doublebond is given by the number of carbon atomsit is from the CH3 terminus. To denote the“counting back” system is in operation thenotation is given as v -x or n-x; thus the twomain isomers of linolenic acid (18:3) are giv-en as 18:3 (v -3) and 18:3 (v -6) or as 18:3(n-3) and 18:3 (n-6). In some systems, theminus sign may be omitted giving, for exam-ple, 18 :3 (v 3) or 18: 3 (n3) and 18:3 (v 6) and18:3 (n6). Respectively, these two isomersare:

18

CH3U

17

CH2U

16

CHu15

CHU14

CH2U

13

CHu12

CHU11

CH2U

10

CHu9

CHF

HOOCUH2

2

CUH2

3

CUH2

4

CUH2

5

CUH2

6

CUH2

7

CUH2

8

C

CH3U

v n

CH2U

v -1n-1

CH2U

v -2n-2

CH2U

v -3n-3

CH2U

v -4n-4

CHuv -5n-5

CHUv -6n-6

CH2UCHuCHFF

HOOCUH2CUH2CUH2CUH2CUHCuHCUH2C

The v -x system will be used in this review.When fatty acids are esterified to glycerol,

they give a series of esters: mono-, di-, andtriacylglycerols (I, II, III). This is the pre-ferred nomenclature to the older mono-, di-,and triglycerides.

CH2UOUCOURFCHUOHFCH2UOH

I

CH2UOUCORFCHUOUCORFCH2UOH

II

CH2UOUCORFCHUOUCORFCH2UOUCOR

III

where R is a long alkyl chain and RCOU is,therefore, the fatty acyl group.

As various isomeric forms are possible, theposition of attached acyl group must be speci-fied in most cases. For this, the stereospecificnumbering ( sn-) system is used so that thetwo prochiral positions of glycerol (IV) canbe distinguished as sn-1 and sn-3.

1 CH2OHF

2 CHOHF

3 CH2OH

IV

With respect to the monoacylglycerols,there are obviously three possible isomersand similarly for the diacylglycerols.

Where different acyl groups are attached tothe glycerol moiety, these can then be individ-ually given. For example, 1-stearoyl-2-oleoyl-3-palmitoyl- sn-glycerol is the major triacyl-glycerol of cocoa butter with stearic, oleic,and palmitic acids on the three OH posi-tions.

Phospholipids possess two fatty acyl groupsat the sn-1 and sn-2 positions of glycerol witha phospho group at sn-3 which is also linked

to a polar head group: choline, serine, etha-nolamine, and inositol are the common ones.For the nomenclature and naming of phos-pholipids and other microbial lipids, the mul-tiauthored treatise Microbial Lipids, editedby RATLEDGE and WILKINSON (1988a, 1989),

may be helpful though there are numeroustext books on lipids that provide similar infor-mation.

2 Accumulation of Lipid

2.1 Patterns of Accumulation

Not all microorganisms can be consideredas abundant sources of oils and fats, though,like all living cells, microorganisms always

contain lipids for the essential functioning of




Fig. 1. Typical lipid accumulation pattern for ayeast (Rhodotorula glutinispR. gracilis) growingon a high C:N ratio medium in batch culture. Bio-mass L, % lipid content l, NH4

c in medium [

(from YOON et al., 1982).

membranes and membranous structures.Those microorganisms that do produce a highcontent of lipid may be termed “oleaginous”in parallel with the designation given to oil-bearing plant seeds. Of the some 600 differentyeast species, only 25 or so are able to accu-mulate more than 20% lipid; of the 60,000fungal species fewer than 50 accumulate morethan 25% lipid (RATLEDGE, 1989a).

The lipid which accumulates in oleaginousmicroorganisms is mainly triacylglycerol (seeSect. 1.1). If lipids other than this type are re-quired then considerations other than thoseexpressed here might have to be taken into

account to optimize their production. Withfew exceptions, oleaginous microorganismsare eukaryotes and thus representative spe-cies include algae, yeasts, and molds. Bacteriado not usually accumulate significantamounts of triacylglycerol but many do accu-mulate waxes and polyesters (see Sects. 5.1and 5.2) which are now of commercial inter-est.

The process of lipid accumulation in yeastsand molds growing in batch culture was eluci-dated in the 1930s and 1940s (see WOOD-BINE

, 1959, for a review of the early litera-ture, and RATLEDGE, 1982, for an updatedreview of these aspects). A typical growthpattern is shown in Fig. 1. This pattern is alsofound with the accumulation of polyester ma-terial in bacteria (see Sect. 5).

