1-s2.0-S0163782709000423-main

Progress in Lipid Research 48 (2009) 375–387

Contents lists available at ScienceDirect

Progress in Lipid Research

journal homepage: www.elsevier .com/locate /p l ipres

Review

Yarrowia lipolytica as a model for bio-oil production

Athanasios Beopoulos a, Julien Cescut b, Ramdane Haddouche a, Jean-Louis Uribelarrea b,Carole Molina-Jouve b, Jean-Marc Nicaud a,*

a Microbiology and Molecular Genetic Laboratory, CNRS UMR2585, INRA UMR1238, AgroParisTech, INRA centre de Versailles-Grignon BP 01, F-78850 Thiverval-Grignon, Franceb Laboratoire d’Ingénierie des Systèmes Biologiques et des Procédés, CNRS UMR5504, INRA UMR792, INSA, 135 Avenue de Rangueil, F-31077 Toulouse, France

a r t i c l e i n f o

Article history:Received 24 July 2009Received in revised form 18 August 2009Accepted 20 August 2009

Keywords:LipidYeastYarrowia lipolyticaTriacylglycerolLipid particleb-oxidationFermentation

0163-7827/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.plipres.2009.08.005

Abbreviations: AA, arachidonic acid; ACAT, acyl-Coadenosine triphosphate; DAG, diacylglycerol; DHAP, dsubstrates; IMP, inosine 50-monophosphate; LB, lipidpolyunsaturated fatty acids; SCO, single-cell oil; SE, s

* Corresponding author.E-mail address: [email protected]

a b s t r a c t

The yeast Yarrowia lipolytica has developed very efficient mechanisms for breaking down and usinghydrophobic substrates. It is considered an oleaginous yeast, based on its ability to accumulate largeamounts of lipids. Completion of the sequencing of the Y. lipolytica genome and the existence of suitabletools for genetic manipulation have made it possible to use the metabolic function of this species for bio-technological applications. In this review, we describe the coordinated pathways of lipid metabolism,storage and mobilization in this yeast, focusing in particular on the roles and regulation of the variousenzymes and organelles involved in these processes. The physiological responses of Y. lipolytica to hydro-phobic substrates include surface-mediated and direct interfacial transport processes, the production ofbiosurfactants, hydrophobization of the cytoplasmic membrane and the formation of protrusions. Wealso discuss culture conditions, including the mode of culture control and the culture medium, as theseconditions can be modified to enhance the accumulation of lipids with a specific composition and to iden-tify links between various biological processes occurring in the cells of this yeast. Examples are presenteddemonstrating the potential use of Y. lipolytica in fatty-acid bioconversion, substrate valorization and sin-gle-cell oil production. Finally, this review also discusses recent progress in our understanding of themetabolic fate of hydrophobic compounds within the cell: their terminal oxidation, further degradationor accumulation in the form of intracellular lipid bodies.

� 2009 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3762. Lipid synthesis and accumulation factors in oleaginous microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376

2.1. Oleaginous yeasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3762.2. Lipid accumulation pathways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3772.3. The de novo lipid synthesis pathway and non-polar lipid synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3772.4. Biochemistry and regulation of lipid accumulation potential in oleaginous yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379

3. Factors affecting lipid accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3804. Modes of culture for ensuring high levels of lipid accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381

4.1. Batch mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3814.2. Continuous mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3824.3. Fed-batch mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382

5. Mastering lipid production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
5.1. The ex novo lipid accumulation pathway. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3835.2. The b-oxidation degradation pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3835.3. Growth conditions and genetic modifications favoring lipid production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
ll rights reserved.

A:cholesterol acyltransferase; ACC, acyl-CoA carboxylase; ACL, ATP citrate lyase; AMP, adenosine monophosphate; ATP,ihydroxyacetone phosphate; FFA, free fatty acids; GLA, c-linolenic acid; G-3-P, glycerol-3-phosphate; HS, hydrophobic

body; MAG, monoacylglycerol; ME, malic enzyme; NADPH, nicotinamide adenine dinucleotide phosphate; PUFA,teryl esters; TAG, triacylglycerols (triglycerides).

(J.-M. Nicaud).

http://dx.doi.org/10.1016/j.plipres.2009.08.005

mailto:[email protected]

http://www.sciencedirect.com/science/journal/01637827

http://www.elsevier.com/locate/plipres

TLiw

376 A. Beopoulos et al. / Progress in Lipid Research 48 (2009) 375–387

6. Potential applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3847. Summary, conclusions and future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386

1. Introduction

The yeast Yarrowia lipolytica is often found in environments richin hydrophobic substrates, such as alkanes or lipids, and has devel-oped sophisticated mechanisms for the efficient use of hydropho-bic substrates (HS) as the sole carbon source [1,2]. One of themost striking features of this yeast is the presence in its genomeof several multigene families involved in these metabolic path-ways. The complexity and multiplicity of these genes enable Y.lipolytica to use and valorize a wide range of hydrophobic sub-strates (HS). Using these mechanisms, this yeast can accumulatelipids to levels exceeding 50% of cell dry weight [3]. Y. lipolyticamay therefore be considered an oleaginous yeast. Lipid accumula-tion is probably enhanced by the many protrusions on the cell sur-face, facilitating HS uptake from the medium [4]. The internalizedaliphatic chains are then broken down to meet needs for growth, oraccumulate in an unchanged or modified form. These lipids formthe storage lipid fraction, which consists mostly of triacylglycerols(triglycerides) (TAG) and steryl esters (SE). In addition to directsubstrate assimilation from the medium, de novo TAG biosynthesisis another energy storage process providing fatty acids for mem-brane phospholipid formation. SE formation and mobilization pro-vide the sterols required for membrane proliferation. Storagemolecules accumulate in a specialized compartment of the cellknown as the lipid body (LB). Yeast lipid bodies consist of a lipidcore encased in a phospholipid monolayer, within which manyproteins with diverse biochemical activities are embedded [5–7].Several of these proteins metabolize lipids and the LB thereforeprobably plays a key role not only in lipid storage, but also in lipidbiosynthesis, metabolism, degradation and substrate trafficking[6]. LB formation and function are tightly linked to the synthesisof TAG and SE. A recently identified lipid-binding protein in Y.lipolytica LB [8,9] has been implicated in lipid trafficking betweenthe cytoplasm and LB, suggesting that free (non-esterified) fattyacids (FFA) probably accumulate in lipid bodies too [4,8,10,11].

A few models have been developed for the study of lipid metab-olism. These models include Saccharomyces cerevisiae, which haslong been used as a genetic model in studies that have greatly im-proved our understanding of lipid metabolism [12]. The enzymesinvolved in TAG biosynthesis, storage and degradation are verysimilar between species, and particularly between yeasts, but S.

able 1pid contents and fatty acid profiles of selected oleaginous yeasts [16]. Lipid contents areight).

Species % Lipid (glip gX�1) Major fatty acid resi

C16:0 C

Cryptococcus curvatus 58 25 TCryptococcus albidus 65 12 1Candida sp. 107 42 44 5Lipomyces starkeyi 63 34 6Rhodotorula glutinis 72 37 1Rhodotorula graminis 36 30 2Rhizopus arrhizus 57 18 0Trichosporon pullulans 65 15 0Yarrowia lipolytica 36 11 6

a T means trace.b 0 means none detected.

cerevisiae is not an oleaginous yeast and accumulates only moder-ate amounts of lipids (less than 15% of its biomass). Furthermore,unlike S. cerevisiae, which produces similar amounts of TAG andSE, Y. lipolytica stores mostly TAGs (>90%). This yeast is also unu-sual in accumulating significant quantities of FFA within the cell.

The unique features of Y. lipolytica, together with the availabil-ity of efficient genetic tools for this species, have stimulated inter-est in the use of this yeast as a model for bio-oil production, withgreat potential for biotechnological applications. Several technolo-gies, including various fermentation configurations, have been al-ready used for single-cell oil (SCO) production by strains of Y.lipolytica grown on various agro-industrial by-products or waste[2,11]. The potential applications of these processes include theproduction of reserve lipids with particular structures (e.g. oils en-riched in essential polyunsaturated fatty acids) and the productionof nonspecific oils for use as renewable starting materials for thesynthesis of bio-fuels. This review aims to provide insight intothe routes of biosynthesis and degradation leading to the forma-tion of oils and an overview of recent advances in the physiologyand genetics of Y. lipolytica relating to the assimilation of HS.

2. Lipid synthesis and accumulation factors in oleaginousmicroorganisms

2.1. Oleaginous yeasts

Few microorganisms are known to accumulate lipids to a signif-icant level. Those species able to do so to a level corresponding tomore than 20% of their biomass are described as oleaginous. Fewerthan 30 of the 600 species of microorganisms investigated in onestudy were found to be oleaginous [13–15]. The best known oleag-inous yeasts include genus of Candida, Cryptococcus, Rhodotorula,Rhizopus, Trichosporon and Yarrowia. On average, these yeasts accu-mulate lipids to a level corresponding to 40% of their biomass.However, in conditions of nutrient limitation, they may accumu-late lipids to levels exceeding 70 % of their biomass (Table 1). Nev-ertheless, lipid content and profile differ between species. Forinstance, Cryptococcus curvatus and Cryptococcus albidus accumu-late lipids to equivalent levels (58% and 65%, respectively), buttheir fatty acid profiles differ significantly. C. curvatus accumulates

e expressed, in terms of mass, as a fraction of dry cell mass (% ½glip g�1X �, weight/dry

dues (relative% w/w)

16:1 C18:0 C18:1 C18:2 C18:3

a 10 57 7 0b

3 73 12 08 31 9 15 51 3 03 47 8 012 36 15 46 22 10 122 57 24 11 28 51 1

A. Beopoulos et al. / Progress in Lipid Research 48 (2009) 375–387 377

large amounts of palmitic acid, whereas oleic acid is the principalfatty acid accumulating in C. albidus. By contrast, in Rhodotorulaspecies, lipid content diverges significantly (Rhodotorula glutinisand R. graminis accumulate lipids at levels corresponding to 72%and 36% of their biomass, respectively), but fatty acid compositionremains similar. Y. lipolytica accumulates lipids to lower levels thansome other oleaginous species, but it is the only yeast known to beable to accumulate such a high proportion of linoleic acid (morethan 50% of the fatty acid residues present – Table 1).

2.2. Lipid accumulation pathways

Lipids may accumulate via two different pathways: (1) de novosynthesis, involving the production, in defined conditions, of fattyacid precursors, such as acetyl and malonyl-CoA and their integra-tion into the storage lipid biosynthetic pathway (the Kennedy path-way, see below) and (2) the ex novo accumulation pathway,involving the uptake of fatty acids, oils and triacylglycerols (TAG)from the culture medium and their accumulation in an unchangedor modified form within the cell. This pathway requires hydrolysisof the hydrophobic substrate (HS), transport of the released fattyacids within the cell, their re-assembly in the TAG and the sterylester (SE) fractions and their accumulation within the LB. The mainenzymes involved in these pathways are summarized in Table 2.

2.3. The de novo lipid synthesis pathway and non-polar lipid synthesis

Non-polar lipid synthesis in yeasts requires a constant supply ofcoenzyme A (CoA)- activated FA for acylation of the glycerol back-bone to synthesize TAG or the esterification of sterols to producesteryl esters (SE). The first two carbon atoms for de novo FA synthe-sis are provided by cytosolic acetyl-CoA, through citrate cleavageby ATP-citrate lyase (ACL) in the TCA cycle. The fatty acid chain

Table 2Genes involved in fatty acid metabolism in Y. lipolytica and S. cerevisiae.a

Gene SC name EC number

GUT1 YHL032c EC 2.7.1.30GPD1 YDL022w EC 1.1.1.18GPD2 YDL059w EC 1.1.1.18GUT2 YIL155c EC 1.1.99.5PAP YMR165c EC 3.1.3.4SCT1 YBL011w EC 2.3.1.15GPT2 YKR067w EC 2.3.1.15SLC1 YDL052c EC 2.3.1.51DGA1 YOR245c EC 2.3.1.20LRO1 YNR008w EC 2.3.1.158TGL3 YMR313c EC 3.1.1.3TGL4 YKR089c EC 3.1.1.3TGL5 YOR081c EC 3.1.1.3ARE1 YCR048w EC 2.3.1.26ARE2sc YNR019w EC 2.3.1.26ARE2yl EC 2.3.1.26TGL1 YKL140w EC 3.1.1.13POX1 YGL205w EC 6.2.1.3POX2POX3POX4POX5POX6MFE1 YKR009c EC 4.2.1.74POT1 YIL160c EC 2.3.1.16ACL1 NPACL2 NPMAE1 YKL029c EC 1.1.1.38ACC1 YNR016C EC 6.4.1.2

Bioinformatic data were obtained from the Saccharomyces Genome Database (http://wwwNP; not present in this microorganism.

a Genes, corresponding S. cerevisiae gene name and EC number, Y. lipolytica ortholog (

then grows through the addition of units of malonyl-CoA (endo-plasmic reticulum) or acetyl-CoA (mitochondria). Mitochondrialacetyl-CoA for elongation is supplied by the breakdown of the lip-ids accumulated via the b-oxidation pathway. Malonyl-CoA is gen-erated by acyl-CoA carboxylase (ACC) and is the principal source ofcarbon atoms for de novo FA synthesis. For each step in the elonga-tion of the growing FA acyl chain, two molecules of NADPH are re-quired. This NADPH is generated principally by malic enzyme (ME).These three enzymes (ACL, ACC and ME) are believed to play a cru-cial role in determining the potential for lipid accumulation and inregulating this process (see below).

TAG synthesis generally follows the Kennedy pathway [17].During the first step of TAG assembly, glycerol-3-phosphate(G-3-P) is acylated by G-3-P acyltranferase (SCT1) to generatelysophosphatidic acid (LPA), which is further acylated by lysophos-phatidic acid acyltransferase (SLC1) to generate phosphatidic acid(PA). Upon dephosphorylation by phosphatidic acid phosphohy-drolase (PAP), diacylglycerol (DAG) is released from PA. A geneencoding PAP, YALI0D27016g, has just been identified in thegenome of Y. lipolytica (Table 1). This gene is 39% identical to thePAH1 gene of S. cerevisiae. Pah1p belongs to the PAP1 enzymefamily, the members of which require Mg2+ [18] as a cofactor forcatalytic activity. The activity of this enzyme in S. cerevisiae isregulated by lipids, nucleotides and phosphorylation [19,20].

In the final step of TAG synthesis, DAG is acylated in the sn-3position, via an acyl-CoA-dependent or acyl-CoA-independentreaction (Fig. 1). The acyl-CoA-dependent reaction is catalyzed bythree enzymes: Dga1p, Are1p and Are2p. DGA1 encodes an acyl-CoA:diacylglycerol acyltransferase (ACAT), but is unrelated to theDGAT1 subfamily encoding enzymes homologous to the acyl-CoA:cholesterol acyltransferases identified in plants and mammals[21]. Instead, it belongs to the same family as the mammalianDGAT2 gene. In Y. lipolytica, Dga1p seems to be the major enzyme

YL orthologYL name

Function

YALI0F00484g Glycerol kinaseYALI0B02948g Glycerol-3-phosphate dehydrogenase (NAD(+))

Glycerol-3-phosphate dehydrogenase (NAD(+))YALI0B13970g Glycerol-3-phosphate dehydrogenaseYALI0D27016g Phosphatidate phosphataseYALI0C00209g Glycerol-3-phosphate acyl transferase

Glycerol-3-phosphate acyl transferaseYALI0E18964g 1-acyl-sn-glycerol-3-phosphate acyltransferaseYALI0E32769g Diacylglycerol acyltransferaseYALI0E16797g Phospholipid:diacylglycerol acyltransferaseYALI0D17534g Triacylglycerol lipaseYALI0F10010g Triacylglycerol lipase

Triacylglycerol lipaseYALI0F06578g Acyl-CoA:sterol acyltransferase

Acyl-CoA:sterol acyltransferaseYALI0D07986g Acyl-CoA:sterol/Diacylglycerol acyltransferaseYALI0E32035g Cholesterol esteraseYALI0E32835g Acyl-coenzyme A oxidaseYALI0F10857g Acyl-coenzyme A oxidaseYALI0D24750g Acyl-coenzyme A oxidaseYALI0E27654g Acyl-coenzyme A oxidaseYALI0C23859g Acyl-coenzyme A oxidaseYALI0E06567g Acyl-coenzyme A oxidaseYALI0E15378g Multifunctional beta-oxidation proteinYALI0E18568g Peroxisomal oxoacyl thiolaseYALI0E34793g ATP-citrate lyase, subunit aYALI0D24431g ATP-citrate lyase, subunit bYALI0E18634g Malic enzymeYALI0C11407g Acetyl-CoA carboxylase

.yeastgenome.org/) and the Genolevures database (http://cbi.labri.fr/Genolevures/).

gene name), and corresponding function. EC number: enzyme commission number.

http://www.yeastgenome.org/

http://cbi.labri.fr/Genolevures/

Acetyl-CoA Malonyl-CoA

Elongation cycle

Sterol Acyl-CoA

SE

TGL1

ARE1ARE2

Sterol

ββ-oxydation

Acyl-CoA

G-3-P

LPA

PA

SCT1

SLC1

PAP

DAG

DHAP

PL Acyl-CoA

TAG

TGL3, TGL4

GlycerolFFA

POX 1-6

MFE1

Acetyl-CoA

GPD1

GlycerolGUT1

Fatty acid synthesis

Neutral lipidsynthesis

Mobilization

Degradation

LRO1 DGA1, ARE1, ARE2

GUT2

THIO1

Glucose

Fig. 1. Overview of the various pathways involved in fatty acid synthesis and in the storage and degradation of non-polar lipids. Synthesis of non-polar lipids (SE and TAG)required sterol, acyl-CoA and glycerol-3-phosphate (G-3-P). Synthesis of fatty acids (Acyl-CoA) is catalyzed by the fatty acid synthase from the basic blocks acetyl-CoA andmalonyl-CoA through elongation cycles. G-3-P can be produced either from glycerol by the glycerol kinase encoded by GUT1 or from glucose via conversion ofdihydroxyacetone DHAP by the glycerol-3-phosphate dehydrogenase encoded by GPD1 gene. G-3-P can be oxidized to DHAP by the glycerol-3-phosphate dehydrogenaseencoded by GUT2 gene. The synthesis of SE is catalyzed by SE synthases encoded by ARE1 and ARE2. For the synthesis of TAG, three acyls are added to the G-3-P backbonethrough enzymatic steps: first, an acyl is added at the sn-1 position of G-3-P by a G-3-P acyltransferase to produce LPA (SCT1 gene), then a second acyl is added at the sn-2position by a 1-acyl G-3-P acyltransferase (SLC1 gene) to produce phosphatidic acid (PA), which is then dephosphorylated by a phosphatidate phosphatase (PAP) yieldingDAG. Finally, the third acyl can be added at sn-3 position either by the acetyl-CoA-dependent pathway (directly from Acyl-CoA) by acyl-CoA:diacylglycerol acyltransferase(DGA1) and by acyl-CoA:diacylglycerol acyltransferase/acyl-CoA:cholesterol acyltransferase (ARE1, ARE2) or by the acetyl-CoA-independent pathway (from a glycerophos-pholipid, PL) by the phospholipid:diacylglycerol acyltransferase (LRO1). Homologs to S. cerevisiae TAG lipases TGL1, TGL3 and TGL4 involved in SE and TAG mobilization havebeen identified (Table 2). The FFA can then be degraded in the b-oxidation pathway which involved acyl-CoA oxidase (POX), multifonctional beta-oxidation protein (MFE1)and the thiolase (THOI1). In Y. lipolytica, six genes (POX1–POX6) coding for acyl-CoA oxidases are involved in the second step of the b-oxidation.


involved in TAG synthesis, accounting for 45% of DAG acylation(Beopoulos et al., manuscript in preparation). In vivo essays haveshown that this enzyme preferentially makes use of oleic and pal-mitic acid as precursors for CoA-mediated acylation. The contribu-tion of Dga1p in Y. lipolytica seems to increase during the growthcycle, being greatest in the stationary phase. The tagging of Dga1pwith the fluorescent protein EYFP showed Dga1p to be locatedmostly at the surface of LB (Beopoulos et al., manuscript inpreparation).

Two steryl ester synthases, encoded by the ARE1 and ARE2genes, have been identified as orthologs of the genes found in S.cerevisiae, and have been shown to contribute to DAG acylationby acting as acyl transferases in an acyl-CoA-dependent mecha-nism [22,12]. YlAre1p has an amino-acid sequence about 30% iden-tical to that of its two orthologs in S. cerevisiae, whereas YlAre2phas a lower level of sequence identity to Ylare1p (17%) and is moresimilar to the diacylglycerol O-acyl transferase (DGAT) from theplant Perilla frutescens (22%), human DGAT1 (28%) and DGAT1 fromthe plant A. thaliana (25%) [23]. All these enzymes belong to theacyl-CoA:cholesterol acyltransferase (ACAT) family, but their genesdisplay higher levels of sequence identity to the members of themammalian DGAT1 gene family [24]. Despite the differences be-tween their sequences and that of Dga1p, Are proteins displayDGAT activity and are the major enzymes required for TAG synthe-sis during the exponential growth phase of Y. lipolytica. The ob-

served DGAT activity may result from the evolution of ARE genesfrom an ancestral DGAT1 gene duplication in yeast. In vivo studieshave suggested that Are1p prefers saturated acyl-CoA substrates,whereas Are2p prefers unsaturated acyl-CoAs and, more specifi-cally, incorporates oleic acid into TAG (Beopoulos et al., manuscriptin preparation).