The key to lipid accumulation lies in allow-ing the amount of nitrogen supplied to the

Fig. 2. Electron micrograph of Cryptococcus curva-tus (pCandida curvatap Apiotrichum curvatum)strain D grown for 2 days on nitrogen-limiting me-dium (viz. Fig. 1) showing presence of multiple lip-id droplets. Total lipid content approx. 40%, mark-er bar: 1 mm (from HOLDSWORTH et al., 1988).

culture to become exhausted within about24–48 h. Exhaustion of nutrients other thannitrogen can also lead to the onset of lipid ac-cumulation (see GRANGER et al., 1993, for arecent reference) but, in practice, cell proli-feration is most easily effected by using a lim-iting amount of N (usually NH4

c or urea) inthe medium. The excess carbon which isavailable to the culture after N exhaustioncontinues to be assimilated by the cells and,by virtue of the oleaginous organism possess-ing the requisite enzymes (see below), is con-verted directly into lipid.

The essential mechanism which operates isthat the organism is unable to synthesize es-sential cell materials – protein, nucleic acids,etc. – because of nutrient deprivation andthus cannot continue to produce new cells.Because of the continued uptake of carbonand its conversion to lipid, the cells can thenbe seen to become engorged with lipid drop-lets (Fig. 2). It is important to appreciate,however, that the specific rate of lipid biosyn-thesis does not increase; the cells fatten be-cause other processes slow down or cease al-

together and, as lipid biosynthesis is not



2 Accumulation of Lipid 141

Fig. 3. Typical lipid accumulation pattern for ayeast (Rhodotorula glutinis) growing on nitrogenlimiting medium in continuous culture. Biomass L,

% lipid content [ (from YOON and RHEE, 1983).

linked to growth, this may continue unabated.The process of lipid accumulation (Fig. 1) canbe seen as a two-phase batch system: the firstphase consists of balanced growth with all nu-trients being available; the subsequent “fat-tening” or “lipogenic” stage occurs after theexhaustion of a key nutrient other than car-bon and, of course O2. The role of O2 duringlipid formation was discussed briefly in the1st Edition of “Biotechnology” (RATLEDGE,1986).

Accumulation of lipid has also beenachieved in single stage continuous culture(RATLEDGE et al., 1984) and a typical accu-

mulation profile dependent upon the dilutionrate (growth rate) is shown in Fig. 3. As withbatch cultivation, the medium has to be for-mulated with a high carbon-to-nitrogen ratio,usually about 50:1. The culture must begrown at a rate which is about 25–30% of themaximum. Under this condition, the concen-tration of nitrogen in the medium is virtuallynil and the organism then has sufficient resi-dence time within the chemostat to assimilatethe excess carbon and convert it into lipid.The rate of lipid production (i.e., g LP1 hP1)

is usually faster in continuous cultures than inbatch ones (EVANS and RATLEDGE, 1983;FLOETENMEYER et al., 1985).

The exact ratio of C to N chosen for themedium was originally considered to be of lit-tle consequence provided N was the limitingnutrient and sufficient carbon remained toensure good lipid accumulation. However,YKEMA et al. (1986) showed that a range of lipid yields in an oleaginous yeast, Apiotri-chum curvatum (originally Candida curvatabut now Cryptococcus curvatus; see BARNETTet al., 1990) were traversed in continuous cul-ture by varying the C:N ratio of the growthmedium. There was a hyperbolic relationshipbetween the C:N ratio and the maximumgrowth (dilution) rate that the organism could

attain: the lowest growth rate was at the high-est C:N ratio of 50:1 and this, in turn, con-trolled the amount of lipid produced and theefficiency of yield (g lipid per g glucose used)with which it was produced. Although thehighest lipid contents of the cell (50% w/w)were obtained with a C:N ratio of 50:1 orover, the optimum ratio for maximum pro-ductivity (g LP1 hP1 lipid) was at a ratio of 25:1 with glucose (YKEMA et al., 1986) and at30–35:1 when whey permeates were usedwith same yeast (YKEMA et al., 1988). Similar

results for describing the optimum C:N ratiofor lipid accumulation have been developedby GRANGER et al. (1993) using Rhodotorula

glutinis.Interestingly, YKEMA et al. (1986) com-

mented that Apiotrichum curvatum simulta-neously accumulated about 20% carbohy-drate in the cells along with the 50% lipid.Such a phenomenon of carbohydrate forma-tion had been conjectured by BOULTON andRATLEDGE (1983a) to be a likely event to ac-count for an observed delay in lipid synthesisafter glucose assimilation had been initiated.This carbohydrate was also recognized inde-pendently by HOLDSWORTH et al. (1988) inthe same yeast and was considered to be gly-cogen. As YKEMA et al. (1986) pointed out, if the biosynthesis of the polysaccharide which,like lipid, is a reserve storage material, couldbe prevented then this would enhance the to-tal amount of lipid producible with a cell.