In yeasts, the acyl-CoA-independent reaction is carried out bylro1p, a protein with a sequence 27% identical to that of the humanlecithin cholesterol acyl-transferase (LCAT) [25]. This enzyme hasboth phospholipase and acyltransferase functions. Unlike its hu-man ortholog, yeast Lro1p cannot synthesize sterols and functionsas a phospholipid: diacyl glycerol acyl transferase (PDAT) [26]. Theacylation of DAG is an esterification reaction involving the sn-2group of glycerophospholipids, preferentially PtdCho and PtdEtn.The deletion of LRO1 in Y. lipolytica results in a 35% decrease inTAG levels in vivo, with Lro1p making a minor contribution duringthe exponential phase and a much greater contribution during thestationary phase (Beopoulos et al., manuscript in preparation).

SE formation involves the reaction of a fatty acid molecule withthe hydroxyl group of sterols [27]. The SE fraction accounts for 50%of storage lipids in S. cerevisiae [28], whereas only small quantitiesof SE are synthesized (2–5%) in Y. lipolytica [29]. In S. cerevisiae, theacyl-CoA:sterol acyltransferases Are1p and Are2p are the only SEsynthases involved in sterol esterification, as a double mutant lack-ing both ARE genes cannot synthesize SE [30]. Similar results have


been obtained for Y. lipolytica (Beopoulos et al., manuscript in prep-aration). In vivo measurements of SE accumulation in Y. lipolyticahave suggested that the Are proteins act in synergy and that theirrelative contribution is probably greater in the exponential growthphase.

Fatty acid (Acyl-CoA) synthesis is catalyzed by the fatty acidsynthase complex, with the basic acetyl-CoA and malonyl-CoAbuilding blocks. The acyl-CoA may be stored as sterol ester (SE)or triacylglycerol (TAG). SE synthesis is catalyzed by an SE synthasehomologous to the human acyl-CoA:cholesterol acyltransferase(ACAT) and SE mobilization is catalyzed by SE hydrolases, whichrelease sterol and FFA. TAG synthesis requires acyl-CoA and glyc-erol-3-phosphate (G-3-P). G-3-P may be produced from glycerolor from dihydroxyacetone (DHAP). GUT1 encodes a glycerol kinasethat converts glycerol to G-3-P in the cytosol. The G-3-P producedis then oxidized to DHAP by the glycerol-3-phosphate dehydroge-nase encoded by the GUT2 gene and this DHAP may enter glycoly-sis or gluconeogenesis. G-3-P may also be used as a skeleton fortriacylglycerol synthesis. Three acyl groups are added to the G-3-P backbone to generate TAG, and this process involves fourenzyme-catalyzed steps: (1) an acyl group is added at the sn-1position of G-3-P by a G-3-P acyltransferase to produce LPA; (2)a second acyl group is added at the sn-2 position by a 1-acyl G-3-P acyltransferase (AGAT) to produce PA; (3) the PA is thendephosphorylated by a phosphatidate phosphatase (PAP), yieldingDAG and (4) the third acyl group is added at the sn-3 position viathe acyl-CoA-dependent pathway or the acyl-CoA-independentpathway. In the acyl CoA-independent pathway, this third acylgroup is supplied by a glycerophospholipid, whereas, in the acyl-CoA-dependent pathway, it is supplied by acyl-CoA. TAG can bemobilized by conversion to FFA and DAG upon hydrolysis by TAGlipase. The FFA generated may then be degraded by the b-oxidationpathway which involved POX, MFE and THIO genes. This pathwayinvolves four enzyme-catalyzed steps. In Y. lipolytica, six genes(POX1 to POX6) encoding acyl-CoA oxidases involved in the secondstep of b-oxidation have been identified [31,2].

2.4. Biochemistry and regulation of lipid accumulation potential inoleaginous yeast

Oleaginous microorganisms begin to accumulate lipids when anelement in the medium becomes limiting and the carbon source(such as glucose) is present in excess. Many elements can inducelipid accumulation. Nitrogen limitation is generally used in lipidaccumulation studies in microorganisms. Nitrogen limitation isthe easiest condition to control and is generally the most efficienttype of limitation for inducing lipid accumulation. During thegrowth phase, the carbon flux is distributed between the four mac-romolecular pools (carbohydrate, lipid, nucleic acid, protein).Nitrogen is essential for the protein and nucleic acid synthesis re-quired for cellular proliferation. This process is therefore slowed bynitrogen limitation. However, in conditions of nitrogen limitation,the catalytic growth rate slows down rapidly, whereas the rate ofcarbon assimilation slows more gradually. This results in the pref-erential channeling of carbon flux toward lipid synthesis, leadingto an accumulation of triacylglycerols within discrete lipid bodiesin the cells. If non-oleaginous microorganisms are placed in thesame nutrient-limiting medium, further cell proliferation tendsto cease, with carbon flux into the cell maintained but, in this case,the carbon is converted into various polysaccharides, includingglycogen and various glucans and mannans.

During the transition between the growth phase (growth withthe production of catalytic biomass) and the lipid accumulationphase (decrease in growth rate due to nutrient limitation and thediversion of excess carbon to lipid production), some pathwaysare repressed (nucleic acid and protein synthesis), whereas others

are induced (fatty acid and triacylglycerol synthesis). This transi-tion is induced by the establishment of nitrogen limitation (see be-low). In addition, during the accumulation phase, precursors(acetyl-CoA, malonyl-CoA and glycerol) and energy (ATP, NADPH)are required for lipid synthesis. We will describe here the role ofthe key enzymes involved in regulating lipid accumulationpotential.

AMP deaminase is activated by the exhaustion of nitrogen inthe medium during the growth of an oleaginous microorganism.AMP deaminase catalyzes the following reaction [32]:

AMP! IMPþ NHþ4

The activation of AMP deaminase decreases mitochondrial AMPconcentration and increases cellular ammonium concentration.The decrease in AMP concentration inhibits isocitrate dehydroge-nase, blocking the citric acid cycle at the isocitrate level. Aconitasemediates the accumulation of citrate in mitrochondria, with exitfrom the mitochondria mediated by the citrate/malate cycle[32,33].

This reaction provides large amounts of acetyl-CoA for fatty acidsynthesis. Acetyl-CoA is provided by the cleavage of citrate comingfrom the mitochondria by ATP-citrate lyase (ACL) in the cytosol.ACL cleaves the citrate to give oxaloacetate and acetyl-CoA.

CitrateþHS-CoAþ ATP! acetyl-CoAþ oxaloacetateþ ADPþ Pi

This enzyme is absent from non-oleaginous yeasts, such as S.cerevisiae, but has been shown to be present in Y. lipolytica.Whereas the human ACL consists of a single protein encoded bya single gene, the ACL enzymes of both Y. lipolytica and Neurosporacrassa consist of two subunits, Aclap and Aclbp, encoded by ACL1(YALI0E34793g) and ACL2 (YALI0D24431g), respectively (Table 2).This enzyme requires an ammonium ion for activation and isdependent on adenosine mono- and diphosphate [16,34]. How-ever, ammonium ions are scarce in the absence of nitrogen, dueto the induction of AMP deaminase [32,33].

In addition to acetyl-CoA, fatty acid synthesis requires a contin-uous supply of malonyl-CoA and NADPH. Malonyl-CoA can also begenerated from acetyl-CoA, in a reaction catalyzed by acetyl-CoAcarboxylase (Acc1p) [35].

Acetyl-CoAþHCO�3 þ ATP!malonyl-CoAþ ADPþ Pi

In mammalian cells, this enzyme is activated in the presence oftricarboxylic acid intermediates, such as citrate [36]. In yeast, how-ever, Acc1p undergoes allosteric activation as a function of citrateconcentration [37]. In Y. lipolytica, this enzyme is encoded by theACC1 gene, and is known as YALI0C11407 g (Table 2).

NADPH is required for the function of the fatty acid synthase(FAS). It is thought that NADPH concentration is controlled bythe activity of malic enzyme (ME). This enzyme catalyzes the fol-lowing reaction:

Malateþ NADPþ ! pyruvateþ NADPH

The first evidence of ME involvement in lipid accumulation wasprovided by the inhibition of this enzyme by sesamol in Mucor cir-cinelloides, resulting in a decrease in lipid accumulation from 25%to 2% of cell biomass [38]. Ratledge et al. subsequently demon-strated a direct correlation between decreasing ME activity duringthe lipid accumulation phase and the extent of lipid accumulation[39]. ME overproduction in M. circinelloides, through the expressionof the gene under the control of the strong constitutive promoter ofthe glyceraldehyde-3-phosphate dehydrogenase gene (gpd1), wasrecently shown to increase lipid accumulation by a factor of 2.5[40]. In Y. lipolytica, this enzyme is encoded by the MAE1 gene


(YALI0E18634 g – Table 2). However, preliminary results for MEoverproduction in Y. lipolytica indicate that NADPH concentrationis not limiting for lipid accumulation in this yeast.

Table 3Lipid content and lipid yield as a function of the nature of nutrient limitation forCandida sp. 107 (a) Rhodotorula glutinis (b) and Yarrowia lipolytica (c).

Limitation Yeast Culturemode

% Lipidðglip g�1

X ÞMaximumyieldðglip g�1

gluÞ

References

Nitrogen a Batch 37% 0.22 [37]Nitrogen c Continuous 28% 0.11 [46]Nitrogen c Fed-batch 38.6% 0.22 [42]Nitrogen c Batch 11% 0.017 [47]Nitrogen b Fed-batch 72% 0.255 [41]Carbon a Batch 14% 0.07 [37]Phosphorus a Batch 31% 0.15 [37]Phosphorus and

nitrogena Batch 35% 0.052 [37]

Magnesium a Batch 32.3% 0.10 [37]Zinc b Batch 34% 0.22 [44]Iron b Batch 45% 0.12 [44]

3. Factors affecting lipid accumulation

Lipid accumulation depends primarily on microorganism phys-iology, nutrient limitation and environmental conditions, such astemperature and pH. It is also affected by the production of sec-ondary metabolites, such as citrate and ethanol.

The transition from catalytic growth to lipid accumulation gen-erally occurs when excess carbon in the medium is associated witha nutrient limitation affecting biomass production. Y. lipolytica andS. cerevisiae have followed different courses of physiological evolu-tion. S. cerevisiae catabolizes glucose efficiently, glycerol to a lesserextent, and fatty acids poorly. S. cerevisiae does not use fatty acidsvery efficiently. Indeed, the doubling time of cell populations onthis substrate is about 4 h. It is widely accepted that this poor lipidutilization efficiency results from a limited capacity for b-oxidation(breakdown of FA to generate acetyl-CoA). S. cerevisiae cannot useoils (mixture of TAG and FA) due to its inability to secrete lipases.For S. cerevisiae (Fig. 2A), the presence of excess carbon (more glu-cose taken up than required for biomass production) induces ametabolic shift from oxidative to oxido-reductive metabolism withethanol production. In aerobic batch culture, S. cerevisiae producesethanol, with a glucose-to-ethanol conversion yield of0:548 Cmoletoh Cmol�1

glc , a maximum specific growth rate of 0.43 h�1

and an average ethanol volumic productivity of 3.3 g l�1 h�1 [41].Y. lipolytica grows as well on glucose (lmax = 0.26 h�1, [41,42])

as on oleic acid (lmax = 0.33 h�1 [43]). In conditions of nutrient lim-itation in the presence of excess carbon, Y. lipolytica produces largeamounts of TCA-cycle intermediates, such as citric (CA), isocitricacid (ICA), 2-ketoglutaric acid and pyruvic acid (for review see[2]). It also converts excess carbon into TAG. By contrast to S. cere-visiae, Y. lipolytica (Fig. 2B) is a strict aerobic yeast unable to pro-duce ethanol. Depending on C/N ratio, different governingmetabolisms can be observed: pure growth, organic acid produc-tion or conversion of excess carbon into lipids (triacylglycerolsand sterol esters). By monitoring growth, carbon excess in the

Fattyacid

Glucose

glycerol

Ethanol

BiomassMetabolite

Fatty Acidtransport

andββββ-oxydation

Respiration capacity

Oil

eaisiverec.SA B

Fig. 2. Schematic comparison of S. cerevisiae and Y. lipolytica metabolisms. S. cerevisiacerevisiae cannot catabolize oils, because it does not secrete lipase. In the presence of exceacids and glycerol. This yeast secretes lipases and is therefore able to grow on oils. In the

anabolism of Y. lipolytica can be either oriented towards organicacids production or lipids production (TAG, SE). Y. lipolytica is anoleaginous yeast able to accumulate lipids to levels exceeding50% of cell dry weight.

Lipid accumulation, in terms of lipid profile, amount, productiv-ity and conversion yield, is influenced by various operating condi-tions, such as the nature of nutrient limitation, pH, aeration andtemperature conditions. Published values for lipid accumulationby the yeasts Candida sp. 107, R. glutinis and Y. lipolytica are pre-sented in Table 3 as a function of the nature of nutrient limitation:nitrogen, magnesium, zinc, iron or phosphorus. However, nitrogenlimitation is most commonly used to induce lipid accumulationand gives the best conversion yield with glucose, reaching0:22 glip g�1

glu [44,45].We will deal here specifically with nitrogen limitation, which

remains the most efficient form of nutrient limitation for theinduction of lipid accumulation. The culture strategy involvestwo phases, as a function of yeast metabolism. The first is a growthphase [48,49]. The growth of the yeast is then slowed by nitrogenlimitation and the lipid accumulation phase begins. The global con-version yield of glucose into lipids in batch culture depends on the

AlkaneFatty acid

Glucose

Ethanol

Lipid

Biomass

No limitation

Glycerol

Metabolite

acitylopil.Y

e catabolizes preferentially sugars (glucose, saccharose) rather than fatty acids. S.ss sugar, S. cerevisiae produces ethanol. Y. lipolytica catabolizes alkanes, glucose, fattypresence of excess carbon, Y. lipolytica either produces metabolites or stores lipids.

0.2

0.25

0.3

0.35

0.4

0.45

0.5

100 150 200 250 300 350 400initial ratio C/N [Cmolglc.molN

-1]

conv

ersi

on o

f gl

ucos

e to

lipi

ds y

ield

[Cm

ollip

.Cm

olgl

c-1]

Fig. 3. Changes in conversion yield for the conversion of glucose into lipids plottedagainst C/N ratio for batches of R. glutinis [44].


duration of the growth phase and the transition to the accumula-tion phase. The duration of the growth phase depends on the C/Nratio. The total substrate-to-lipid conversion yield therefore de-pends on the initial C/N ratio of the batch culture. Glucose-to-lipidconversion yield increases from 0.25 to 0.40 ðCmollip Cmol�1

gluÞ as C/N ratio increases from 150 to 350 ðCmolglu mol�1

N Þ for the oleagi-nous yeast R. glutinis (Fig. 3) [44]. However, a C/N ratio exceeding350 gC g�1

N (data not shown) creates a severe nitrogen deficiency,leading to a rapid decrease in cell viability before the cells are ableto enter the lipid accumulation stage.

The immediate precursor of cellular lipid accumulation in ole-aginous microorganisms is citric acid [15,50–52]. In Y. lipolytica,if the initial C/N molar ratio is high [80–120 Cmol Nmol�1], cellgrowth may be followed by significant citric acid production, withlow levels of lipid accumulation within cells. Papanikolaou et al.assumed that ATP-citrate lyase was inactive in the presence of ex-cess glucose, resulting in low levels of lipid accumulation [53].However, if a cosubstrate was used, optimal C/N ratios of 35 CmolNmol�1 for lipid production from glucose and 180 Cmol Nmol�1 forlipid production from glycerol and glucose were obtained for batchcultures [54,55]. In continuous culture conditions, Aggelis andKomaitis used a medium with a C/N ratio of 66 Cmol Nmol�1 [46].

In general, the maximum C/N ratio suitable for lipid accumula-tion can be estimated from the ratio YX=S

q ½CmolX Cmol�1sub� , where q is

the nitrogen content of biomass [Nmol Cmol�1], YX/S is the ratio oftheoretical biomass production to substrate, expressed as Cmol ofbiomass/Cmol of substrate in conditions of carbon limitation in theabsence of by-product production [56].

If carbon is not limiting (present in large excess), the uptake ofcarbon is limited only by the substrate transport system of the cell.In this case, limiting concentrations of nitrogen in the medium leadto the induction of lipid accumulation. The critical nitrogen con-centration for lipid induction in Y. lipolytica has been found to beabout 10�3 mol l�1[42]. It is important for nitrogen concentrationto exceed this threshold value to prevent the production of second-ary metabolites (citric acid) that will otherwise affect lipidaccumulation.

Conversely, if the extracellular carbon supply is exhausted,stored lipids may be mobilized. Thus, lipid accumulation is alwaysdependent on the influx of carbon substrates. This complex regula-tion makes it difficult to achieve high rates of lipid accumulation inbatch culture. In such conditions, lipid accumulation is always fol-lowed by citric acid production [47]. These findings highlight theneed to manage both carbon and nitrogen flows to optimize lipidaccumulation and reduce citric acid production (see below).

Lipid profile can be modified by adjusting the temperature [57].Fatty acid composition is dependent on culture temperature, be-cause the degree of saturation generally decreases with decreasing

temperature. For example, R. glutinis produces 0:44 gAGI g�1AG unsat-

urated lipids at 15 �C and 0:27 gAGI g�1AG at 30 �C [44]. This modula-

tion of the lipid profile probably results from an increase indesaturase stability at low temperatures, with no such increasein stability observed for the other enzymes. However, low temper-atures do not favor lipid production, because they also lead to largedecreases in cellular activity and metabolism. There are few alter-natives to genetic modification for the modulation of fatty acidprofile. One of these alternatives is the use of growth inhibitors,such as cerulenin [58], or natural antimicrobial compounds, suchas Teucrium polium extracts [46].

4. Modes of culture for ensuring high levels of lipidaccumulation

Bioprocesses for lipid production may be designed on the basisof our knowledge of Y. lipolytica metabolism, taking into accountthe ability of this yeast to produce large amounts of intermediatesand to accumulate large amounts of lipids and to break them downby b-oxidation, even on a glucose substrate [42]. Four metabolicstates can be defined as a function of differences in C/N flux ratiofor a constant nitrogen flux (Fig. 4). The first corresponds to a car-bon flux lower than that required for growth. In this case, if thecells have storage lipids, they make up the carbon deficit by mobi-lizing these storage lipids (Fig. 4, state a). The second state corre-sponds to the maximum growth rate obtained with optimalcarbon influx from the medium, at a defined nitrogen flux rate. Thisstate results in maximal biomass production (Fig. 4, state b). Thethird state corresponds to an influx of excess carbon from the med-ium, which has a high C/N ratio, resulting in a decrease in biomassproduction and high levels of lipid production (Fig. 4, state c). Thefourth state results from a further increase in C/N ratio (Fig. 4, stated), leading to the repression of lipid accumulation in favor of sec-ondary metabolite production. If the desired end product of theprocess is lipid, then the process must be designed so as to ensurethe maximal conversion of the carbon taken up into lipids, by min-imizing by-product (citric acid) production and maximizing lipidsynthesis, holding the cells in state c.

Three different modes of culture are commonly used: batch,fed-batch and continuous mode. Most of the processes describedin previous publications relate to batch mode [45,47,59].

4.1. Batch mode

In batch cultures, minerals and carbon substrates are initiallymixed in the bioreactor, with a high initial C/N ratio to boost lipidaccumulation. As nitrogen is actively consumed right from thestart of culture, the rC/rN ratio (residual carbon to residual nitrogenratio) continually increases, tending to infinity. In this mode,growth remains exponential whilst nitrogen is not limiting(Fig. 4, state b). Granger showed that, in R. glutinis grown at30 �C with glucose, nitrogen limitation led to a decrease in the rateof substrate consumption by a factor of four and an increase in li-pid production rate by a factor of two to three [44]. Microbialmetabolism then shifts into phase c (Fig. 4). Nevertheless, citricacid production is induced as a function of rC/rN ratio, resultingin a shift of microbial metabolism into phase d (Fig. 4). Citric acidproduction decreases the total conversion yield for the productionof lipids from carbon substrate. Thus, this conversion yield de-pends mostly on the ratio of biomass constituted during thegrowth phase to lipids accumulated during the accumulationphase in batch culture. Control of the ratio of carbon consumptionto nitrogen consumption is therefore essential to prevent citricacid secretion, hence the importance of monitoring rC/rN ratio incontinuous and fed-batch cultures.

0%

50%

100%

150%

200%

250%

a b c d

carbon substrate

nitrogen catalyticbiomass

lipids citricacid

Car

bon

flow

/gro

wth

car

bon

flow

nee

d

Fig. 4. Yeast activity as a function of carbon flow rate for a fixed nitrogen flow rate. State a: pure growth and mobilization of stored lipids; state b: pure growth; state c:growth with lipid accumulation; state d: growth, lipid accumulation and citric acid production. The size of arrow is proportional to flow. The carbon substrate and nitrogenarrows correspond to specific influx into cells. Catalytic biomass, lipids and citric acid arrows correspond to specific production rates (consumption for lipids, state a).

Fig. 5. Volumetric productivity of Y. lipolytica biomass P(X) and lipid P(lip) plottedagainst dilution rate on industrial glycerol. Cells were grown in single-stagecontinuous cultures, with a fixed C/N ratio at pH 6 ± 0.05; T = 28 �C and aerationrate = 1.8 VVM. [55].