Although most studies on microbial lipidaccumulation have been conducted usingbatch cultivation and, for accuracy, in contin-uous culture, other growth systems have also

been explored. In particular, fed-batch cul-




ture has proved effective in increasing boththe cell density and lipid contents of oleagi-nous yeasts: YAMAUCHI et al. (1983) usedethanol as substrate with Lipomyces starkeyiand achieved a biomass density of 150 g LP1

with a lipid content of 54%. Similarly, PANand RHEE (1986) achieved 185 g (dry wt.) of Rhodotorula glutinis per liter with a lipid con-tent of 43% using glucose as the fed-batchsubstrate. In this latter case, O2-enriched air(40% O2 c 60% air) had to be used to sus-tain the cells. At the density recorded, thepacked cell volume was 75% of the total vol-ume of the fermentation medium. Without

using additional O2, it seems likely that celldensities of up to 100 g LP1 could beachieved with most oleaginous yeasts (see,e.g., YKEMA et al., 1988) though filamentousmolds may pose other problems. Economicconsiderations, however, would probably beagainst the use of O2-enriched air for anycommercial process. Interestingly, it is sug-gested that higher rates of lipid formationmay occur with fed-batch techniques thanwith batch- or continuous-culture approaches(YKEMA et al., 1988).

At the end of the lipid accumulation phase(see Fig. 1), it is essential that the cells arepromptly harvested and processed. If glucose,or other substrate, has become exhausted onthe end of the fermentation, then the organ-ism will begin to utilize the lipid as the role of the accumulated material is to act as a reservestore of carbon, energy, and possibly evenwater. HOLDSWORTH and RATLEDGE (1988)showed with a number of oleaginous yeaststhat after carbon exhaustion following lipidaccumulation, the lipid began to be utilizedwithin 1.5 h thus indicating the dynamic stateof storage lipids in these organisms.

2.2 Efficiency of Accumulation

The efficacy of conversion of substrate tolipid has been examined in some detail inboth batch and continuous culture. In gener-al, the latter technique offers the bettermeans of attaining maximum conversions asthe cells are operating under steady state con-ditions and carbon is not used with different

efficiencies at each stage of the growth cycle.

Conversions of glucose and other carbohy-drates including lactose and starch, to lipid upto 22% (w/w) have been recorded with a vari-ety of yeasts (RATLEDGE, 1982; YKEMA etal., 1988; DAVIES and HOLDSWORTH, 1992;HASSAN et al., 1993) which compares favora-bly with the theoretical maximum of about31–33% (RATLEDGE, 1988). Somewhat loweryields appear to pertain with molds (WOOD-BINE, 1959; WEETE, 1980). The reason forthis difference is not obvious though it maybe due to a somewhat slower growth rate of molds than yeasts. It should be said, however,that there has not been the same amount of

detailed work carried out with molds as withyeasts. Claims that microorganisms haveachieved higher conversions of glucose orother sugars to lipid should be treated withcaution: either there will be found to be addi-tional carbon within the medium and not tak-en into the mass balance or, as may occasion-ally happen, the “lipid” has been improperlyextracted and may contain non-lipid material.However, if experimental data are calculatedso that the yield of lipid or fatty acids can bebased on the fraction of glucose being used

solely for lipid biosynthesis, then values closeto the theoretical value have been attained inpractice (GRANGER et al., 1993).

When ethanol is used as substrate, the the-oretical yield of lipid is 54% (w/w) (RAT-LEDGE, 1988). Though only a 21% conversionof ethanol to lipid was recorded by YAMAU-CHI et al. (1983) in the fed-batch culture of Lipomyces starkeyi, higher conversions wererecorded by EROSHIN and KRYLOVA (1983)also using a fed-batch system for the cultiva-tion of yeasts on ethanol: conversions of 26%,27%, and 31% were obtained using, respec-tively, Zygolipomyces lactosus (Lipomyces te-trasporus) and two strains of Cryptococcus al-bidus var. aerius. The reason for these veryhigh values lies in the efficiency by whichethanol can be converted to acetyl- CoA, thestarting substrate for lipid biosynthesis (seebelow). With glucose, the maximum yield of acetyl-CoA can only be 2 mol per mol utilizedwhereas with ethanol the yield is 1 mol permol. On a weight-to-weight basis, therefore,ethanol (MW 46) is almost twice as efficientas glucose (MW 180) in providing C2 units.

GRANGER et al. (1993) have recorded direct

4 microbial lipids

Documents