4.2. Continuous mode

Lipid production in continuous culture has been modeled byYkema et al. [56]. In a continuous culture, the C/N ratio in the cul-ture medium and rC/rN are constant for a given dilution rate. Forlow dilution rates, with intermediate C/N ratios promoting lipidaccumulation ð40 gC g�1

N Þ, the lipid and biomass concentrations ob-tained are higher than those obtained for higher dilutions. Indeed,at similar rC/rN ratios, low specific growth rates promote lipid accu-mulation. The optimization of the process therefore involves deter-mining the optimal dilution rate with an optimal intermediate C/Nratio (see Fig. 5).

4.3. Fed-batch mode

The production of lipid by yeast may ensure that lipids are effi-ciently and reproducibly produced [60]. If lipid production is to becontrolled, regulation of the environmental variables is require tomaximize the stability of the metabolic state. This stability canbe achieved through the precise control of nutrient flow rate.Greater accuracy in the control of nutrient flow rate is associatedwith greater control of metabolic state and more optimal produc-tion, which is the case for protein production processes [61] (for re-view see [62]).

In fed-batch culture, nitrogen and carbon flows are monitoredto control the specific growth rate and the rC/rN ratio. As shownin Fig. 6, a fed-batch culture of Y. lipolytica at 28 �C with glucose

as the substrate goes through three phases: a pure growth phase(i), a transition phase (ii) and a lipid accumulation phase (iii).

During the growth phase, yeast metabolism results in the bal-anced distribution of carbon between the four main macromolecu-lar pools (carbohydrate, lipid, nucleic acid, protein), with catalyticbiomass production (biomass without the accumulation of a mac-romolecular compound). This phase corresponds to state b in Fig. 4.

255

260

265

270

275

280

285

290

295

300

305

0 10 20 30 40 50 60 70 80time [h]

-202468101214161820

glucose [g] NH4 [g] lipids [g] biomass [g] citric acid [g]

growth transition lipid accumulation

gluc

ose,

bio

mas

s, li

pid,

citr

ic a

cid

conc

entr

atio

n [a

rbitr

ary

unit]

nitr

o gen

con

cent

ratio

n [a

rbitr

ary

unit]

Fig. 6. Modeling and prediction for a fed-batch culture of Y. lipolytica. The first part of the growth (from 0 h to 27 h) corresponds to a pure growth phase, with a C/N flux ratioequal to catalytic biomass production requirements. The second phase (transition from 27 to 40 h) corresponds to the establishment of nutrient limitation (nitrogen),followed by the establishment of lipid accumulation. The last step (from 40 to 70 h) corresponds to the lipid accumulation phase, during which nutrient limitation iscontrolled by optimizing the C/N ratio [around 20 Cmol mol�1

N to favor lipid accumulation.


The transition phase corresponds to the establishment of nitrogenlimitation, with excess carbon leading to the accumulation of lip-ids. This phase corresponds to the transition between state b andstate c. Lipid accumulation phase is the most extensive, corre-sponding to the establishment lipid production under constantnitrogen limitation conditions, with a C/N ratio of about20 Cmol Nmol�1, preventing citric acid production.

5. Mastering lipid production

Lipid accumulation in oleaginous species is upregulated in cul-ture media containing fatty acids or oils. Y. lipolytica has developedsophisticated mechanisms for taking up and assimilating hydro-phobic substrates. However, this process is counterbalanced bybeta oxidation, which mobilizes the accumulated lipids. Neverthe-less, with a suitable culture medium and appropriate genetic mod-ifications, high levels of lipid accumulation can be achieved,exceeding 60% of cell dry weight in some cases, with modificationof the profile of accumulated fatty acids.

5.1. The ex novo lipid accumulation pathway

Y. lipolytica has evolved elaborate strategies and adaptationmechanisms for the efficient use of hydrophobic substrates, suchas n-alkanes, fatty acids and triacylglycerols. It has adapted tothe use of these substrates by evolving genes encoding surfactantsfor their solubilization (liposan), by modifying its cell surface tofacilitate adhesion of hydrophobic droplets (cell surface protru-sions), thereby maximizing cell-substrate contact and by develop-ing complex transport mechanisms for the incorporation of thesecompounds into the cell (for reviews see [2], [63], [64]). This adap-tation of Y. lipolytica may have involved genome amplification andsuccessive evolution of the genes required for the utilization of di-verse hydrophobic substrates (with respect to the chain lengths ofalkanes and fatty acids). Indeed, Y. lipolytica has at least 13 familiesof genes involved in hydrophobic substrate utilization, reflectingthe vast expansion of its genome with respect to those of otherascomycetous yeasts [65,66]. For example, the lipase gene familyof this yeast has 16 members and the acyl-CoA oxidase familyhas six members.

A fine example of the complex mechanisms developed by Y.lipolytica is in the hydrolysis and incorporation of oil substrates.The yeast secretes an extracellular lipase called lip2p, encoded bythe LIP2 gene. This gene encodes a precursor pre-pro-mature pro-tein with a Lys-Arg (KR) cleavage site [67]. It simultaneously pro-duces other intracellular lipases, such as Lip7p and Lip8p, whichmay be released into the medium, depending on the substrate.These lipases have different chain-length specificities, with Lip2,Lip7 and Lip8 displaying maximal activity with oleate (C18), capro-ate (C6) and caprate (C10), respectively [68,69]. The released fattyacids must then be transported into the cell. Kholwein et al. carriedout the first study of FA transport in Y. lipolytica. They showed thatan energy-free transporter was required below a threshold of10 lM, whereas at higher concentrations, lauric or oleic acid dif-fused freely [70]. This model also suggested that Y. lipolytica hadtwo different chain length-selective transporters. Papanikolaouand Aggelis analyzed fatty acid uptake in Y. lipolytica duringgrowth on lipid-containing media with different substrate compo-sitions [55]. Various mixtures of stearin and hydrolyzed rapeseedoil were used for growth, and fatty acid uptake and profiles wereanalyzed. Regardless of the external fatty acid composition, allfatty acids displayed similar incorporation constants for uptake(0.023 h�1), with the exception of C16:0 and C18:0, which had low-er constants (0.01 and 0.013 h�1, respectively) (Fig. 4). However, ifY. lipolytica is grown on saturated fatty acids alone, the incorpora-tion constants for C16:0 and C18:0 were significantly higher(0.018 h�1) [71] (see Fig. 7).

5.2. The b-oxidation degradation pathway

In Y. lipolytica, FA are broken down in peroxisomes, via the four-step b-oxidation pathway. Six different acyl-CoA oxidases (Aox1-6,encoded by the POX1 to POX6 genes) catalyze the first and rate-lim-iting step of b-oxidation. These acyl-CoA oxidases (Aox) have dif-ferent activities and substrate specificities, as shown by genedisruption in Y. lipolytica [31] and by expression and purificationin E. coli. Aox2p is very active and highly specific for long-chainfatty acids, whereas Aox3p preferentially acts on short-chain fattyacids. Thevenieau et al. [63] recently showed that Aox1p andAox6p are involved in breaking down dicarboxylic acids (DCA).

0.000

0.005

0.010

0.015

0.020

0.025

0.030In

corp

orat

ion

cons

tant

(h-

1)

Fatty acid

C12:0

C16:0

C18:1

C18:2C18:3C14:0

C18:0

Fig. 7. Fatty acid uptake in Y. lipolytica. Incorporation constant (k, h�1) of fatty acidsduring the growth of Y. lipolytica on lipid-containing medium (mixture ofhydrolyzed rapeseed oil and stearin). (Adapted from [71]).


Gene disruption analysis have also shown that the mechanismof peroxisome entry depends on POX genotype and that these en-zymes have no peroxisome targeting sequence (PTS). By contrast,the acyl-CoA oxidase (Aox) complex of Y. lipolytica has been shownto be assembled in the cytosol before its import into peroxisomesas a heteropentameric, cofactor-containing complex, with Aox2pand Aox3p playing a key role in this process [72]. Aox proteinshave also been shown to play a major role in peroxisomal division.During peroxisome maturation, a membrane-bound pool of Aoxinteracts with a membrane-associated peroxin, Pex16p. Pex16pdownregulates membrane fission, thus preventing the excessiveproliferation of immature peroxisomal vesicles. The Dpox4 andDpox5 mutants form giant peroxisomes [73].

The second and third steps of b-oxidation are catalyzed by themultifunctional enzyme encoded by the MFE gene, which displayshydratase and dehydrogenase activities (Fig. 1). The fourth step iscatalyzed by the 3-ketoacyl-CoA-thiolase encoded by the POT1gene. A decane-inducible peroxisomal acetoacetyl-CoA thiolase(encoded by PAT1) has recently been identified and, together withthe thiolase encoded by POT1, this enzyme is thought to catalyzethe last step of b-oxidation. Changes in b-oxidation flux or thecomplete abolition of this process might increase lipid accumula-tion (see below).

5.3. Growth conditions and genetic modifications favoring lipidproduction

The ability of Y. lipolytica to utilize and degrade HS has led to itsuse in bioremediation processes and in fermentation techniquesfor the production of intermediate metabolites, enzymes andadded-value oils. The capacity of this yeast to break down HS hasmade it possible to decrease chemical oxygen demand (COD) sig-nificantly in oil mill wastewater containing fats, sugars, phosphate,phenols and metals [74]. Similar methods have been used for selec-tive medium degradation [75] and fat separating processes inwastewater purification. The production of intermediate metabo-lites was highlighted by Papanikolaou et al. [53], who used Y.lipolytica to produce citric acid from raw glycerol, the major by-product of biodiesel production units. A high C/N ratio, combinedwith a buffered pH, resulted in rates of citric acid production of35 g.l�1, and low rates of lipid accumulation, probably due to thedownregulation of ATP-citrate lyase in the experimental condi-tions used. Similar results were obtained by Stottmeister et al.[76], who used sunflower oil or alkanes as a substrate for the pro-duction of up to 250 g l�1 citric acid, in fed-batch cultures. Marty

et al. developed a mineral medium fulfilling the nutritionalrequirements of Y. lipolytica for the production of the extracellularlipase Lip2 which is known to have a high enantiomeric resolutioncapacity and catalyzes re-esterification reactions [77], yieldingover 60,000 U l�1 Lip2.

The ability of Y. lipolytica to utilize and accumulate HS has ledto the use of this yeast for the production of specific added-valueoils from raw materials. Lipid composition and accumulationcapacity in Y. lipolytica depend strongly on growth and cultureconditions, providing a broad range of choice for substrate/finalproduct combinations. For example, replacing glucose by oleicacid as the carbon source increases lipid accumulation capacityand the size of lipid particles and also modifies the compositionof lipids and proteins [78]. Depending on the composition of thesubstrate used, this yeast selectively removes or assimilates FAfrom the substrate to produce fats with a predetermined compo-sition [43]. Papanikolaou et al. [79] used these properties to pro-duce cocoa butter-like substances from mixtures of inexpensivesubstrates, such as saturated fat and raw glycerol. In a similar pro-cedure, Schrader et al. obtained high yields of c-lactone aromafrom castor oil [80].

Several procedures for increasing lipid production in Y. lipolyticahave been proposed. Fine adjustments of culture conditions can beused to upregulate lipid metabolism: Bati et al. reported a majoreffect of dissolved oxygen, nitrogen/carbon ratio, pH and amountof oil substrate on lipid accumulation, resulting in yeasts contain-ing 37–70% lipid [81]. Aggelis et al. obtained yeasts that accumu-lated lipids to 43% of dry biomass, with a volumetric productivityof 0.12 g lipid l�1 h�1, using industrial fats or glycerol – both ofwhich are cheap industrial carbon side-products – as the substrate[46]. In addition to the regulation of nutrient levels and cultureconditions, genetic engineering approaches have been developed,based on the potential of yeasts as cell factories for the productionof large amounts of oil with a particular lipid composition andadded-value metabolites. Redirection of carbon flux toward TAGassembly mediated by deletion of the glycerol-3-phospate dehy-drogenase gene (GUT2), an enzyme situated at the crossroad of li-pid metabolism and DHAP production for glycolysis, increases lipidaccumulation to levels three times those observed in wild-typestrains [22]. The additional deletion of the POX1 to POX6 genesencoding the acyl-CoA oxidases, completely abolishes b-oxidation,preventing lipid mobilization. The resulting mutant strain had lipidlevels four times those of the wild-type strain. In addition, the POXmutant genotypes affected lipid profile, as each POX gene has a dif-ferent substrate specificity. POX2 deletion blocks the use of long-chain fatty acids, whereas POX3 deletion decreases the uptake ofshort-chain fatty acids [82]. In mutants producing only Aox4p, orAox1p and Aox6p, only a few, small intracellular LB are formed,resulting in a so-called ‘‘slim yeast” phenotype. Slim yeasts cannotaccumulate the substrate, probably due to the feedback regulationof oxidation, affecting HS transporters and downregulating the dif-fusion process. By contrast, strains producing either Aox2p alone orboth Aox2p and Aox4p form fewer, but larger, LB than the wildtype. The overexpression of Aox2p in the Dpox2–5 quadruple mu-tant restores lipid accumulation, by increasing non-polar lipid stor-age in the LB, resulting in an ‘‘obese yeast” phenotype. Thesefindings suggest a regulatory mechanism preventing or enabling li-pid storage as a function of the ability of the yeast to assimilate thesubstrate from the medium. The LB phenotype also depends on theacyltransferase profile of mutant strains, with strains lacking ARE2sterol-acyltransferase being able to form large LB.

6. Potential applications

A potential application of Y. lipolytica or of oleaginous microor-ganisms in general is the production of single-cell oils (SCO).


Single-cell oils may be defined as edible oils obtained from micro-organisms and of a similar type and composition to the oils andfats obtained from plants or animals. SCO are now accepted asbiotechnological products fulfilling key roles in the supply of majorpolyunsaturated fatty acids (PUFA), which are known to be essen-tial for human nutrition and development. The commercial nichetargeted by SCO is that of dietary supplements enriched in docosa-hexaenoic acid (DHA), arachidonic acid (AA) and c-linolenic acid(GLA) [3]. Picataggio et al. (pending US patent application Ser. No.10/840,579) recently demonstrated the feasibility of engineeringY. lipolytica for the production of x-3 and x-6 fatty acids (e.g.18:3, x-3, ALA, a-linoleic acid; C18:3, x-6, GLA, c-linoleic acid;18:4, x-3, STA, stearidonic acid; 20:3, x-3, ETrA, eicosatrienoic acidand x-6, DGLA, dihomo-c-linoleic acid; 20:4, x-6, ARA, arachidonicacid; 20:5, x-3, EPA, eicosapentaenoic acid and 22:6, x-3, DHA,docosahexaenoic acid), by introducing and expressing heterologousgenes encoding the proteins of the x-3/x-6 biosynthetic pathwayin the oleaginous host. Damude et al. expressed in Y. lipolytica abifunctional D12/x3 desaturase from Fusarium moniliformis. Thisstrain produced a-linolenic acid (ALA, 18:3 D9,12,15) to levels corre-sponding to 28% of its cellular dry mass [83]. Similarly, GLA produc-tion was obtained by overproducing the D6 and D12 desaturasesfrom Mortierella alpina in Y. lipolytica under the control of a strongconstitutive promoter [84]. Furthermore, Dupont de Nemours haveidentified and isolated genes encoding acytransferases suitable forthe transfer of these newly synthesized PUFA into TAG. These mod-ified strains have been patented (US Patents 7267976, 7238482,7256033) for the marketing of dietary supplements based on PUFAto protect against cardiovascular disease.

One potential application of a Y. lipolytica strain with an alteredPOX genotype grown on a medium with a specific compositionwould be the production of dicarboxylic acids (DCA) or diacids[2]. Y. lipolytica strains are also regarded as very promising agentsfor the treatment of mineral oil pollution and plant oil waste. Oilmill wastewater (OMW), in particular, is a major source of waterpollution as it contains fats, sugars, phosphate, phenols and metals.Scioli and Volaro [74] reported an 80% decrease in the COD ofOMW after treatment for 24 h treatment with Y. lipolytica, whereasOswal [85] found this yeast to be capable of decreasing COD by 90%in palm oil mill wastewater.

Efforts are currently being made to generate strains with ex-treme oil accumulation capacities and specific fatty acid composi-tions. These studies are based on the regulation of substratetransport mechanisms, the overproduction or elimination of desat-urases involved in TAG synthesis and the strict control of genes in-volved in fatty acid metabolism. The preliminary results obtainedare very promising and potential applications have been identifiedin the areas of SCO and biodiesel production.

Re-engineering microbial metabolism to favor oil production forfuel use seems to be the way forward to a third generation of bio-fuels: oils will be produced from more appropriate materials (suchas cellulose, glycerol, or even oil waste) in better ways. The capac-ity of Y. lipolytica to produce long-chain molecules and to adjust itsmetabolism for the production of specific oils could lead to the pro-duction of more energy-efficient fuels.

7. Summary, conclusions and future directions

Studies of lipid accumulation are currently focusing on maxi-mizing oil production in unicellular organisms, such as yeast or al-gae, and in plants, for the generation of bio-fuels, includingbiodiesel. Like fossil hydrocarbons, TAG oils are highly concen-trated stores of saturated hydrocarbons that can be oxidized togenerate energy. The metabolism of Y. lipolytica, oriented towardthe accumulation of lipids, and the unique capacity of this yeastto use HS efficiently, make this microorganism a prime candidate

for use in the production of bio-oils. The use of this yeast as a mod-el organism for investigating the mechanisms underlying thesemetabolic pathways has already provided insight into substratetransport processes, the function and use of the diverse organellesin metabolic processes and the identification and regulation ofgenes involved in these processes.

However, although research on yeast has generated a number ofimportant results concerning TAG and SE synthesis, many ques-tions remain unanswered. One of these questions concerns theredundancy of the enzymes involved in lipid metabolism. The mul-tiple functions of these proteins, which may act as acyltransferasesor hydrolases, generating similar products from similar substrates,raises questions about the existence of multiple copies of enzymesto serve as a back-up for important metabolic pathways. These en-zymes may be subject to multiple regulation procedures, includingclassical regulation at the transcriptional and translational levels,post-translational modifications of enzymes, and additional coor-dination at the organelle level. In addition, certain enzymes maybe involved in protein–protein interactions or subject to differen-tial regulation by auxiliary proteins, through processes that remainto be clarified [86].

For example, LB-associated proteins characteristically have PATdomains in their primary sequences. The PAT domain is defined bya conserved amino-acid sequence present in perilipin, adipophilin(also named adipocyte differentiation-related protein, ADRP), andthe tail-interacting protein of 47 kDa (TIP47) [87]. The PAT proteins(perilipin, adipophilin/adipose differentiation-related protein,TIP47, and other related proteins) have structural and regulatoryfunctions and are thought to target lipid bodies through differentmechanisms [88]. Some PAT proteins are found exclusively on lipiddroplets (e.g. adipophilin), whereas others are found both in thecytosol and on lipid droplets. The structural properties underlyingthe binding of PAT proteins to lipid droplets provide a large numberof possibilities for regulating this distribution. Indeed, the localiza-tion of several proteins, such as TIP47, to lipid droplets is highlyregulated and can be induced by adding fatty acids to the mediumto trigger droplet formation [89]. However, the mechanismunderlying this regulation remains largely unclear.

It has been suggested that lipase access to the core of the lipiddroplet is regulated in part by PAT proteins, and the phosphoryla-tion of perilipin may indeed be crucial for the regulation of lipaseaccess to substrates [90]. It has been estimated that PAT proteinscover 15–20% of the droplet surface, possibly resulting in sterichindrance, restricting lipase access. However, the hydrolysis andmobilization of sterol esters is less well understood. In S. cerevisiae,sterol esters and TAG are hydrolyzed by three consecutive steps(Yehl, Yeh2, and Tgl1 for SE, Tgl3, Tgl4, and Tgl5 for TAG) catalyzedby enzymes present on the surface of lipid bodies [91,92]. Little isknown about the regulation of the activities of these enzymes, butPAT proteins may be involved in the limiting step of hydrolysis. Inaddition to the small number of lipid droplet-specific proteinsidentified, such as PAT proteins, many cellular proteins withknown functions have been identified in lipid body fractions inproteomic studies. Some of these proteins are involved in dynamiccellular processes, suggesting a possible similar role in lipid dropletbiology [78,89,93]. However, both the localization of most of theseproteins to droplets and their functional roles remain to be vali-dated. Another unanswered question relating to lipid metabolismconcerns LB biogenesis. Insight into this process might also provideclues to the role of auxiliary proteins in the early steps of LP forma-tion at the molecular level.

In conclusion, too few detailed studies of the functional andstructural properties of enzymes involved in lipid metabolism havebeen carried out. In one such study, a polymorphism in DGAT wasrecently identified as responsible for a quantitative trait locuscontrolling high levels of oil production in maize. A single


amino-acid substitution boosted oil content by up to 41% and oleicacid content to 107% [94]. An understanding of the molecular pro-cesses governing oil storage in lipid bodies might lead to novel ap-proaches to the engineering of unicellular microorganisms.Improvements in our understanding of the regulatory and struc-tural properties of the enzymes involved in lipid metabolism andthe use of genetic techniques for the engineering of these organ-isms would make it possible to maximize cellular lipid storageand, therefore, oil production per cell. A combination of modifica-tions to the culture conditions and the regulated expression of keyenzymes (native or indigenous) could be used to produce specificlipid compounds with added value and of potential interest inthe oleochemical and petrochemical field.

The study of de novo lipid accumulation as a result of the con-version of carbohydrate substrates, such as glucose, provides in-sight into the regulation of the entire lipid synthesis pathway.The accumulation of lipids produced from these substrates is trig-gered by nutrient limitation. Nitrogen limitation is the principaltype of limitation studied and the C/N ratio is the key factor gov-erning lipid accumulation. Most ex novo lipid accumulation studieshave been carried out in batch mode. However, the fed-batch modeseems to be the best culture system, because carbon and nitrogenflows can be perfectly controlled, making it possible to dissociategrowth and lipid accumulation and the absence of by-productsecretion. Novel approaches combining lipidomic, metabolomicand genetic approaches and making use of fed-batch cultures willundoubtedly provide a wealth of information about the regulationof lipid metabolism.

Acknowledgements

A. Beopoulos was supported by the Lipicaero French nationalresearch program (ANR-PNRB) No. 0701C0089. J. Cescut was sup-ported by CNRS and Airbus. R. Haddouche was supported by theEuropean research program (LIPOYEAST). This work was finan-cially supported in part by Aerospace Valley. We thank Julie Sappaof Alex Edelman and Associates for her excellent help in correctingthe English version of the manuscript.

References

[1] Barth G, Gaillardin C. Yarrowia lipolytica. In: Wolf K, editor. Nonconventionalyeasts in biotechnology. Berlin, Heidelberg, New York: Springer-Verlag; 1996.p. 313–88.

[2] Fickers P, Benetti PH, Wache Y, Marty A, Mauersberger S, Smit MS.Hydrophobic substrate utilisation by the yeast Yarrowia lipolytica, and itspotential applications. FEMS Yeast Res 2005;5:527–43.

[3] Ratledge C. Single cell oils for the 21st century. In: Cohen, Ratlege, editors.Single cell oils. Champaign: AOCS Press; 2005. p. 1–20.

[4] Mlickova K, Roux E, Athenstaedt K, d’Andrea S, Daum G, Chardot T, et al. Lipidaccumulation, lipid body formation, and acyl coenzyme A oxidases of the yeastYarrowia lipolytica. Appl Environ Microbiol 2004;70:3918–24.

[5] Zweytick D, Athenstaedt K, Daum G. Intracellular lipid particles of eukaryoticcells. Biochim Biophys Acta 2000;1469:101–20.

[6] Brown DA. Lipid droplets: proteins floating on a pool of fat. Curr Biol2001;11:446–9.

[7] Fujimoto T, Ohsaki Y, Cheng J, Suzuki M, Shinohara Y. Lipid droplets: a classicorganelle with new outfits. Histochem Cell Biol 2008;130:263–79.

[8] Athenstaedt K, Jolivet P, Boulard C, Zivy M, Negroni L, Nicaud JM, et al. Lipidparticle composition of the yeast Yarrowia lipolytica depends on the carbonsource. Proteomics 2006;6:1450–9.

[9] Ferreyra RG, Burgardt NI, Milikowski D, Melen G, Kornblihtt AR, Dell’ AngelicaEC, et al. A yeast sterol carrier protein with fatty-acid and fatty-acyl-CoAbinding activity. Arch Biochem Biophys 2006;453:197–206.

[10] Beckman M. Cell biology. Great balls of fat. Science 2006;311:1232–4.[11] Papanikolaou S, Chevalot I, Komaitis M, Marc I, Aggelis G. Single cell oil

production by Yarrowia lipolytica growing on an industrial derivative of animalfat in batch cultures. Appl Microbiol Biotechnol 2002;58:308–12.

[12] Czabany T, Athenstaedt K, Daum G. Synthesis, storage and degradation ofneutral lipids in yeast. Biochim Biophys Acta 2007;1771:299–309.

[13] Ratledge C. In: Ratledge C, Wilkinson SG, editors. Microbial lipids, vol.2. London: Academic Press; 1988.

[14] Rattray JBM. In: Ratledge C, Wilkinson SG, editors. Microbial lipids, vol.1. London: Academic Press; 1988. p. 552–687.

[15] Ratledge C. Yeasts, moulds, algae and bacteria as sources of lipids. In: Kamel,editor. Technological advances in improved and alternative sources oflipids. London: Blackie Academic and Professional; 1994. p. 235–91.

[16] Ratledge C, Wynn JP. The biochemistry and molecular biology of lipidaccumulation in oleaginous microorganisms. Adv Appl Microbiol 2002;51:1–51.

[17] Kennedy EP. Biosynthesis of complex lipids. Fed Proc 1961;20:934–40.[18] Carman GM, Han GS. Roles of phosphatidate phosphatase enzymes in lipid

metabolism. Trends Biochem Sci 2006;31:694–9.[19] Wu WI, Lin YP, Wang E, Merrill Jr AH, Carman GM. Regulation of phosphatidate

phosphatase activity from the yeast Saccharomyces cerevisiae by sphingoidbases. J Biol Chem 1993;268:13830–7.

[20] Wu WI, Carman GM. Regulation of phosphatidate phosphatase activity fromthe yeast Saccharomyces cerevisiae by nucleotides. J Biol Chem 1994;269:29495–501.

[21] Oelkers P, Cromley D, Padamsee M, Billheimer JT, Sturley SL. The DGA1 genedetermines a second triglyceride synthetic pathway in yeast. J Biol Chem2002;277:8877–81.

[22] Beopoulos A, Mrozova Z, Thevenieau F, Le Dall MT, Hapala I, Papanikolaou S,et al. Control of lipid accumulation in the yeast Yarrowia lipolytica. ApplEnviron Microbiol 2008;74:7779–89.

[23] Beopoulos A, Chardot T, Nicaud JM. Yarrowia lipolytica: a model and a tool tounderstand the mechanisms implicated in lipid accumulation. Biochimie2009;91:692–6.

[24] Cases S, Smith SJ, Zheng YW, Myers HM, Lear SR, Sande E, et al. Identification ofa gene encoding an acyl CoA:diacylglycerol acyltransferase, a key enzyme intriacylglycerol synthesis. Proc Natl Acad Sci USA 1998;95:13018–23.

[25] Oelkers P, Tinkelenberg A, Erdeniz N, Cromley D, Billheimer JT, Sturley SL. Alecithin cholesterol acyltransferase-like gene mediates diacylglycerolesterification in yeast. J Biol Chem 2000;275:15609–12.

[26] Dahlqvist A, Stahl U, Lenman M, Banas A, Lee M, Sandager L, et al.Phospholipid:diacylglycerol acyltransferase: an enzyme that catalyzes theacyl-CoA-independent formation of triacylglycerol in yeast and plants. ProcNatl Acad Sci USA 2000;97:6487–92.

[27] Garbarino J, Sturley SL. Homoeostatic systems for sterols and other lipids.Biochem Soc Trans 2005;33:1182–5.

[28] Daum G, Lees ND, Bard M, Dickson R. Biochemistry, cell biology and molecularbiology of lipids of Saccharomyces cerevisiae. Yeast 1998;14:1471–510.

[29] Mlickova K, Roux E, Athenstaedt K, d’Andrea S, Daum G, Chardot T, et al. Lipidaccumulation, lipid body formation, and acyl coenzyme A oxidases of the yeastYarrowia lipolytica. Appl Environ Microbiol 2004;70:3918–24.

[30] Sandager L, Gustavsson MH, Stahl U, Dahlqvist A, Wiberg E, Banas A, et al.Storage lipid synthesis is non-essential in yeast. J Biol Chem 2002;277:6478–82.

[31] Wang H, Le Dall MT, Wache Y, Laroche C, Belin JM, Nicaud JM. Cloning,sequencing, and characterization of five genes coding for acyl-CoA oxidaseisozymes in the yeast Yarrowia lipolytica. Cell Biochem Biophys 1999;31:165–74.

[32] Ratledge C. Regulation of lipid accumulation in oleaginous micro-organisms.Biochem Soc Trans 2002;30:1047–50.

[33] Botham PA, Ratledge C. A biochemical explanation for lipid accumulation inCandida 107 and other oleaginous micro-organisms. J Gen Microbiol1979;114:361–75.

[34] Pfitzner A, Kubicek CP, Rohr M. Presence and regulation of ATP:citrate lyasefrom the citric acid producing fungus Aspergillus niger. Arch Microbiol1987;147:88–91.

[35] Al-Feel W, Chirala SS, Wakil SJ. Cloning of the yeast FAS3 gene and primarystructure of yeast acetyl-CoA carboxylase. Proc Natl Acad Sci USA1992;89:4534–8.

[36] Goodridge AG, Fantozzi DA, Klautky SA, Ma XJ, Roncero C, Salati LM.Nutritional and hormonal regulation of genes for lipogenic enzymes. ProcNutr Soc 1991;50:115–22.

[37] Gill CO, Ratledge C. Effect of n-alkanes on the transport of glucose in Candidasp. strain 107. Biochem J 1972;127:59–60.

[38] Wynn JP, Kendrick A, Ratledge C. Sesamol as an inhibitor of growth and lipidmetabolism in Mucor circinelloides via its action on malic enzyme. Lipids1997;32:605–10.

[39] Wynn JP, Bin Abdul Hamid A, Ratledge C. The role of malic enzyme in theregulation of lipid accumulation in filamentous fungi. Microbiology1999;145(Pt 8):1911–7.

[40] Zhang Y, Adams IP, Ratledge C. Malic enzyme: the controlling activity forlipid production? Overexpression of malic enzyme in Mucor circinelloidesleads to a 2.5-fold increase in lipid accumulation. Microbiology 2007;153:2013–25.

[41] Alfenore S, Molina-Jouve C, Guillouet SE, Uribelarrea JL, Goma G, Benbadis L.Improving ethanol production and viability of Saccharomyces cerevisiae by avitamin feeding strategy during fed-batch process. Appl Microbiol Biotechnol2002;60:67–72.

[42] Cescut J. Accumulation d’acylglycérols par des espèces levuriennes àusage carburant aéronautique: physiologie et performances deprocédés. Toulouse: Université de Toulouse; 2009. p. 283.

[43] Papanikolaou S, Aggelis G. Modeling lipid accumulation and degradation inYarrowia lipolytica cultivated on industrial fats. Curr Microbiol 2003;46:398–402.

[44] Granger L. Caractérisation cinétique et stoechiometrique de la synthèse d’acidegras chez Rhodotorula glutinis. Toulouse: Institut National des sciencesappliquées de Toulouse; 1992. p. 247.


[45] Rupcic J, Blagovic B, Maric V. Cell lipids of the Candida lipolytica yeast grown onmethanol. J Chromatogr A 1996;755:75–80.

[46] Aggelis G, Komaitis M. Enhancement of single cell oil production by Yarrowialipolytica growing in the presence of Teucrium polium L. aqueous extract.Biotechnol Lett 1999:747–9.

[47] Papanikolaou S, Galiotou-Panayotou M, Chevalot I, Komaitis M, Marc I, AggelisG. Influence of glucose and saturated free-fatty acid mixtures on citric acid andlipid production by Yarrowia lipolytica. Curr Microbiol 2006;52:134–42.

[48] Briffaud J, Engasser J. Citric acid production from glucose. I. Growth andexcretion kinetics in a stirred fermentor. Biotechnol Bioeng 1979;21:2083–92.

[49] Engasser JM, Hisland P. Diffusional effects on the heterogeneous kinetics oftwo-substrate enzymic reactions. J Theor Biol 1979;77:427–40.

[50] Botham PA, Ratledge C. A biochemical explanation for lipid accumulation inCandida 107 and other oleaginous micro-organisms. J Gen Microbiol1979;114:361–75.

[51] Boulton CA, Ratledge C. Regulatory studies on citrate synthase in Candida 107,an oleaginous yeast. J Gen Microbiol 1980;121:441–7.

[52] Klasson T, Clausen E, Gaddy J. Continuous fermentation for the production ofcitric acid from glucose. Appl Biochem Biotechnol 1989;21:491–509.

[53] Papanikolaou S, Muniglia L, Chevalot I, Aggelis G, Marc I. Yarrowia lipolytica asa potential producer of citric acid from raw glycerol. J Appl Microbiol2002;92:737–44.

[54] Papanikolaou S, Chevalot I, Komaitis M, Aggelis G, Marc I. Kinetic profile of thecellular lipid composition in an oleaginous Yarrowia lipolytica capable ofproducing a cocoa-butter substitute from industrial fats. Anton Leeuw2001;80:215–24.

[55] Papanikolaou S, Aggelis G. Lipid production by Yarrowia lipolytica growing onindustrial glycerol in a single-stage continuous culture. Bioresour Technol2002;82:43–9.

[56] Ykema A, Verbree EC, van Verseveld HW, Smit H. Mathematical modelling oflipid production by oleaginous yeasts in continuous cultures. Anton Leeuw1986;52:491–506.

[57] Kendrick A, Ratledge C. Lipid formation in the oleaginous mouldEntomophthora exitalis grown in continuous culture: effects of growth rate,temperature and dissolved oxygen tension on polyunsaturated fatty acids.Appl Microbiol Biotechnol 1992;37:18–22.

[58] Georgopapadakou NH, Dix BA, Smith SA, Freudenberger J, Funke PT. Effect ofantifungal agents on lipid biosynthesis and membrane integrity in Candidaalbicans. Antimicrob Agents Chemother 1987;31:46–51.

[59] Kim JW, Park TJ, Ryu DD, Kim JY. High cell density culture of Yarrowia lipolyticausing a one-step feeding process. Biotechnol Prog 2000;16:657–60.

[60] Dyal S, Narine S. Implications for the use of Mortierella fungi in the industrialproduction of essential fatty acids. Food Res Int 2005;38:445–67.

[61] Chang C, Ryu D, Park C, Kim J, Ogrydziak D. Recombinant bioprocessoptimization for heterologous protein production using two-stage, cyclicfed-batch culture. Appl Microbiol Biotechnol 1998;49:531–7.

[62] Nicaud J, Madzak C, van den Broek P, Gysler C, Duboc P, Niederberger P, et al.Protein expression and secretion in the yeast Yarrowia lipolytica. FEMS YeastRes 2002;2:371–9.

[63] Thevenieau F. Metabolic engineering of the yeast Yarrowia lipolytica for theproduction of long-chain dicarboxylic acids from renewable oil feedstock. PhDthesis, in Institut National Agronomique Paris-Grignon; 2006.

[64] Beopoulos A, Chardot T, Nicaud JM. Yarrowia lipolytica: a model and a tool tounderstand the mechanisms implicated in lipid accumulation. Biochimie 2009.

[65] Dujon B, Sherman D, Fischer G, Durrens P, Casaregola S, Lafontaine I, et al.Genome evolution in yeasts. Nature 2004;430:35–44.

[66] Thevenieau F, Gaillardin C, Nicaud JM. Applications of the non-conventionalyeast Yarrowia lipolytica. In: Satyanarayana T, Kunze G, editors. Yeastbiotechnology: diversity and applications. Netherlands: Springer; 2009.

[67] Pignede G, Wang H, Fudalej F, Gaillardin C, Seman M, Nicaud JM.Characterization of an extracellular lipase encoded by LIP2 in Yarrowialipolytica. J Bacteriol 2000;182:2802–10.

[68] Pignede G, Wang HJ, Fudalej F, Seman M, Gaillardin C, Nicaud JM. Autocloningand amplification of LIP2 in Yarrowia lipolytica. Appl Environ Microbiol2000;66:3283–9.

[69] Fickers P, Fudalej F, Le Dall MT, Casaregola S, Gaillardin C, Thonart P, et al.Identification and characterisation of LIP7 and LIP8 genes encoding twoextracellular triacylglycerol lipases in the yeast Yarrowia lipolytica. FungalGenet Biol 2005;42:264–74.

[70] Kohlwein SD, Paltauf F. Uptake of fatty acids by the yeasts, Saccharomycesuvarum and Saccharomycopsis lipolytica. Biochim Biophys Acta1984;792:310–7.

[71] Papanikolaou S, Aggelis G. Selective uptake of fatty acids by the yeast Yarrowialipolytica. Eur J Lipid Sci Technol 2003;105:651–5.

[72] Titorenko VI, Nicaud JM, Wang H, Chan H, Rachubinski RA. Acyl-CoA oxidase isimported as a heteropentameric, cofactor-containing complex intoperoxisomes of Yarrowia lipolytica. J Cell Biol 2002;156:481–94.

[73] Guo T, Kit YY, Nicaud JM, Le Dall MT, Sears SK, Vali H, et al. Peroxisome divisionin the yeast Yarrowia lipolytica is regulated by a signal from inside theperoxisome. J Cell Biol 2003;162:1255–66.

[74] Scioli C, Vollaro L. The use of Yarrowia lipolytica to reduce pollution in olivemill wastewater. Wat Res 1997;31:2520–4.

[75] Alkasrawi M, Nandakumar R, Margesin R, Schinner F, Mattiasson B. A microbialbiosensor based on Yarrowia lipolytica for the off-line determination of middle-chain alkanes. Biosens Bioelectron 1999;14:723–7.

[76] Stottmeister U, Behrens U, Weissbrodt E, Barth G, Franke-Rinker D, Schulze E.Utilization of paraffins and other noncarbohydrate carbon sources formicrobial citric acid synthesis. Z Allg Mikrobiol 1982;22:399–424.

[77] Guieysse D, Sandoval G, Faure L, Nicaud J-M, Monsan P, Marty A. New efficientlipase from Yarrowia lipolytica for the resolution of 2-bromo-arylacetic acidesters. Tetrahedron: Asymmetr 2004;15:3539–43.

[78] Athenstaedt K, Jolivet P, Boulard C, Zivy M, Negroni L, Nicaud JM, et al. Lipidparticle composition of the yeast Yarrowia lipolytica depends on the carbonsource. Proteomics 2006;6:1450–9.

[79] Papanikolaou S, Muniglia L, Chevalot I, Aggelis G, Marc I. Accumulation of acocoa-butter-like lipid by Yarrowia lipolytica cultivated on agro-industrialresidues. Curr Microbiol 2003;46:124–30.

[80] Schrader J, Etschmann MM, Sell D, Hilmer JM, Rabenhorst J. Appliedbiocatalysis for the synthesis of natural flavour compounds – currentindustrial processes and future prospects. Biotechnol Lett 2004;26:463–72.

[81] Bati N, Hammond E, Glatz B. Biomodification of fats and oils: trials withCandida lipolytica. J Am Oil Chem Soc 1984;61:1743–6.

[82] Mlickova K, Luo Y, D’Andrea S, Pec P, Chardot T, Nicaud J. Acyl-CoA oxidase, akey step for lipid accumulation in the yeast Yarrowia lipolytica. J Mol Catal B:Enzym 2004;28:81–5.

[83] Damude HG, Zhang H, Farrall L, Ripp KG, Tomb JF, Hollerbach D, et al.Identification of bifunctional delta12/omega3 fatty acid desaturases forimproving the ratio of omega3 to omega6 fatty acids in microbes and plants.Proc Natl Acad Sci USA 2006;103:9446–51.

[84] Chang L, Chen D, Chen Y, Nicaud J, Madzak C, Huang Y. Production offunctional c-linonlenic acid (GLA) by expression of fungal D12- and D6-desaturase genes in the oleaginous yeast Yarrowia lipolytica. In: Ching, Jeu-Fu,editors. Biocatalysis and agricultural biotechnology. Boca Raton: CRC Press,Taylor & Francis Group; 2009. p. 164–79.

[85] Oswal N, Sarma PM, Zinjarde SS, Pant A. Palm oil mill effluent treatment by atropical marine yeast. Bioresour Technol 2002;85:35–7.

[86] Rajakumari S, Grillitsch K, Daum G. Synthesis and turnover of non-polar lipidsin yeast. Prog Lipid Res 2008;47:157–71.

[87] Londos C, Sztalryd C, Tansey JT, Kimmel AR. Role of PAT proteins in lipidmetabolism. Biochimie 2005;87:45–9.

[88] Zimmermann R, Lass A, Haemmerle G, Zechner R. Fate of fat: the role ofadipose triglyceride lipase in lipolysis. Biochim Biophys Acta 2009;1791:494–500.

[89] Welte MA. Proteins under new management: lipid droplets deliver. Trends CellBiol 2007;17:363–9.

[90] Sztalryd C, Xu G, Dorward H, Tansey JT, Contreras JA, Kimmel AR, et al. PerilipinA is essential for the translocation of hormone-sensitive lipase during lipolyticactivation. J Cell Biol 2003;161:1093–103.

[91] Koffel R, Tiwari R, Falquet L, Schneiter R. The Saccharomyces cerevisiae YLL012/YEH1, YLR020/YEH2, and TGL1 genes encode a novel family of membrane-anchored lipases that are required for steryl ester hydrolysis. Mol Cell Biol2005;25:1655–68.

[92] Daum G, Wagner A, Czabany T, Athenstaedt K. Dynamics of neutral lipidstorage and mobilization in the yeast. Biochimie 2007;89:243–8.

[93] Cermelli S, Guo Y, Gross SP, Welte MA. The lipid-droplet proteome reveals thatdroplets are a protein-storage depot. Curr Biol 2006;16:1783–95.

[94] Zheng P, Allen WB, Roesler K, Williams ME, Zhang S, Li J, et al. A phenylalaninein DGAT is a key determinant of oil content and composition in maize. NatGenet 2008;40:367–72.

1-s2.0-S0163782709000423-main

Documents

lipid accumulation3804

lipid accumulation pathways

mastering lipid production

novo lipid synthesis

lipid body mag

nonpolar lipid synthesis

lipid researchemail

accumulation factors